Plasmodium knowlesi infecting humans in Southeast Asia: What's next?

Plasmodium knowlesi, a simian malaria parasite, has been in the limelight since a large focus of human P. knowlesi infection was reported from Sarawak (Malaysian Borneo) in 2004. Although this infection is transmitted across Southeast Asia, the largest number of cases has been reported from Malaysia. The increasing number of knowlesi malaria cases has been attributed to the use of molecular tools for detection, but environmental changes including deforestation likely play a major role by increasing human exposure to vector mosquitoes, which coexist with the macaque host. In addition, with the reduction in human malaria transmission in Southeast Asia, it is possible that human populations are at a greater risk of P. knowlesi infection due to diminishing cross-species immunity. Furthermore, the possibility of increasing exposure of humans to other simian Plasmodium parasites such as Plasmodium cynomolgi and Plasmodium inui should not be ignored. We here review the current status of these parasites in humans, macaques, and mosquitoes to support necessary reorientation of malaria control and elimination in the affected areas.


Introduction
For many years, it has been accepted that only 4 species of Plasmodium namely Plasmodium falciparum, Plasmodium vivax, Plasmodium malariae, and Plasmodium ovale cause natural malaria infection in humans. However, this assumption was challenged when the first case of naturally acquired Plasmodium knowlesi infection was reported in an American army surveyor in the state of Pahang, Malaysia in 1965 [1].
The long-tailed macaque (Macaca fascicularis) was identified as a natural host of P. knowlesi, Plasmodium fieldi, Plasmodium cynomolgi, Plasmodium coatneyi, and Plasmodium inui, among others [2,3]. However, no mosquitoes were found to harbor sporozoites of these Plasmodium species in the area where the first case was reported. In contrast, Anopheles hackeri was incriminated as the vector of P. knowlesi in the state of Selangor and found to be attracted to nonhuman primates rather than humans, biting especially at canopy level [4]. Thus, it was postulated that human infection with simian malaria parasites is a rare event [5].
However, this scenario was greatly challenged when a large focus of human P. knowlesi infection was reported among the local population in Sarawak, Malaysia in 2004 [6]. It is a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 believed that P. knowlesi infections in humans had been there much earlier but were only detected when molecular tools were adopted [7]. Under experimental conditions, both P. inui [8] and P. cynomolgi [9,10] can infect humans through mosquito bites, while natural P. cynomolgi infection had been reported recently [11][12][13][14][15]. On this background, we here review the current occurrence of P. knowlesi in human, mosquito, and macaque hosts, examining also the possible emergence of other simian Plasmodium species as zoonoses in Southeast Asia.

Methodology
A literature search was carried out using PubMed/Web of Science, Google Scholar, and other sites to find relevant materials related to simian malaria in humans, macaques, and vectors. The following search terms were used singly or in combination: simian malaria, primate malaria, zoonotic malaria, non-human primate, Anopheles, malaria vector, macaques, and monkeys. The most relevant publications related to the current topic were selected. All relevant early to present publications were included. Since the focus of the review article is on Southeast Asia, other simian Plasmodium species infecting humans in other parts of the world are not discussed in detail.

Epidemiology
Human infection by P. knowlesi had been reported from all Southeast Asia countries except Timor Leste [16] (Table 1).

Knowlesi malaria in Malaysia
Malaysia, notably Sabah and Sarawak, reported the highest numbers of P. knowlesi cases in Southeast Asia [43] following the drastic reduction of malaria cases caused by the human malaria parasites. Knowlesi malaria now accounts for all local cases reported [44].
In Sabah, it was demonstrated that knowlesi malaria spread gradually from areas with no transmission of human malaria to other areas, as human malaria cases were reduced [45]. Most cases were found in the Southwest interior region, gradually spreading to the West Coast and then on to the northern area and finally to the East Coast where P. vivax was still present in substantial numbers [45]. In Sarawak, there were a total of 9,364 P. knowlesi cases from 1992 to 2014 [46]. From 73 cases in 1992, the number increased dramatically to 1,095 in 2014, far exceeding the incidence in Peninsular Malaysia [46]. Environmental changes especially associated with deforestation and land exploration bring human population in close contact with Anopheles mosquito vectors over time, which inevitably increases the risk [47,48].

