Do plant viruses attack animals? examples?

Do plant viruses attack animals, if yes please give an example of the virus.

I feel both plant and animal viruses are different, and they cannot attack each other hosts.

Numerous viruses infect plant, however, none of them so far is known as pathogen to animal and human beings. Only three families, Bunyaviridae, Rhabdoviridae and Reoviridae contain viruses known to infect plant, animal and human.

Plant viruses attacking animals or vice versa would mean that an animal serves as a biological vector to a plant or that a plant serves as a biological vector to an animal (depending on whose prospective you take by considering it as the host). This is highly unlikely

  1. due to very different properties of the plant and the animal cells, while the virus has to be adapted to replicating in both
  2. due to the nature of the interactions between the plants and the animals which are unfavorable to transmission between viral habitats (unlike, e.g., direct blood exchange between an animal and a stinging insect).

On the other hand, plants and insects routinely serve as mechanical vectors for transmitting viruses: an insect feeding on one plant may carry virus particles to another plant, and similarly a plant may mediate viral particles exchange between insects.

See Viral Ecology by Hurst et al.

Do plant viruses attack animals? examples? - Biology

As you’ve learned, viruses often infect very specific hosts, as well as specific cells within the host. This feature of a virus makes it specific to one or a few species of life on Earth. On the other hand, so many different types of viruses exist on Earth that nearly every living organism has its own set of viruses trying to infect its cells. Even prokaryotes, the smallest and simplest of cells, may be attacked by specific types of viruses. In the following section, we will look at some of the features of viral infection of prokaryotic cells. As we have learned, viruses that infect bacteria are called bacteriophages (Figure 1). Archaea have their own similar viruses.

How do viruses affect animals? -Gianni, 10, Cayman Islands

Our planet is home to more than seven million amazing animal species. While we have our differences, we also have something in common: We are all made up of a bunch of cells. My friend Jeb Owen told me all about it.

He’s a scientist at Washington State University who is really curious about how insects eat blood and interact with animal hosts. He’s even been called a disease detective, tracking down viruses transmitted by insects.

We can think of animal cells as water balloons, Owen said. Of course, cells hold more than just water. Inside cells are different parts—almost like a little working kitchen—making things our bodies and cells need.

The cell also holds the animal’s genetic material, or DNA, that acts like a little cookbook. The cookbook is the genetic material that has all the instructions for what makes you, well, you. The outsides of cells have small openings to move things in and out of the cell.

But viruses don’t have all the parts cells have. In fact, a virus is really just a bit of genetic material wearing a protective coat. It’s like a little cookbook without a kitchen. So, viruses can’t make anything on their own. A virus needs a cell to make more viruses using the cell’s kitchen. In a way, viruses are a bit like burglars. They’ve got special “keys” on their coats they use to get through the openings on the outsides of cells.

Once a virus breaks in, it can trick the cell into making more of the virus. The cell makes so much virus the cell bursts like an over lled water balloon, releasing all the new virus copies. When the cells burst, it can make it hard for the body to work, which causes sickness. The immune system, which defends the animal’s body against infection, may recognize something unusual is up.

Sometimes if enough virus gets in, the immune system that works to protect you also ends up causing harm. Cells that protect you and kill off the virus end up killing healthy cells in the process. This can also make us sick.

Some viruses put their genetic material into the genetic material of the animal’s cells. This can make animal cells misbehave and become cancerous. Cancer cells cause your tissues, or the community of cells working together, to fail. This can make you very sick, too.

Most viruses only infect one kind of animal. Even though animals are related, there are small differences in the cells of each kind of animal. It is like the cells of different animals have specific doors and locks on the outsides of the cells. Viruses open those “locks” and can only use that kind of animal as a host. When viruses develop “keys” that work on more than one kind of cell, they can move between different kinds of animals.

It’s a great question you ask, Gianni. Scientists are curious about it, too. After all, knowing more about viruses can help us understand how they move around and how to prevent them. In that way, we can help our animal—and people—friends live even better lives.

Part 2: Genetic Engineering for Virus Resistance in Plants: Different Viruses Demand Different Strategies

