AUSTRALIAN FRONTIERS OF SCIENCE, 2005
Walter and Eliza Hall Institute of Medical Research, Melbourne, 12-13 April
Molecular ecology of the coral-algal symbiosis
Dr Madeleine van Oppen, Senior Research Scientist, Australian Institute of Marine Science
Introduction by Dr Janice Lough (Session chair) In the words of Monty Python, ‘Now for something completely different.’ We have been at the cellular level, the DNA level. We are going to start talking about ecosystems and coral reefs. In fact, if this session had been given, say, 10 to 15 years ago, you might have been asking, ‘Can coral reefs survive humans?’ The answer for the Caribbean reefs is clearly no; they have virtually gone. Now we are asking, ‘Can coral reefs survive humans and human-induced climate change?’
We have been hearing much about technological advances in experimental methods. The subjects of the two talks this morning are the result of technological advances 300 years ago that allowed us to start using fossil fuels and increasing the amount of greenhouse gases in the atmosphere. We are now seeing the results of that. There is 30 per cent more carbon dioxide in the atmosphere than there was at the start of the Industrial Revolution; 30 per cent of that has actually gone into the oceans and is changing the oceans’ chemistry – which is another story.
Temperatures are now warmer than they were: in 2004 temperatures were about 0.7 °C warmer than at the end of the 19th century, when extensive instrumental records began. Nine of the 10 warmest years have occurred in the past 10 years. Temperatures could be 2° to 6° warmer by the end of this century.
We are already seeing the impacts of these climate changes, particularly in high latitudes where the warming is amplified. The Arctic sea ice is melting, the rate of melting of the Greenland Ice Sheet is increasing, the Inuit are losing their habitats, in temperate latitudes we are seeing changes in species distributions and seasonal cycles, and in the tropical oceans, where temperatures have only warmed about 0.5°, we are seeing an increased frequency of coral bleaching events. In fact, in 1998 and 2002, about 50 per cent of the Great Barrier Reef, in Australia, bleached and, I should say, a large amount of it recovered.
Can coral reefs survive the projected changes for the tropical oceans of 1° to 3° warming by the end of this century? The answer may well lie in the coral-algal symbiosis that allows the tropical coral reefs to form these big structures and to be centres of marine biodiversity. The next two speakers are very well placed to discuss this and present new results for you.
(Click on image for a larger version)
First I will give you a little bit of background, because I assume that most people are not very familiar with coral reefs.
All reef corals form an obligate symbiosis with dinoflagellates of the genus Symbiodinium, and we usually refer to those as zooxanthellae.
The photograph on the left shows the tips of a coral colony. Here you can see the little skeletal structure, with the tentacles of the coral polyp sticking out. The other picture is a magnification of the tentacles of the coral polyps, and you can clearly see the zooxanthellae sitting inside the host tissues, embedded in the endodermic cells.
(Click on image for a larger version)
The zooxanthellae occur at very high densities, more than a million per square centimetres of coral surface. The coral host is highly dependent, in terms of its energy budget, on the translocation of photosynthates from the symbionts to the host. Zooxanthella photosynthesis also results in an increased rate of calcification by the host. In fact, without the symbiosis we wouldn’t have coral reefs.
(Click on image for a larger version)
Coral bleaching is a breakdown of the symbiosis between the zooxanthellae and the coral host. Basically, the zooxanthellae are expelled by the host. It usually also involves the loss of some photosynthetic pigments by the algae.
(Click on image for a larger version)
The zooxanthellae give the coral its typical brown appearance. At the left you see a healthy coral branch; it looks brown because its tissues are full of zooxanthellae. And at the right is a recently bleached piece of coral. It is still alive, but you can’t really see the tissue because it is transparent. So the white colour is due to its skeleton.
(Click on image for a larger version)
During mass bleaching events, whole reefs can bleach white, as you can see in this picture. If bleaching conditions persist for long enough, many of these corals may actually die.
