top of page

The banana is dying. The race is on to reinvent it before it's too late

Credit: Wilson Hennessy

[WIRED] The world's most popular fruit is facing extinction, and scientists are racing to use gene-editing to save it. To succeed, they'll need to overcome an even bigger problem: opposition to GMO crops.

During the summer of 1989, Randy Ploetz was in his laboratory just south of Miami, when he received a package from Taiwan. Ploetz, who had earned his doctorate in plant pathology five years earlier, was collecting banana diseases and regularly received mysterious packages containing pathogens pulled out of the soil from far-flung plantations. But gazing down his microscope, Ploetz realised this Taiwanese pathogen was unlike any banana disease he’d encountered before, so he sent the sample for genetic testing. It was Tropical Race 4 (TR4) – a strain of the fungus Fusarium oxysporum cubense that lives in the soil, is impervious to pesticides, and kills banana plants by choking them of water and nutrients. It was a pathogen that would go on to consume the next three decades of his professional life.

TR4 only affects a particular type of banana called the Cavendish. There are more than 1,000 banana varieties in the world, but the Cavendish, named after a British nobleman who grew the exotic fruit in his greenhouses on the edge of the Peak District, makes up almost the entire export market. The Brazilian apple banana, for example, is small and tart with firm flesh, while the stubby Pisang Awak, a staple in Malaysia, is much sweeter than the Cavendish. But no banana has become as ubiquitous as the Cavendish, which accounts for 47 per cent of all global production of the fruit. According to the Food and Agriculture Organisation of the United Nations, this amounts to 50 billion tonnes of Cavendish bananas every year – 99 per cent of all global banana exports.

The UK, which imports five billion bananas every year, has become used to this seemingly endless supply of cheap and nutritious fruits shipped from plantations thousands of kilometres away across the Atlantic. But the high-volume, low-margin banana industry has been balancing on a knife edge for decades. “It looks very stable because we’re getting bananas, but the environmental and social costs that allow that to happen have been high,” says Dan Bebber, a researcher at the University of Exeter who works on a UK government-funded project aimed at securing the future of the banana. If one part of this tightly-wound supply chain snaps, the whole export industry could come tumbling down.

Despite its ubiquity, the Cavendish is something of a genetic outlier among crops: because it has three copies of each chromosome, it is sterile and can only reproduce by creating clones of itself. This makes the Cavendish an ideal crop to grow at scale – farmers know how a plantation of Cavendish bananas will respond to pesticides, how fast its fruit will ripen, how many bananas each plant will yield. “You know what’s going to happen to a Cavendish banana when you pick it,” says Bebber. “When you put it in a refrigerated container, you know exactly what’s going to come out of the other end most of the time.” Cavendish plants are short, so they don’t blow over easily in a hurricane, are easy to spray with pesticides, and reliably produce lots of bananas.

By concentrating all their efforts on the Cavendish, banana exporters have built a system that allows a tropical fruit grown thousands of kilometres away to appear on supermarket shelves in the UK for less than £1 per kilo – undercutting fruits like apples which are grown in dozens of varieties much closer to home. “People want cheap bananas,” says Bebber. “The system is set up for a very uniform crop.” To put it bluntly – uniformity equals higher profits-per-plant for banana producers. “They are addicted to Cavendish,” says Ploetz, today a 66-year-old professor at the University of Florida’s Tropical Research and Education Centre. It is this genetic uniformity that lays the foundation for an $8 billion-a-year export industry.

A TR4-infected banana plantation near Darwin, Australia. To try and prevent its spread, the region is subject to bio-quarantine rules.

Credit: Jeff Daniels

The Cavendish hasn’t always been popular. Before the 1950s, Europe and America’s banana of choice was the Gros Michel – a creamier, sweeter banana that dominated the export market. Unlike the Cavendish, which needed to be transported in boxes to protect its fragile skin, the robust and thick-skinned Gros Michel was ideally suited to long, bumpy journeys across the Atlantic. At the time, the thin-skinned and slightly bland Cavendish was seen as a second-rate banana.

However, Gros Michel had one weakness. It was susceptible to Tropical Race 1 (TR1), an earlier strain of the Fusarium fungus. TR1 was first detected in Latin America in 1890 and, in the 60 years that followed, it tore through banana plantations in Latin America, costing the industry $2.3 billion in today’s terms. Faced with no other choice, the major banana firms switched production to their backup banana: the Cavendish. In 1960, the world’s biggest banana exporter, United Fruit Company (now called Chiquita) began switching to the Cavendish, following the lead of its smaller rival, Standard Fruit Company (now called Dole) which switched in 1947. Despite all its shortcomings, the Cavendish had one huge advantage over the Gros Michel, which disappeared from US supermarket shelves forever in 1965: it was completely resistant to TR1.