00:00:03.04 Hi and welcome to the lecture. I'm Roger Beachy,
00:00:05.23 the president of the Donald Danforth plant science center in St. Louis, Missouri.
00:00:09.06 Some of you saw the previous lecture in which I talked about the replication and the spread cell to cell
00:00:17.11 of tobacco mosaic virus, a model virus.
00:00:21.12 In this lecture, I want to talk about how we have used that information
00:00:26.12 and previous information to develop strategies for transgenic plants
00:00:33.05 that would be resistant to virus infection.
00:00:35.27 I'm going to show you several examples and the bottom line, I guess the take-home message is
00:00:40.17 first, if you're going to do the studies that are hopefully have applications in biotechnology,
00:00:47.13 you need to know your pathogen, know how it works, how it's transmitted from cell to cell,
00:00:51.24 how it's transmitted from one plant to the next, so that you can use genetic intradictions, or gene therapy approaches
00:00:58.28 to improve the resistance of plants to infection or replication or a spread,
00:01:04.14 and therefore reduce the impact of disease in a plant.
00:01:09.16 We're going to talk about several viruses in this talk
00:01:13.08 and first I'm going to review some of the strategies that have been used for developing virus-resistant plants.
00:01:19.25 I'll talk about tobacco mosaic virus as a model of how we use this information to develop what we call coat-protein mediated resistance
00:01:28.13 and a real live example of how we've applied this, or are applying it now,
00:01:32.15 in combination or collaboration with our colleagues in India,
00:01:37.13 to develop resistance against a very severe disease in groundnuts, or peanuts,
00:01:42.25 a crop which affects many small holder farmers in India.
00:01:48.03 If we can solve this problem, we will be very pleased of course.
00:01:51.25 And then I want to talk about a very different example, rice tungro bacilliform virus, it's a pararetrovirus
00:01:57.11 and you'll see the strategies that we used here are very different than the ones that we used up above in tobacco mosaic virus.
00:02:02.25 And lastly, another challenge that a colleague of mine, Claude Fauquet, and his lab are working on:
00:02:08.08 so, single stranded DNA virus, a very different type, and the strategies used there will be different
00:02:13.25 than the strategies used in number 1, number 2 and number 3.
00:02:17.11 So, let's go ahead. It's now been more than twenty years since we had the first field trial
00:02:23.16 of a genetically engineered, disease resistant food crop.
00:02:28.00 My lab was involved in the collaboration with the Monsanto company
00:02:31.20 because very early on, we didn't have all the technology necessary to develop transgenic plants
00:02:36.22 in our own labs and we collaborated with scientists at Monsanto to develop the plants that you see here.
00:02:43.18 These are tomato plants, they carry a capsid of protein of a simple virus, tobacco mosaic virus.
00:02:50.16 Tobacco mosaic virus and tomato mosaic virus are cousins.
00:02:53.20 And these are, this is the result of infection of these ten plants on this slide with the virus in our greenhouses
00:03:03.06 and you can see that the plants here are growing much taller and are much more healthy looking than the ones below.
00:03:09.25 These are resistant to virus infection and what we found in 1985 was that we could reproduce the work from tobacco
00:03:17.11 to tomato to petunias and other plants.
00:03:21.02 We then were able to, through applications made to the US Department of Agriculture
00:03:25.26 and the Environmental Protection Agency, to plant these in the field.
00:03:28.23 And it's been now more than twenty years, my hair is not brown anymore,
00:03:34.18 it's more light colored, but I've grayed because of a lot of things.
00:03:40.29 And one of those challenges that I find still curious is how technology like this one and others
00:03:47.11 have still found it difficult to be released into the commercialization stream.
00:03:55.27 To date, we have only one success from an academic laboratory
00:04:00.08 of virus resistant plants that have been developed through these kind of technologies.
00:04:05.15 That's a papaya that was developed at the University of Hawai'i.
00:04:10.01 And there's now a squash variety that's also on the market from the public sector, from the private sector.
00:04:16.25 So, why hasn't there been more success?
00:04:18.19 Is it because we don't know enough about the virus?,
00:04:21.22 or is it that we don't know enough about how to do regulations?,
00:04:25.11 is it what we don't know about commercialization,
00:04:28.12 why has the public sector not been more involved in this science and technology?
00:04:33.01 Because as you'll see, what we can contribute is significant.
00:04:36.13 And we need to learn how to use our knowledge to enhance the contribution of our science
00:04:44.16 to food production in this world where population will continue to grow,
00:04:50.09 where ever more pressures of food production will continue to grow as population grows,
00:04:55.25 where there's still only a certain amount of land that can be planted,
00:05:00.02 only a certain amount of fresh water that can go onto agriculture lands, and yet population continues to grow.
00:05:05.22 The challenges of plant sciences include those, the studies of basic mechanisms of disease,
00:05:12.00 and finding ways to interdict those diseases, either through classical genetics and plant breeding
00:05:17.09 or through, perhaps, through biotechnology and what I want to talk about today are specific examples
00:05:22.14 where biotechnology can be used, or has been used, to develop plants that have resistance to the pathogen.
00:05:29.13 Our next challenge is to get these tested under confined field trial situations
00:05:35.29 to evaluate them not only in laboratories, in greenhouses, but in actual field situations,
00:05:41.11 as has been done in this first case, now more than twenty years ago.
00:05:44.21 What I hope is that there will be more and more examples of the public sector, such as we are,
00:05:50.13 to developing products that will be useful in release for people to increase their food production.
00:05:58.14 So, first a little bit of review of what has been done in scientific laboratories to develop resistance to viruses.
00:06:08.07 There's been a variety of strategies. Some of these are called pathogen derived.
00:06:12.06 That is, we use genes and open reading frames or small RNAs
00:06:16.26 to develop strategies to stop the virus in its process of virus replication.
00:06:22.24 And there are a variety of those,
00:06:24.06 in some cases, it's possible to use the genes that come from the coat protein sequences,
00:06:29.10 as we did for the tomato mosaic virus case.
00:06:31.21 Or, using replicase-related proteins, those enzymes that are required for replicating the viral genome.
00:06:39.13 In other cases, their strategies have been to block the function of the movement protein,
00:06:43.11 the latter has not been very successful to date, but I hope through the last lecture,
00:06:48.11 you gained a sense of how much we know about movement proteins,
00:06:52.02 and the more we know, the possibility of blocking their action, perhaps having a wider array of effects on virus infection.
00:07:00.28 The complementary strategies are to use sequences from the viral genome that will induce the formation of small RNAs
00:07:08.07 that can then silence, and lead to the degradation of viral RNAs during the replication process.
00:07:14.24 This seems to work quite well in some cases, but not in others.
00:07:17.24 Now, the challenge here is that these, it's imagined or expected that the RNA mediated resistance events
00:07:24.26 will work in those cases where there's very little divergence in viral sequence.
00:07:29.29 As you know, silencing complexes are formed because of the close sequence similarity between one RNA and the target RNA.
00:07:40.12 If those diverge because the divergence in viral sequences, it might be a challenge for this kind of approach.
00:07:47.25 Nevertheless, RNA mediated resistance can give very high levels of resistance in transgenic plants.
00:07:53.08 That will not be the topic of my talk today, we've been working primarily on protein mediated events.
00:07:59.23 There are other examples that are being used in which resistance is derived not from the pathogen sequences,
00:08:06.06 but from other sequences.
00:08:07.21 For example, I'm going to show you a case in which we've used transcription factors
00:08:11.22 to restrict the replication and disease caused by rice tungro disease.
00:08:16.15 In some cases, there are single-stranded DNA binding proteins.
00:08:20.03 Other strategies have re-created interferon like strategies in plants or using, developing plants that have monoclonal antibodies
00:08:29.00 that would bind specific viruses or virus proteins and slow down their virus replication by doing such.
00:08:35.17 There is a whole series of data that now indicate that some of the translation factors that are involved
00:08:43.17 in translating viral RNAs can be used to enhance disease resistance in certain cases, classes, of viruses.
00:08:50.27 So, again, the more we know about the virus and how it replicates,
00:08:54.18 and the more we know about the host and how it responds to virus invasion,
00:08:58.18 the more successful we'll be to develop strategies for resistance against infection.
00:09:03.09 So, I want to talk about a model system, the coat protein mediated resistance.
00:09:09.03 How does it work? Why does it work? And how can we make it more effective?
00:09:13.25 I want to give you a real-life example of a virus where we've used a coat protein to protect against tobacco streak virus in India,
00:09:22.03 it's a real world application of the technology that's now more than twenty years old.
00:09:25.21 Rice tungro disease, using novel approaches because the approaches up here didn't work in the one on rice tungro,
00:09:32.00 we found out about ten or twelve years ago.
00:09:33.29 And second, then lastly, I want to talk about a protein that we call G5 protein,
00:09:37.22 it's a protein from an inovirus, a bacteriophage, to control gemini virus replication.
00:09:43.25 Now, there are lots of plant viruses, lots of animal viruses, lots of viruses of phage,
00:09:49.25 and in this virosphere, developed through the International Committee on Virus Taxonomy,
00:09:55.16 and Claude Fauquet and his colleagues, developed this virosphere, I won't go through it,
00:10:00.05 but I want to talk about this group of viruses at this point. The virus I'm going to talk about is tobacco mosaic,
00:10:05.21 it belongs to this class of single stranded RNA viruses that are the subject, was the subject of my previous lecture.
00:10:15.13 A little primer on tobacco mosaic virus. As I indicated earlier, we know quite a bit about how it's replicated
00:10:21.25 from the molecular perspective. We've learned a lot about how the virus replicates in the cell,
00:10:27.14 where it works in the cell, how it forms these replication complexes and talked about how the virus might move from cell to cell.
00:10:34.05 Now, in 1985 we discovered that when we introduced one of the genes from the virus, the capsid protein gene,
00:10:41.28 into tomato plants, we had a population of plants that were resistant.
00:10:46.23 In fact, the level of resistance was correlated with the level of capsid protein.
00:10:50.22 The more capsid protein, the more resistant.
00:10:53.14 As we did our early studies, in the 80's and early 90's, our model that we developed, supported by extensive data,
00:11:01.18 indicated that the coat protein, if the coat protein is in large enough amounts in the plant
00:11:07.07 and if it could self-assemble to make these trimers that I also show here,
00:11:11.25 and these two, or this one, were responsible for binding to virus as the virus entered the cell.
00:11:20.04 And sort of re-capped the virus and not letting the viral RNA out of its shell.
00:11:24.29 If it can't get out its shell, of course, it can't bind to ribosomes, it can't start the infection process.
00:11:30.18 So, in this, in the first few years, as we came to realize that coat protein mediated resistance,
00:11:38.04 was working by blocking the disassembling of the virion and therefore stopping the virus in its tracks.
00:11:43.26 One of the ways that we showed that this might be the case is that when we inoculated these plants that were transgenic
00:11:49.05 with the, that had the coat protein, with the naked viral RNA without the capsid protein,
00:11:54.06 we could overcome the resistance. And I'll come back to that a little bit later,
00:11:57.26 because we used that strategy to study virus replication in cells as well.
00:12:02.08 So, that led us to believe that at least the earliest events in virus disease blocking was caused by a re-capping of the virion.
00:12:11.05 Because Gerald Stubbs and a large number of other laboratories have worked on the structural biology of the virus
00:12:17.07 we were at quite a positive advantage in our studies.
00:12:21.05 What we said is, if virus, if interaction of proteins with each other, the capsid proteins with each other,
00:12:27.02 is important for virus disease resistance, perhaps if we block the ability of these capsid molecules to assemble,
00:12:35.24 or we made mutations that would increase the affinity of the proteins, one for the other,
00:12:40.22 we would increase the resistance level.
00:12:42.19 And I want to show you the outcome of those studies.
00:12:46.05 Now, we used the residues in the structure in this ribbon diagram of two capsid molecules,
00:12:55.09 we know how they're interacting, you see here two acidic amino acids,
00:12:59.27 aspartic acid at position 77 and glutamic acid at position 50, on two different capsids.
00:13:06.00 These should be repulsive, but there's a calcium that lies in between and neutralizes those negative charges
00:13:14.15 and allows it to stay together. Actually, one of the points in the early stage of virus infection, the calcium is released,
00:13:20.26 allowing these to separate from each other and thereby releasing the viral RNA.
00:13:25.16 We're also fortunate to have the quaternary structure of the assembled virion,
00:13:30.01 so we know how they interact, the subunits interact with each other, side to side and top to bottom.
00:13:35.15 It's been really helpful as we've looked at sort of molecular structures and the effect of structure on resistance.
00:13:41.03 So, here are the results of a very simple experiment.
00:13:44.14 What we did was make a number of mutations in these specific amino acids of the capsid protein
00:13:51.22 and then put the gene for the capsid protein back in the virion and asked if the virion would assemble,
00:13:58.09 if it would, if the virus could move cell to cell, if it could go from leaf to leaf,
00:14:02.25 as I talked in the last lecture.
00:14:05.08 But we also put those mutated proteins into tobacco plants and made populations of transgenic plants
00:14:13.10 that had the virus protein.
00:14:15.09 We made something like fifteen independent plant lines for each of the mutations,
00:14:22.06 so you can see we made a lot of mutants,
00:14:24.10 and then identified out of that population those lines that accumulated the same amount of the protein,
00:14:31.07 the wild-type protein, or the mutant protein. And then used these in a plant pathology study,
00:14:36.22 simply challenged these lines with virus and watched them get sick.
00:14:40.28 So, these are non-transgenic lines, you can see that these plants all became diseased in a certain period of time.
00:14:48.22 And the more sick the plants they look, of course, the more virus they have in them.
00:14:53.05 So, these are two independent plant lines that have two, I mean, two examples of two lines that have two different mutations in them,
00:15:00.03 in the coat protein. And you see that there might be a little bit of delay between the onset of disease
00:15:06.19 in this line, compared to this line, but clearly they're not happy and they show disease.
00:15:13.20 But now, jump down to this other group.
00:15:17.14 Here's a group of transgenic lines that you can see have the mutant coat proteins,
00:15:25.01 this line 3646, is the historical line that we've used for very many years.
00:15:29.00 And you see here that these plants are different than the ones up above.
00:15:35.00 In fact, they show some disease, but the onset of disease is later, it's delayed,
00:15:40.19 and it's not as severe. Then there are another broad set of mutations in the coat protein
00:15:47.02 that either don't get sick or get sick very, very late in the infection process.
00:15:52.07 How did they manage to escape the infection? What's the nature of these mutations
00:15:55.26 and how might they delay the infection process?
00:15:59.03 Now, we've not done all the experiments that I'll tell you about with all of them,
00:16:03.00 we've focused on T42W, one of the lines that we've studied the most, but it's emblematic of the others
00:16:10.00 and we have other data that indicated this group really acts as a group and I'll talk about those studies.
00:16:15.12 In this process, we also have made transgenic cell lines in the cell line BY-2, it's a tobacco cell line.
00:16:24.01 Now, the importance of those cell lines, then, is that we could make single cells, or protoplasts, from them
00:16:29.13 and study single cell dynamics of virus infection.
00:16:33.13 And you see that here, in this case, we did the infection of the cells
00:16:37.19 and you see that there's, they were inoculated at zero time, then we took samples at different times after infection
00:16:45.11 and looked for the accumulation of virus proteins.
00:16:48.00 So, you see that in this case, these sets of mutants, there was no, there was replication and accumulation
00:16:55.08 of the coat protein, just as in the non-transgenic cell line.
00:16:59.28 Here's another set of cell lines in which there was no infection or very low level of infection,
00:17:06.06 as shown by the accumulation of coat protein.
00:17:09.22 Basically, this subset of cell line lines, which showed less accumulation of coat protein
00:17:16.10 also showed less accumulation of movement protein or of replicase.
00:17:20.00 They can be divided into two categories those where there was high susceptibility, high susceptibility,
00:17:25.29 and less susceptibility, or greater resistance.
00:17:29.01 Why was that? How did that work?
00:17:30.16 And why does the coat protein function differently in different cases?
00:17:35.02 So, then we went back to the cell biology of how the virus infection and replication complexes is set up.
00:17:41.23 As I indicated in the previous lecture, the red color represents where the virus replicase is located,
00:17:49.00 the green, where the movement protein is and the blue where the coat protein is localized.
00:17:55.18 This is relatively early in the infection cycle, you can see that these green bodies representing some of the virus replication factories
00:18:02.23 are relatively modest in size and have not coalesced, as at later stages.
00:18:07.20 But, if this is what happens, we then looked at these different cell lines that we had produced
00:18:13.28 and we went through the whole process, we can chart the accumulation of the movement protein
00:18:20.23 as an early stage of infection, a mid-stage of infection and later stages of infection.
00:18:26.04 You notice that here, the puncta are very small and that represents the accumulation of the movement protein
00:18:33.07 using a movement protein fusion with green fluorescent protein.
00:18:36.19 In this case, we've challenged each of these cell lines with virus RNA,
00:18:42.06 so we could overcome the resistance reaction from the coat protein and now we're just now following infection.
00:18:47.24 So, you can see that these bodies grow larger in size,
00:18:51.02 they really get larger, and then down below, in the later stages, bodies look quite large,
00:18:56.02 and then, through a proteolysis of the movement protein, targeting degradation,
00:19:01.08 and the bodies are not visible anymore through the green fluorescent protein.
00:19:06.11 They can only be seen by electron microscopy and localization of the replicase.
00:19:10.16 So, we then began to categorize these cell lines in certain ways
00:19:15.06 and say, do they all progress from early, mid to late stages at the same rate?
00:19:21.27 So, a very talented research assistant looked at these cell lines at different hours after infection,
00:19:28.12 at ten hours after inoculation, at different time points,
00:19:32.26 and then decided if they were in early stages, mid stages or late stages of virus infection.
00:19:39.07 And what you see here is that there are a set of cell lines that all act as if they are non-transgenic.
00:19:47.01 They go through progression of disease and accumulation of these bodies
00:19:52.07 just as a non-transgenic cell line would do.
00:19:54.27 There's another set of mutants that are demonstrated here that look like they progress through the infection cycle
00:20:02.23 in much the same way as does a cell line that makes wild-type coat protein.
00:20:07.15 This is the first time we had noted that this process is different in this cell line than in this cell line up here.
00:20:14.08 Notice how it goes through that progression very rapidly.
00:20:18.05 What that indicated to us is if you get virus infection in these cells
00:20:21.25 by avoiding the disencapsidation, or the release of viral RNA, the infection process can even go faster than normal, than expected.
00:20:29.20 That indicated something very special and that is that the capsid protein was acting, perhaps,
00:20:34.16 as a positive regulator during the replication cycle. That's something we hadn't discovered before.
00:20:39.18 And others of these protein mutants act also as positive regulators,
00:20:45.21 but then down below you see another whole group of cell lines.
00:20:50.23 In those cell lines, they mostly get trapped in the early stages of infection, this stage,
00:20:58.02 or the one just down below.
00:21:00.03 Why was that?
00:21:01.14 That indicated that these, to us, or suggested to us, that these mutant coat proteins
00:21:07.14 were actually acting as negative regulators, not positive regulators.
00:21:11.01 And so we went back to our cell biology.
00:21:13.06 Now look at these two slides and what you're seeing on the left hand slide is an infected cell
00:21:20.05 that has one of these mutants that act as a negative regulator.
00:21:23.07 Notice you find almost no, there's almost no movement protein there.
00:21:27.02 It's making mostly capsid protein and viral replicase, a little bit of movement protein, but not a lot.
00:21:35.04 And only on the other side of this slide is the non-transgenic cell line, which you see very, very large replication bodies,
00:21:43.05 just as I showed you in the previous lecture. These bodies are able to move around rapidly within the cell,
00:21:49.24 they are able to move cell to cell in an infected leaf.
00:21:52.11 So, here was the first indication that this is a negative regulator in some way,
00:21:57.13 and blocking the production of movement protein. And here's the clue:
00:22:02.13 if there's less movement protein, there's less ability of the virus to move from cell to cell.
00:22:07.08 So, you can see that you can have blocking of a process that results in disease resistance.
00:22:13.10 So, what's the structural biology of that whole, what makes the structure of that high-resistance conferring mutant
00:22:22.18 different from the structure of the capsid protein, and how might that structure affect function.
00:22:28.22 I told you earlier, I showed you earlier, that the virus is really a helical array of about 2,140 of these capsid protein molecules.
00:22:39.00 They all stack together and wrap around the viral RNA to produce the virion
00:22:44.00 and they do so so it looks like a helical stack.
00:22:46.24 Here's a, this stack, however, makes a view in the electron microscope that looks a lot like this.
00:22:55.02 These are cryo-EM reconstructions, or pictures, of virions.
00:22:59.22 Notice that this part here is notably different than this. And in fact, if you look at the Fourier transform
00:23:07.05 of the incidence of reflection after, you find that this one is different than this one, they're spaced differently.
00:23:15.17 That spacing was, that information was then used by Tom Smith at the Danforth Center
00:23:20.28 to help us model this structure and what Tom concluded in his modeling is that this structure is different than this one.
00:23:32.10 Essentially, the top, if you represent this as the top of the protein and the bottom of the protein,
00:23:36.17 the top and bottom, you have an A/B, A/B arrangement up here.
00:23:41.07 But what's happened here is that you're having an A/B, B/A arrangement.
00:23:45.29 And the suggestion is that this may be a negative regulator, whereas this is not a negative regulator.
00:23:52.23 Interestingly, Gerald Stubbs, nearly 20 years ago, in his electron microscopic view of some of these virions,
00:24:01.16 in the way he did his science, suggested that there might, some of these occurred in natural virus infection.
00:24:08.14 So, I think we suggest that what's happened is that the mutations that we developed
00:24:13.22 create a, change the incidence of this and this, and the more of this, the more resistance there is.
00:24:25.20 We know that these globs, these assemblies, these platforms, can interact with these.
00:24:32.29 You see it in this structure here, some of these are A/B, A/B and some are A/B, B/A, co-assembled.
00:24:42.25 That was an interesting, that model was very interesting and important, if you were a structural biologist
00:24:49.17 these two images in C and in G can be viewed as stereo pictures. So, here's the model.
00:24:58.06 During virus infection, the replicase binds to the, first part of the infection process produces the enzyme, the replicase.
00:25:08.01 That replicase then structure, binds to this end of the viral RNA and makes the complementary RNA.
00:25:16.16 It then goes to points on that complementary RNA where there are appropriate structures,
00:25:23.04 they're internal promoters, they're shown here as stem loops,
00:25:27.14 and the consequence is that this enzyme then makes these smaller RNA molecules that act as messenger RNAs
00:25:34.27 to produce the movement protein.
00:25:37.14 Other such genomic RNAs make more capsid protein. In the presence of the wild-type capsid protein,
00:25:44.27 you have an enhancement of this promoter region and make lots more movement protein and coat protein messenger RNAs.
00:25:52.20 However, we think that, we propose that when the mutant coat protein and the wild-type protein bind together,
00:26:00.12 they block that receptor, or that promoter, or the binding of the replicase protein to that structure.
00:26:08.16 In any event, you get much less movement protein being made. Some made, but not very much.
00:26:12.27 As a consequence, you don't have as much movement protein. As I talked about in the last lecture,
00:26:17.24 those movement proteins are absolutely essential for moving these virus replication factories around the cell.
00:26:24.22 Here's a virus replication factory that's produced, but it doesn't have movement protein.
00:26:28.09 It also doesn't move over here unless there's a little bit of movement protein stuck into it.
00:26:33.23 So, these factories move around because of the availability of the movement protein in right place, at the right time,
00:26:41.00 to bind to these, to the transport mechanisms, over to plasmodesmata.
00:26:46.15 So, that's the model of how resistance might work.
00:26:49.11 So, over the last fifteen, twenty years, we've learned a lot about coat protein mediated resistance,
00:26:55.08 as these fly-ins show. As you read through them, I think, I hope I've covered and given some,
00:27:02.16 covered each of these topics and hope you can understand where we are in our understanding of how the coat protein can act as a positive regulator,
00:27:09.28 or as a negative regulator. Now, those of us who are interested in protein structure and function
00:27:14.18 can take heart from this, because there are some virus proteins, some virus capsid proteins
00:27:20.08 are only positive regulators.
00:27:22.03 The more we know about those positive regulators, it'd hold, that we could learn more about how to make a negative regulator out of that positive
00:27:29.05 and use that knowledge to strategize, to develop approaches to develop disease resistance,
00:27:34.28 so that we can have more coat-protein mediated resistance that's more durable.
00:27:40.01 I'm going to show you a brief example of an application of this technology, now twenty years old,
00:27:46.11 this is a collaboration between my laboratory and a laboratory in Hyderabad, India,
00:27:51.23 this is ICRISAT, it studies the crops that are important in southern, tropical areas of the world
00:27:59.28 and in south India, of course, is quite tropical, very arid, and that's their focus,
00:28:06.05 as well as focusing in Africa. Here's an example of a disease in groundnuts that is really very devastating,
00:28:13.03 it can be very devastating and it's transmitted by an insect called, I mean an arthropod, thrips.
00:28:20.02 And it carries the pollen from one plant to the next, and the pollen, on the outside of the pollen,
00:28:25.10 is the virus and as the insect chews on the leaves, it transmits the virus from the pollen's surface into the leaf structure,
00:28:32.29 into the cells and off goes the disease.
00:28:35.16 But it's very devastating and the impact can be quite remarkable, as indicated by some of these headlines in the newspaper.
00:28:43.18 The disease was first identified in the mid-90's, it recurs depending upon weather and humidity,
00:28:54.16 and rainfall and you can see that it affects the area of India that's sort of the middle, downward,
00:29:00.09 the state of Maharashtra is up here, Mumbai, and then on down through Tamilnadu.
00:29:05.29 And what was important in our realization was that this disease not only affects groundnuts, or peanuts,
00:29:12.29 but it also affects other crops. And we found if we could have a strategy that helped develop resistance against this virus in peanuts,
00:29:20.06 we could also apply that in other crops or other institutions can do so.
00:29:24.08 So, this is an example, the virus affects sunflower and if you know Indian agriculture,
00:29:31.11 you know that sunflower and groundnuts are main sources of vegetable oils that are used in the cooking in India.
00:29:38.04 These two diseases have dramatically reduced the amount of cooking oil, and they have to import cooking oil
00:29:43.25 from other sources, including palm oil and soy oil.
00:29:47.20 The disease also affects melons and cotton, and marigolds, it affects the floriculture industry as well.
00:29:57.25 What was really interesting to us is that these viruses that are in these different groups of plants
00:30:03.08 are very similar to each other in sequence. We're very optimistic, then, that if we could solve the disease in peanuts,
00:30:09.22 we could maybe have some impact on these other diseases.
00:30:13.26 This virus is very different from tobacco mosaic virus, but there was a history of study of this virus
00:30:20.22 soon after we described coat protein mediated resistance against the tobamoviruses,
00:30:27.11 TMV and tomato mosaic, colleagues in Europe asked the question of whether or not the coat protein resistance
00:30:35.29 would work against these classes of virus, this is a group of viruses called Ilar, a group to which other important diseases belong.