(Click on image for a larger version)
Coral mass bleaching events have greatly increased in frequency and intensity over the past 30 years. The worst worldwide bleaching event was the one in 1997–98, when coral reefs in over 50 countries were affected. This slide lists the countries where some of the worst mortalities occurred, and up to 70–99 per cent of the corals died on some reefs. On the Great Barrier Reef (GBR) we saw that approximately 42 per cent of the reefs bleached to some extent, and in 2002 the GBR experienced an even worse bleaching event, with more than 50 per cent of its reefs being affected by bleaching.
(Click on image for a larger version)
I would like to give you a little bit of background about the zooxanthellae. They are morphologically very simple organisms, just single celled. They are golden-brown in colour, and relatively few morphological characters that you can to distinguish between species. But genetically it is an extremely diverse group.
You see here a phylogenetic tree based on ribosomal DNA of zooxanthellae, of the whole genus Symbiodinium, and so far eight distinct lineages or clades, as we tend to refer to them, have been distinguished. Each of those clades comprises many species and strains. The genus is old – it is estimated to be approximately 65 million years old, so it is a very old group.
On the Great Barrier Reef, scleractinian (stony) corals are usually dominated by species of zooxanthellae in the C clade, although species in the D clade are not uncommon and occasionally we will find associations with Symbiodinium strains in clades A, B and G.
(Click on image for a larger version)
It is not surprising that this enormous taxon diversity in Symbiodinium is matched by physiological diversity. There is a range of published studies, mainly on cultured zooxanthellae, that show that zooxanthellae can differ in their response to light and to temperature, and they can differ in the production of certain compounds, such as microsporine-like amino acid synthesis. This suggests that corals within a species may not be physiologically uniform, but that the physiology may actually be affected by the taxonomic identity of the symbiont.
(Click on image for a larger version)
Although some corals transmit those symbionts from one generation to the next – from the maternal colony to the eggs – most corals don’t. Most corals have to acquire zooxanthellae from the environment every generation, and they do so early on in life, either as a juvenile polyp – here you can see a single-polyp coral colony, whose brown appearance is due to the zooxanthellae inside its tissues – or as a larva.
The photograph at the bottom left shows a light microscopy squash map of a larva, and inside the endodermis you can see brown balls which are the zooxanthellae that it has already acquired.
This uptake of zooxanthellae every generation creates an enormous opportunity for the coral host to establish an association with a range of symbionts. Indeed, there are several published studies that show that coral species can associate with a range of symbionts. Sometimes those different symbionts occur in different colonies, but sometimes even within a single colony we find more than one strain of Symbiodinium. And sometimes the distribution of those zooxanthellae is correlated to environmental factors such as light or depth.
(Click on image for a larger version)
I would like to talk to you today about two research questions. Firstly, how flexible is this coral-algal symbiosis? Secondly, is the physiological performance of the holobiont, which is the host and the symbiont together, affected by the type of endosymbionts harboured?
(Click on image for a larger version)
For a lot of our work we use juvenile corals, because it is relatively easy to obtain large numbers of individuals and also we can control both symbiotic partners. We can, basically, create different combinations, and that is very useful for experiments.
A lot of our work happens close to either Magnetic or Orpheus Islands, or one of the mid-shelf reefs, Davies Reef. What we usually do is to collect mature coral colonies just before the mass spawning event, and spawn them in bins, collect the gametes, mix in the fertiliser and raise coral larvae.
(Click on image for a larger version)
When we were trying to figure out how flexible the symbiosis actually is, we settled some of those coral larvae on terracotta tiles. (All the white dots you see here are juvenile corals that have not yet taken up zooxanthellae; that is why they still look white.) We then placed them back into the field and sampled a subset of the juveniles over time and genetically identified the zooxanthellae that they harboured inside their tissues.
(Click on image for a larger version)
This slide shows the results of that study for one particular species, where we followed zooxanthella succession over a period of approximately six months. Four weeks after settlement, in some of the juveniles we could detect a C1 strain, in others a D strain, and in some a combination of both. We knew that at that particular location C1 and D were the two most common types of zooxanthellae. So this suggests that initial uptake is not very selective.
However, over time we noticed a marked increase in juveniles that were dominated by clade D zooxanthellae. From other work that we have done, we know that mortality rates do not differ between C- and D-juveniles, so this must be due either to competition between those zooxanthella strains inside the host tissues or to some host-mediated mechanism that up- or down-regulates the relative abundance of one or other strain.