Knowlesi malaria in other Southeast Asian countries
Besides Malaysia, large numbers of knowlesi malaria cases were recorded in neighboring countries, Brunei, Indonesia, and Thailand (Table 1). The low reported incidence of knowlesi malaria in other Southeast Asia countries might be due to misdiagnosis through microscopy and a scarcity of specific studies [49]. Thus, the knowlesi malaria cases reported from these countries may be just the tip of the iceberg. This underlines the importance of multinational collaboration in reducing or eliminating knowlesi malaria in future.

Knowlesi malaria among travelers
There are also increasing numbers of cases of knowlesi malaria imported from Southeast Asia to Europe, Asia, America, and Oceania (Fig 1) [50,51]. The countries outside Southeast Asia which had reported imported P. knowlesi malaria in international travelers are listed in S1 Table. Other simian malarias in humans The first reported case of a natural transmission of P. cynomolgi was from Hulu Terengganu, the east coast of Peninsular Malaysia in 2011 [11]. Following that, natural infections of P. cynomolgi in humans became increasingly evident in Sarawak [13], northern Sabah [12], and western Cambodia [14]. A P. cynomolgi case was reported in a tourist from Denmark who had visited Peninsular Malaysia and Thailand in 2018 [15].
In addition to human Plasmodium species, more infections of P. knowlesi, P. cynomolgi, P. coatneyi, and P. inui were detected in Malaysian samples through a PCR approach, some of which comprised mixed infections of P. knowlesi + P. cynomolgi and P. knowlesi + P. coatneyi [52]. These were detected from hospital samples as well as from community surveys [52] and highlight the importance of using diagnostic tests specific for these species.
in Table 2. The compiled data from year 2004 until 2017 show that Malaysian Borneo has the highest prevalence of simian malaria in their wild macaques followed by Cambodia, Singapore, and Indonesia. In Malaysia, M. fascicularis has the highest prevalence of simian malaria. In Peninsular Malaysia, this species was positive for simian malaria in forested areas, but not in urban areas [54]. Overall, P. knowlesi and P. inui infections have similar rates in Peninsular Malaysia, followed by P. cynomolgi [54]. In contrast, in Sabah, P. inui infection is the most prevalent infection in long-tailed macaques, followed by P. knowlesi and P. cynomolgi with similar rate [48], while in Sarawak, P. knowlesi and P. inui have similarly high prevalence, followed by P. coatneyi and P. cynomolgi [55]. Most of the malaria-positive macaques in Peninsular Malaysia and Sarawak have mixed infections [55,56]. Besides, nonhuman primates in Southeast Asia harbor other simian Plasmodium parasites namely Plasmodium eylesi, P. fieldi, Plasmodium fragile, Plasmodium hylobati, Plasmodium jefferyi, Plasmodium pitheci, Plasmodium simiovale, Plasmodium silvaticum, and Plasmodium youngi [54]. Although these have not shown infection in humans, further studies are needed to determine if any of these species can cause human infections.
P. knowlesi has not been detected in wild macaques in Cambodia and Indonesia even though indigenous cases of knowlesi malaria in humans have been reported there [57,58]. Moreover, P. knowlesi is not the most prevalent malaria parasite found in macaques in each region/country in Southeast Asia except for Malaysian Borneo. The most prevalent simian malaria parasites in the other countries is either P. inui or P. cynomolgi. Both can reach more than 70% prevalence in some regions. Based on studies done thus far (Table 2), P. inui has the widest geographical distribution in Southeast Asia followed by P. cynomolgi. This is expected since these 2 parasites can infect a wide range of Old and New World monkeys (naturally or experimentally). Besides, P. inui is the widest-ranging parasite, likely due to its high chronicity in the hosts [65], followed by P. cynomolgi. It is also noteworthy that there is a generally high multispecies infection rate among the wild macaques. This leads to a bigger reservoir and higher risks of zoonotic transmission of P. inui and P. cynomolgi to humans.