00:30:45.07 The genome structure of this virus is quite different than TMV,
00:30:48.10 there are three different RNA molecules packed into this circular particle
00:30:53.04 and the coat protein is actually made as a sub-genomic RNA that comes off of RNA3.
00:30:58.14 We worked to develop a gene construct which was then made available to our colleagues in India.
00:31:06.27 I should say that before we did the work, we had free, full freedom to operate with all inventiveness
00:31:13.21 and we had agreements with those who help the relevant, important patents on the processes and the technology, per se,
00:31:21.12 that if we were successful, these products would be released, free of charge,
00:31:25.10 without any royalties or license fees to the farmers in India who needed them.
00:31:30.02 Kind of a philosophy that we in the academic sector can apply under a variety of conditions.
00:31:38.18 Here the slides show that the effect of infecting transgenic peanut plants with this virus in greenhouse conditions.
00:31:47.29 Our colleagues in India have developed a large number of transgenic lines
00:31:52.01 and now are evaluating them for the genetic structure and for the resistance.
00:31:56.03 They have a number of lines. This is a line that has a delay of infection, but not a high level of disease resistance.
00:32:05.25 These lines, on the other hand, are highly resistant to infection,
00:32:10.21 these cell lines, these plant lines also have higher level of coat protein than those that are less resistant.
00:32:16.08 So, it's a coat protein mediated resistance phenomenon, we're hoping that we'll be able to extent this work
00:32:23.17 into more greenhouse studies and eventually into field trials in the next year or two.
00:32:29.19 So, I'm pleased that the technology that's now more than twenty years old can actually be used in crop situations,
00:32:35.04 in India and we think in other parts of the world as well.
00:32:38.24 So, how does it work for this virus? Is it like TMV?
00:32:41.29 That we don't know, it's not been studied to the same detail at all, in the case of the resistant lines.
00:32:49.01 But some work that was done by Lee Gehrke's lab and published earlier in Science,
00:32:53.23 provided a structural model for how the capsid protein of this virus binds to,
00:32:59.16 or a virus that's like tobacco streak/alfalfa mosaic virus, ALNV, is an alfalfa virus,
00:33:05.28 and with a similar strategy for replication. And Lee and his colleagues have looked at the structure
00:33:11.22 of the complex of coat protein with the viral RNA and found that, yes indeed,
00:33:17.07 as we have known from the biological experiments that were done over the last twenty years,
00:33:21.18 that in fact, the capsid protein binds to the 3' end of the viral RNA
00:33:26.05 to some of these hairpin loop structures and important for regulating replication.
00:33:30.07 The challenge is that we know the coat protein is a positive regulator, it's essential for virus replication.
00:33:36.26 So, how does it work as a negative regulator? That we don't know.
00:33:40.18 Whether it creates specific structures, or intermediates that block its function as a positive regulator
00:33:47.04 we simply don't know and we need some ongoing work to figure out how this works.
00:33:51.16 I want to talk about a very different virus.
00:33:54.16 This is a virus that's a pararetrovirus. The disease is actually caused by co-infection of two viruses,
00:34:01.05 one that's spherical, contains RNA, one that's bacilliform.
00:34:04.20 These two together can be transmitted by the insect vector, a green leaf hopper,
00:34:10.03 and if transmitted to a young rice plant, that's shown in this case,
00:34:14.10 that the result is that it causes a dwarfing of the plant, it stunts its growth,
00:34:21.02 and accumulates large amounts of virus particles in the phloem tissue and causes this severe discoloration.
00:34:28.13 In severe cases, the infection is so strong that it prevents development to the point there is no panicle made
00:34:35.17 and if no panicle, there are no rice seeds.
00:34:37.25 I've seen fields in India where the disease is in the range of only, it reduces the output, the yield of the crop
00:34:46.29 by as much as 80 percent. It's periodic, it depends on the availability of the insect
00:34:54.16 and whether or not there are infected plants in the region.
00:34:57.18 This virus is responsible for making the proteins that allow the insect vector to acquire this virus,
00:35:07.04 as well as this virus. Now, for a number of years, we've studied coat protein mediated resistance against this virus,
00:35:13.22 and this one's an interesting virus because it contains three capsid molecules,
00:35:18.04 it looks more like some of the animal viruses than any other plant virus that we know about.
00:35:23.10 But the level of resistance provided by coat protein, or capsid proteins, in transgenic rice lines, was low.
00:35:30.11 It did reduce the amount of replication and disease, but in fact, we came to know that this virus is the responsible agent for the stunting.
00:35:39.09 We then began to attack this, to learn more about how this virus works.
00:35:43.15 And through some work of a number of outstanding colleagues, we learned, we've been tearing this virus apart
00:35:53.17 and learning how the promoter works and how replication goes on.
00:35:57.06 This is a virus that doesn't live in the US and so, all experiments related to it need to be confined to controlled environments
00:36:06.24 and locked chambers and so forth. It belongs in a class of viruses called pararetroviruses.
00:36:11.29 And the life cycle of a pararetrovirus is very different than the life cycle of tobacco mosaic virus.
00:36:17.14 The virus enters the cell, and by, again, in this case, by feeding from the insect,
00:36:21.19 and then you can see that after disassembly, the genetic information, the double-stranded DNA,
00:36:29.07 goes into the cell nucleus. Unlike other pararetroviruses, this one, there's no evidence that this one integrates into the viral genome.
00:36:36.04 But here it forms a minichromosome and that results in the product of more double-stranded DNA,
00:36:43.20 which is transported by an unknown mechanism, goes into the cytoplasm
00:36:47.13 where gene expression leads to the production of capsid proteins and movement proteins and so forth
00:36:51.19 and then the assembly of more virus particles.
00:36:54.22 So, it's very different than tobacco mosaic virus or other RNA-containing viruses.
00:36:59.14 The genome of the virus is a circular double-stranded DNA virus.
00:37:05.10 The promoter region, which is responsible for making this full-length transcript, shown in this dark band here,
00:37:11.15 is in this region. Now, in studying that in rice plants and in other plants,
00:37:17.15 we found, we came to know that the promoter that is in this region,
00:37:20.26 when linked to a reporter gene, shows expression of that promoter only in vascular tissue.
00:37:25.29 That's in line with the observation made by plant pathologists that the virus accumulates in vascular tissue.
00:37:32.20 In our studies and subsequently of the promoter region, we came to identify a unique region in this promoter sequence
00:37:40.28 that was unlike other sequences that had been described as promoter related sequences.
00:37:47.13 And we called it Box II, it's proximal to the TATA binding protein site, binding site.
00:37:55.15 Through the standard single hybrid, and two hybrid kinds of procedures, we identified proteins that bound to that promoter sequence.
00:38:06.04 These we call RF2a, rice factor two a, rice factor two b.
00:38:12.07 These can form binding, these can bind to that Box II sequence I talked to you about earlier,
00:38:17.12 either as homodimers or as heterodimers. Both of these are activators of transcription.
00:38:22.20 We characterized the function of each of these domains, proline rich, glutamic acid rich and this glutamine rich domain,
00:38:30.28 and the acidic domain and so forth.
00:38:32.26 Down here is an additional protein, this is RF, this is a protein that we call RLP1.
00:38:40.02 It also is a bZIP protein, but it does not bind to the DNA binding site alone
00:38:46.15 it can only bind in the presence of either RFIIa or RFIIb.
00:38:50.10 But it's a negative regulator.
00:38:52.26 So, our model is that, as you know, these bZIP domains recognize a unique binding sequence
00:39:00.09 and as that sequence is bound and does its function,
00:39:03.13 and one the functions, apparently is that, in binding, it can interact with TATA binding protein,
00:39:08.18 we showed that there is some specific interaction between the glutamic acid domain of these proteins
00:39:14.07 and the TATA binding protein that presumably plays some role in bringing the holoenzyme to bind on that promoter.
00:39:21.03 Once it binds, it makes the messenger RNA and off goes the translation and then process of replication and packaging that are necessary.
00:39:30.13 But the characterization of these transcription factors was an interesting part of the exercise,
00:39:36.09 but we then had some ideas about how we might use that information.
00:39:40.03 So, here's a hypothesis: if the replication strategy involves these transcription factors,
00:39:49.10 how do these, what are the roles of these transcription factors in plant development?
00:39:53.09 So, we placed the experimental emphasis on the study of the promoter
00:39:58.23 and we, and overexpression of the transcription factors in transgenic lines.
00:40:03.09 For example, here's a transgenic line that has, in which we've reduced the amount of the one transcription factor
00:40:11.06 by using an antisense gene construct. So, a transgenic line that doesn't look very happy, it's stunted.
00:40:16.01 It doesn't have many roots either.
00:40:18.10 But it can outgrow it with time. And eventually, it will produce a small amount of seeds.
00:40:24.28 In contrast, this is another in which we've overexpressed RFIIb, the second transcriptional activator.
00:40:34.09 And you see that those plants can get through this early process, but they don't continue further.
00:40:40.10 They get stunted. In fact, some of the coloration of the leaves look interesting,
00:40:44.02 in that they lack some of the pigments and they have a little bit of red tinge on them.
00:40:50.04 This is what happens when we overexpress RLP1, the negative regulator.
00:40:55.11 When we knock that gene down, that amount of RLP1 down with an antisense gene construct,
00:41:01.13 we saw no phenotype. We did see a phenotype when we overexpressed the amount of RLP1.
00:41:07.21 Now, if we compare and contrast the symptoms of disease caused by the virus,
00:41:15.19 and any of these phenotypes of the transgenic lines, it looks most like this one, stunted and a little discoloration.
00:41:21.29 And we began to have the hypothesis that maybe virus infection was having an effect on development
00:41:28.01 by reallocating, or repositioning or reusing those transcription factors for replication of the virus
00:41:34.17 and not allowing the host to do its own development.
00:41:37.25 And this was further supported by this experiment. Shunhong Dai and his colleagues developed transgenic lines
00:41:48.06 that overexpressed the DNA binding protein of one of those transcription activators, RF2a.
00:41:54.18 Now those of you who know gene regulation know that when you overexpress the DNA binding domain,
00:42:02.05 it acts as a dominant negative repressor and it binds to the sites in the chromosomal DNA of the organism,
00:42:09.07 binds those sites to which the transcription factor would normally bind, and thereby not allow the transcription factor to bind there.
00:42:16.23 Now, when we expressed that dominant negative protein under the control of a promoter that would work in all cell types,
00:42:25.14 the ubiquitin promoter, we got a very strong phenotype of stunting.
00:42:30.07 When we use a different promoter, this one from the virus, rice tungro virus,
00:42:36.11 and overexpress this guy, we got also some stunting, not quite as severe, as you would imagine if you think about it.
00:42:44.06 It's not quite as severe. If we overexpress the dominant negative protein using promoters that are known to be expressed in the vascular tissue,
00:42:55.01 this is the shrunken 1 promoter from maize that we've used in this case,
00:42:59.07 and this is from a gene called phenylalanine ammonia lyase, also both of these are vascular tissue,
00:43:05.14 we got stunting. On the other hand, if we expressed that dominant negative protein in cells which are green,
00:43:13.06 these are the ones that are not vascular tissue by and large, they are the mesofiller tissue in the leaves,
00:43:18.09 finally, the plants are not stunted, and that, again, was a validation that the function of this, of proteins that have this domain,
00:43:30.19 are important for plant growth and development through growth and influence on the vascular tissue.
00:43:36.21 Remember that the virus is infecting the vascular tissue as well.
00:43:42.07 So, we then took these cell lines and asked whether or not they were resistant to virus infection or not.
00:43:49.10 And I'm going to show you two sets of studies. I want to show you next the result of a study
00:43:55.06 that was done in the Philippines by some colleagues and PhilRice, an institute that's funded by the government.
00:44:01.18 And they had the capability of feeding these plants on insects, with insects, that have been previously fed on virus infected plants.
00:44:10.29 And then we just looked at whether or not the plants got sick or not,
00:44:15.14 whether they were stunted.
00:44:16.18 And Dr. Alfonso and his colleagues showed that the plant was stunted, as expected,
00:44:24.01 and it's very similar to a susceptible variety that they use in the Philippines.
00:44:29.03 Now, the plant line for this pot labeled number 3 is a line of rice, a cultivar of rice, that's infected, resistant to infection, or eating,
00:44:41.20 by the insect. It's not necessarily resistant to the virus per se, but the insect doesn't like to feed on it,
00:44:48.01 so it's got a resistant capability because it's not fed on by the insect.
00:44:53.29 And then beside it, lines pot 4 and 5 are those plant lines that we have overexpression of the factor RF2a.
00:45:02.02 And those that have overexpression of the factor RF2b, and they're very much more like this line.
00:45:07.06 So, our plants didn't get sicker faster, rather they had some disease resistance.
00:45:14.16 When we thought about these experiments, talked about them in the lab,
00:45:18.27 we thought, well, maybe the virus would just use more of these replication factors,
00:45:22.21 or these transcription factors and replicate faster. It seems not to be the case.
00:45:26.27 So, we needed to show that, and with additional studies.
00:45:30.02 We then went on and learned how to do the virus inoculation, not with insects,
00:45:37.03 but in the laboratory with a special tool that we had, using agrobacterium to mediate the infection process.
00:45:44.08 And of course keep these in locked chambers.
00:45:47.24 And this is the result of those experiments that we had in our laboratory
00:45:53.13 and I want to show you them in contrast to what happened in the Philippines.
00:45:57.12 So, in this line, so, this represents the population of thirty plants that were inoculated.
00:46:04.17 In the case of plants that are control line, you see that after inoculation, down at the bottom corner,
00:46:13.04 and plotting this, by about 35 or 40 days or so, the plants cease growth.
00:46:20.12 They're stunted, they stop growing, just as we would have expected them to do.
00:46:24.20 The lines up above are a series of controls. This red line is a non-inoculated partner, this is the same line as this one,
00:46:33.10 but not infected, you can see how the plants continued to grow.
00:46:36.26 And then we show the RF2b and RF2a plant lines, and you see that the plant lines that are in the blue,
00:46:44.16 these are the two lines that are having the, that have the overexpression of RF2a, the transcription factor.
00:46:52.22 And they grow as well as, if not slightly better, than the controls that were not inoculated.
00:46:57.05 Likewise, the population of RF2b plant lines is indicated in this panel.
00:47:04.01 In this panel, we're looking at disease incidence in populations of plants.
00:47:09.25 This was developed by our colleagues in the Philippines. And in this case, you have a mathematical formula for disease incidence
00:47:16.22 and you can see that there's less disease incidence over a two month period by this plant line,
00:47:22.04 which has resistance to the insect. One of our plant lines, two of our plant lines are very similar in its resistance
00:47:27.22 to this model line that they're using as their internal control.
00:47:31.29 Here you see a bubble, indicating that these plants that are highly resistant do show a burst of disease symptoms in young leaves,
00:47:41.09 but then as the plants grow, they tend to outgrow it and the disease severity index is lower.
00:47:46.18 Other plant lines that contain the other transcription factor are a little bit different than these, but much less than the control.
00:47:53.14 Now, one of the possibilities that we entertained was that if we would make more of the transcription factor,
00:47:58.29 the virus would use it and you'd replicate more. And you'd get more severe virus, disease, than in the control.
00:48:07.20 And that clearly has not happened here.
00:48:09.18 But we didn't know about, anything about the molecular processes so Shunhong (Dai) and his assistants
00:48:17.22 worked to create cell lines from these transgenic lines and non-transgenic lines.
00:48:22.24 These cell lines could then be inoculated with the virus and we could study single cycle growth curves.
00:48:29.23 Now, I'll remind you that these are non-differentiated cells, they don't have phloem, they don't have vascular tissue
00:48:35.21 but they are de-differentiated for production of these transcription factors.
00:48:41.21 In other words, we expect that the transcription factors are made in these cell lines.
00:48:45.29 And then, of course, the transgene that we've introduced in these cases would introduce the over-production of RF2a,
00:48:55.08 as well as RF2b. What we show here in these panels are the accumulation of viral RNA in non-transgenic cells
00:49:03.06 and in transgenic cells.
00:49:05.25 It's clear that this band is lighter, there's less viral RNA in this, in these two lines then there is in the control.
00:49:13.10 That was a surprise. We had not anticipated that that would be the case.
00:49:17.25 Shunhong went on with the studies and this is really a place to say if you're a graduate student or a postdoc
00:49:25.03 and you have a great idea, go ahead and do it if you can.
00:49:27.28 I advised him not to do the study and he went ahead and really to my great surprise,
00:49:34.22 has revealed more detail about the resistance than what I had anticipated.
00:49:41.15 So, this is now, back using plants, not cell lines, but plants.
00:49:46.09 And these are plant lines that have been selected
00:49:49.14 and in this group there are thirty individual plants that are, the leaves are taken from thirty plants
00:49:54.20 and combined together, so if you count across here, you can imagine how many plants were grown in this experiment.
00:50:01.15 And you can see in the dark blue line that in the non-transgenic cells, or plants,
00:50:08.21 the amount of virus RNA accumulates to a high level at about twenty to twenty five days after inoculation
00:50:15.19 and then drops off and stays high. The transgenic cell lines that overexpress, overproduce the amount of RF2a and RF2b
00:50:23.01 have considerably less viral RNA than the controls and that this group has far less than this.
00:50:29.17 Well, how about virus itself? So, he did the experiment to do straight PCR instead of real-time RT PCR up here.
00:50:37.04 He then did the accumulation of viral RNA and true to fact, it's this line, this one set of lines that you see with the short bars
00:50:45.08 are highly resistant, not only to replication, not only to gene expression, but also to virus accumulation.
00:50:51.01 So, these lines not only look good, they have less virus in them.
00:50:55.11 Now, that might, you might anticipate then that these factors, these transcription factors
00:51:00.17 are not only affecting growth and development of the plant,
00:51:03.21 but they might have something to do with this transition from the susceptibility of a juvenile plant to the resistance of the more adult plant.
00:51:11.28 Now, perhaps by affecting the innate immunity pathway or other pathways that provide innate resistance to the plant
00:51:19.10 as it grows. The model that we have for the resistance reaction now is quite different than we had imagined
00:51:26.25 or that we saw in the case of tobamoviruses, tobacco mosaic virus. In this case, the insect feeds,
00:51:34.04 carries a virus to a young plant and as a virus invades, invades in the phloem tissue, as I talked about earlier,
00:51:43.01 where these transcription factors are, and we suggest that the transcription factors are re-allocated
00:51:47.14 to the replication and gene expression of the viral genome, therefore leading to stunting and disease
00:51:53.14 and that what we've done is shift that balance so that we now are able to outgrow the virus infection.
00:52:00.12 Now, the last group of viruses I want to talk about very briefly is very different
00:52:05.00 than either of the two examples I've shown before. This is a group of viruses that are single-stranded DNA viruses.
00:52:11.27 They're important in tropical, sub-tropical regions of the world, they cause severe disease.
00:52:19.10 They are characterized by a great deal of genetic variation
00:52:23.08 and a lot of sort of promiscuous crossing between the viral species.
00:52:30.03 They mix, they recombine, produce more pathotypes and disease can be more severe or less severe
00:52:35.26 depending upon the outcome of the genetic crossing, or mixing of the genetics.
00:52:42.00 This is a disease caused in cassava and this is a severely affected plant line, clearly it's very stunted.
00:52:50.17 The cassava plant produces large roots full of starch and is a sole source of sugar, or carbon,
00:52:58.20 for many hundreds of millions of people in Africa alone.
00:53:02.02 But normally, it would grow to a height of someplace above two meters,
00:53:06.21 if it's well fed, if there are no diseases. This one is a very short stem that's perhaps less than 20 centimeters.
00:53:17.16 The virus that invades, that's causing this disease, is a gemini. Gemini geminate because it has two particles,
00:53:24.21 one on top of the other and carry, this is a group of viruses that are called the old-world gemini viruses,
00:53:32.06 have two DNA components. And I won't go through the replication strategy of this virus,
00:53:37.00 it's quite different than those I've talked about earlier in this lecture.
00:53:40.13 One of the strategies that Dr. Fauquet and his colleagues have taken is using an siRNA strategy
00:53:47.02 based on specific sequences, determining what sequences are susceptible to siRNAs
00:53:52.12 and then engineering plants through genetic transformation to create plant lines
00:53:57.12 that can then be tested in greenhouses and growth chambers
00:54:00.11 and you that he was able to develop some virus resistant plants that have an siRNA strategy.
00:54:06.13 Now another strategy that he's taken is based on work that Dr. Bruce Alberts and his colleagues did
00:54:11.19 in the mid-1970's, when Bruce was in the biochemistry department at UCSF.
00:54:17.05 And in that case, we were studying an inovirus, a bacteriophage and identified a DNA binding protein, what's called G5,
00:54:26.10 and a post-doc in our lab a number of years ago showed that the G5 protein could be produced in E. coli
00:54:33.27 and could be shown to bind to gemini virus DNA.
00:54:36.21 He then did experiments to develop plant lines that would overexpress the G5 protein
00:54:43.02 and found, surprisingly, the ability to slow down the replication of gemini viruses.
00:54:48.26 That led, then, to additional studies in which we've improved the gene expression by variety of well-known tools
00:54:57.07 that are used by molecular biologists and then developed transgenic plants that carry and produce larger amounts of this G5 protein
00:55:06.04 and have some resistance. Now, the interesting part about this strategy is that the G5 protein does not bind based on sequence.
00:55:13.00 It's non-sequence specific. So, one would think that it would bind any single-strand DNA.
00:55:18.02 On one hand, you might think the protein would affect plant growth and development.
00:55:22.05 So far, we've seen none of that. But we have seen that the plant lines that contain this protein are resistant to a variety of gemini viruses.
00:55:31.25 For example, this is a cassava mosaic virus from Sri Lanka and that's quite removed from Africa
00:55:41.07 and the sequence of this virus is quite different from others in Africa.
00:55:45.08 Yet, the expression of that G5 protein gives resistance against this strain in comparison to the control.
00:55:53.28 It also works very well against a cassava mosaic disease strain from Uganda.
00:56:01.16 And one from Kenya, as shown by these slides.
00:56:07.03 Now, the, that gives us some hope that perhaps a protein like this would have pan-typic resistance against many kinds of gemini viruses
00:56:17.19 and that work is continuing and the current status of this project is that it's now moved from experimental phase
00:56:24.16 to one that we hope will lead to the development of material that can be tested in Africa by our colleagues in Uganda and Kenya
00:56:33.10 and other parts of the continent.
00:56:35.17 We have established strong partnerships in which their scientists come to the Danforth Center for research into gemini viruses
00:56:43.02 and developing biotechnology tools to detect the virus and then to develop lines that are resistant.
00:56:48.29 So, we're learning a little bit about what it means to be pre-commercial.
00:56:53.29 How do you do that? Academic institutions aren't very good at it, we've tried to learn from the private sector.
00:56:59.12 We have applications in place at the Institutional Biosafety Committees in Uganda and Kenya
00:57:08.27 with a cultivar that we call 60444. Now, the problem is that 60444 is a nice laboratory model
00:57:15.22 and it's used in some settings as a commercial variety for starch production,
00:57:20.18 but in fact it's not farmer preferred and if you've eaten some potatoes and yams and cassava,
00:57:28.21 you know that the varieties are different and you will like a certain variety, but not another.
00:57:32.18 So, the challenge now is to transfer the technology that apparently works in this model cultivar
00:57:37.13 into farmer preferred cultivars. So, our colleagues have identified those that are preferred most by the farmers
00:57:43.23 and we're now doing the construction and the work to transform that material, the best gene constructs, into those varieties that would be used by the farmers.
00:57:53.14 The project directors here are Claude Fauquet and Nigel Taylor.
00:57:56.25 Nigel is the specialist in transformation and regeneration of tissue and Claude the gemini virologist.
00:58:03.26 And we've worked now on this project, or they have worked on this project, for nearly twenty years,
00:58:07.18 with, now, some success. And with many colleagues that have come from Africa and Latin America to help.
00:58:14.08 But the job's not done and this is, I think, the penultimate slide.
00:58:19.15 This reminds us that the projects that happen in laboratories are a long way from the projects that end up in the field.
00:58:30.03 In the development of drugs that would have pharmaceutical application,
00:58:34.25 academics normally pass off their best leads to the private sector to help them develop them for commercial purposes.
00:58:42.19 And that's largely the case in agriculture biotechnology, we know the large companies have been very successful
00:58:48.25 in corn and soybeans and canola and cotton and so forth,
00:58:53.19 but the project that we're talking about here is a project that has little or no commercial value.
00:58:58.04 It's a crop that's important for those who live on 1 to 2 dollars a day.
00:59:03.27 They can buy enough for a dollar a day to feed themselves and their families, a family of four or five.
00:59:09.13 But that's all they might eat. And so this has little or no commercial value,
00:59:15.15 so we've done this in partnerships that have started in the US and will, we hope, end in fields in Africa or Asia.
00:59:25.14 But this is a long process. Our basic work, actually, started in 1988
00:59:31.14 and it wasn't until 1999 that we have the first transgenic line that looked promising.
00:59:36.10 But it failed under more stringent conditions.
00:59:40.18 Started all over again, re-engineered the plants, now those have been working for a number of years.
00:59:47.00 We're ready for our first, initial, confined field trials. These are field trials that are in a confined space
00:59:52.15 that no one has access to, that no material leaves.
00:59:55.19 It gets burned and trashed in that field.
00:59:58.09 If they're successful there, then it goes on to the next stage of trialing, in farmers' fields.
01:00:02.26 That can take more time. While you do this, in this re-iteration, you must work with the regulatory agencies.
01:00:09.12 Remember, the transgenic crops are regulated differently than plants that are developed through classical cross-breeding
01:00:16.00 or through mutation breeding, high energy neutron bombardment, and so forth,
01:00:20.29 breaking and rearranging chromosomes is something that's approved by the processes,
01:00:26.12 but genetic engineering at this point is not, it's regulated differently, so we as academics have to learn how to do this.
01:00:32.04 This is a very time consuming process and very expensive.
01:00:35.06 And that in part is what limits, I think, the involvement of the public sector,
01:00:40.18 such as the Danforth Center scientists or scientists, perhaps at your institution, from moving into real contributions in agriculture biotechnology.
01:00:48.26 But it takes time and it takes dedication and specific funding that will allow it to go on.
01:00:54.11 So, what have we learned and what have I told you today?
01:00:57.11 I've tried to describe a number of strategies that are being used to develop disease resistant crops.
01:01:03.27 We've made great process in virus resistance.
01:01:06.16 Other laboratories are making progress in developing resistance to fungal infections or nematodes, parasitic nematodes, and bacterium.
01:01:16.23 In our case, we've focused mostly on protein-based resistance reactions,
01:01:22.09 a lot of other laboratories are working on RNA mediated events.
01:01:25.05 Through this, we've learned a lot about what works and what doesn't work.
01:01:28.15 And as we've done so, we've learned more about the host, more about the virus
01:01:33.12 and hopefully have a greater chance of contributing to the production of food, and nutrition,
01:01:39.22 around the globe, in this, where plant sciences, basic sciences and the animal side can make contributions,
01:01:48.02 if we focus on the important of basic science and the translation to applications such as I've talked about today.
01:01:54.20 I hope you've found this interesting and if you want to contact me, please feel free to do so,
01:02:01.00 at the website that's listed on the first slide.
01:02:04.03 Thank you and I hope you have a good rest of the day.
01:02:06.11 Good luck.

