This content is current only at the time of printing. This document was printed on 21 March 2019. A current copy is located at https://apvma.gov.au/node/19446
You are here
Transcript for Andrew Negri, Australian Institute of Marine Science
Australian Models for aquatic risk assessment—seagrass study and development of alternative test methods
This presentation was delivered at the APVMA’s science feature session on 15 October 2015. The full video is available on our YouTube channel.
Thanks everybody for hanging around until this late stage in the day. Thanks to the APVMA for getting me up to talk about the new tests that we're developing for particularly seagrasses and photosystem II herbicides. Just to give you a bit of a background, I'm from Townsville at the Australian Institute of Marine Science. We're a Commonwealth authority. We're publicly funded. We do get some external funding from elsewhere. We have around 200 staff and we basically work around the top end of Australia so everywhere from the Great Barrier Reef right around there to the Ningaloo Reef on the other side. We're essentially a little bit the same as CSIRO.
We've got a range of facilities up there in Townsville. We've got some great lab facilities. We're about to get out on the reef both on the west coast and over on the east coast here. We've got the world's most advanced experimental aquarium system in the world, which was just commissioned about two years ago. We call that the National Sea Simulator.
With today's talk I'm going to be talking about herbicides and the Great Barrier Reef to start off with. Just as an introduction, why and how we became involved as a research agency rather than why do we do it, rather than consultants et cetera. I'll talk then about seagrass and PS2 herbicides and some of the background to the techniques that I'll be discussing later. I'll then get into a couple of acute eco‑toxicology tests that we've developed for seagrass and they are based around the effect on photosynthesis. What I'd like to do then is discuss a longer‑term experiment where we take the effects on the photosystem and then have a look at how that progresses to affect the whole plant. We'll look at a few ways of applying these new acute tests and that will be it.
Herbicides are detected in the Great Barrier Reef. They are applied on farms up and down the coast and if you look here you can see sugar cane in yellow around the Mackay Whitsunday area and up in the wet tropics. We also have a lot of grazing down in around the Fitzroy area and we get a lot of photosystem II herbicides from these coastal zones getting into the marine environment during the wet season when we have these monsoonal rainfalls. I think Chris will be talking a lot about these various scenarios in the next talk.
The photosystem II herbicides are the ones that we're most interested in. The reason for that is that they are the ones that we find in greater concentrations in the Great Barrier Reef. They work by blocking electron transport in photosystem II. They are an excellent pre‑emergent herbicide but the issue with that is that photosystem II is conserved right across pretty much all plants and so they can have some negative effects on non‑target species.
Here is a typical flood plume in the wet season. You can see there's a lot of turbid water heading out there into the Great Barrier Reef. This water contains a lot of particulates, it contains nutrients, and this is where the herbicides get into the near‑shore coastal regions. This is a bit of a close up but when you look from space you can sometimes see flood plumes, which go from literally hundreds to thousands of kilometres—well, the GBR is over 1000 kilometres long and almost the entire length of the GBR can be turbid from these particular flood plumes when you have very heavy rainfalls. You can see here with the turbid water it actually decreases the light that gets down to the seagrass. This is the major threat for seagrass on the Great Barrier Reef. It's actually this reduction in light. The reason we see PS2 herbicides there more than most other pesticides is that they are relatively mobile.
I want to talk today about a few of the key organisms that we're interested in that aren't normally looked at in terms of regulatory testing. Tropical micro‑algae. There's certainly been a lot of eco‑toxicology tests of temperate micro‑algae but is there anything special about tropical micro‑algae? I'm not going to talk much about that today except to say that we are interested in corals, seagrasses, et cetera, but it's the bottom of the food chain that's also very important.
Mangroves. Again, I'm not going to be talking much about mangroves today except that photosystem II herbicides could certainly affect mangroves. They're difficult to work with because they're so large but they are a really key habitat‑forming species in the Great Barrier Reef and in fact all the way around the coast. We shouldn't ignore them.