Distribution of P. knowlesi vectors
Most of the vector studies have been done in Malaysia [16] followed by Vietnam [42,66]. Studies in Malaysia first identified An. hackeri as the vector of P. knowlesi [4]. However, this species is so highly zoophilic that it has been considered impossible that it could infect humans [5]. After the recognition of the importance of P. knowlesi malaria in humans in Sarawak, vector studies were conducted in different ecological niches. In Kapit district, Anopheles latens was identified as the main natural vector of P. knowlesi for both macaques and humans [67], while in Sabah, Anopheles balabacensis was incriminated [68,69]. In both states, these 2 species, belonging to the Leucosphyrus group, are also vectors of human malaria [16].
In Pahang, Peninsular Malaysia, Anopheles cracens, a more anthropophilic member of the Leucosphyrus group was identified as the vector of P. knowlesi [70], while in Selangor, An. hackeri and Anopheles introlatus were found to be the vectors [4,71]. It seems that different geographical locations have a different predominant species of Leucosphyrus group. Within the same area, different ecological conditions play an important role. In Kapit, Sarawak An. latens and Anopheles watsonii were predominant in forested ecotypes, while in the farm areas, Anopheles donaldi and An. latens were the main species [67]. However, dissection and further examination of those mosquitoes revealed only An. latens to be positive with sporozoites and to be attracted to both humans and macaques. Thus, An. latens had been incriminated as the main vector for P. knowlesi in Kapit, Sarawak. The role of ecology was also observed by Wharton and Eyles, who incriminated An. hackeri as a vector for P. knowlesi at the coastal region of Selangor [4].
Other species of Anopheles like An. watsonii and An. donaldi found in sympatry with An. latens in Kapit were positive for oocysts with PCR-positive for the genus Plasmodium, but species identification by using primers for simian and human malaria parasites was not successful [67]. Anopheles kochi in Kuala Lipis, Pahang was found in considerable numbers in monkey baited traps but was not positive for oocyst or sporozoites by dissection [70]. A recent study by Hawkes and colleagues in Sabah found a non-Leucosphyrus group species, An. donaldi, to be positive for DNA of P. knowlesi, P. cynomolgi, and an unknown Plasmodium species [72]. Unfortunately, in this study, the whole mosquito was processed, so it is not known whether oocysts or sporozoites were present, and therefore, whether these mosquitoes were vectors.
In the forest and forest fringes of Khanh Phu commune, Vietnam, Anopheles dirus, the main vector of human malaria in this region [73], was identified as the primary vector of P. knowlesi [66]. It was the only Anopheles species found with sporozoites, and the highest number infected was detected in the forest compared to the forest fringes. P. knowlesi was the second most common Plasmodium species identified in the vectors using nested PCR assay followed by other simian plasmodia [66]. P. knowlesi was detected in the salivary glands of An. dirus as was the case in the previous study by Marchand and colleagues in the same locality [42].
In the other countries of the Mekong region (Cambodia, Laos, Myanmar, and Thailand), An. dirus is the main vector for human malaria besides Anopheles minimus [73]. Thus, based on results from Vietnam, it can be extrapolated that An. dirus could play a role in the transmission of knowlesi malaria in that region. In Palawan, Phillipines, Tsukamoto and colleagues demonstrated the presence of Plasmodium oocysts and sporozoites in An. balabacensis in an area where macaques were present. At that time, molecular tools were not available, so they carried out experiments to expose An. balabacensis, also a member of the Leucosphyrus group, to Plasmodium-positive macaques and showed that An. balabacensis could develop sporozoites PLOS NEGLECTED TROPICAL DISEASES [60]. In conclusion, it can be stated that, at present, only the Leucosphyrus group of Anopheles are confirmed vectors for P. knowlesi in Southeast Asia.