  • Part 1: Biology of Plant Virus Infection


There are a large number of bacterial toxins that could be used as a biological weapon. They have high mortality rates, are very toxic and are easily produced, as it would be the case of the toxin of Clostridium botulinum. These toxins produce botulism. Another interesting toxin is ricin (extracted from the shrub Ricinus communis) which has already been used as a biological weapon, has no antidote, and according to the CDC is one of the most powerful poisons that are known.

Using genetic modification has been achieved that bacteria such as Escherichia coli (that do not produce these toxins) can generate them. By inserting special genes in non-pathogenic bacteria, is becoming easier to produce large amounts of toxins.

DO NOT PANIC! Current Nations have extensive prevention and Biodefense programs. The research and knowledge of these microorganisms are the solutions to a possible biological attack.

Biological Attack Fact Sheet: Human Pathogens, Biotoxins, and Agricultural Threats

A biological attack is the intentional release of a pathogen (disease causing agent) or biotoxin (poisonous substance produced by a living organism) against humans, plants, or animals. An attack against people could be used to cause illness, death, fear, societal disruption, and economic damage. An attack on agricultural plants and animals would primarily cause economic damage, loss of confidence in the food supply, and possible loss of life. It is useful to distinguish between two kinds of biological agents:

  • Transmissible agents that spread from person to person (e.g., smallpox, Ebola) or animal to animal (e.g., foot and mouth disease).
  • Agents that may cause adverse effects in exposed individuals but that do not make those individuals contagious to others (e.g., anthrax, botulinum toxin).

The U.S. Department of Homeland Security and The National Academies teamed up in 2003 to produce fact sheets on chemical, biological, radiological, and nuclear attacks designed to help better prepare the media for the types of threats facing the nation.

Each fact sheet provides clear, concise information to the media and the public on the characteristics, dangers, and consequences associated with various types of attacks. Each fact sheet has been through a rigorous peer review process evaluated by independent members of the National Academies, many of whom are recognized as the nation's foremost experts in their field.


All living organisms have evolved ways to protect themselves against abiotic and biotic assaults. For example, microbes utilize DNA restriction/modification systems to protect against foreign DNA they also contain systems to detoxify and/or extrude xenobiotics or excessive reactive oxygen species (ROS). Multicellular organisms use other systems, and participation of one or more levels of immunity is often involved. The best studied and most appreciated in jawed vertebrates is the acquired/adaptive immune system with its well-known B and T cells and antigen-specific antibodies. This level of immunity is super-imposed on the much more fundamental, evolutionarily-ancient innate immune system, which is present not just in mammals but also in other animals and in plants. Only in the last several decades has the importance of innate immunity for the survival of multicellular organisms begun to be appreciated. It protects humans, other animals, and plants from the thousands of potentially-harmful microbes encountered daily. The development of innate immunity in multicellular organisms required the evolution of cell surface receptors that could recognize/bind molecules whose chemical structure/pattern is generally conserved within various classes of foreign organisms but is absent in “self” molecules. These conserved foreign (non-self) molecules are termed Microbe-Associated Molecular Patterns (MAMPs), also referred to as Pathogen-Associated Molecular Patterns (PAMPs), and their presence is detected by members of a large family of pattern recognition receptors (PRRs). PRRs activate one or more signaling pathways, often with the aid of co-receptors, to induce downstream defense responses. Examples of MAMPs include bacterial lipopolysaccharide, flagellin, EF-Tu, DNA, lipoproteins, peptidoglycans, and fungal chitin. Several excellent reviews of MAMPs are available [1–4].

In addition to biotic assault, organisms must cope with a variety of abiotic assaults such as mechanical or cellular damage, as well as environmental stresses like drought and salinity. Some endogenous molecules activate the innate immune system when they are released into the extracellular space (including plant apoplast) from their normal location due to damage (trauma) these molecules are referred to as Damage-Associated Molecular Patterns (DAMPs [3, 5]). DAMPs are passively released from dying cells due to damage, trauma, ischemia, or infection-induced necrosis. In addition, they can be actively secreted by certain immune cells or severely stressed cells (e.g. certain cancer cells [3]). While MAMPs are derived from microorganisms and activate the innate immune system, DAMPs are host cell derived and both initiate and perpetuate innate immune responses. It is generally accepted that these defenses help protect the damaged tissue, which is vulnerable to infection due to the disruption of physical barriers that would otherwise prevent microbial ingress. In mammals, inflammation is another component of the innate immune response it not only helps to prevent/suppress infection, but also aids in healing.

This review will focus on DAMPs, particularly those of plants. DAMPs will be compared to MAMPs and to a newly-identified class of innate immunity activators termed Nematode-Associated Molecular Patterns (NAMPs [6]) since all three classes induce many of the same defense responses and share some signal transduction components.

Animal DAMPs

We begin our discussion with animal DAMPs since they were first recognized and most extensively studied. The term DAMPs was coined by Seong and Matzinger in 2004 [7]. Table 1 lists 26 DAMPs, including purines, pyrimidines, DNA (unmethylated CpG), oxidized low-density lipoproteins, N-formyl peptides, and a variety of proteins. Cognate receptors for most have been identified (Table 1). In addition, some DAMPs form complexes with partner molecules/interactors to enhance or facilitate signaling. Among these is High Mobility Group Box 1 (HMGB1), which is one of the first identified and best characterized DAMP. HMGB1 is a highly abundant, chromatin-associated protein that is present in all animal cells [8]. It consists of two basic DNA-binding domains, designated HMG boxes A and B, and a highly acidic C-terminal tail that participates in specific intra-molecular interactions [9]. In the nucleus, HMGB1 binds the minor groove of DNA to facilitate DNA condensation, nucleosome formation, and transcription factor binding [10]. When it is released into the extracellular milieu from necrotic, damaged, or severely stressed cells, it functions as a DAMP with chemo-attractant and cytokine-inducing activities [11].

Extracellular HMGB1 mediates a range of biological responses in association with multiple receptors, such as the Receptor for Advanced Glycation End products (RAGE), Toll-like receptor 2 (TLR2), TLR4, TLR9, C-X-C chemokine receptor type 4 (CXCR4), Siglec-10, and T-Cell Immunoglobulin Mucin Receptor 3 (TIM3) [11, 12]. Notably, specific heterocomplex formation between HMGB1 and a variety of interactors, such as adaptor MD-2 or pro-inflammatory ligands lipopolysaccharides, and CpG oligodeoxynucleutides, enhances or facilitates signaling and in some cases is critical for HMGB1’s recognition by distinct receptors (Table 1). The specific heterocomplex formation appears to be at least partially regulated by the different redox states of HMGB1, which in part depend on a reversible intra-molecular disulfide bond formed between cysteine residues 23 and 45 [12, 13]. Recent studies showed that reduced HMGB1 forms a heterocomplex with CXCL12, which promotes the recruitment of inflammatory cells to damaged tissue through recognition by the CXCR4 receptor [14]. Disulfide bond-containing HMGB1 specifically binds MD-2, which facilitates recognition by TLR4, leading to induction of the NF-κB-mediated transcriptional activation of pro-inflammatory cytokines [13, 15]. HMGB1 also interacts with several other receptors, including RAGE and TLR2 it is presently unclear whether specific redox states are required for its recognition by these receptors [11]. HMGB1’s diverse activities, partner molecules, and receptors likely account for its multiple roles in many prevalent, devastating human diseases.

We recently discovered that HMGB1 binds salicylic acid (SA) this suppresses both reduced HMGB1’s chemo-attractant activity and disulfide bond-containing HMGB1’s ability to induce the expression of pro-inflammatory cytokine genes and COX-2 [16]. The SA-binding sites on HMGB1 were identified in the HMG-box domains by NMR studies and confirmed by mutational analysis. A HMGB1 protein mutated in one of the SA-binding sites retained chemo-attractant activity, but lost binding of and inhibition by SA, thereby firmly establishing that SA binding to HMGB1 directly suppresses its pro-inflammatory activities. Natural and synthetic SA derivatives with much greater potency for inhibition of HMGB1 also were identified, thereby providing proof-of-concept that new SA-based molecules with high efficacy are achievable.

Plant DAMPs

In contrast to animals, many fewer DAMPs have been identified in plants to date (Table 2). The largest and arguably the best-characterized class are polypeptides/peptides produced from larger precursor proteins. These include three families discovered by Ryan and his colleagues during their studies to identify systemin – a term “used to describe polypeptide defense signals that are produced by the plant in response to physical damage and that induce defense genes, either locally or systemically” [17]. An 18 amino acid (aa) polypeptide was isolated from 60 lb of tomato seedling and shown to induce the synthesis of wound-inducible proteinase inhibitor proteins [18]. This tomato systemin is generated by wound-induced processing of a 200 aa prohormone prosystemin, which is located in the cytoplasm of vascular phloem parenchyma cells. Systemin induces the neighboring companion cells and sieve elements of the vascular bundle to synthesize jasmonic acid (JA), which in turn systemically activates the expression of proteinase inhibitor genes [19–21].

While systemin is present in many other Solanaceous species, including potato, pepper and nightshade [22], it is not found in tobacco. This finding prompted Ryan’s group to search for another type of systemin. Ultimately, two hydroxyproline-rich 18 aa polypeptides, that are processed from a 165 aa preproprotein but share no sequence homology with the tomato systemin, were identified [17].

A third family of peptide-based DAMPs was discovered in Arabidopsis [23]. These 23 aa plant elicitor peptides (Peps) are derived from a 92 aa precursor. Two receptors have been identified for AtPepl, PEPR1, and PEPR2 [24, 25]. AtPeps induce a variety of innate immune responses and enhanced resistance, and a form of precursor ProPep3 was recently shown to be released into the extracellular space upon infection of Arabidopsis with hemi-biotrophic Pseudomonas syringae [26]. A maize (Zea mays) ortholog, ZmPep1, was subsequently identified and shown to enhance resistance to microbial pathogens, just like AtPepl [27]. For a more in-depth discussion of endogenous peptide elicitors, see Yamaguchi and Huffaker [28].

Another class of DAMPs found in plants, as well as animals, is derived from the extracellular matrix. In vertebrates fragments of hyaluronan, a simple linear polysaccharide consisting of repeating D-glucuronic acid and D-N-acetylglucosamine, induce innate immunity when released by mechanical damage or hydrolytic enzymes [29]. These fragments are perceived by the leucine-rich repeat-containing TLR2 and TLR4 receptors [29, 30]. Similarly, plants contain the pectic polysaccharide homogalacturonan, a linear polymer of 1, 4-linked α-D galacturonic acid, which helps maintain cell wall integrity. Fragments of this polymer, called oligogalacturonides (OGs), can be released mechanically or more commonly by pathogen-encoded hydrolytic enzymes. OGs induce innate immune responses, including MAPK activation, callose deposition, ROS production, elevated cytosolic Ca 2+ , and defense gene activation [31, 32]. The wall-associated kinase 1 (WAK1) has been identified as a likely receptor for OGs [33, 34].

Extracellular ATP (eATP) comprises yet another class of plant DAMPs found in both plants and animals. Despite decades of mounting evidence that eATP acts as a signaling molecule, this function was largely discounted/discredited, probably because of ATP’s ubiquitous nature and central role as the universal energy currency in all living organisms from bacteria to humans [35, 36]. Only with the identification of its plasma membrane-localized receptors, first in animals (see [35]) and then in plants [37], was its signaling function accepted in both kingdoms. In animals eATP acts as a neurotransmitter and signaling molecule that participates in muscle contraction, cell death, and inflammation [35]. Two types of receptors are involved: a G protein-coupled P2Y receptor and a ligand-gated ion channel P2X receptor. In plants eATP’s signaling role was more recently confirmed with the identification of its receptor, Does not Respond to Nucleotides 1 (DORN1 [37]). eATP’s designation as a plant DAMP is based on the combined observations that i) the dorn1 mutant displays suppressed transcriptional response not only to ATP but also to wounding, ii) most of the genes induced by application of eATP are also wound-inducible [36], and iii) eATP treatment induces typical innate immune responses, including cytosolic Ca 2+ influx, MAPK activation, and induction of dense-associated genes, including some involved in the biosynthesis of JA and ethylene [36, 38, 39]. However, it is not yet known whether it contributes to resistance to pathogens.