Corals, which are actually animals but they may be affected by herbicides because they are a symbiotic organism that host micro‑algae within their cells, they host dinoflagellates and these provide about 70 to 90% of the energy that the coral needs for growth and reproduction. If there's an effect of the herbicides on the dinoflagellates this may have a flow on effect to the coral itself. Finally seagrass, which is another very important species. It creates a very important habitat and it's an important food for various organisms on the reef.
Why did we become interested? It was about in the year 2000 when photosystem II herbicides were first detected in the Great Barrier Reef Marine Park. Around the same time there was a new instrument called the pulse amplitude modulated fluorometer which held great promise in being able to detect in a non‑invasive way effects of various contaminants or other pressures on plants. The Reef CRC at the time asked us whether or not this could be a useful technique to look at the effects of photosystem II herbicides that had just been discovered on corals. Here you can see a diver with a diving PAM, it's an underwater pulse amplitude modulated fluorometer. You can put it up against a plant or a coral and you can measure a proxy for the efficiency of photosynthesis in that plant or in the coral.
I'm just going to give you a very brief background on how it works. When light hits photosystem II in a plant, the energy can be either used for photochemistry or it can be emitted as fluorescence or dissipated as heat. The PAM fluorometer measures this fluorescence in two different ways. You use this ratio to describe the photosynthetic efficiency. It has a very good correlation in most scenarios with photochemistry.
When PSII herbicides get into the system they block this electron transport in photosystem II. Once this happens, if there's relatively high light, you get oxygen radicals formed within photosystem II and this causes damage. The other thing is that you get an increase in fluorescence, and this changes the ratio, it drops the ratio and indicates a drop in photosynthetic efficiency. You can measure the amount of binding of the herbicide to photosystem II, and you can measure damage to photosystem II using different measurement techniques with the PAM fluorometer.
The advantage of the PAM fluorometer is that it is non‑destructive and it's rapid. It has very high precision. It doesn't matter how many chloroplasts you've got there, it's the ratio of the fluorescence that's important so it's very precise. Importantly it measures the effects of PSII herbicides at the site of action in the plant.
I've shown you there a probe but you can also use it in a weld plate format. The imaging PAM takes, a two‑dimensional picture across surfaces as well. The disadvantage is that it's not as sensitive to other contaminants that don't have direct effect on photosystem II. However, if a contaminant affects photosystem II and damages photosystem II then it may be applicable in that case. It's used in heat stress studies in corals a lot before they bleach. Finally and we'll be talking about this, when you use a technique like this you need some supporting evidence to indicate whether this technique has any ecological relevance or not.
The initial focus of the work back in about 2003, we exposed corals to herbicides and we saw certainly there was reduced photosynthesis, there was damage to photosystem II as indicated by the PAM and we saw coral bleaching. You can see that this coral over here, although it's alive, the coral has actually ejected the symbiotic micro‑algae to reduce the amount of photo‑oxidated stress in the coral itself. It can live for a little while like that but if it doesn't get a population of symbiotes back it will run out of energy. The other thing that we found and particularly Ross Jones's work back then who works with us now was that diuron was certainly one of the most potent of the herbicides that were detected in the Great Barrier Reef.
It was around this time that the APVMA were doing their preliminary review on Diuron and we were working on our corals. We hadn't published anything at that point. We got a visit from DuPont at that time. It was really interesting because they knew a hell of a lot more about Diuron than we ever would and still do. They had several eco‑toxicologists that came along and there was some really good information transfer there. They had two questions, why the focus on Diuron and AIMS is a federal government agency and we see ourselves as the honest broker so we certainly weren't picking on Diuron. It was just that Diuron was detected in the marine environment much more often than other herbicides and its high potency made it really the pesticide that we really wanted to use as a reference. That one was easy to answer.
The next question was, are the effects of PAM fluorometry at all environmentally relevant? That was a tough question. What we did after that was put a couple of PhD students onto this problem. The first one was Marie Magnusson and she did a lot of work in a range of areas. One of the things that she looked at was whether or not there was any correlation between traditional micro‑algal growth assays and biomass assays over 72 hours with the results you get from PAM fluorometry. Here you can see a series of flasks using a typical regulatory‑type test. They've been exposed to different concentrations of herbicides that increase as you head from left to right.