But the Cavendish has no defence against TR4. When Ploetz first encountered the new pathogen, there had been just a handful of suspected infections reported. In 1992, Ploetz received packages containing TR4 from plantations in Indonesia and Malaysia. “At the time all we knew was that it was a new pathogen,” he says. “We didn’t know what to expect as far as its broader implications. The more samples we got from these export plantations, the more we began to realise that this was a bigger issue than we had ever anticipated,” he recalls. His prediction proved to be eerily accurate.

In 2013, TR4 was found for the first time in Mozambique. Ploetz thinks it had been carried on the boots and equipment of banana planters from southeast Asia. The pathogen has now travelled to Lebanon, Israel, India, Jordan, Oman, Pakistan and Australia. In 2018, it was found in Myanmar. “Then in southeast Asia,” Ploetz says. “It’s everywhere.”

When TR4 hits, the destruction is near-total. “It looks like somebody’s gone to the plantation with a herbicide,” Ploetz says. “There are big areas that no longer have any plants at all.” The fungus, which can live undetected in the soil for decades, enters banana plants through their roots and spreads to the water- and nutrient-conducting tissue within, eventually starving the plant of nourishment. Two to nine months after being infected, the plant – hollowed out from the inside – collapses in on itself. The soil it grew in, now riddled with the fungus, is useless for growing bananas.

As TR4 creeps across the globe towards Latin America, the Cavendish’s genetic uniformity is starting to look like a curse. Ploetz estimates that TR4 has already killed more Cavendish bananas than Gros Michel plants killed by TR1, and, unlike the previous epidemic, there is no TR4-resistant banana ready to replace the Cavendish. And time to find a solution is rapidly running out. “The question is, ‘when is it going to come over here?’,” Ploetz says. “Well, it may already be here.”

So far, Latin America, which grows almost all of the world’s export bananas – including those for the US and Europe – has escaped TR4. But, Ploetz says, it’s only a matter of time. “Our concern in Central America is that if somebody has an outbreak on their property, they are going to keep their mouths shut, and then it’ll have spread widely by the time people realise it’s there,” he says.

Faced with a crisis that could see the Cavendish gone forever, a handful of researchers are racing to use gene-editing to create a better banana and bring the world’s first TR4- resistant Cavendish to the market. To get there, they will butt up against not only the limitations of technology, but resistance from lawmakers, environmentalists and consumers wary of GM crops. But as TR4 closes in on Latin America, gene-editing may be the last chance we have to save the one banana we have chosen above all others.

Christina Pignocchi, senior scientist at Tropic Biosciences, inspects a crop of Cavendish banana plants growing in its glasshouse on a research park in Norwich.

Credit: Dan Burn-Forti

In a field outside a small town called Humpty Doo in Australia’s sparsely-populated Northern Territory, one solution to the TR4 epidemic has been growing for the last six years. “In the Northern Territory, [TR4] is in virtually all the banana growing areas,” says James Dale, a professor at Queensland University of Technology in Brisbane. “Most plantations are still shut down.” But in that one field, the world’s only TR4-resistant Cavendish bananas have been thriving, while all around them, plants have succumbed.

For eight years, the key to creating TR4-resistant bananas remained locked within Dale’s laboratory. In 2004, he isolated a single gene from a wild banana called Musa acuminata malaccensis. Unlike its distant offspring, Musa acuminata malaccensis is unlikely to ever find itself as a cereal-topper. Its small, thin fruits are filled with upwards of 60 hard seeds, each about half a centimetre in diameter. But the inedible plant has something else going for it. It is naturally resistant to TR4.

After isolating the resistance gene – RGA2 – from the wild banana, and inserting it into a Cavendish plant, Dale hit a roadblock. “We weren’t allowed to take the fungus from the Northern Territory into our glasshouses,” he says. Australia’s strict bio-quarantine rules prevented any TR4-infected soil traveling from the blighted Northern Territory into Queensland, where most of the country’s bananas grow.

It wasn’t until he received a call from an Australian plantation owner that Dale got the chance to put his edited bananas to the test. Robert Borsato opened his banana plantation just outside Humpty Doo in 1996 – a year before TR4 was detected in Darwin, 40km away. By the late 2000s, Borsato’s farm was overrun with the disease. Desperate, he turned to Dale for help.

