Plasmodium knowlesi malaria infections in Malaysia: The last parasite standing?
This essay was written by Rob Moon and was first published in the 2013 Mill Hill Essays.
Malaysia has one of the oldest malaria control programmes in the world, initiated in 1901 by the pioneering work of Sir Malcolm Watson. More than 100 years on, malaria finally appears to be on the back foot and Malaysia is now targeting elimination by 2020. However, in 2004 Balbir Singh and colleagues discovered a new and unwanted complication in this grand plan. They discovered that a species of simian malaria parasite, Plasmodium knowlesi, was causing a significant number of cases of malaria in humans. In the summer of 2011 I received a Winston Churchill Memorial Fellowship to travel to Malaysia on the trail of this emergent parasite and to discover how it will affect the country to which it has become inextricably linked.
A short history of Plasmodium knowlesi
P. knowlesi was first described in 1932 by Dr. Robert Knowles and his assistant Biraj Mohan Das Gupta in Kolkata (Calcutta), after studying an infected Cynomolgus macaque (Macaca fascicularis) from Singapore. Further work on the parasite was published later in the year by Sinton and Mulligan (1932) who took the liberty of naming the parasite in honour of Robert Knowles. Even within the first few years of its discovery it was clear that P. knowlesi was of particular interest among the many primate malaria parasites. Malaria parasites are single celled animals known as protozoa, which have a complicated lifecycle spanning two hosts: one mosquito and one vertebrate. Like many parasites their success depends on their exquisite adaptation to invading and growing in their respective hosts and as such each species of malaria parasite will only infect a very narrow range of vertebrate hosts. So whilst hundreds of species of malaria parasites exist, affecting lizards, birds and mammals, each one only affects a few closely related animal species and only four were thought to affect humans: Plasmodium falciparum, P. vivax, P. ovale and P. malariae. The early researchers duly tested which other primates could be infected by P. knowlesi and, surprisingly, found that a broad range of primates were susceptible, including various macaque species, leaf-eating Presbytis monkeys and even African Baboons. This broad host range was not the only surprising thing about P. knowlesi. Whilst it caused a mild chronic infection in the natural Cynomolgus macaque host, it caused severe and deadly infections in most other species. Finally, the early malariologists discovered that human volunteers were also susceptible to infection with P. knowlesi.
For some time this was seen as an interesting, but artificial, observation; an event limited to laboratory conditions. Then in 1965, P. knowlesi came back to the fore when the first case of a natural infection in man was identified. The patient was a surveyor for the US Army just returned from working for days alone in the jungles of Pahang province in Malaysia. The reasons for his sojourn in the jungle were unclear, but at a time of heightened cold war tensions in the region, rumours of espionage and clandestine operations were popular explanations. The man was diagnosed with malaria and treated, but the species was twice misdiagnosed, first as P. falciparum, then as P. malariae. It was only when samples were sent on to malaria researchers, that they finally identified the parasite as P. knowlesi. Further investigations revealed that cases in humans appeared to be rare, and only occurred in people spending time in the jungle canopy at night, exposing them to bites from mosquitoes which would normally prefer macaques.
P. knowlesi remained an important parasite in the lab, enabling the first discovery of antigenic variation in malaria parasites, and of the Duffy receptor. The latter was the first malaria red blood cell invasion pathway discovered and an absolute requirement for both P. vivax and P. knowlesi parasites to invade red blood cells. But in 2004 it was another discovery back in Malaysia that returned P. knowlesi to the spotlight.