We recently identified a fourth class of plant DAMPs, the Arabidopsis HMGB protein AtHMGB3 [40]. All eukaryotic cells, including plants, have HMGB1-related proteins. In Arabidopsis, 15 genes encode HMG-box domain-containing proteins. They have been subdivided into four groups: (i) HMGB-type proteins, (ii) A/T-rich interaction domain (ARID)-HMG proteins, (iii) 3xHMG proteins that contain three HMG boxes, and (iv) the structure-specific recognition protein 1 (SSRP1) [41]. Based on their nuclear location and domain structure, the eight HMGB-type proteins (HMGB1/2/3/4/5/6/12/14) are thought to function as architectural chromosomal proteins, similar to mammalian HMGB1. Notably, AtHMGB2/3/4 are present in the cytoplasm and as well as the nucleus [41–43]. The cytoplasmic function of these proteins is not known. However, the cytoplasmic subpopulations should have greater access to the extracellular space (apoplast) after cellular damage as compared to the AtHMGBs located exclusively in the nucleus [41–43], since they are not bound to DNA and need only cross the plasma membrane to enter the apoplast. Given the well-established role of mammalian HMGB1 as the prototypic DAMP, the presence of a cytoplasmic subpopulation of AtHMGB3 raised the possibility that this protein serves a similar function. Indeed, when recombinant AtHMGB3 was infiltrated into Arabidopsis leaves, it exhibited DAMP-like activities similar to those of AtPep1. Treatment with either protein induced MAPK activation, callose deposition, defense-related gene expression, and enhanced resistance to necrotrophic Botrytis cinerea [40].

In contrast to mammalian HMGB1, which can be actively secreted following post-translational modification, there is no evidence for secretion of AtHMGB3. It probably enters the extracellular space passively when cells are damaged mechanically, such as by insects, or during infection by necrotrophic pathogens. Indeed B. cinerea infection caused release of AtHMGB3 into the apoplast within 24 h after inoculation. Such rapid release during the early phase of cellular necrosis induced by necrotrophs could enhance resistance by activating immune responses [40].

Additional analyses revealed that AtHMGB3, like HMGB1, binds SA, and that this interaction, which is mediated by conserved Arg and Lys residues in AtHMGB3’s single HMG box, inhibits its DAMP activity [40]. This finding appears to conflict with SA’s well-known role as a positive regulator of immune responses [44–47]. However, while SA-induced defense responses are critical for resistance to biotrophic and hemi-biotrophic pathogens, the main hormone responsible for activating defenses against necrotrophic pathogens and insects is JA [44, 45]. The JA and SA defense signaling pathways are generally mutually antagonistic [48]. SA-mediated inhibition of AtHMGB3’s DAMP activity may therefore provide one mechanism through which these pathways crosstalk. In this scenario, cellular damage caused by infection with necrotrophic pathogens would lead to the release of AtHMGB3 into the extracellular spaces this would activate JA/ethylene-associated defenses to help neutralize this threat. In contrast, infection by biotrophic pathogens induces SA biosynthesis [44, 45]. Increased SA levels could then antagonize the activation of JA-associated defenses by suppressing AtHMGB3’s DAMP activity, as well as promote the activation of SA-associated defenses that are more effective against this type of pathogen [40].

The discovery that extracellular AtHMGB3 is a plant DAMP whose immune response-inducing activity is inhibited by SA binding provides cross-kingdom evidence that HMGB proteins function extracellularly as DAMPs in both plants and animals. Moreover, it highlights the existence of common targets and shared mechanisms of action for SA in plants and humans. Interestingly, the majority of plant DAMPs identified to date have counterparts in animals. Our studies have further indicated that plants and animals share common targets of SA beyond the HMGBs [46]. For example, the glycolytic enzyme glyceraldehyde 3-phosphate dehydrogenase (GAPDH) in both plants and humans binds SA and as a result has altered activity. SA suppresses GAPDH’s roles in replication of Tomato Bushy Stunt Virus in plants and may have similar effects on hepatitis C virus replication in humans [49]. It also suppresses GAPDH-mediated neuronal cell death in animals [50]. Preliminary analyses of high-throughput screens suggest the existence of many more SA targets in both plants and humans. Perhaps the presence of multiple SA targets in animals evolved in response to either ingestion of low levels of SA that are naturally present in plant material, or endogenous synthesis of SA from benzoates [46]. Future studies will be required to assess whether these novel plant and animal SA-interacting proteins function as DAMPs.


Nematodes, one of the most abundant animals in nature, parasitize both plants and animals. Several studies indicated that plants could perceive infection by nematodes [51–53], but the identity of the perceived nematode-derived signal was unknown. We recently identified a group of defense signaling molecules from several genera of plant-parasitic nematodes, including both root-knot and cyst nematodes [6]. They are an evolutionarily conserved family of nematode pheromones called ascarosides. Ascr#18, the most abundant ascaroside in plant-parasitic nematodes, induces hallmark innate immune responses including activation of i) MAPKs, ii) defense genes, and iii) the SA and JA defense-signaling pathways, as well as, enhanced resistance to viral, bacterial, fungal, and oomycete pathogens and root-knot nematodes in several dicot and monocot plant species.

Recent candidate viral vaccines produced in plants

Hepatitis B virus vaccines

Hepatitis B virus (HBV) vaccines are one of the blockbuster vaccine success stories of modern times: since identification of the virus in the 1960s, it took less than 20 years for a subunit vaccine to get to market. However, this was in the form of 22 nm subviral particles purified from the serum of human carriers of HBV, and although highly effective, was expensive to produce and of limited supply – to say nothing of the ever-present risk associated with a blood product isolated from HBV carriers who may carry any number of other, as yet undetected viruses. It was a triumph of modern molecular biology, therefore, when a very similar virus-like particle (VLP) vaccine derived from expression of the HBV small surface antigen (S-HBsAg) was developed in 1984 [12]. While this was initially still expensive – US$40/dose, with three intramuscular doses being necessary – prices have come down very significantly, to the point that more than 110 countries now routinely immunise infants as part of the Extended Programme of Immunisation (EPI) [13]. The recombinant vaccines are highly effective and safe, and have helped set the standard for later introductions, such as of the recombinant VLP-based Human papillomavirus (HPV) vaccines.

However, there is still space for improvements in HBV vaccines, both in terms of cost of goods, and in specific antigen content. An increasing desire worldwide for “needle-free” vaccine delivery, for example, would require cheaper production of larger amounts of antigen for oral delivery, which current production modalities would not be able to meet. Problems with non-response of certain groups of people to the current vaccines have also necessitated the development of third generation products, containing the middle (M-HBsAg) and/or large (L-HBsAg) surface antigens, which contain the strongly immunogenic preS1 and/or preS2 domains. However, these vaccines are more expensive and less readily available [11]. Accordingly, plant production of HBsAg-based HBV vaccines has gone on for over 20 years, with a variety of products being made pre-clinical testing of oral delivery of transgenic potato-delivered products for almost as long [14], with a preclinical trial of an orally-delivered product in 2001 [15], and human clinical trial in 2005 [16].

Oral delivery of HBsAg in transgenic plant material has not proved to be particularly effective, however, with immunogenicity generally being low [7]. This has effectively led to the general curtailment of the transgenic-plant-as vaccine efforts [9]. It is interesting in this regard that despite the concept of “vaccination via banana” having been hyped in the popular press since the 1990s (eg:, it was not until 2005 that HBsAg was first expressed in transgenic banana fruit in India, albeit at relatively low yield [17]. A recent review has proposed a combination of approaches, with parenteral vaccination with purified plant-produced HBsAg followed by oral boosting with less well purified antigen as tablets or capsules [18]: preliminary studies in mice using lyophilised HBsAg VLP-producing transgenic lettuce converted into tablets appear to add weight to the proposal [19].

The highest plant yield of conventional (=S, or small) HBsAg was achieved via use of deconstructed Tobacco mosaic virus-based cDNA MagnICON vectors (Icon Genetics, Halle, Germany): this was around 300 mg/kg wet weight in Nicotiana benthamiana, and the recombinant protein was full-length, formed disulphide-linked dimers, displayed the conformationally-determined ‘a’ antigenic determinant, and assembled correctly into VLPs. Interestingly, vacuum-mediated agroinfiltration of whole plants led to a better product, as determined by presence of the ‘a’ determinant [20].

Plant expression of HBV antigens other than the standard S HBsAg has been attempted in recent years by a number of groups. The Arizona Biodesign Institute group, who pioneered much of the biofarming of HBV vaccines, showed in 2004 that the S form of HBsAg could be used via transient agroinfiltration-mediated expression as a useful fusion partner, with an N-terminal fusions of up to 239 amino acids being tolerated without alteration of its particle-forming ability or major antigenic properties [21]. They followed this with the demonstration that the HBsAg middle protein (M protein) - with the highly immunogenic 55 amino acid pre-S2 region fused at N-terminus of the S protein – could be successfully produced in plants, and elicited stronger humoral immune responses than the S protein when injected into mice [22]. The same group used MagnICON vectors to express over 2 g/kg wet weight of plants of VLP-forming and highly immunogenic HBcAg, the “core” antigen [23]: HBc has considerable potential both as a therapeutic vaccine and as a display vehicle for other peptides, including the HBV preS region [11]. A subsequent paper from the Biodesign Institute group detailed the use of a ssDNA Bean yellow dwarf mastrevirus-based vector system (reviewed here [24]) to produce 800 mg/kg of HBcAg in N benthamiana[25].

It is safe to say that the potential for production of HBV vaccines in plants has been well realised: while wholly edible/orally-administered vaccines may remain a pipedream, levels of antigen production for both HBsAg and HBcAg have been achieved that are some of the highest for any plant-produced molecules, and the products have been shown to be antigenically appropriate and highly immunogenic.

Hepatitis C vaccines

Between 130 and 150 million people worldwide are chronically infected with Hepatitis C virus (HCV), and around 350 000 people die annually from HCV-related liver disease [26]. While a significant proportion of those infected will clear their infections naturally, the majority will go on to develop chronic infections – with a 15-30% risk of cirrhosis of the liver within 20 years. The only treatment presently available is antivirals, with the most common regime being combination antiviral therapy with interferon and ribavirin, which are effective against all viral genotypes. Newer drugs are becoming available, such as the very promising sofosbuvir [27] however, vaccines will be the only effective means of preventing infection from occurring.

Plant production of candidate HCV vaccines has a reasonably long history, given that the virus was only described in 1989 [28],[29]. It is interesting that the first products have all been intended as therapeutic vaccines: in 2000, Nemchinov and colleagues described the use of a Tobacco mosaic virus (TMV)-derived vector in N benthamiana to express a synthetic hypervariable region 1 (HVR1)-derived peptide called R9, a potential neutralising epitope of HCV derived from the envelope protein E2, fused to the C-terminal of the B subunit of cholera toxin (CTB). The plant-derived HVR1/CTB reacted with immune sera from individuals infected with four of the major genotypes of HCV, and mice immunised intranasally with crude plant extract produced anti-HVR1 antibodies which specifically bound HCV VLPs [30]. This was followed by a succession of reports on plant expression of chimaeric plant virus coat protein molecules containing the same (R9) epitope: these included the use of Cucumber mosaic virus (CMV) CP, with immunoreactivity of sera from chronically infected patients with the recombinant CP [31] leading to use of purified R9-CMV to elicit in vivo responses in rabbits, and in vitro cellular responses as measured by interferon gamma production for lymphocytes from human patients [32],[33]. Subsequently, the same group showed that purified R9-CMV particles were stable under simulated gastric and intestinal conditions, and could elicit a humoral immune response in rabbits fed with R9-CMV infected lettuce plants [34].

The R9 epitope has also been expressed as a fusion product with the Alfalfa mosaic virus (AMV) CP, via transfecting plants transgenic for AMV RNAs 1 and 2 with recombinant CP-encoding RNA3: the R9/ALMV-CP reacted with HVR1-specific monoclonal antibodies and immune sera from individuals infected with HCV [35].

In another approach, use of an E2-derived epitope as a plant-produced C-terminal fusion to Papaya mosaic virus (PMV) CP led to formation of flexible rodlike VLPs which were actively internalised in bone-marrow-derived antigen presenting cells (APCs). C3H/HeJ mice injected twice with the VLPs exhibited a humoral response lasting more than 120 days against both the CP and the E2 epitope, with the production of IgG1, IgG2a, IgG2b and IgG3 suggesting a Th1/Th2 mixed response [36].

Another interesting recent development was the filing of a USPTO patent application on plant production of the whole E2 protein, as apparently this has not been particularly successful because of misfolding. The patent covers agroinfecting a N. benthamiana cell expressing the recombinant E2 protein with one or both of the molecular chaperones calnexin and calreticulin, as this leads to an improvement in the yield of recombinant E2 protein [37]. This adds to what appears to be a successful expression, albeit at low level, of HCV E1 protein in plants [38].

It appears as though plant production of candidate therapeutic vaccines for HCV is quite well covered however, the prophylactic vaccine space is far less well investigated: hopefully, this will change with optimisation of plant production of full-length E2 protein as detailed above.

Influenza virus vaccines

Human seasonal influenza is currently mainly caused by viruses from two distinct genera of Influenzavirus: these are three Influenza A viruses (H1N1, H1N1pdm09, and H3N2), and two lineages of Influenza B virus (Yamagata and Victoria). Annual attack rates globally are estimated at between 5–10% in adults and 20–30% in children, with about 3–5 million cases of severe illness, and about 250 000–500 000 deaths [39],[40]. The conventional chicken egg-based inactivated whole-virus split vaccine technology has a production capacity as of 2011 of 1.42 billion doses of trivalent vaccine, and a production level of 620 million doses - albeit with a six-month lead-in period every year [41]. This is manifestly obviously not capable of dealing with pandemics, when

“…the potential vaccine supply would fall several billion doses short of the amount needed to provide protection to the global population” [42].

It is worth noting that this report was issued just three years before the 2009 pandemic – and was prophetic in its predictions of vaccine shortages, and of the overly long interval between identification of the pandemic agent, and availability of vaccines. The report went on to say:

A vaccine cannot be developed with certainty until after the pandemic virus emerges.

Current global capacity to manufacture influenza vaccines is limited.

Two doses may be needed for a pandemic vaccine because of the absence of pre-existing immunity. This could further delay time to achieve protection and add operational challenges to delivery.

High antigen content may be required this would limit the total number of doses that can be made available with the current egg-based technology for inactivated vaccines. [my emphasis]

The target population for vaccination could potentially be the entire global population of over six billion this would require comprehensive resources to support operational and logistic demands in many countries.

In respect of the last point, it is worth noting that in 2001, while northern hemisphere influenza vaccine production capacity was rated at 1,069 million doses, with actual production of 534 million doses, southern hemisphere capacity was 352 million doses, with a production of only 86 million [41]. Thus, provision of pandemic vaccines for the “Global South” would be - and was in fact, in 2009 [43] - gravely lacking with current technology. I am indebted to one of the referees of this review for the observation that “It is important that all the eggs are literally not placed in one basket – or all of the vaccines put in eggs”.

Perhaps fortunately, then, influenza vaccines have been the major success story for plant-expressed antigens, largely due to two major factors: first, the haemagglutinin (HA) protein is the main determinant of neutralisation, and the only essential component of a vaccine second, it appears to be well expressed in plants, and to fold properly. Perhaps the most significant factor of plant expression of HA, however, is the fact that expression of this alone is sufficient for the efficient formation at high yield of highly immunogenic VLPs, which bud from the plant cell outer membrane [44]. This is especially relevant to influenza vaccine development in light of the fact that HA-containing VLPs produced in insect cells elicit broader cross-neutralising immune responses than whole virion inactivated influenza virus or recombinant HA [45]. These properties have been exploited by a number of groups in work reported on in detail elsewhere [46]-[48] accordingly, this review will be limited to recent developments of specific interest to pandemic and seasonal human vaccine development.

Plant-based technology lends itself very well to “orphan” or “niche” vaccines, because of what is in effect infinite scalability of production. It is also very well suited for manufacture of “rapid response” vaccines, such as those directed against pandemic influenza and bioterror agents: for this reason the US Defense Advanced Research Projects Agency (DARPA) in 2005 launched an Accelerated Manufacturing of Pharmaceuticals (AMP) programme that included an investigation of the suitability of plants as a manufacturing platform for the purpose. In 2009, as a response to the H1N1 “swine flu” virus pandemic, they initiated the “Blue Angel” effort, which was:

“…an accelerated and integrated effort to deliver effective interventions for pandemic influenza”

Part of this involved using US$100 million to fund a challenge for four companies to demonstrate a capability to use plants to produce 100 million doses of influenza vaccine a month. These were the Fraunhofer USA Center for Molecular Biotechnology in Delaware, Kentucky Bioprocessing in Owensboro, the Project GreenVax consortium with partners from Texas A&M University system and G-Con from Texas, and Medicago USA in North Carolina [49]. As of July 2012, Medicago Inc. had produced, as part of a “rapid fire” milestone, more than 10 million doses of an H1N1 VLP-based influenza vaccine candidate in one month, by Phase 1-appropriate current good manufacturing practices (cGMP) [50].

Latest developments from contenders in this challenge include preclinical and human clinical trials of a number of HA-based products. The first was of Medicago’s H5N1 A/Indonesia/5/05 HA VLP vaccine candidate, which constituted “…the first ever report of administration of a plant-made VLP vaccine to humans” [47]. In preclinical work reported in this paper, two low doses of Alhydrogel® alum-adjuvanted plant-made VLPs (1.8 μg) in ferrets prevented pathology and reduced viral loads following heterotypic (A/Vietnam/1203/04 H5N1 clade 1 virus) lethal challenge. Interestingly, this protection occurred despite the fact that HAI titres to the A/Vietnam/1203/04 challenge virus were detectable in only 75–87.5% of the challenged ferrets: this prompted the comment that the correlates of protection for influenza virus infection are not fully understood, and that the VLPs are probably stimulating innate immune responses not seen for conventional vaccines.

