One of the most significant natural events in the ecology of influenza viruses occurs each May on the beaches of Delaware Bay, New Jersey. Tens of thousands of horseshoe crabs come ashore on the first full moon in May to mate and lay their eggs in the sand. At just this time, migrating shorebirds (red knots and ruddy turnstones) are arriving in their tens of thousands, having flown non-stop from South America. The migration of these birds is timed so that they can refuel on horseshoe crab eggs, gaining up to 30 per cent of their body weight before setting out on the next flight, to Churchill Bay in Canada, en route to the far north of Canada, where they in turn mate and breed. In Delaware Bay, the birds deliver influenza viruses to beaches that are shared with the resident gulls and wading birds that also feed on horseshoe crab eggs.35 The horseshoe crab is known as the ‘keynote species’ in this amazing sequence of events.36
The horseshoe crab (Limulus polyphemus) is an extremely ancient creature, having evolved even before dinosaurs. It lives on the floor of oceans and bays, feeding on worms and other invertebrates. These crabs can be found on the beaches of the east coast of North America from Maine to Mexico, but by far the highest population density occurs at Delaware Bay.
At the first full moon in May, the male crabs arrive. The females, nearly twice the size of the males, arrive soon after. The male selects a mate and attaches himself to her with a special limb. The female digs holes in the wet sand and deposits her eggs in them for the attached male – and additional unattached males – to fertilise. Each female may lay as many as five nests of eggs – up to 80,000 eggs in all.
In the 1980s the beaches at Delaware Bay were covered with horseshoe crabs, and the waterline was marked by a layer of green crab eggs washed from the sand.
These eggs are just the fuel that the migratory shorebirds need after their four-day flight from Lagoa de Peixe National Park in southern Brazil (a distance of 4828 kilometres). Up to 25 species of seabird converge on Delaware Bay in May to feast on horseshoe crab eggs.37 In addition to the red knot (Calidris canutus) and the ruddy turnstone (Arenaria interpres), the most abundant are the sanderling (Calidris alba), the semi-palmated sandpiper (Calidris pusilla) and three species of gull: the great black-backed gull (Larus marinus), herring gull (Larus argentatus) and laughing gull (Leucophaeus atricilla). The most amazing are the red knots, which migrate from Tierra del Fuego, at the tip of South America, in three hops to Delaware Bay (Figure 5.1). Before this long-distance migration they gorge themselves, consuming up to 14 times their body weight in mussel sprats and converting them to fat. At that point, the physiology of the bird changes: those organs not needed for flight (the liver, leg muscles and gut) shrivel in size to accommodate more stored fat. This means the birds cannot digest solid food en route; the gelatinous horseshoe crab eggs at Delaware Bay are the perfect fare.
The shorebird that turned out to be of most interest from our perspective was the ruddy turnstone. These birds (and the other species) arrive from the Atlantic east coast of the United States and from the northern coast of South America, often having joined red knots on the final leg of their migration. The two species compete for food on the way to their breeding grounds in the Canadian Arctic.
Figure 5.1 The migration routes of red knots and ruddy turnstones. The red knots travel in three hops from Tierra del Fuego, at the tip of South America, to Delaware Bay, and from there head to northern Canada to breed. The ruddy turnstones join the migration from the north of South America. Some turnstones follow the coastline to Delaware Bay. At the first full moon in May the horseshoe crabs come ashore at Delaware Bay (gold star) to lay their eggs in the sand (A). The resident gulls and shorebirds (B) are joined by red knots and ruddy turnstones (C, D) which have migrated long distances to refuel on the horseshoe crab eggs. The ruddy turnstone, the bird of greatest interest, is shown separately in D, while panel C shows a mixture of mainly red knots and ruddy turnstones. The leg band on the ruddy turnstone (D) was added during the annual counting and swabbing of the birds. Photos courtesy of Jere Parobek, St Jude Children’s Reasearch Hospital
Since our two attempts to detect influenza viruses in American seabirds (Guano Islands, Peru; Dry Tortugas, Florida) had not been fruitful, our team from St Jude Children’s Research Hospital was looking for the right time and place to try again. Research results presented at a scientific meeting in 1983 by Larry Graves about the detection of several influenza viruses in gulls at a Baltimore landfill in 1977–79 provided the first clue. The written report appeared many years later.38 The next clue came from an English ornithologist, William (Bill) Slayden, who was aware of the Delaware Bay shorebird migration in May and suggested sampling at that time.