When reviewing cases of malaria occurring in the Kapit region of Sarawak, in Malaysian Borneo, they noticed a very odd thing. Besides the cases of P. falciparum, the most deadly form of malaria and P. vivax, which is dangerous but less severe, they found clusters of patients with severe malaria diagnosed with P. malariae. This was strange because of the four human malaria parasites, P. malariae was considered benign with very rare cases of hospitalisation and mild symptoms. Further surveys combined with genetic testing confirmed that more than half the cases identified over the study period (120 of 208) were diagnosed by smear as P. malariae, but genetic testing revealed that all were in fact infected with P. knowlesi. What had fooled the US doctors diagnosing the army surveyor more than forty years before had continued to fool clinicians since. Whilst genetically similar to P. vivax, the morphology of the P. knowlesi parasites on stained blood slides (the gold standard for diagnosis) look very similar to P. malariae, and technicians were not trained to look out for monkey malaria in humans. In the years that followed the same team identified further foci of transmission in Sabah in the north of Malaysian Borneo as well as in Pahang province in peninsular Malaysia. Furthermore, not only was this parasite widespread and causing severe disease but it was also causing death. Four fatal cases were identified amongst the 312 cases examined in this follow up study. P. knowlesi cases in humans have since been discovered in Indonesia, Vietnam, Cambodia, Thailand, the Philippines and on border areas of Burma, in a range covering the whole of South-East Asia.
The Institute for Medical Research, Malaysia
Malaysia has remained central to the story of P. knowlesi and research to understand it, largely due to the progress of both research and disease surveillance in the country. The nation is on track for its aim of developed nation status by 2020 and thanks to a rigorous control program its malaria cases are down to around 6000 per year, resulting in about 30-40 deaths per year. My journey to study P. knowlesi in the field started in Kuala Lumpur, capital city of Malaysia and home to the Institute for Medical Research (IMR). The IMR was one of the first research institutes to be established in the tropics, built in 1900 in what was then British Malaya. Its remit was to address the local problem of infectious diseases like malaria and cholera, but it has since been expanded, modernised and has extended its focus to other serious public health threats, including HIV and cancer.
At the IMR I joined the lab of Dr Noor Rain Abdullah, whose research focuses on both the screening and development of new antimalarial drugs, and field surveys of malaria. The bulk of the fieldwork is carried out on the basis of active case detection, where the survey teams will actively seek out people with malaria infection by screening whole villages. This provides a more accurate picture of transmission rates as well as facilitating early diagnosis- important for nipping new outbreaks in the bud.
I joined the team on a fieldwork trip to Sabah, Borneo, to survey the area around a place called Kota Marudu for cases of malaria. The study was specifically setup to look at P. falciparum and P. vivax infections, but P. knowlesi infections were also known to be common in this area. After a short flight from the mainland we were met at the airport in Kota Kinabalu by the local Ministry of Health workers who monitor and help the villages in malaria hotspots. These teams knew the area intimately so could provide invaluable local knowledge- and they were superb off- road drivers, an important skill as I would later learn. Although malaria is fairly common in this area of Borneo, its transmission is highly localised and predominantly occurs in the small villages, called Kampungs in Malay, situated in the jungle or jungle fringe. From the central base in a clinic in Kota Marudu, the team could drive out to each of these villages to screen as many people as were willing.
Borneo is home to a wide variety of ethnic groups and this area was largely populated by the Dusun, a race indigenous to Borneo, who in the Kampung predominantly work as rubber tappers or as small scale subsistence farmers in forest clearings. The housing was highly variable in the villages, with some large sturdy wooden houses, but many simple constructions made of bamboo. The latter often were open on at least one side, and so, in the absence of a bednet, offered no protection from the night-time biting mosquitoes which can transmit malaria. In most cases the Kampung could only be reached by driving for an hour along tarmac roads followed by a further hour or so along narrow and hilly mud tracks through the jungle. Whilst skilled drivers could navigate these roads when dry, if the weather turned to rain, which it did almost every day, the roads quickly became impassable. Rain in Malaysia is almost exclusively torrential so the mud roads could be turned into foot-deep rivers in a matter of minutes. The speed of this transformation was brought home to me when a downpour brought a premature end to our first white-knuckled expedition into the jungle. Whilst an unfortunate delay for us, it is an everyday problem for the public health teams that work here. Limited access causes both delays in treatment and increased opportunities for further transmission of the disease.