Bionomics of the vectors
With extensive deforestation and land exploration in Malaysia over the past decades, there have been changes in the bionomics of the mosquitoes. Deforestation has caused the migration of the long-tailed macaques from forested area to farms and semi-urban areas where they usually scavenge for food. This may have triggered mosquitoes to follow the host and adapt to forest fringes and farm areas [16]. Fruit orchards that were propagated in those areas were ideal sites for the Anopheles vectors [74].
Most of these vectors tend to bite humans early, between 1900 and 2100 hours, but this varies with the geographical locations and the mosquitoes species [16,[67][68][69][70][71]. An. latens prefers to bite macaques at 3 to 6 m above ground. It was demonstrated that the periodicity of the gametocyte stage in the macaques' host coincides with the biting times of the vector mosquito [75]. This raises the possibility that these vector mosquitoes become infected by biting macaques late at night and transmitting the infection to humans by biting them in the early part of the night, when they are not protected.

Other simian malaria parasites in vectors
Many studies have investigated the presence of other simian malaria parasites in salivary glands and oocysts from Anopheles mosquitoes using PCR. Based on the compiled data (Table 3), P. cynomolgi and P. inui were more prevalent in the vectors compared to P. knowlesi. These parasites were present as monoinfections and multiple infections. This observation is in parallel with the prevalence of simian plasmodia recovered from macaques [55,56,58]. The overlapping distribution of the vectors and the macaques in forested areas might explain the high prevalence of P. cynomolgi and P. inui in both the macaques and in the Anopheles mosquitoes. Anopheles mosquitoes identified as natural vectors for simian plasmodia in Southeast Asia are shown in Table 4. There are studies showing coinfections of these simian malaria parasites with human Plasmodium in mosquitoes [42, 66,76] indicating possible simultaneous transmission.

Increasing case numbers of knowlesi malaria
With the reduction in human malaria cases, in Southeast Asia, more and more cases of simian malaria, especially due to P. knowlesi, are detected in humans as seen in Malaysia and Sumatra in Indonesia [23]. One factor could be simply increased awareness and the availability of molecular diagnosis. Besides, increasing exposure of humans to mosquitoes that transmit simian malaria may also be a factor. This is compounded by human encroachment into areas where deforestation and environmental changes are taking place.
However, this hardly explains the continued steep increases in Malaysia over the last 25 years. With reduction of human malaria, control measures may be relaxed, increasing the risk of simian malaria. In addition, diminishing cross-species protective immunity may play a role. P. knowlesi and P. vivax share important antigenic properties [84][85][86]. In vitro, specific antibodies against P. vivax antigens from animals and malaria patients inhibit erythrocyte invasion by P. knowlesi [86,87]. Data from neurosyphilis malariotherapy demonstrated that patients previously infected with P. vivax had lower susceptibility to P. knowlesi [88].
In addition, are we beginning to see human-to-human transmission occurring? At least in some areas, in Malaysia and Vietnam, the main vectors of simian malaria are also the main vectors of human malaria. While monkey-to-human transmission currently remains the main route of transmission for knowlesi malaria [89], human-to-human transmission could well occur, and may become more likely as prevalence continues to increase.

Possibilities of other simian malaria infecting humans
As noted above, cases of other simian malaria parasites infecting humans have been detected with increasing frequency over the last 10 years. With reports of P. cynomolgi in humans [11][12][13][14][15], the results of the study demonstrate the importance of being proactive to prevent outbreaks. Previous studies have indicated that antisera to P. falciparum, P. malariae and P. ovale from humans can cross-react with the P. cynomolgi antigen [90][91][92]. Therefore, if P. knowlesi emerged in humans due to the waning cross-immunity [45], this could occur also for other simian malarias.
As P. inui and P. cynomolgi are the most prevalent simian malaria parasites in the macaque hosts and vectors in Southeast Asia, humans are exposed to these parasites by bites of infectious mosquitoes [66]. The vectors of simian malaria are now known to bite both macaques and humans [67,68,70] unlike in the 1960s when An. hackeri was the only verified vector of simian malaria [4]. As human infections with P. inui and P. cynomolgi result in milder infections which do not require hospitalization compared to infection caused by human malaria or   Indonesia, Peninsular Malaysia, and Thailand [73,79] An. hackeri Inoculation of sporozoites into rhesus macaque. Selangor, Peninsular Malaysia [4] Malaysian Borneo, Peninsular Malaysia, Philippines, and Thailand [73,79] An. latens Detection of parasite DNA in sporozoite-infected salivary gland. Sarawak, Malaysian Borneo [67] P. knowlesi [93], these infections could go unnoticed as asymptomatic infections or be misdiagnosed. Under these circumstances, those parasites could adapt to humans in the years to come. Indeed, some studies have shown that host switching is common in the evolution of malaria parasites [94,95]. Thus, we believe that the surveillance and investigation of naturally acquired infections with other simian plasmodia should be given high priority.