“I told him, ‘we’ve got this possible solution, but we have no idea whether these plants are resistant – would you work with us?’” recalls Dale, who is 68 and wears rimless glasses and a scruffy grey beard. “And we went up there and that really was bingo,” he says, grinning.

The three-year trial finished in 2015, but it would be two more years before Dale published his results in the journal Nature Communications. By the end of the trial, between 67 and 100 per cent of the plants without the resistance gene had been killed or infected with TR4. Of the five plant lines with the added RGA2 gene, four had much lower infection rates – below 30 per cent – and one line showed no signs of the disease at all. Another set of plants edited with a TR4-resistance gene from a roundworm showed similar survival rates.

CRISPR-edited cells become banana seedlings. The hope is that they will grow into TR4-resistant trees.

Credit: Dan Burn-Forti

After the success of the initial field trial, Dale is launching another study in Humpty Doo, encompassing an area more than ten times larger than the original site. He hopes to see the edited Cavendish on sale by 2021 – the first genetically-modified (GM) bananas ever sold in Australia. They would be the first GM bananas sold anywhere, but another trial Dale is running, a Bill and Melinda Gates Foundation-funded plan to engineer vitamin-A enriched Cavendish bananas in Uganda, will likely pip the Australian bananas to the post.

But Dale’s TR4-resistant bananas are still to pass a vital test. He hasn’t eaten a single one – not even on the sly, he insists, as the terms of his trial license prohibit anyone from tasting the fruit. “We actually have to squish them up and use them as mulch,” Dale says. Instead, all of his TR4-resistant bananas – the only ones of their kind anywhere in the world – are turned into fertiliser.

The problem is that Dale’s plants are classified as genetically-modified organisms (GMOs). His bananas contain genetic information from two organisms – the gene from Musa acuminata malaccensis is transplanted into the Cavendish genome by using bacteria as a “shuttle”. And under the Australian Office of the Gene Technology Regulator, experimenting with GMOs is only permitted under strict conditions designed to prevent any potential harm to humans and to minimise the chance that GM plants will breed with naturally-occurring plants and introduce genetic changes. A worry that, in the case of the sterile Cavendish, is unnecessary.

Dale recalls a field trial of GM bananas hit by a cyclone in North Queensland. “All of the bananas were on the ground – they were just blown down,” he says. The next morning he received a call from the Office of the Gene Technology Regulator asking whether there was GM banana material blown all over Australia. “I suspect so,” Dale told the regulator. But because Cavendish bananas are sterile, there was zero chance that any stray GM-banana DNA would end up in another plant. “Bananas are, probably of all the crops, the absolute safest to do both glasshouse and field trials on GM material. There’s no chance of escape.”

If his next trial is successful, Dale plans to apply for a tasting license and then bring the bananas to market. “During the next four to five years that it’s going to take to get these bananas through the regulation process, TR4 is going to become a really, really significant factor in the Australian industry,” Dale says. And since Australia bans the import of fresh bananas, the government may be forced to choose between accepting GM bananas or lifting its import restrictions. “My bet is they’ll have a GM Cavendish,” Dale says.

Outside of Uganda and Australia, the future for the GM banana looks bleak. In the EU, only 64 GM crops are approved for sale – all of them versions of cotton, maize, oilseed rape, soybean or sugarbeet – with the vast majority of them going into animal feed. Only one GM crop is cultivated in the EU – MON 810 – a form of maize genetically-engineered to be resistant to a moth that bores holes into the plant. Despite being relatively common in the US, GM fruit and vegetables have never been sold in the EU, and banana companies, too, have shunned GM fruit. “We’re a completely natural company,” an executive from Del Monte told me on the phone when I raised the question of gene-edited crops.

Dale knows that his TR4-resistant bananas are unlikely to ever leave Australia. “If the world accepted GM, then they’d be ready to go,” he says. Although scientists have been unable to find any long-term health impacts linked to any consumption of genetically modified food – a stance endorsed by the World Health Organisation and the American Medical Association – consumer and environmental groups have long opposed the technology.

Dozens of countries, including China, Russia, Japan, Australia, Brazil and the European Union, legally require GM food to be labelled. In the US, where many food companies place voluntary “No GMO” labels on their products, a law requiring the labelling of GM foods was signed by President Obama in July 2016, but food manufacturers have to date been slow to respond to the new regulations.