The human trial of the H5 HA VLPs was performed in healthy adults 18–60 years of age who received 2 doses 21 days apart of 5, 10 or 20 mg of alum-adjuvanted H5 VLP vaccine or placebo (alum). Immunogenicity was evaluated using Haemagglutination-Inhibition (HAI), Single Radial Hemolysis (SRH) and MicroNeutralisation (MN) assays: results from all three assays were highly correlated, with clear dose-responses with all measures of immunogenicity. Almost 96% of those in the 2×10 or 2×20 μg dose groups mounted detectable MN responses, indicating promising immunogenicity. In their words, “These data are particularly encouraging in light of the fact that traditional, egg-based split H5N1 vaccines showed only modest immunogenicity in humans at doses as high as 45 mg with or without alum as an adjuvant”.

One particularly encouraging aspect of this study was their testing of plant-specific glycan responses in the study group. One of the purported potential hazards of plant-made vaccines is the possibility that they might exacerbate pre-existing allergies to plant N-glycans, or even elicit such responses de novo [51]. In a clear negation of this potential for this vaccine, this study found no IgEs to plant-specific glycans in any of the 48 subjects, and no significant statistical difference in increase of IgGs to plant glycans between placebo and vaccine groups. That this may be important in influenza vaccines made in different expression systems is shown by evidence that glycosylation of recombinant A/California/04/09 HA made in insect cells and plants is different, with high mannose type glycans in plant-expressed HAs, and complex type glycoforms for the insect-expressed HA [52]. Further evidence that glycosylation of H5 HA may affect immunogenicity includes the fact that while antibody titres are higher with Sf9 insect cell-derived HA, neutralisation and HAI titres are much higher with CHO mammalian cell-produced HA [53] – although this does not seem to have been a problem with plant-produced H5 HA VLPs.

An illustration of the speed and scalability of transient expression in N benthamiana for emergency response to novel influenza virus outbreaks was shown recently by Medicago Inc., who produced grams of cGMP-grade plant-made H7N9 vaccine, as HA-only VLPs, in response to the outbreak in humans of that very severe influenza virus in China in 2013. The first vaccine lots were available only 19 days after the company accessed the H7 HA gene cDNA sequence, and as little as 3 μg in one dose of the H7 VLP vaccine administered with or without GLA (glucopyranosyl lipid A) adjuvant elicited high antibody titres in mice [54].

In 2012, accounts of Phase 1 trials of both H1N1pdm HA-derived (A/California/04/09) and HPAI H5N1 (A/Indonesia/05/05) HA-derived products were published by researchers from what is now the Fraunhofer Center for Molecular Biotechnology in Delaware, USA (, following proof that they could use a TMV-based transient expression system to produce both proteins [48]. Recombinant HA sequences both included KDEL ER retention and 6xHis affinity purification sequences at their C-termini and were purified using detergent-containing buffers, Ni-Sepharose and hydrophobic interaction and anion exchange chromatography columns, at scales up to 50 kg of N benthamiana biomass under cGMP. Mouse immunogenicity studies demonstrated that intramuscular (IM) injection of 1 μg of H1 HA or 5 μg of H5 HA protein adsorbed to 0.3% Alhydrogel as adjuvant elicited serum HAI antibody responses with titres ≥1:40 in at least 67% of vaccinated mice, with significant enhancement of immunogenicity and consequent dose-sparing afforded by use of the adjuvant. In rabbits, two IM doses of 45 μg of H1 HA plus Alhydrogel elicited HAI titres of ≥1:40 in 100% of animals, while two doses of 90 μg of H5 HA were required for the same titres in 80% of rabbits. Results in ferrets were similar, with lower responses to the H5 HA. Conclusions from this study were that both antigens were safe and immunogenic, and suited to human trial.

In contrast to the preclinical results, however, testing of the H5 HA vaccine in human volunteers showed no enhancement of immunogenicity by Alhydrogel: 15 and 45 μg doses with Alhydrogel adjuvant, and at 90 μg dose with and without Alhydrogel, were administered in a two-dose regimen three weeks apart the highest responses were in the 90 μg unadjuvanted group [55].

The H1 HA vaccine antigen was tested with sera derived from human volunteers vaccinated with a conventional AS03 adjuvanted pdmH1N1 vaccine in order to assess its vaccine potential [56]. It was recognized strongly by serum antibodies and antibody-secreting cells from vaccines. Additionally, there was good correlation between results obtained using the plant-made antigen and the conventional vaccine antigen both by ELISPOT and by intracellular cytokine staining assays. The conclusion was that the candidate H1 HA had good vaccine potential, but needed a good adjuvant for use in a clinical trial. This has been done recently [57], in a first-in-human Phase 1 study with H1 HA given twice at dose levels of 15, 45 and 90 μg with and without Alhydrogel, in healthy adults 18–50 years of age. The highest seroconversion rates, measured by HAI (78%) and virus MN assays (100%), were in the 90 μg non-adjuvanted vaccine group after the second vaccine dose.

Thus, results from these two plant-made monomeric candidate HA vaccines are similar, with high doses being required to obtain suitable responses, and – unlike the Medicago trial - no obvious effect of using Alhydrogel adjuvant. It is also questionable whether vaccine antigens containing 6xHis tags would be licenced in a final product.

Another product from this group, however, has more promise: this is an engineered soluble trimeric HA (H1N1pdm) with a heterologous trimerisation motif, which induced serum antibodies active in HAI assays as well as protective immunity in mice given a lethal virus challenge. As may have been expected, the effective doses of the trimeric HA were much lower than previously required for the monomeric HA described above [58]. Further exploration of enhancing the immunogenicity of the H1N1pdm monomeric HA was also promising: this entailed doubly-adjuvanting the antigen with silica nanoparticles (SiO2) and the mucosal adjuvant candidate bis-(3′,5′)-cyclic dimeric guanosine monophosphate (c-di-GMP) [59]. Mice were vaccinated intratracheally, which resulted in systemic humoral immune responses, and induction of a strong mucosal immune response with antigen-primed T-cells in the lungs.

While biofarming is often touted as being ideal for vaccine and other pharmaceutical production for the developing world, there is very little actually published from developing countries on use of the technology for their own purposes. One object example comes from my own lab in South Africa, where we explored the prospects of making emergency response pandemic influenza vaccines [43]. We were successful in making high-yield candidate subunit vaccines via transient agroinfiltration-mediated expression in N benthamiana from both a full-length (up to 130 mg/kg FW) and a soluble version (up to 675 mg/kg FW) of the H5N1 A/Viet Nam/1194/2004 HA, via a human codon-use optimization of the HA gene, and showed that while both proteins elicited usable titres of antibodies in mice and chickens by HAI assays, the full length protein was significantly better. Subsequent work showed that the full-length H5 HA could be extracted as semi-pure VLPs from the apoplastic space of unmacerated N benthamiana leaves via buffer infiltration and gentle centrifugation of cut leaves rolled into strips (H Inderthal, II Hitzeroth and EP Rybicki, unpublished) (Figure 1).

Transmission electron micrograph of virus-like particles composed of Influenzavirus A H5N1 haemagglutinin (HA) and plant lipids, produced by transient expression of the H5 HA via agroinfiltration of Nicotiana benthamiana leaves with recombinant Agrobacterium tumefaciens . Full-length HA was produced using the pTra-ERH vector modified for apoplastic secretion, as described by Mortimer et al. [43]. Particles were eluted by gentle centrifugation from leaves that had been vacuum infiltrated with phosphate buffered saline pH 7.4, then cut into strips and rolled up for insertion into centrifuge tubes.

A number of other different plant-produced influenza vaccine candidates are at earlier stages of development, but show promise: these include HA protein attached to a number of different partner molecules, as well as candidate “universal vaccines” based on the highly conserved M2e or M2 ion channel protein ectopic domain.

Chemical conjugation of plant-made monomeric H1N1 A/California/07/09 HA to the surface of Tobacco mosaic virus (TMV) virions produced a vaccine which elicited a potent antibody response as measured by HAI, and allowed dose-sparing in mice, with a single 15 μg IM dose being sufficient to protect all mice tested, if given with either alum (Alhydrogel) or a squalene oil-in-water based adjuvant (Addavax) adjuvant [60]. The TMV carrier is also unique in that it harnesses dendritic cell (DC) uptake and subsequent activation to stimulate effective antigen presentation, which influences the quality of the subsequent protective immune response. Using TMV as an antigen carrier allows repeated boosting without loss of immune activation to the partner antigen, even when anti-TMV antibodies were already present: in fact, prior exposure to TMV potentiated the response to HA. This group was able to produce 500 g of monomeric HA protein at pilot scale in less than a month – a good advertisement for the potential of this technology for an emergency response vaccine capability.

Another interesting approach was to make an elastin-like polypeptide fusion with a stabilised soluble trimer-forming H5N1 HA (ELPylated H5 HA, or H5-ELP): production was enhanced by ELPylation, and the molecules could be easily purified by the characteristic inverse transition cycling technique, and elicited neutralising antibodies in mice which bound plant-produced H5 HA VLPs and inactivated egg-produced virus [61].

Production of other influenza virus antigens has been limited, with the only other protein investigated to any extent being the M2e peptide, or ectopic domain of the M2 ion channel protein: this has considerable potential as a universal vaccine candidate, as it is highly conserved between all influenza A viruses. The insertion of a M2e2–24 peptide in two different locations in the Human papillomavirus type 16 VLP-forming L1 protein has been reported [62], as has the use of surface display on a filamentous VLP-forming potexvirus CP [63], albeit without preclinical evaluation. However, a more recent investigation that used surface display on TMV virions was more noteworthy in that synthetic M2e peptides derived from a H1N1 virus inserted in between TMV U1 CP residues 155 and 156 were well displayed on virions. The candidate vaccines elicited high antibody titres in IM immunised mice, and were 100% protective against 5 LD50s of homologous challenge virus (A/PR/8/34), and 70% of heterologous (A/California/04/2009) challenge [64].

While plant-made influenza viruses in general, and HA-based vaccines in particular seem to be products with significant potential, there are still some avenues to explore – such as the expression and testing of neuraminidase (NA), matrix protein (M1) and nucleoprotein (NP), all of which can play a role in immunity to and recovery from influenza virus infections, and which could constitute valuable additions to the plant-made influenza vaccine armoury.

Papillomavirus vaccines

Human papillomavirus (HPV) vaccines based on VLPs made by recombinant expression and assembly of the major capsid protein L1 are the latest blockbuster viral vaccines, with annual sales of both Merck’s yeast-made Gardasil and GSK’s insect cell-made Cervarix VLP-based vaccines close to or higher than US$1 billion as of 2012 [65]. Gardasil targets the two most prevalent HPVs found in 70% of cervical cancer cases – HPV-16 and HPV-18 – as well as the two most common genital wart-causing viruses, HPVs −6 and −11, while Cervarix contains only HPVs −16 and −18. While the global impact of the target disease is in the developing world – in 2001 women in developing countries accounted for

85% of both annual cases of cervical cancer (

500 000 cases worldwide) and annual deaths from cervical cancer (

300 000 worldwide) [66] – the main market for the vaccines is in the developed world, largely due to their expense. While this is changing to some extent, with a number of governments negotiating favourable tenders for supply of HPV vaccines for EPI programmes, it remains a problem – and the imminent release of next-generation products with wider coverage – mixtures of more types of HPV L1, or chimaeric L1 vaccines including all or part of the minor capsid protein L2, or L2-based vaccines - will do nothing to fix this problem [67],[68].

Another problem, as yet unaddressed by available vaccines, is therapeutic vaccination for those already infected with high-risk HPVs, or who have severe genital warts: neither Gardasil nor Cervarix has any effect on established infections, and neither will the new wider coverage VLP-based vaccines if they elicit the overwhelmingly humoral antibody responses seen so far [68]. Experimental vaccines based on the high-risk HPV oncogenes E6 and E7 have shown some promise in animal models however, given that most of the burden of cervical disease as well as cancer is in developing countries [69], the problem of cost and small market size will again be a problem should any licenced vaccines be developed.

The plant-based production of Human papillomavirus vaccines has had quite a long history, that has been well reviewed up to 2010 for all HPV vaccines [10], and 2013 specifically for VLPs [9]. Briefly, the important early landmarks in farming of prophylactic papillomavirus vaccines are the proof that transgenically-expressed HPV-16 or HPV-11 L1 can assemble into immunogenic VLPS, albeit at low yield [70]-[73] proofs of efficacy of a plant-made Cottontail rabbit papillomavirus (CRPV) L1-based vaccine [74] and Rabbit oral papillomavirus or CRPV L2 peptides surface-displayed on rTMV vaccines [75] very high yields of HPV-16 L1 and VLPs via agroinfiltration-mediated transient expression [76] or via transplastomic (chloroplast) expression [77]. An interesting approach was shown in a study that used transplastomic plants to express a HPV-16 L1 mutated so as to form only pentameric capsomers rather than particles, fused to the Escherichia coli heat-labile enterotoxin subunit B (LTB) as a built-in adjuvant [78]. The resultant protein accumulated to 2% of total soluble protein (TSP), and had all the correct epitopes and biochemical activity of both parent molecules, indicating that it could be a viable vaccine candidate.

One of the few investigations to look at non-genital HPV L1s was a study of plant expression of the L1 of HPV-8, a high-risk cutaneous papillomavirus associated with epidermodysplasia verruciformis and non-melanoma skin cancer in immunocompromised people [79]. The expressed protein formed VLPs in planta, albeit at relatively low yield - and only if the 22aa C-terminal nuclear localisation signal was removed. This is similar to what was found for HPV-11 L1 [73], and in contrast to results for HPV-16, where removal lowered yield [76]. In only the third investigation of expression of the L1 of a non-human papillomavirus, reasonably high levels of protein (183 mg/ml biomass) were achieved for transient agroinfiltration-mediated expression of Bovine papillomavirus type 1 L1 [80]. The protein formed VLPs, albeit only T = 1 30 nm particles rather than the 50 nm T = 7 native structure, which were nevertheless highly immunogenic in rabbits.

The only documented expression of the minor papillomavirus capsid protein L2 was from our laboratory (reported in [81]): this was HPV-16 L2, human codon optimised, and expressed to levels of

30 mg/kg – which we noted at the time could be useful, as the protein is present in virions at a maximum ratio of 1 L2 molecule per L1 capsomere, and given L1 yields that are

10–20-fold higher, co-expression of the proteins could result in efficient L1 + L2 VLP formation.

Candidate therapeutic vaccines based on the HPV-16 oncoprotein E7 have been investigated in some detail too: the first expression of the protein in plants – transient expression in N benthamiana via a recombinant Potato virus X-derived vector - was accompanied by a proof of efficacy in a mouse model, with protection from tumour development caused by the HPV-16 E7-expressing C3 cell line accompanied by strong cytotoxic T-cell responses [82]. The same group went on to show 5-fold enhanced expression of E7 by targeting of the protein to the secretory pathway [83]. The possibly immune-enhancing properties of N benthamiana extracts indicated in previous work were further explored, with proof of immunomodulatory activity on human monocyte derived dendritic cells (MDDCs) [84]. Possibly in response to concerns about using an unmodified oncoprotein as a vaccine, they engineered mutations that abolished interaction with cell cycle governing proteins (E7GGG), and showed that a plant-produced fusion product of this with the beta-1,3-1,4-glucanase (LicKM) of Clostridium thermocellum elicited the same protective response [85],[86]. Experimental scale-up of production of this fusion protein as a therapeutic vaccine was achieved with 50% final yield (at 100 mg/kg biomass) of protein at 99% purity, by means of metal-ion affinity chromatography and gel filtration [87].

An interesting fusion of prophylactic and potentially therapeutic HPV vaccines was an investigation of the feasibility of producing chimaeric HPV-16 VLPs with L1 fused to a string of cytotoxic T-lymphocyte epitopes from HPV 16 E6 and E7 proteins, in transgenic tomato plants (Lycopersicon esculentum) [88]. While expression levels were low, VLPs were formed in planta and elicited both anti-L1 neutralising antibody and anti-E6/E7 CTL responses. A development from this work was the use of the chimaeric VLPs made in tomato plants for detection of HPV-16 specific antibodies in patients with grade 1 cervical intraepithelial lesions (CIN 1) [89], pointing up the potential use of plant-made products as inexpensive reagents.

New developments in the area of mainly prophylactic or VLP-based vaccines are limited to just a few studies, largely involving chimaeric L1 molecules or fusions of L1 to other partners. However, one recent interesting variant to the classic HPV L1 VLP approach was one in which HPV-16 L1 was co-expressed with E coli LTB in doubly transgenic tobacco (N tabacum) plants to reasonable yield (

0.3% TSP), and extracts were used to orally immunise mice (4x over 52 days) [90]. Combination with LTB dramatically increased intestinal mucosal IgA responses to L1 as well as L1-specific splenic cell proliferation and the expression of IFN-γ and IL-4 in CD4+ T cells and spleen cell supernatants.