This was a bit unexpected, because adults of these species at the same reef are always dominated by C1 zooxanthellae. So somewhere between the age of six months and adulthood another change must take place, and we suspect that that may be related to reproductive maturity.
(Click on image for a larger version)
The second question is: is physiological performance affected? We looked at two traits, growth and heat tolerance. To look at growth we worked with two species from Magnetic Island, Acropora millepora and Acropora tenuis, because at Magnetic Island these corals show different specificities for zooxanthellae, where Acropora millepora associates with D zooxanthellae and Acropora tenuis with C zooxanthellae.
So again we raised larvae, we settled them onto tiles, and then we took them into the lab under sterile conditions and offered them either D or C zooxanthellae, to create four experimental treatments. We then placed them back in the field, and followed growth over time.
(Click on image for a larger version)
This slide shows the results from that experiment. You can see that both species grew faster when they associated with the C1 zooxanthellae, compared with the D-juveniles, even if this combination was not the natural one at that location. As for the A. millepora species, it was not a homologous combination but still they grew faster.
The top two photographs show those juvenile corals at the end of the experiment, and you can quickly see what an enormous size difference there is between the two treatments.
(Click on image for a larger version)
So this shows that growth rates are indeed determined by the zooxanthella strain in both coral species, with C-juveniles growing two to three times faster than D-juveniles.
[SLIDE: Is heat tolerance affected by type of zooxanthellae harboured?] (first of two slides)
Perhaps a more pertinent question is: is heat tolerance affected by the type of zooxanthellae harboured? A few studies have been published, some from my own lab but also some from other labs, that show correlations between the occurrence of certain symbiont types and water temperatures. But the study I will present to you today I think provides more compelling evidence that certain symbiont types give the coral better heat tolerance.
What we did at the Australian Institute of Marine Science (AIMS) was to conduct a transplantation experiment, where we transplanted conspecific colonies from the Keppel islands and from Davies Reef to Magnetic Island. We also transplanted Magnetic Island corals to Magnetic Island as a control. Mean summer temperatures in the capitals are more than 2° lower than the ones at Maggie, and mean summer temperatures on Davies Reef are approximately 1° lower than at Magnetic Island.
We transplanted those colonies between six and nine months before summer. During summer the Magnetic Island transplants didn’t bleach – this was not a bleaching summer, it was a normal summer – but the other transplants bleached, probably because they were acclimatised to colder waters.
[SLIDE: Effect of transplantation on zooxanthella type]
What I haven’t told you yet is that the Magnetic Island populations are dominated by clade D zooxanthellae. The Davies Reef populations and the Keppels population are both dominated by a C2 symbiont – another species in the C clade – but they have a different strain. There are some sequence differences between those strains, so they are different.
The Magnetic Island population didn’t bleach and it also didn’t change the symbionts in its tissues. Of the Davies Reef transplants, all colonies bleached pale to white, some of them died, and the ones that recovered, recovered with their original strain of Symbiodinium. All the Keppels transplants bleached white, and again some of them died. And the ones that recovered, actually recovered with a different type of symbiont, clade D Symbiodinium. From other data that we have, we suspect that there was a very low abundance of this clade D Symbiodinium already present in those tissues, but it was below the detection levels of our methods.
We were curious to find out whether this change, this ‘shuffling’ from one strain to another, had any effect on heat tolerance.
[SLIDE: Heat stress experiments]
So we took those transplants, as well as the native populations, back into the lab. At AIMS we have a temperature-controlled indoor aquarium facility where we do these kinds of experiments. We assessed a range of traits, basically to assess the heat stress response. One of those traits is the photosynthetic capacity of the zooxanthellae, which you can measure with a PAM fluorometer, as is shown here. I won’t say more about that, because Peter Ralph will talk about that a bit more in his talk. But this photosynthetic capacity is basically used as a measure of stress in corals.