Dale suspects that – outside of a few unique cases – the world will never accept his GMO bananas. “We have lost the GM discussion,” he says. But, in 2016, as he was poring over the results from his field trial of TR4-resistant crops, Dale spotted an announcement that reignited his hopes for a superior Cavendish. In April, the United States Department of Agricultural (USDA) approved a mushroom that had been engineered to resist browning using a new gene-editing tool called CRISPR. In March 2018, the USDA clarified its position, saying that it would not regulate “a set of new techniques that are increasingly being used by plant breeders to produce new plant varieties that are indistinguishable from those developed through traditional breeding methods.”

The USDA’s logic is simple. If you’re using gene-editing to make a simple tweak – say, a single deletion in a gene that changes only one small aspect of the whole plant – then that’s just what can happen in nature anyway. Precise gene-editing, the regulator argues, is just accelerating the natural breeding process. To the USDA, a gene-edited banana is just a banana.

In July 2018, Dale published results of an experiment where he used CRISPR to modify the Cavendish genome so plants grew up to be white and shrunken. Although this proved that it’s possible to use CRISPR to edit banana cells, Dale’s albino bananas were technically still GMOs as they all contained a fraction of bacterial DNA inserted to make it easier to find the five to ten per cent of edited cells in a solution containing as many as a million embryogenic cells. Ultimately, the CRISPR-edited bananas won’t contain DNA from any other organism: they’ll be Cavendish through and through. “I had to go way back and start again,” says Dale, shaking his head ruefully. Dale might have been the first to create a GM-version of the Cavendish that was immune to TR4, but in the race to create the first gene-edited version, he’s no longer the only competitor.

Gilad Gershon, CEO of Tropic Biosciences, with a Cavendish plant. his initial goal is to use CRISPR to create slower ripening fruit – and then take on the challenge of a TR4-resistant banana.

Credit: Dan Burn-Forti

In a lab just outside of Norwich, Ofir Meir, the CTO of Tropic Biosciences, is holding the future of the banana in his hand: row up upon row of greyish clusters of cells arranged in a Petri dish. It will be months before these clusters grow shoots and are ready to join the neat lines of plants, each no more than a couple of centimetres tall, growing inside test tubes. From there, a handful of specimens will make their way into the greenhouses on the other side of the research park. Meir, 40, raises his voice to be heard over the low thrum of the growth chambers keeping the plants at 28.3°C: “One day, these shoots will become a field in South America.”

Genetically speaking, the plants in Meir’s test tubes are almost identical to every other Cavendish plant on the planet. The difference comes down to a couple of genes. Meir’s bananas have been edited using CRISPR-Cas9, a DNA-editing molecule co-discovered in 2012 by geneticists Emmanuelle Charpentier and Jennifer Doudna. CRISPR can, with a few molecular snips, deactivate a gene within an organism. This technique allowed the browning-resistant mushrooms to sidestep the USDA’s GMO regulations.

“CRISPR is precise, it’s relatively easy to use, and it allows a young company like us to start doing real genetic editing,” says Gilad Gershon, Tropic’s CEO. Gershon, who founded the company in July 2016, was working for the Californian agricultural investment firm Pontifax AgTech when he became convinced that CRISPR was about to blow open the agricultural industry.

“This really marks a revolution for the industry,” says Gershon, 36. For decades, the field had been dominated by a handful of agrochemical firms – Monsanto, Syngenta, Bayer and DuPont – who channelled their GMO efforts into blockbuster crops like corn, soya, cotton and rape seed. “It was just so expensive – you needed to spend $100 million on them, so you were obliged to work on corn,” he says. “Now, when the costs are a fraction of that, the field of opportunities is much bigger.”

In an industry where margins are razor–thin, a small tweak to make a better banana could have huge implications. The tiny cell clusters in Meir’s Petri dish are embryogenic banana stem cells that have been edited to grow into full-sized plants with fruit that ripens more slowly than a typical Cavendish. When bananas ripen, they release a gas called ethene, which prompts other fruit to follow suit and ripen more quickly. One rogue yellow banana aboard a container ship can cause a chain reaction that may wreck as much as 15 per cent of a shipment. If Gershon can tweak the genomes of bananas so they ripen more slowly, it could stop millions of tonnes of bananas spoiling, and save exporters a fortune.

Yet slow-ripening bananas are just the prelude to Gershon’s plans. His firm is also using the gene-editing technique to create naturally-decaffeinated coffee and stop the flesh of bananas from browning so quickly. But the real prize for Gershon? TR4-resistant bananas.