A study of chimaeric HPV-16 L1 molecules from our group [91] investigated neutralising antibody epitopes derived from HPV-16 L2 comprising amino acid residues 108–120, 56–81 or 17–36 substituted into the C-terminal helix 4 (h4) region of L1, from amino acid 414, following evidence that these were the most effective in terms of eliciting a wider spectrum of cross-neutralising antibodies to high-risk HPVs [92]-[95]. Following previous experience with L1, we used a human codon-use optimised L1 gene with similarly-optimised inserts, transiently expressed in N benthamiana via agroinfiltration, and targeted for import into chloroplasts. All chimaeras were very highly expressed, with yields of up to 1.2 g/kg plant tissue. The L1 chimaera containing L2 amino acids 108–120 (L1:L2108–120) was the most successful candidate in that it assembled into small VLPs (

30 nm), and elicited anti-L1 and anti-L2 responses in IM immunised mice, and immune sera neutralised homologous HPV-16 and heterologous HPV-52 pseudovirions. The other chimaeric L1s did not form distinct particles, and elicited significantly weaker humoral immune responses for the same dose of protein. The L1:L2108–120 particle formation was in contrast to repeated previous expression of this and the other chimaeras in insect cells, where only loose aggregates of capsomers were formed for L1:L2108–120[95],[96].

Another investigation of the potential of the L2108–120 epitope was undertaken using a fusion of the peptide to the N-terminus of the Potato virus X coat Protein (PVX CP) for surface display, expressed via a recombinant PVX vector in N benthamiana: expression levels of up to 170 mg/kg biomass were obtained, and the chimaeric protein elicited anti-L2108–120–reactive antibodies after SC injection or tattoo administration [97].

E7-based vaccines have continued to be investigated, with a number of approaches yielding promising results. A useful crossover approach mixing potentially prophylactic and therapeutic vaccines was that of the same group who developed the HPV-16 L1 chimaera with a string of T-cell epitopes from HPV 16 E6 and E7 fused to its C-terminus [98]. These workers immunised C57BL/6 mice with the antigen, and could demonstrate persistent anti-L1 IgG antibodies for over 12 months, with good neutralising activity. There was also efficient long-term protection from tumour growth induced by HPV-16 E6/E7-expressing TC-1 tumour cell challenges, and significant tumour reduction (57%) in animals with established tumours given therapeutic vaccination.

In another use of PVX CP fusions, this time with the E7GGG protein mentioned previously [99], it was shown that while both N- and C-terminal fusions with PVX CP formed long filamentous VLPs when expressed in E coli, and expression in N benthamiana was at high levels, the latter did not apparently form VLPs. Another production modality for E7GGG was investigated in the form of transplastomic Chlamydomonas reinhardtii, a well-characterised unicellular alga [100]. E7GGG was expressed alone or as a His6 or FLAG fusion protein, to levels of 0.12% of TSP. Affinity purification was performed for both tagged forms, and C57BL/6 mice were vaccinated SC with C reinhardtii extract or purified protein mixed with QuilA adjuvant. Both specific anti-E7 IgGs and E7-specific T-cell proliferation were shown in mice vaccinated with either inoculum, and tumour protection was shown after challenge with TC-1 tumour cell line expressing E7. The authors note that this was the first successful expression of a soluble E7 derivative in plants, as previously the protein had to be expressed as a fusion product in order to get a usable yield.

Another recent paper that explored the expression of E7 in transplastomic tobacco determined that targeting the protein by means of a transit peptide to the thylakoid lumen - a chloroplast inner compartment with different redox potential from the stroma, among other characteristics – increased production by more than 80-fold [101]. However, these authors did not inactivate the oncogenic potential of the protein, which means it presently has value only as an illustration of a useful yield enhancement technique.

Our group has recently explored the potential of a novel HPV-16 E7-derived gene construct – a synthetic shuffled HPV-16 E7 (16E7SH) that has lost its transforming properties, but retains all naturally-occurring CTL epitopes – for expression in plants as a candidate therapeutic vaccine [102]. The E7SH gene has previously successfully been used as a DNA vaccine in a mouse tumour model, eliciting potent cellular and humoral immune responses, including tumour protection and regression [103]. We fused the gene translationally to one encoding the Zera® peptide, a self-assembly domain of the maize gamma-zein seed storage protein that induces the accumulation of recombinant proteins into protein bodies (PBs) within the endoplasmic reticulum in a variety of eukaryotic expression systems [104]. This generally allows stabilisation of recombinant proteins, as well as enhanced accumulation and far easier purification. While E7SH alone was expressed at only a low level by agroinfiltration-mediated transient expression in N benthamiana, high-level expression of E7SH-Zera was achieved, with a maximum of 1.1 g/kg biomass, and the resultant protein bodies could be easily purified. Immune responses comparable to the E7SH DNA vaccine were demonstrated in mice, with specific humoral as well as cell-mediated immune responses, and significant tumour regression in vaccinated mice with pre-existing tumours. Interestingly, simply mixing Zera-only PBs and 16E7SH also enhanced immune responses, indicating an independent adjuvant activity for the Zera® component. Moreover, use of the E7SH-Zera gene as a DNA vaccine also resulted in increases in IFN-γ levels of mouse splenocytes compared to mice inoculated with the 16E7SH gene, a trend that was also observed for Granzyme B ELISPOT assays as well as in chromium release assays. We feel we have demonstrated proof of efficacy in a mouse tumour model of a novel HPV therapeutic vaccine candidate, which should be easy and cheap to produce and purify. For further development of this vaccine, a DNA vaccine prime followed by matched protein boost might be ideal in order to achieve further enhancements in immunogenicity.

The overall prospect for plant production of both prophylactic and therapeutic vaccines against HPV infections and disease appears bright: gold-standard VLP-based vaccine candidates can be made at high yield via transient expression, as can viable therapeutic vaccine candidates efficacy has been demonstrated in animal models for both types of vaccine early pipeline research appears to show the feasibility of making dual-purpose vaccines that may both prevent infection and hasten recovery from infection.

HIV vaccines

There is little need to either reiterate the need for vaccines for Human immunodeficiency viruses (HIV) in general, or of the predominant HIV-1 in particular there is also little need, in the face of a plethora of literature, to detail the approaches to, or problems inherent in, the development of said vaccines [105],[106]. As could be expected, HIV vaccines were an early target for plant expression studies, with a variety of targets: indeed, HIV-1 p24 capsid protein was expressed successfully in transgenic tobacco, albeit at low yield (0.35% of TSP), as long ago as 2002 [107] the suitability of transgenic maize as a production platform for oral delivery of HIV vaccines in seed extracts was tested using SIV major surface glycoprotein gp130 [108] a novel gp41-derived molecule incorporating a CTB fusion as polymerising agent and adjuvant was produced via agroinfiltration-mediated expression in N benthamiana[109] and shown to produce mucosal and serum anti-membrane proximal region (MPR) antibodies in mice after mucosal prime-systemic boost immunisation [110] a Tat monomer was produced in spinach via rTMV as a potential oral vaccine [111] epitopes derived from HIV-1 Gag and Env were fused to HBsAg and expressed as VLPs in transgenic tomato fruit [112] various Gag-derived antigens (p24, p41, p55) were produced in transgenic tobacco and via rTMV in N benthamiana for investigation of their utility as CTL-inducing immunogens by our group [113] HIV-1 Nef was produced in N benthamiana using an agroinfiltration-mediated transient expression system [114]. The field was comprehensively reviewed recently [115], so this review will be limited discussion of major advances, and more recent work.

While many and varied antigens have been produced in various types of plant using a variety of expression systems, there have been few systematic investigations of antigens regarded as central to mainstream HIV vaccine production – that is, full-length Gag and Env – and no clinical trials of any candidate vaccine products, although plant-produced anti-HIV MAbs have gone to Phase 1 trial [116]. The most successful expression of full-length Pr 55 Gag precursor polyprotein to date was done with subtype B HIV-1 Gag in transplastomic tobacco, with yields of enveloped

100 nm diameter VLPs of up to 400 mg/kg biomass [117]. This was close to 10 000-fold more than a previous best by our group, with VLPs produced in transgenic tobacco [81], and represents a viable yield for a product with serious vaccine potential: recent evidence that lack of progression to AIDS is linked to strong CTL responses to Gag [118]-[120] reinforces the need for vaccines that can target such responses, and Pr 55 Gag VLPs are potent elicitors of CTL, especially when used as a protein boost to a heterologous prime in primate models [121],[122]. A recent paper details how HIV-1 p24 antigen expressed transgenically in either Arabidopsis thaliana or Daucus carota showed a priming effect in mice fed whole plant material, eliciting humoral immune responses detected as serum anti-p24-specific IgG after an intramuscular purified p24 protein boost [123]. It is interesting that dose-dependent antigen analyses using transgenic A. thaliana showed that low p24 antigen doses were superior to high doses.

A new plant-made HIV multiantigen that makes use of VLPs is “…enveloped particles… consisting of Gag and a deconstructed form of gp41 comprising the membrane proximal external, transmembrane and cytoplasmic domains (dgp41)” [124]. This combines the proven qualities of Gag VLPs as self-adjuvanting potent humoral and cellular response-inducing immunogen with an envelope protein comprising the membrane proximal external region (MPER) of HIV gp41, which is important in infection processes and in eliciting broadly neutralising anti-HIV antibodies. Both the gag and gp41 genes were extensively deconstructed in terms of removal of potential methylation sites, and cryptic splice and polyadenylation sites, and plant codon-use optimised. Gag was expressed constitutively in transgenic N benthamiana plants under control of the CaMV 35S promoter, to levels of

22 mg/kg biomass. Transgenic plants were then infiltrated with Agrobacterium transformed with a deconstructed MagnICON replicating vector expressing the dgp41 envelope protein that was targeted to the apoplast by means of a barley alpha-amylase signal peptide. Coexpression allowed >2-fold greater accumulation of both proteins, with dgp41 reaching

9 mg/kg. Iodixanol density gradient centrifugation indicated that the coexpressed proteins cosedimented in a denser fraction than for pgp41 when expressed alone, and similarly to Gag-only fractions that were known to contain VLPs – and indeed, characteristic

100 nm diameter enveloped VLPs were seen in extracts and in situ in plant leaf sections, and could be shown to bud into the medium from protoplasted co-transfected cells. I believe these authors are not exaggerating in their claim that “These findings provide further impetus for the journey towards a broadly efficacious and inexpensive subunit vaccine against HIV-1”, given that their vaccine candidate includes both the whole of Gag, and a popular Env-derived target antigen.

The expression of full-length Env or even of gp120 in plants has been an elusive target: there is only a single published account of plant-based expression of HIV-1 Env, and despite this being highly successful, it was only as an adjunct to the demonstration of rapid and high-level transient production of functional HIV broadly neutralising monoclonal antibodies [125]. The Env was the HIV 89.6.P gp140ΔCFI envelope developed by the Vaccine Research Centre, NIH, and was expressed both in stable transgenic SR1 N tabacum and transiently via agroinfiltration in N benthamiana, as KDEL-tagged and native forms, for ER retention or secretion and consequent differential glycosylation. The gp140 was successfully expressed – to

80 mg/kg biomass – and purified by leaf homogenization and clarification, followed by Galanthus nivalis agglutinin (GNA) lectin column affinity and DEAE ion exchange chromatography. As expected, ER-retained and secreted forms of gp140 were differentially glycosylated, with the former containing only OMT glycans (no complex glycans), and the latter with only low percentages of complex-type N-glycans. This low percentage of complex glycans is similar to the high mannose glycans on virion-associated HIV-1 Env, which differs from the higher percentage of complex glycans on Env produced in mammalian cell lines. The plant-made Env also bound HIV-1-specific MAbs produced either in CHO cells or also produced in plants, indicating no significant difference between it and mammalian cell-produced Env. This is the first convincing evidence for production of a HIV-vaccine-relevant Env molecule in plants, which, together with the evidence for Gag-based VLPs including Env derivatives from the previous work, indicates that true HIV VLPs could be produced in plant systems.

One other Env-related production modality in plants that could be of commercial interest for HIV as well as for other viruses of humans and animals is revealed in a patent filed by Medicago Inc.: this details the use of chimaeric constructs of “ectodomains” from enveloped virus trimeric surface proteins – such as retroviruses, rhabdoviruses, herpes-, corona-, paramyxo-, pox- and filoviruses - fused to an influenza virus HA protein transmembrane domain and cytoplasmic tail, in order to produce VLPs similar to the HA-only VLPs previously mentioned [126]. Their HIV Env construct was a fusion of various portions of the ConS ΔCFI gp145 - an engineered Env lacking the gp120-gp41 cleavage site, the fusion peptide, an immunodominant region in gp41 and the cytoplasmic tail (CT) domain [127] – and the transmembrane (TM) and CT domains of the HA2 portion of either the H3 or the H5 HA molecule. In the words of the patent, “Although native HIV Env protein poorly accumulates in plants, a chimeric HIV Env protein, fused to a transmembrane (TM) and cytoplasmic tail (CT) domains from influenza HA accumulates at high level, and buds into HIV VLPs in absence of core or matrix protein, in plants”. This could provide a genuinely novel source of HIV Env antigens of enhanced immunogenicity, especially for use as a boost vaccine for heterologous prime-boost vaccination regimes.

Bluetongue virus vaccine

One of the more exciting success stories in veterinary biofarming in recent years is undoubtedly the proof of efficacy of a plant-made VLP-based vaccine for Bluetongue virus (BTV), a muticomponent dsRNA-containing orbivirus in the family Reoviridae[128]. This project was a part of the EU FP7 Plant Production of Vaccines (PlaProVa) initiative (, whose activities to do with VLP-based vaccines are partly reported on here [129]. Bluetongue disease is a relatively newly emerged problem in sheep and goats in northern Europe [130], and is almost certainly a result of climate change affecting the distribution of the Culicoides midges that transmit it [131]. Vaccines are seen as an essential part of disease control: however, unlike the case in endemic areas such as South Africa where attenuated live vaccines are used routinely, concerns about vaccine safety and the possible emergence of new strains of virus because of genomic reassortment with live vaccine strains, have resulted in a push for the development of recombinant protein-only vaccines [132].

It has been established that recombinant expression of BTV proteins in insect and other animal cell culture systems results in a variety of structures being formed: expression of VP3 alone produces subcore-like particles (SCLPs) VP3 and VP7 together produce core-like particles (CLPs) expression of these plus VP5 and VP2 produces authentic VLPs [133],[134]. Building on evidence that these VLPs are protective in sheep challenged with live virus [135], the PlaProVa group investigated whether it was possible to do the same using plant-produced antigens. Gene constructs were based on the Netherlands NET2006/04 strain of BTV-8, and sequences for VP2, VP3, VP5 and VP7 genes were Nicotiana codon-use optimised. The genes were cloned for expression into pEAQ-HT, an expression vector containing the Cowpea mosaic virus (CPMV)-derived HT or HyperTrans translational enhancer sequence from RNA2 [136]. Interestingly, infiltration of N benthamiana plants with either VP3 and VP5 alone produced necrotic symptoms if these were combined with the other proteins, no necrosis was observed. Simple screening of clarified extracts by density gradient ultracentrifugation showed that expression of VP3 alone, or of VP3 + VP7, or of all four together, resulted in virion proteins co-sedimenting as high-MW aggregates. Some problems with stoichiometry resulting from relative overexpression of VP3 – an over-accumulation of SCLPs - were addressed by use of vectors expressing more than one protein: VP2 and VP5 were expressed via a dual HT vector, with VP3 and VP7 together on another vector with VP3 expression being detuned by use of a non-HT-containing 5′ UTR. This had the result of down-regulating the formation of CLPs, and a significant shift in the equilibrium towards VLP formation. Total BTV-8 protein yield by use of these vectors was >200 mg per kg biomass the final yield of gradient-purified VLPs was

70 mg per kg – gratifyingly high, given a vaccine dose (see below) of 50 μg VLPs/dose.

Testing of the plant-made antigens was done in sheep: immunogenicity was assessed by injecting two sheep with 20 μg of VLPs mixed 1:1 (v/v) with Freund's incomplete adjuvant, and boosting at 21 and 42 days. Serum collected 18 days after first boost was positive for BTV antibodies using a commercial ELISA test kit, and serum from day 56 final bleed reacted with all four structural proteins in western blots, with strongest reactivity toward the major immunogenicity determinant (VP2) and the most abundant structural protein (VP7). Efficacy was tested by injecting four groups of five sheep with either 50 μg VLP or 200 μg CLP mixed 1:1 (v/v) with Montanide ISA70 VG adjuvant, or 5 × 10 4 TCID50/mL commercial live attenuated BTV-8, and boosting on day 28. Animals were challenged with 1 mL infected sheep blood containing live BTV-8 on day 63, and clinical reactions monitored for 2 weeks. Plant-produced VLPs had an identical protective efficacy profile as assessed by clinical reaction index (CRI) as the live attenuated, BTV-8 vaccine, while plant-produced CLPs were poorly protective. Sera from both the VLP and the live attenuated vaccine group showed high serum neutralization titres after day 28 however, VLPs induced high antibody levels only after booster injection, whereas neutralising antibodies were elicited by the attenuated vaccine as soon as 7 days after vaccination. Plant-produced CLPs offered partial protection against live virus challenge, similar to insect cell–produced CLPs.

The results of this investigation are significant for biofarming and vaccinology in particular for a number of reasons. First, as a general finding, it showed it is possible to reasonably easily produce a multi-protein complex with varying stoichiometry: such an approach could also be applied to other complex viruses, such as the related African horsesickness virus, human rotavirus, or picornaviruses such as poliovirus or foot and mouth disease virus. Second, and for BTV in particular, the ability to produce high levels of properly-assembled VLPs that are protective in sheep is a valuable proof of concept and of efficacy for a vaccine against an important emerging disease. Additionally, the timescale of mere days for production following preparation of the agroinfiltration inoculum means that it is possible to respond quickly and potentially locally to outbreaks of emerging disease.