[SLIDE: Is heat tolerance affected by type of zooxanthellae harboured?] (second slide, showing four graphs)
This slide shows the results of this photosynthetic capacity during the heating experiments. We had one control temperature and three heating temperatures. On the y-axis is the photosynthetic capacity – healthy corals usually have a value of between 0.6 and 0.8 – and as the coral stresses, the value drops. It is probably good to just look at the graph for 32° heating temperature. The two most dominant populations are the red and yellow ones. The red is the Magnetic Island populations, with clade D zooxanthellae, and the yellow one is actually the Keppel islands transplant that shuffled from a C type symbiont to a D type symbiont.
If we look at the Keppels native population, which still is dominated by C2 zooxanthellae, we see that it is the most sensitive population of all of them. So by changing from C to D this population has increased its heat tolerance.
The Davies native and the Davies transplanted colonies had exactly the same heat stress response, even though the transplant had been sitting at Magnetic Island for over a year by the time we did these experiments.
[SLIDE: Is heat tolerance affected by type of zooxanthellae harboured?] (second slide again, but with a statement superimposed)
So this shows that shuffling from C to D symbionts increased the thermal tolerance of coral colonies by 1°C to 1.5 °C.
(Click on image for a larger version)
The next question is then: what is the potential in corals for this shuffling of algal endosymbionts? The most commonly used genetic techniques usually detect a single-cell zooxanthella strain in a coral colony, but those techniques are not very sensitive, because any relative abundances below 5 to 10 per cent go undetected.
At AIMS we developed a more sensitive technique, using quantitative real-time PCR, and we find a completely different picture.
(Click on image for a larger version)
We looked at four common corals on the GBR here I have plotted five locations, some mid-shelf and some inshore and the y-axis shows the ratio of clade D to clade C Symbiodinium. The dotted line indicates the point where there are equal amounts of clades C and D inside the host tissues.
You can see that in many cases there is very low abundance of the D symbionts. For example, the point at the left of the top left-hand graph shows that approximately one in 10,000 cells is a clade D Symbiodinium cell, whereas the rest is clade C.
We had previously genotyped the zooxanthellae in all these coral colonies and we had found, based on less sensitive techniques, that all of them harboured only a single symbiont strain. But 78 per cent of those corals actually turned out to have a background strain of a different clade, when we used the more sensitive technique. And 90 per cent of those were dominated by C but had the more heat tolerant clade D strain as a background.
(Click on image for a larger version)
So this shows that not a few, but most corals carry their own parachutes in the form of background symbiont strains that confer better heat tolerance to the coral colony.
(Click on image for a larger version)
To summarise those experiments: we showed that initial uptake of zooxanthellae by corals is non-selective, and we propose that that is an adaptive trait, because it gives the coral the capacity to reshuffle zooxanthellae. We have also shown that corals can indeed reshuffle those zooxanthellae, and that many corals maintain an intrinsic potential to shuffle symbionts by maintaining those heat tolerant strains at very low background levels.
We showed that zooxanthellae do affect growth and heat tolerance. This suggests that the reshuffling of zooxanthellae represents a trade-off, because if you shuffle to a heat tolerant strain, that comes at the cost of reduced growth. And possibly it also has effects on competition and reproductive output.
(Click on image for a larger version)
Now the big question: will this symbiont shuffling save the reefs? Well, we saw that shuffling from C to D symbionts is likely to increase thermal tolerance, but only by 1°C to 1.5°C. Although that is a huge ecological improvement, with huge ecological benefits, it might not be enough. Predictive rates of warming over the next 100 years vary between about 1°C to 2°C and 6°C. Moreover, we don’t fully understand the consequences of shufflings. New growth will be reduced and reproduction might be reduced, so it is a true trade-off and we need to further understand what effect that will have on coral reefs.
Moreover, I think we really need to understand what the potential is for corals to adapt to increasing seawater temperatures by selection on genetic variation that is present in coral populations. I think there is quite good evidence that there is variation, because there is clear evidence for geographical differences in the heat stress response. So that, in a way, is hopeful. But we haven’t quantified what the potential rates of adaptation would be.
(Click on image for a larger version)
I would like to end by acknowledging my collaborators: Bette Wills, from James Cook University, and Ray Berkelmans, from AIMS. Some of the work I presented involved work conducted by students Angela Little and Jos Mieog, and lots of volunteers have helped out with the coral spawning field work.