Control of simian malarias
While human malaria control strategies mainly rely on insecticide-treated nets or indoor residual spraying and chemotherapy, the control of simian malaria poses a serious challenge, as neither the nonhuman primate reservoir nor the mainly forest-dwelling vectors are much affected by the human-focused interventions. A study in Sabah, Malaysia [96] indicated that mosquito nets have less protective effect against P. knowlesi infection mainly due to the bionomics of the local vector, An. balabacensis. New vector control tools like repellents and insecticide treated clothing need to be investigated. Besides, more robust vector studies should be conducted to enable the design of new tools for vector control. Given the importance of environmental change for the epidemiology of these infections, Forest, Agricultural, and Public Health Departments should work together to identify and investigate possible infections thoroughly and seek to curb transmission [74]. Certification of malaria-free status is granted by WHO if the transmission of human malaria in an entire country has been interrupted for at least 3 consecutive years [97]. For Southeast Asian countries aiming to obtain WHO certification of malaria-free status, it would be helpful if WHO could state whether P. knowlesi is to be considered a human malaria parasite in the context of certification. More generally, we call on WHO to spearhead the recognition of simian malarias in humans as a threat, for example, (but not only) by including simian malaria in the annual World Malaria Reports. This could spur increased surveillance and research to the benefit of the populations at risk and the ultimate goal of malaria eradication.

Case management of simian malarias
A rapid diagnostic test with high specificity and sensitivity is yet to be developed for P. knowlesi. In moving forward, it is also crucial to initiate the development of diagnostic kits for both P. inui and P. cynomolgi. In a recent case reported in Kelantan, Malaysia [98], a patient infected with P. cynomolgi was misdiagnosed as P. vivax through microscopic examinations. When later reexamined molecularly, it was found that the patient was infected with P. cynomolgi, not P. vivax. Furthermore, the knowledge of clinicians about simian malarias is insufficient [98]. It is not generally known, for example, that P. cynomolgi produce hypnozoites [99] and thus requires primaquine treatment.

Conclusions
Many countries in Southeast Asia are rapidly reducing indigenous human malaria transmission. However, the ongoing increase in P. knowlesi cases poses a major challenge to malaria control. Besides P. knowlesi, priority should be given to surveillance and control of other simian Plasmodium species, especially P. inui and P. cynomolgi, which have the potential to infect human. Simian malaria could emerge and spread as a major public health problem, as the 4 classical human malarias are reduced and as environmental change brings human into increased contact with simian malaria. The early recognition and containment of transmission of these simian malaria among humans should be given high priority.

Key Learning Points
• The reduction in human malaria cases may have exposed human populations in some areas of Southeast Asia to a greater risk of being infected by Plasmodium knowlesi due to diminishing cross-species immunity.
• Increased number of imported cases of simian malaria in countries outside Southeast Asia shows that simian malaria is now an international concern.
• Plasmodium cynomolgi and Plasmodium inui are more prevalent in Anopheles vectors and macaques hosts in Southeast Asia than P. knowlesi, raising the possibility of increased natural transmission of these simian malaria parasites to human in future.
• The possible emergence of other simian parasites (besides P. knowlesi) necessitates rapid diagnostic tests, appropriate treatment regimes, and new control strategies for malaria caused by these parasites.