Dual-use or “One Health” vaccines

The “One Health Initiative” ( is a “…worldwide strategy for expanding interdisciplinary collaborations and communications in all aspects of health care for humans, animals and the environment”, which it hopes to achieve by, inter alia, “Joint efforts in the development and evaluation of new diagnostic methods, medicines and vaccines for the prevention and control of diseases across species”. There are a number of obvious viral vaccine targets for this initiative: these are all of the zoonotic viruses that affect domestic and farmed livestock for which there are either unsatisfactory human vaccines or therapeutics, or no human vaccines at all, and high-impact recently emerging viruses with no vaccines for animals or humans. A good example in the former category would be rabies virus examples for the second would be agents such as Severe acute respiratory syndrome (SARS) and Middle Eastern respiratory syndrome (MERS) coronaviruses (CoVs), West Nile virus (WNV), and Ebola and Marburg filoviruses. Thomas Monath has also recently defined a “one health paradigm” [137] which specifies three frameworks for development and use of vaccines to control zoonoses:

Framework I vaccines target dead-end human and livestock hosts.

Framework II vaccines target [mainly arthropod-vectored] infections of domesticated animals as a means of preventing spread to humans.

Framework III vaccines target wild animal reservoirs.

His example for Framework 1 vaccines would be West Nile virus for Framework II, Rabies virus, Rift Valley fever, Venezuelan equine encephalitis, and Hendra viruses for Framework III, reservoir-targeted agents such as oral bait rabies.

Given the obviousness of the targets, it is surprising that very few of them have been explored for their potential as plant-made vaccines. Rabies is one of these, and relatively early on: a chimaera of Alfalfa mosaic virus CP and epitopes derived from glycoprotein G and the nucleoprotein was successfully expressed in plants via two different plant virus-based systems. Parenterally-injected extracts protected mice from challenge, and human volunteers who had ingested rCP-containing plant material produced rabies virus-neutralising antibodies [138]. A full-length synthetic G protein gene – with a plant secretion signal peptide, and ER retention signal – was found to express reasonably well (0.4% TSP) in transgenic tobacco, and purified protein elicited complete protection against virulent intracerebral challenge in intraperitoneally-immunised mice [139].

The development of an effective oral vaccine against rabies has also been reported, with single doses of 50 μg Vnukovo strain rabies glycoprotein G in transgenic maize seed containing the antigen at 1% of TSP, protecting all vaccinated mice from lethal vampire bat rabies virus challenge [140]. Further testing of this vaccine was done in sheep: maize kernels containing different doses of G protein (0.5, 1, 1.5 and 2 mg) were given in a single dose by the oral route, and 2 mg doses elicited a degree of protection comparable to that conferred by the injected commercial vaccine [141]. The authors state that “…this is the first study in which an orally administered edible vaccine showed efficacy in a polygastric model”, which is an important landmark in veterinary and potentially One Health vaccinology.

Two important disease agents that are prime candidates for consideration as One Health vaccines and occur almost exclusively in developing countries, are the tick-borne Crimean-Congo haemorrhagic fever virus (CCHFV) and mosquito-borne Rift Valley fever virus (RVFV), both familial bunyaviruses. CCHFV has the wider distribution, including Africa, Asia, southern and eastern Europe and the Middle East [142], while RVFV is largely restricted to sub-Saharan Africa. However, both viruses are regarded as having emerging potential, with RVFV in particular regarded as having the potential to spread to Europe, Asia, and the Americas [143]. While attenuated live vaccines are available for RVFV, these are regarded as having significant side effects there are no accepted vaccines for either virus for general use in humans.

A recent study investigated the expression of the two neutralizing epitope-rich CCHFV envelope glycoproteins Gc and Gn in hairy root cultures and in leaves derived from transgenic tobacco plants [144]. Proteins accumulated to levels of 1.8 mg/kg biomass in hairy roots and 1.4 mg/kg in leaves. Separate groups of mice were fed transgenic leaves or roots, or fed the plant material and injected SC with the plant-made proteins, or vaccinated with an attenuated CCHFV vaccine as a positive control. Mice in all the immunised groups had a consistent rise in anti- Gc and Gn IgG and IgA antibodies in serum and faeces, respectively. Mice in the group that was fed and parenterally boosted, however, exhibited a significant rise in anti-CCHFV IgG (titre of 1/32 000 compared to 1/256 for oral-only) after a single boost. Additionally, the plant-purified Gc and Gn proteins reacted with human immunoglobulins in serum from a patient who had recovered from CCHFV infection. The potential for recombinant protein production in plants as a CCHFV vaccine appears obvious, although yields seem low for routine production.

The production of Rift Valley fever virus antigens in plants is much less well studied: there is one PhD thesis that reports expression of a truncated Gn or soluble ectodomain construct and of the nucleoprotein (N) gene in transgenic Arabidopsis thaliana[145]. This study chose these two antigens because Gn is the more effective at eliciting neutralising antibodies of the two envelope glycoproteins and the recombinant Gn ectodomain (tGn) is known to be protective, and the N antigen elicits non-neutralising antibodies but also a strong cellular immune response that is partially protective. The N protein accumulated to high levels while truncated Gn protein did not accumulate to levels detectable by western blot. Feeding groups of mice transgenic plant material three times (0, 2 and 4 weeks) containing tGn or

70 μg of N protein per 4 mice resulted in seroconversion after the second feeding for both antigens, with higher titres (10 3 – 10 4 ) for N compared to tGn (10 2 – 10 3 ). While these results are very preliminary as far as a vaccine goes, they indicate the feasibility of first, using transgenic or transiently-produced antigen in plant tissue as an oral vaccine in animals, and second, of producing antigen for complete or partial purification processes that could result in a human vaccine.

Anti-viral therapeutic antibodies

A useful recent review details how plant-based antibody products may “…provide lower upfront cost, shorter time to clinical and market supply, and lower cost of goods (COGs)…[and] improvements in pharmacokinetics, safety and efficacy” [146]. As an object example, an important aspect of rabies disease prevention is therapy, given the many people in developing countries who get bitten by suspected rabid animals annually – and a major development in this area is the production of potent rabies-neutralising antibodies in plants, given the prevailing situation of mainly equine-produced sera being in short supply and of variable quality. A consortium of researchers that includes South Africans have recently described the engineering and production in transgenic N tabacum plants of both a humanised IgG version of the broadly neutralising murine MAb E559, and the murine version [147]. Purification via agarose protein A/G affinity chromatography yielded 1.8 mg/kg biomass (0.04% TSP) for the chimaeric Mab, and 1.2 mg/kg biomass (0.03% TSP) for the murine Ab. Both antibodies assembled properly, and were equivalent to hybridoma-produced MAbs in neutralising a panel of lyssaviruses that included all the phylogroup I viruses classical RABV, Duvenhage virus, European bat lyssavirus types 1 and 2, and Australian bat lyssavirus, although no neutralisation was seen for the phylogroup II viruses Lagos bat virus and Mokola virus. The efficacy in post-exposure prophylaxis of the humanised Ab was tested in hamsters injected with a lethal dose of CVS-11 strain of virus: the plant-produced antibody was apparently more effective than the commercial human rabies immunoglobulin (HRIG Rabigam), in that survival for both treatment groups was >50% after 14 days, and zero and 11% for HRIG and plant Ab groups, respectively, after 28 days.

The production of anti-Ebola virus antibodies has also recently been explored in plants: this could yet become an important part of the arsenal to prevent disease in healthcare workers, given that at the time of writing an uncontrolled Ebola haemorrhagic fever outbreak was still raging in West Africa (, had spread from Guinea to Sierra Leone, Liberia, Nigeria, Spain and to the USA (, and the use of experimental solutions was not only being suggested [148], but was being put into practice.

As background, use of a high-yielding geminivirus-based transient expression system in N benthamiana that is particularly suited to simultaneous expression of several proteins had previously allowed expression of a MAb (6DB) known to protect animals from Ebola virus infection, at levels of 0.5 g/kg biomass [149]. The same group also used the same vector system (described in detail here [24]) in lettuce to produce potentially therapeutic MAbs against both Ebola and West Nile viruses [150].

A more comprehensive investigation was reported recently, of both plant production of Mabs and post-exposure prophylaxis of Ebola virus infection in rhesus macaques [151]. Three Ebola-specific mouse-human chimaeric MAbs (h-13 F6, c13C6, and c6D8 the latter two both neutralising) were produced in whole N benthamiana plants via agroinfilration of magnICON TMV-derived viral vectors. A mixture of the three MAbs – called MB-003 – given as a single dose of 16.7 mg/kg per Mab 1 hour post-infection followed by doses on days 4 and 8, protected 3 of 3 macaques from lethal challenge with 1 000 pfu of Ebola virus. The researchers subsequently showed significant protection with MB-003 treatment given 24 or 48 hours post-infection, with four of six monkeys testing surviving, compared to none in two controls. All surviving animals treated with MB-003 experienced insignificant if any viraemia, and negligible clinical symptoms compared to the control animals. A significant finding was that the plant-produced MAbs were three times as potent as the CHO cell-produced equivalents – a clear case of plant production leading to “biobetters”. A follow-up of this work investigated efficacy of treatment with MB-003 after confirmation of infection in rhesus macaques, “according to a diagnostic protocol for U.S. Food and Drug Administration Emergency Use Authorization” [152]. In this experiment 43% of treated animals survived, whereas all controls tested here and previously with the same challenge protocol died from the infection.

An article published during the Ebola outbreak [153] detailed the therapeutic use in rhesus macaques of a cocktail of anti-Ebola Mabs called ZMapp. This is described as a successor to MB-003, incorporating components of the ZMAb cocktail developed by the National Microbiology Laboratory of the Public Health Agency of Canada and another antibody mixture, and was developed by Mapp Biopharmaceutical of San Diego with Defyrus of Toronto and manufactured by Kentucky BioProcessing. The cocktail proved able to rescue 100% of macaques when administered up to 5 days post live virus challenge. It was noteworthy that advanced disease in many animals - indicated by elevated liver enzymes, mucosal haemorrhages and generalised petechia - was reversed, leading to full recovery. ZMapp was also found to bind virions of the Guinean variant of Ebola implicated in the present outbreak. The authors claim that “ZMapp exceeds the efficacy of any other therapeutics described so far, and results warrant further development of this cocktail for clinical use”.

In news received during consideration of this article that may vindicate this view, a report quoted as coming from the National Institute of Allergy and Infectious Diseases states that two US healthcare workers who contracted Ebola in Liberia were treated with ZMapp [154]. Despite being given up to nine days post-infection in one case, it appears to have been effective [155]. In later developments, the therapy was also given to another five people, two of whom died. The US government was negotiating at time of writing to produce large amounts of ZMapp as a first-line therapy [156],[157]. As an illustration of the scale-up problem facing the manufacturers, in the macaque trial referred to above [151], animals were given three doses of ZMapp at 50 mg/kg intravenously at 3-day intervals. For an adult 70 kg human given the same dose, this would mean a total of 10.5 (3 × 3.5) grams of ZMapp: assuming optimal purified yield of each of the three MAbs individually at 100 mg/kg plant biomass, this would mean 105 kg of N benthamiana would be needed to dose just one person optimally. While biofarming may be the most scalable technology for producing MAbs and other therapeutics, this sort of scale requires resources far larger than currently exist.

A novel application of the same technology was also used to produce an Ebola immune complex (EIC) in N benthamiana, consisting of the Ebola envelope glycoprotein GP1 fused to the C-terminus of the heavy chain of the humanised 6D8 MAb, which binds a linear epitope on GP1. Geminivirus vector-mediated co-expression of the GP1-HC fusion and the 6D8 light chain produced assembled immunoglobulin, which was purified by protein G affinity chromatography. The resultant molecules bound the complement factor C1q, indicating immune complex formation. Subcutaneous immunisation of mice with purified EIC elicited high level anti-GP1 antibody production, comparable to use of GP1 VLPs [158]. This is the first published account of an Ebola virus candidate vaccine to be produced in plants.

Future prospects for plant-produced vaccines

The prospects for the increasing use of plants for the manufacture of pharmaceuticals in general look increasingly bright, especially with the recent approval and licensure of Protalix’s carrot cell-produced Gaucher disease therapeutic enzyme glucocerebrosidase, traded as ELELYSO™ (taliglucerase alfa) ( This product was approved by the US Food and Drug Administration for injection in May 2012 for long-term replacement therapy for adults with a confirmed diagnosis of Type 1 Gaucher disease. While it is marketed as a biosimilar to the mammalian cell-produced enzyme, it is in fact a biobetter, given that it is naturally mannosylated – which aids uptake by macrophages, for example – unlike the conventional product, which has to be chemically altered. Other biologics such as therapeutic MAbs for use as anti-HIV agents in microbicides are also getting very close to registration – and the recent experience with anti-Ebola MAbs may well speed the process.

However, while the prospects for animal vaccines in general and possibly human therapeutic vaccines also seem favourable, given an increasing number of proofs of efficacy in this sphere, for prophylactic human vaccines specifically they are not as bright – in the short term, at least. The length and rigour of the human prophylactic vaccine developmental and clinical testing path compared to animal and human therapeutic vaccines are a huge obstacle to commercial production of novel biofarmed vaccines, as is the entrenched investment in conventional technology by Big Pharma. It may be that the technology will find a niche on the fringes of conventional vaccinology, where there is a need for small-scale production of vaccines for orphan diseases or for rapid responses to bioterror-related or emerging viral disease outbreaks. One area where biofarmed vaccines could break through soon could be rapid-response vaccines to novel influenza virus outbreaks: the capacity for such a response has already been demonstrated (see above) it remains to put it to the test in a real-life scenario. Once this happens, there is an increased likelihood of the technology being employed for other agents, such as MERS-CoV, RVFV and West Nile and Chikungunya viruses, where “One Health” principles are important.

Do plant viruses attack animals? examples? - Biology

From the smallest insect to the largest mammal, most animals eat plants. These animals are called herbivores. At first, you may think that plants just lay there and get eaten. They certainly can't get up and run away! However, plants have many defenses to help them survive.

Two Types of Defenses

  • Constitutive - A constitutive defense is one that is always present in the plant. Most plant defenses are constitutive.
  • Induced - An induced defense is a temporary defense that is targeted to defend against an area of the plant where it has been attacked or injured.

Just like us, plants can get diseases that can make them sick and die. In order to keep pathogens and small bacteria from getting inside, plants have rigid cell walls. They also have a waxy cuticle on the outside of their leaves that protects them.

Plants also have to defend against insects. Many trees and bushes have a thick bark on their branches and stems that keeps insects outside. Bark has many layers and the outside of the bark is dead and hard. This keeps all but the most determined insects from boring into the trunk of the tree.

Some plants use thorns to protect themselves from being eaten by larger animals. Thorns can poke and bother an animal enough to get it to move on to another plant. Some examples of thorns include the thorns on the stem of a rose bush and the spines on a cactus. Certain types of cactus spines can be especially dangerous as they have barbs that stick to the skin and are not easy to remove.

Plants often develop chemicals that act as poisons making an animal sick or even killing it. Over time, animals learn not to eat the poisonous plants. Some common poisonous plants include daffodil bulbs, poison ivy, wisteria, foxglove, and chrysanthemums.

Sometimes plants are able to detect when they are being attacked by certain insects. They will emit chemicals that attract predators to the animals that are attacking it.

One way to keep from being eaten is to taste bad. Many plants use chemicals to give them a bitter taste. If a better tasting plant is nearby, then the animal will move on.

Some plants have actually turned the tables on insects and not only defend against them, but eat them. One example is the venus flytrap which has a trap that looks like leaves. If a fly, or other insect, happens upon its leaves, it will quickly snap the trap close and then release enzymes to digest the insect.


Plants have evolved in an environment rich with microorganisms that are eager to capitalize on the plants' biosynthetic and energy-producing capabilities. There are approximately 450 species of plant-pathogenic viruses, which cause a range of diseases. However, plants have not been passive in the face of these assaults, but have developed elaborate and effective defence mechanisms to prevent, or limit, damage owing to viral infection. Plant resistance genes confer resistance to various pathogens, including viruses. The defence response that is initiated after detection of a specific virus is stereotypical, and the cellular and physiological features associated with it have been well characterized. Recently, RNA silencing has gained prominence as an important cellular pathway for defence against foreign nucleic acids, including viruses. These pathways function in concert to result in effective protection against virus infection in plants.

Plant and Animal Pathogens

With respect to farm animals, only a small number of viral diseases are capable of inflicting major economic damage. Examples include foot-and-mouth disease (FMD) in cattle and pigs, classical swine fever and African swine fever in pigs, and avian influenza and Newcastle disease in poultry. Some livestock diseases are “zoonotic,” meaning that they cause illness in humans as well as animals examples include anthrax, tularemia, brucellosis, avian influenza, and Rift Valley fever, caused by a mosquito-borne virus. 2

The World Organization for Animal Health classifies five non-endemic livestock pathogens as “List A” agents because of the severity of the illnesses they produce, their ease of dissemination, and their high level of transmissibility: FMD, bluetongue, Rift Valley fever, bovine spongiform encephalitis (“mad cow disease”), and avian influenza. A second group of pathogens on “List B” are moderately easy to disseminate and cause moderate diseases with low fatality rates they include brucellosis, salmonella, glanders, typhus fever, viral encephalitis, and two non-living toxins (ricin and Staphylococcus enterotoxin B). In 2011, a major pathogen of cattle, rinderpest, was eradicated worldwide, a major achievement of international animal health. 3

Terrorist attacks against crops or livestock could be carried out with a variety of harmful agents, including viruses, bacteria, fungi, nematodes, and insect pests. Most serious plant diseases are caused by fungal pathogens, such as wheat smut, rice blast, brown stripe mildew of corn, and karnal bunt of wheat. Fungal spores can be grown in large quantities, are stable under different weather conditions, and are naturally transmitted through the air. These agents dramatically reduce the yields of corn, rice, and wheat. Small amounts of fungal pathogens could spread to large areas of cropland, and their long incubation period makes them hard to detect at an early stage 2 .