This is part of Marie's results. She used micro‑algae that were isolated from the Great Barrier Reef. You've got the inhibition of growth on this axis and on this axis you've got the inhibition of photosynthetic efficiency from the PAM fluorometer. You can see that there's a really nice correlation. The PAM fluorometry data is much tighter than the growth data but you've essentially got a one‑to‑one correlation over that 72‑hour period. The reason for this is that when the herbicide affects the photosynthesis, there's very little energy storage in the diatoms and in the green algae that she looked at and there's a pretty much immediate effect on growth, which isn't the case with corals and seagrass. They're much more complicated. We would consider that in many situations PAM fluorometry can be a very good and ecologically relevant indicator for eco‑toxicology for micro‑algae at least.
More recently there's been a consensus that if any organisms are at risk from herbicides, it's much more likely to be seagrass than corals. We've moved into that area a bit more in recent times. To recap, seagrass are habitat‑forming. They are food for dugongs and turtles and I mentioned before that they are under threat from light limitation from those flood plumes I showed earlier on. When you have flood plumes, and the turbidity can be increased for very long periods of time, for up to months and recent data indicates for up to six months, you get an attenuation of light. Over years we have very heavy rainfall in the Great Barrier Reef because these sediments keep re‑suspending over time and eventually they flush out of the system. This is when you find dead turtles and dead dugongs up on the beach, after this period of time. Seagrass are really important for the people in the north that are interested in them.
There was some very early work that had been done on exposing diuron or exposing seagrasses to diuron, work by Haynes, and they showed that there was an effect on photosynthesis of 0.1µg per litre which is pretty low. It's lower than concentrations that have been found in the marine park which is up to around 1µg per litre in near‑shore environments. But the work wasn't that well designed to answer an eco‑toxicological problem. They didn't do concentration response curves, for instance. We needed to do something a little bit more robust.
In about 2011 the National Environmental Research Program (NERP), we were lucky enough to get some funding in that to do some work on pesticides. We did some work on persistence. We also did some work on toxicology. Two of the questions were to figure out how toxic were a range of pesticides to a range of seagrass species using PAM fluorometry. The other one was, as I mentioned before, if there's an effect on photosynthesis how does this translate to whole plant effects in the long term?
I'll just start with one of our acute tests. Here's some seagrasses that are potted. We bring them in from the environment. We acclimate them in the lab. They've got their root rhizome complex there so everything's intact. We looked at two species and four herbicides. We did a 72‑hour flow through experiment. I'll explain why in a minute. We measured the inhibition of photosynthesis and damage to photosystem II. We also measured growth but there was no impact on growth over that period. That's just an image of what the lab looked like before the Sea SIM was developed. It was still a pretty good system.
Here are some of the results from that initial study. Here you can see the inhibition of photosynthesis is measured by the PAM fluorometer and as the herbicide concentration increases you get an increase, you get inhibition of photosynthesis and diuron is quite a lot more potent than tebuthiuron, for instance. From concentration response curves like this we can drive the IC50 values, that's the concentrate, the inhibition concentration. That can then be used to define how potent one toxin is against another or how sensitive one organism is against another. We can also derive IC10s for that which we would often consider as the threshold for toxicity.
We certainly found effects of diuron at concentrations that can be found in the Great Barrier Reef. I think one of the important things is that we found that as you add diuron to the seagrass the effect happens relatively quickly and by 24 hours it's reached a maximum. It doesn't matter if you run that experiment for seven days. You still get a similar inhibition of photosynthesis. These are results from 72 hours but they are identical to the results we had at 24 hours and identical to some of the results we had over an 11‑week period in terms of inhibition of photosynthesis. It means we can probably get away with shorter assays.
This is just an example of one of the data sets from that paper we published a couple of years ago. Two different seagrass species. The IC50s, the IC10s are pretty close to each other. It's a very precise method. Tebuthiuron up at the other end of the scale, atrazine somewhere in the middle.