A researcher walks in carrying a crate filled with large flasks. Meir picks one out. It is filled with a yellowish liquid and inside there are thousands of white clumps, swirling within the murky solution. This is CRISPR in action. Within that flask containing millions of banana cells, CRISPR molecules are being guided to specific parts of each cell’s DNA and cutting out genes. “You want to take one cell and deliver the machinery to that one cell,” Meir says. “Then, the goal is to generate this cell into a full banana plant.

Sandra Lazauskaite, tissue culture specialist at Tropic, checks CRISPR-edited cells being cultivated in petri dishes

Credit: Dan Burn-Forti

But CRISPR doesn’t edit every cell it comes into contact with, so the challenge is in sifting edited cells from a solution containing millions. Conventionally, researchers insert small bits of foreign DNA to make edited cells stick out, but that’s not an option for Tropic. “Once you’re using a selection marker, it’s regarded as a GMO, you’ve introduced foreign DNA,” says Meir. At Tropic, Meir says he is developing tools so he won’t need to trawl through hundreds of thousands of cells looking for an edited handful. And crucially, he says, this technique doesn’t involve the use of any extraneous DNA at all.

Two Israeli companies, Evogene and Rahan Meristem, are using a similar approach to tackle Black Sigatoka – a fungal banana-leaf infection that can halve the amount of fruit a plant produces. As the joint trial enters its third year of field tests, the companies are hoping the end product won’t be classified as a GMO, making it quicker and cheaper to bring to market. “Hopefully, public acceptance will be there, and the cost to develop an improvement won’t be crazy like it was [with] GMO,” says Ofer Haviv, Evogene’s CEO.

But on July 25, 2018, Europe’s highest court threw the future of CRISPR-edited bananas into doubt. After being asked in 2016 by the French government to clarify how a 15-year-old directive on genetically-modified crops applied to ones created using modern gene-editing techniques, the European Court of Justice ruled that CRISPR-edited crops would not be exempt from existing regulations limiting the cultivation and sale of GM organisms. In the eyes of the EU, there was not much difference between Dale’s transgenic bananas and a CRISPR-edited banana after all.

“Disappointed,” says Johnathan Napier, a plant biotechnologist at Rothamsted Research in Hertfordshire, of the EJC ruling. “I’m disappointed for plant sciences and agriculture research in Europe. I’m disappointed for the innovators and the people trying to actually do good. I think it’s going to be really, really tough for them now to use this technology in Europe.”

The day after the ruling, I revisit Tropic. In the boardroom, Gershon is mulling over the ECJ’s decision. “I think this could have been handled better,” he says. Later, as Tropic’s researchers unwrap their lunches, the conversation circles around the idiosyncrasies of regulators’ thinking. Bombarding seeds with radiation to engineer new crop varieties falls outside of the EU’s GMO rules, they point out, but CRISPR – touted as a more precise way of inducing changes in a plant’s genome – doesn’t. But Gershon is undeterred. Europe is only one market, he says, and the US has already proven itself much more accepting of CRISPR-edited food. By 2050, half of the world’s population are projected to live in the tropics, and it is there that people will really need help to produce more food from the same amount of land. In rural parts of Uganda, Rwanda and Cameroon, bananas can provide up to 25 per cent of people’s average daily calorie intake. “Today there is real necessity, but it’s not spread uniformly,” he says.

Those of us outside of the tropics are walking into a culinary cul-de-sac of our own creation. “We got used to having an endless supply of this really cheap food,” Gershon says. “This economic reality will come to an end. We need to find good solutions in order to keep having people eating this fantastically healthy fruit.” Faced with choosing between giving up bananas altogether or accepting bananas that have been given an evolutionary leg-up in the lab, we might have to rethink our attitude to buying gene-edited fruit.

After more than a month with no rain, Norwich’s driest June since 1962, the grass in the research park is almost completely yellow. But dotted among the parched blades, Meir points out tiny patches of green. Plants that, because of an entirely random mutation in their genome, are able to keep growing, even when they’re starved of water. The Cavendish is not so lucky. Thanks to its sterility, it will never pick up a useful mutation through breeding. Yet, for all its flaws, this is the one banana out of the thousands of varieties out there that we have chosen to grow at such a vast scale. And now, as scientists race to find a way to save it that will please consumers, regulators and the food industry, it is facing the fight of its life. “TR4 is happening,” Gershon says. “It’s just a question of time.”

Featured Posts
Recent Posts
Search By Tags
bottom of page