On arrival the villagers quickly gathered around and after providing consent, lined up for screening. Malaria infections are initiated with a bite from an infected mosquito and the parasite then travels through the blood to the liver where it forms its beachhead by invading and multiplying inside a liver cell. This cell eventually bursts releasing thousands of the tiny parasites into the blood stream which then undergo cycles of invasion, growth and replication in red blood cells. It is this stage that causes all the symptoms of disease, mostly a result of destruction of red blood cells and immune reactions to the parasites. The blood stages are also the first stages that can be readily detected and so patients were screened by making a blood smear, which would later be stained and screened in the lab, and by adding blood to a rapid diagnostic test (RDT). The RDTs are a recent development that not only detect malaria but identify which species of parasite is present. Working in a similar way to pregnancy tests, they are less sensitive than thick smears but produce results within five minutes, meaning that action can be taken before returning to the lab. In this way we screened families of 15 people, or whole villages of 150 people, which in the extreme heat and humidity of the jungle was very hard work!Very few westerners would have visited these villages, so I was subject to open curiosity. The guides explained “he is from England, Britain, UK” before settling on “near Manchester United” to finally produce satisfied looks of approval. But whilst I was a welcome sideshow, particularly when trying my hand at announcing the patient numbers in Malay, it was clear that the villagers well understood the importance of screening for malaria. The level of awareness and engagement visible in many of the communities appeared to be central to the success of control strategies in Sabah.
In a few days we screened 359 people, and found a total of 24 infected people, with 13 cases of P. falciparum, six P. vivax and three P. malariae. Interestingly all three of the P. malariae cases were diagnosed as such by blood smears. The RDT’s however, which detect parasite protein with antibodies, indicated that they were P. vivax. But they were neither. They were in fact the elusive parasite I was searching for; a parasite that looks similar to P. malariae but is genetically similar to P. vivax. It was previously thought that P. knowlesi was only common in those who spent large amounts of time in the forest canopies, such as loggers who often slept in the forest and even in the tree tops. The patients suspected to have caught P. knowlesi in our case were the perfect example of how this restriction clearly does not apply- the patients were a 76 year old man, a 13 year old girl and a nine year old boy- none of whom were likely to be in the canopy at night time. This would rather suggest transmission is likely occurring near to or within the village, and possibly within their homes. On our last day we were joined by the doctor in charge of public health in the area. Based in Kudat, just to the north of Kota Marudu, he told us that there was an extremely high rate of cases of P. knowlesi in the area. In 2009 there were two deaths from malaria in Kudat, and both of those were caused by P. knowlesi, with 125 cases occurring that year in total, so clearly it has become a highly significant threat in this region.
Mosquito vectors – the definitive factor
Control of the mosquito vector has long been a cornerstone of the fight against malaria and this is largely due to its essential role in the parasite’s lifecycle. Rather than simply being a hypodermic syringe as many people imagine, once a mosquito bites an infected human the parasite must undergo a complex series of growth and differentiation steps within the mosquito. In doing so, the parasite moves from the bloodmeal taken from an infected person, through the mosquito gut wall, before eventually invading their salivary glands ready to infect the next person. Mammalian malarias only infect a subset of mosquitoes belonging to the Anopheles genus and vector preference depends on the malaria species. Knowing which mosquito species is transmitting malaria in an area is essential as the behaviour and ecology of the vector determines when and where people get bitten, as well as the best ways of controlling transmission.
In my second foray into the jungle I was able to join Dr Rohani and her team from the Entomology department on a vector survey of a remote area of peninsular Malaysia near Lipis. This area is home to the Senoi, one of several indigenous people known in Malay as the Orang Asli (or aboriginals). The Senoi live a simple life primarily as hunter gatherers with some small scale farming, and are much further removed and isolated from modern life than the villagers in Borneo. Despite being a protected people, they sadly also rely quite heavily on rations provided by the government, as forests are encroached by plantations and many of the animals they hunt become endangered. As in Borneo, with isolation comes the risk of malaria, and so transmission rates remain high in the Orang Asli villagers. Interestingly whilst other human malaria is prevalent, there is thought to be little or no transmission of P. knowlesi. One explanation proposed to me for this was that this is because there are few macaques in the area as they are hunted by the Senoi. If true, this is positive evidence that infections are still tied to the macaque hosts.