The next thing we did was to see whether or not we could develop a miniature test. In this test we used a different species which had leaves which were about a centimetre long. We wondered because we do this with corals. We don't put entire corals into an assay. We actually snip off branches of corals, we heal them and we put them in assays, so we wondered if we could do the same with seagrass. We got another student on board, Adam Wilkinson. He would snip these leaves off and put them into a 12‑well plate to see if he could develop an assay that was as valid as the previous one that I described with the potted plants. We wanted to see whether the response and sensitivity was similar and whether or not we got those results in a relatively high throughput compared to these big pots in a shorter duration.
This is a series of images of two leaves starting at zero hours. This leaf is in controlled conditions and you can see by the blue colour that it's got a high level of photosynthetic efficiency over that 24 hours. In fact, we can run it for longer. We can run it for 48 hours and the photosystem is still intact over that time. When we add 10µg per litre of diuron just to give a dramatic effect, you can see that it takes up the herbicide and it increases the fluorescence as indicated by the colour changing from green through the dirty red on the right‑hand side. You can see that the maximum effect is happening somewhere between eight and 12 hours.
One of the helpful things is that as opposed to weeds on a farm which often take up the herbicide through the root system, the seagrass and many other aquatic plants take up many of these herbicides actually straight through the leaves. It was great they didn't take it up through the vascular system where we would snip them there so it looked to be a pretty good system.
This is just a graph of the potted plants that I mentioned before, of the species we used, of the same species we used in the weld plates versus the weld plate data for inhibition as we increased the concentration of diuron. We also used hydroponic set ups. It was pretty simple though. We just took the seagrass, we removed the soil and sediments from the roots, we let them acclimate for a few days and we put them into the same system. They all had almost identical responses to diuron using the PAM fluorometer. The IC50s that we got from that were very similar to what we got from the other species, the potted plants. It's a very robust system. It's idea for comparing one plant against each other or herbicides against each other.
What we wanted to do next was to try and link the effects on photosynthesis with the effects on the whole plant and to do that we needed to run the experiments for longer. Again, we used two species. We used diuron as the reference, 11 weeks and a flow‑through system. We looked at the inhibition of photosynthesis and damage to photosystem II, but in this case we looked at the effects on the energetics of the plant which is where we expected the other sub‑lethal effects to be and also the growth and survival over that 11‑week period.
I'm just going to show you very much a summary of the results. Or the numbers that you see are the effect sizes when we had a significant result. If there was no number there was no significance or we didn't show it. I just want to try and keep things simple here. You can see that we've got some different concentrations of diuron here. We've got effects on the photosynthesis. Significant effects at 0.3µg per litre which increases when we increased the herbicide. The same goes with damage to photosystem II although it was not quite as sensitive.
If we look at the CN rations and delta‑13 in the leaves, when these increase it indicates that there is less carbon fixation through photosynthesis over that period of time. You can see decreases there as a bit of an indicator that there's less carbon from photosynthesis. What's really important is the starch though. The starch is how the plants store their energy. They store it in the root rhizomes which are really big on seagrasses. You can see here that over that period the energy level or the energy had dropped right down. Even though we didn't see an effect on growth and mortality until the highest concentration, that those seagrasses weren't in good nick at that particular time.
Here is just a diagram of seagrass and you can see that there's a lot of space here for storage of starch in that root rhizome scenario. It's a lot different to something like duck weed which is often used in aquatic toxicology. It's very difficult actually to use the shorter assays for a week or two weeks and expect to see changes in growth in these leaves because they just keep pumping the energy into the leaves to let them grow more. If you use very high concentrations of herbicides which are unrealistic you probably wouldn't get that effect, they would probably just die straight away but we wanted to keep things relatively environmentally sound.
Also these different leaves all grow at different rates. These ones are not growing at all. These ones are growing superfast and these guys are growing somewhere in the middle so it's very difficult to get accurate growth measurements. What that allowed us to do was come up with a conceptual model for the effects of PSII herbicides on seagrass. Started off with a sustained lowering of the photosynthetic efficiency and damage to photosystem II. This led to less carbon fixation. The energy in the root rhizomes was then mobilised into the leaves so the leaves could capture more light and the stored energy was then consumed and at that point you finally got mortality and effects on growth rates in the leaves.