On the first visit to Delaware Bay, in May 1985, we went on our own and with no idea where to find the birds. We eventually found them at Reeds Beach. The sight was quite amazing: the beach was littered with overturned horseshoe crab shells and packed with shorebirds in a feeding frenzy as they dug in the wet sand for crab eggs. As some birds left, more came in. Most were red knots and ruddy turnstones. The collection of fresh faecal samples was easy. We followed the birds along the waterline, collecting freshly deposited droppings on a Dacron swab and putting them into vials containing 50 per cent glycerol with antibiotics to suppress bacterial growth. The collection vials went directly into a cooler containing ordinary ice blocks. Within three days of collection, the samples were air-freighted to the laboratory in Memphis.
There, additional antibiotic was added to each sample and a small amount of the total was injected into 10-day-old developing chicken eggs and held for two days at 35°C. The very first tray of eggs yielded many samples that agglutinated chicken red blood cells, indicating that an influenza or parainfluenza virus was present. (A parainfluenza virus is a virus that comes from wild birds and can infect and kill chickens.) Some 20 per cent of the samples tested positive for influenza viruses.39 We were ecstatic. In the following two years, we isolated influenza viruses belonging to 10 of the 12 different haemagglutinin (H) subtypes of influenza viruses known at that time. These included influenza viruses of the H1N1 group, related to the virus that caused the 1918 Spanish influenza pandemic, and the H3N2 group, responsible for the 1968 Hong Kong pandemic. We also found viruses belonging to the H7N3 subtype, which can evolve into lethal influenza in chickens and turkeys.
We realised that we had stumbled on a goldmine of influenza viruses, and we have mined that source every year since. Most of the viruses came from ruddy turnstones. Sampling of the resident gulls, sanderlings, terns and other shorebirds at monthly intervals over three years detected high levels of influenza virus in May and June but very low levels in September and October; there were no detectable viruses in other months. The small incidence in September and October may come from occasional arrivals by red knots and ruddy turnstones on their return migration. The zero readings in other months illustrate why single-shot sampling of waterfowl can fail to isolate viruses. It really is a case of being in the right place at the right time.
After our findings, influenza virologists in Europe, Asia and Australia began sampling the same species in their countries. They did find influenza viruses, but the incidence was much lower than what we found in May and June in Delaware Bay. Some of our international colleagues found it hard to believe our results and even came to Delaware Bay on the quiet to test for themselves. They confirmed our results.
We now know that Delaware Bay is a hotspot for influenza viruses during migration of the red knots and other shorebirds,40 but we still don’t know why. We can speculate that the migrating birds are stressed after their long migration and therefore may be susceptible to infection. The ruddy turnstones arriving from coastal regions of the United States have no detectable antibodies to any influenza viruses, and the huge congregations of these birds with other species provide optimal conditions for influenza to spread. But what is the actual source of the viruses? One possibility is that the ruddy turnstones, which are voracious scavengers, pick up viruses from other animals or birds or even human waste from the coastal towns of northern South America.41 But we don’t know for sure, showing that there are still important mysteries to solve.
Over the years, many volunteers have helped collect samples at Delaware Bay, including researchers’ granddaughters. One of mine, at the age of three, helpfully pointed out, ‘Look, Grandad, bird poop!’ Other helpers have included National Geographic documentary crews who have documented influenza surveillance at Delaware Bay. Senior molecular biologists who visit cannot believe that these beautiful, healthy birds, which have flown all the way from South America, could be carrying influenza viruses.
From our 30-plus years of surveying influenza viruses at Delaware Bay, we have assembled a huge collection of influenza viruses. Currently, there are 16 subtypes of influenza A virus identified from aquatic birds in nature. Fifteen of these subtypes are found in both ducks and shorebirds. H15 has not been found in the Americas to date but has been found in Eurasian waterfowl.
The different subtypes have different cycles of dominance. One subtype will dominate for one year in one region and be replaced by another subtype the next year. These viruses from ducks and shorebirds provided the background information that enabled us to propose in the late 1990s the following ecological principles of the ecology of influenza A viruses:
•Wild aquatic birds are the natural reservoirs of most of the influenza A viruses found in other species (including humans) (Figure 5.2). Two subtypes of influenza virus have recently been found in bats, so there is more work to be done on identifying virus reservoirs.
•The viruses replicate predominantly in the intestinal tracts of aquatic birds.
•They cause no apparent disease signs and are considered ‘low-pathogenic’.
•There is geographical separation into Eurasian and American lineages.
•Two subtypes, H5 and H7, are unique and can become highly pathogenic after they spread to domestic chickens and turkeys.
•Only the H1, H2 and H3 subtypes are known to have caused pandemics of influenza in humans.