The survey teams employ several tactics to detect both adult and larval mosquito stages in and around the bamboo villages. Some mosquitoes prefer biting animals, whereas others have a taste for humans. This means the best bait to capture mosquitoes spreading malaria…. was us! So our first evening’s work was the infamous BLC or bare leg catch. After rolling up your trousers you simply wait with a torch for the mosquitoes to land and catch them in a glass tube to take back and identify in the lab- with any luck before they bite. Much to the disappointment of the rest of the team I didn’t get attacked by a single mosquito, but using techniques like this, it is possible to identify not only which species bite humans, but also at which locations (i.e. inside or outside homes) and at which time peak biting occurs. Surveys in the 1960s had implicated Anopheles hackeri in transmitting P. knowlesi amongst macaques, a mosquito with a distaste for human biting. More recent work has identified A. latens as a vector in Malaysian Borneo and A. cracens in Peninsular Malaysia. Worryingly, both of these vectors are much more likely to bite humans than A. hackeri with monkey and human baited traps revealing slight preference for biting humans when at ground level. But what is really unclear is whether this has always been the case or whether something is indeed changing in Malaysia.
Emerging or only just detected?
On our return from the Orang Asli village, we made one further stop before Kuala Lumpur, at a small Malay village that was rural but not in any way isolated. With tarmac roads, and manicured tropical flower beds it was an odd place for a malaria field trip. This was to be the site of a future survey, as there had been several cases of malaria here. Strikingly, for the first time, only one species of malaria was present in the area. No P. falciparum, no P. vivax, only P. knowlesi. Like many villages of its type a large fruit orchard was central to the village life. Monkeys are also attracted to the bountiful fruit and it is thought that this close contact between monkeys, humans and an as yet unknown vector is creating a new setting for transmission. Environmental change is having a similar effect elsewhere in Malaysia. Kudat in north Sabah in Borneo has undergone massive deforestation and recent surveys have shown that P. knowlesi accounts for more than 50% of the malaria infections in this region. Macaques are also becoming more common in urban environments as development destroys the small pockets of forest amongst the prime development land. As always, a suitable vector is the only thing that is missing to cause a resurgence of urban transmission.
Sequencing genes from monkey- and human-derived P. knowlesi infections have revealed that whilst cases have only recently been detected in humans, such infections are unlikely to be newly emergent and it is still likely occurring only as a disease spread indirectly from macaques, which act as a reservoir host. However, there is at least anecdotal evidence that human infections are increasing. Many veterans of malaria control in Malaysia doubt that there was anywhere near enough P. malariae diagnosis occurring in the 1970s and 1980s to account for the numbers of P. knowlesi infections that are seen today, even if every one of them was a misdiagnosis. As well as environmental changes it is also possible that the success of malaria control may be part of the reason for the increase in P. knowlesi. Earlier work showed that people previously infected with P. vivax were less susceptible to P. knowlesi infection and severe disease. We now know that due to the genetic similarity and shared invasion pathways of P. knowlesi and P. vivax, antibodies generated to one parasite are cross-protective against the other. As fewer people contract the other human malarias, resistance in the population to P. knowlesi may also decrease. As the reservoir for infection in macaques remains unaffected, elimination of other malaria parasites may actually increase the risk of infection and severe disease caused by P. knowlesi.
Whilst infections currently appear to be passed only from macaques to humans (via mosquitoes), increased infection rates and changing mosquito vectors could provide an opportunity for the parasites to fully leap the species barrier and begin spreading from human to human. Significantly, this would mean infections in areas without macaque populations, allowing the parasite to spill beyond the bounds of its natural host and outside of South-East Asia. Even without a host switch to humans, a malaria parasite with an animal reservoir is a unique challenge. Only further research into vectors and epidemiology of this parasite will reveal how significant a threat it is and may become. It is clear that alongside imported cases of malaria, P. knowlesi will be the focal point of Malaysia’s plans in the lead-up to elimination. So will the latest malaria to leap over to humans be the last to be eliminated in Malaysia? We will have to wait until 2020.