Interesting thing about this is that's a pretty good news story for seagrass in some respects but in other respects these effects are almost identical to the effects you see when you have light attenuation, from sediments in the water, from those turbid events when you have the river water flowing into the marine park. It could be, and we suspect that if there are any effects of herbicides on seagrass in the Great Barrier Reef, it's most likely that they're just adding to the effect in a very similar way to the light attenuation because their modes of action are very similar from the effects on photosynthesis.
I'm just going to quickly show you some of the applications for some of the new tests. For the miniature test that I explained, it's great because you can just apply it so quickly to a range of different scenarios. In this case Adam exposed the seagrass to different temperatures at the same time as different diuron concentrations. You've got high inhibition of photosynthesis up here and you can see that there's this surface of effect and you actually get a great effect of the herbicide at the extreme temperatures. We've seen exactly the same thing with corals but we haven't been able to do this many treatments all at once because this is about 36 different treatments all combined together.
The other thing that you can do is to do matched data sets. Under identical conditions we looked at 10 different photosystem II herbicides and you can look at their IC50s and you can rank them in terms of potency against seagrass. We also looked at an emerging contaminant, an emerging PSII herbicide as well.
I think this is a really important one. It is about ground‑truthing some of the guidelines that were developed for water quality. This is a species sensitivity distribution for diuron that's being developed or proposed for the new ANZECC guidelines. You can see that as you increase the herbicide more species become affected as you go up that curve. Down the bottom here you can see that if your concentration of herbicide is down here, you would expect 95% of species to be protected. The ANZECC guidelines or the new guidelines, the proposed guidelines, might have a 85% species protection set at 0.3µg per litre for diuron. In fact, in the marine park where it is high conservation value world heritage area, I expect that the Marine Park Authority will apply the 99% protection level which is 0.08µg per litre which is quite a lot lower than what it is at the moment.
If we just compare where our effects on photosynthesis from the PAM fluorometry fit here, we're seeing a 10% effect at about 0.3µg per litre, and we're seeing a 50% effect up here on the curve as well. I'd be quite happy to say that you'd get a reasonable amount of protection of marine organisms down here and that seagrass would be well protected down there as well.
I think days like today are fantastic where researchers get to meet end users. Up in the north we've been holding a pesticide working group for the Great Barrier Reef because it's quite an issue up there. It's considered one of the main threats to the Great Barrier Reef in terms of pollution after nutrients and sediments. We've been holding working groups up there as part of the NERP project up until 2014 and we hope to continue something like that into the future as the new research program NESP comes on board.
I'd just like to finish up by making some conclusions about PAM fluorometry. Remember it measures PSII herbicide effects at the site of action, which is handy for that group of pesticides, herbicides. It reaches a maximum after two hours for micro‑algae, up to four hours for some herbicides with some seagrass species. We have to test that and make sure that we're at the maximum. But you don't really need to do experiments for longer if effects on photosynthesis are your end point. Measures damage to photosystem II. The effects on micro‑algae as measured by the PAM fluorometer can be ecologically relevant but it's a much more complicated scenario if you're talking about corals and seagrass but you can see effects on reproduction in corals and you can see effects on the health of the seagrass and on their growth and survival.
Just some general conclusions now. I think that it's important, the reason we get involved in this is we think it's important that regulations and guidelines take into account the organisms that add value to world heritage areas like the Great Barrier Reef, these species that we really want to protect like corals and seagrasses, et cetera. The effects of PSII herbicides on seagrass are probably not going to effect seagrass by themselves in the marine environment. They probably add to the pressures so when we do laboratory tests the results we get really probably underestimate what happens in the real world because there are other effects of low salinity, et cetera. Maybe the formulations increasing toxicity as well. It hasn't been tested. I think that some of these, if we have new photosystem II herbicides then techniques like this are ideal for measuring their relative potency. We need to continue to communicate because by communication we can tailor the research that we do to better meet the needs of the end users.
I'd just like to thank everybody for turning up today but also our research team, our collaborators at James Cook University and UQ, NERP who funded much of the work and also the APVMA. Thanks very much for flying me down for this. Thank you.
Errors and omissions excepted; check against delivery.