Figure 5.2 Results of surveillance for influenza viruses among shorebirds and gulls at Delaware Bay, 1985–2016. As in migratory ducks, the dominant viruses changed from year to year. A wide variety of combinations was isolated, including counterparts of human H1N1 and H3N2, but no H2N2 was detected. It is noteworthy that the H7N3 virus was isolated quite frequently and was the precursor of influenza viruses that became killer strains in domestic poultry in Chile and Mexico. No H14 or H15 viruses were isolated, but H13 and H16 viruses not detected in Alberta wild ducks were present in these shorebirds and gulls. Table courtesy of Scott Krauss, St Jude Children’s Research Hospital
For many years some of these conclusions were treated with considerable scepticism, especially the idea that human pandemics could emerge from aquatic birds. This changed after the emergence of the H5N1 virus, dubbed ‘bird flu’, in Hong Kong in 1997. Since then there has been gradual acceptance of the ‘one world, one health’ concept that viruses that are benign in non-human animal reservoirs – such as influenza, Zika and SARS (severe acute respiratory syndrome) – can become killers when they spread to other animals, including humans (Figure 5.3).
ONE WORLD, ONE HEALTH
The one world, one health concept is based on the realisation that the health of humans and animals (domestic and wild) is intimately linked to the ecosystems in which they live. As many as 60 per cent of known human diseases originate from domestic or wild animals and birds. Influenza is one of the diseases that exemplifies this concept.
Before 1980, influenza viruses were divided into four separate groups according to the species they were believed to infect: one for humans, one for swine, one for equines and one for avians. But after influenza scientists recognised that influenza viruses from different species were intimately interconnected, a uniform nomenclature system was developed in 1980.
Figure 5.3 A diagram showing the natural reservoirs of influenza A viruses in wild aquatic birds (in the circle) and transmission through intermediate hosts to mammals including humans. The solid black lines indicate possible inter-species transmission. The boxes indicate the species that serve as intermediate hosts and are most likely involved in the emergence of zoonotic viruses with pandemic potential. (Influenza viruses have been isolated from bats but their role in interspecies transmission is not known.)
The huge abundance of horseshoe crabs led to their exploitation by humans in the mid-twentieth century. Since they contain little edible tissue, they were ground up for agricultural fertiliser and chicken food. Because the crabs exude a strong attractant odour when cut into sections, they were also used as bait by conch and eel fishermen.
The horseshoe crab has also contributed to our understanding of human vision. The crab has a huge optic nerve that allowed Haldan Keffer Hartline to explain how the receptors in the eye enable us to see. His research led to his receiving the 1967 Nobel Prize in Physiology and Medicine (jointly with Ragnar Gramit and George Wald).
In addition, studies on the blood of the horseshoe crab have identified a particular component that protects the crab from infection in its murky environment. The component, known as limulus amebocyte lysate (LAL), is extracted to form the basis for a clinical test for bacterial endotoxin contamination (a toxin produced in a bacterial cell that can cause disease symptoms in humans). This test is now required by the United States Food and Drug Administration for testing of all instruments, drugs and vaccines (including the influenza vaccine) for biological contamination, and it is also used in Europe and Japan. Consequently up to 250,000 crabs are caught each year and blood samples collected for the LAL preparation by pharmaceutical firms. The crabs are returned to Delaware Bay after the procedure, but 15–30 per cent of them die.42
Unsurprisingly, these human uses of horseshoe crabs have led to their severe depletion. Grinding the crabs up for fertiliser ended after cheaper artificial fertilisers became available, but the demand by conch and eel fishermen increased. By the 1990s the crab population had dropped tenfold or more, with disastrous effects on the migratory shorebirds that feed on the crab eggs. The numbers of red knots and ruddy turnstones fell by 86 and 75 per cent respectively. Many birds failed to gain the threshold weight they needed for the Arctic flight and perished on their trip to their Arctic breeding grounds. By 2006 the numbers of red knots had fallen by another 86 per cent from their 1980s level.
However, thanks to coastal conservation and resource management efforts by the American Littoral Society and the Atlantic States Marine Fish Commission, the harvesting of horseshoe crabs is now banned between 1 May and 7 June. Each state has capped the harvest at 150,000, and conch fishermen are required to use bait-saving devices like bait bags, reducing crab use. Beach access by humans is restricted in late May.
Once horseshoe crabs are flipped onto their backs they have great difficulty turning back over, which is one reason the beaches were littered with crab shells on our first visits. Many of our visiting colleagues and family members spend most of their time turning the crabs right side up again so they can make it back to the sea. This rescue of horseshoe crabs is vital.
The numbers of red knots stabilised and now are gradually increasing, but the numbers of ruddy turnstones remain low. Neither the frequency nor the diversity of influenza viruses has changed with the decline in migrating shorebirds, but collecting samples is more difficult. Influenza virologists work closely with the conservation agencies protecting the birds to ensure minimal disturbance of feeding birds while collecting faecal samples from the beaches, and when the birds are caught by wildlife experts for banding in population studies, virologists take the opportunity to collect samples at the same time.