Do climate changes alter the distribution and transmission of malaria? Evidence assessment and recommendations for future studies

Malaria, a mosquito-borne infectious disease, is considered a significant global health burden. Climate changes or different weather conditions may impact infectious diseases, specifically those transmitted by insect vectors and contaminated water. Based on the current predictions for climate change associated with the increase in carbon dioxide concentrations in the atmosphere and the increase in atmospheric temperature, the Intergovernmental Panel on Climate Change predicts that in 2050, malaria may threaten some previously unexposed areas worldwide and cause a 50% higher probability of malaria cases. Climatebased distribution models of malaria depict an increase in the geographic distribution of the disease as global environmental temperatures and conditions worsen. Researchers have studied the influence of changes in climate on the prevalence of malaria using different mathematical models that consider different variables and predict the conditions for malaria distribution. In this context, we conducted a mini-review to elucidate the important aspects described in the literature on the influence of climate change in the distribution and transmission of malaria. It is important to develop possible risk management strategies and enhance the surveillance system enhanced even in currently malaria-free areas predicted to experience malaria in the future.


INTRODUCTION
Malaria is a mosquito-borne infectious disease caused by the parasitic protozoans of the genus Plasmodium (P. vivax, P. falciparum, P. ovale, P. malariae, P. knowlesi, P. cynomolgi, and P. simium) and transmitted by female mosquito vectors of the Anopheles genus [1][2][3][4][5] . It is considered a significant global health burden at both global and regional levels 6,7 . The spatial distribution of malaria is largely influenced by climatic factors and ecological and socioeconomic factors, such as urban planning, land use, malaria treatment, and capacity of the healthcare system [8][9][10] . The influence of climate and socioeconomic changes on malaria's prevalence and distribution are being studied by researchers worldwide via projections that use different mathematical models (MMs) that consider different variables (e.g., altitude, temperature, humidity, Gini index, rainfall, humidity, and direction and speed of the winds) and predict the conditions for malaria distribution, incidence, transmission, and the costs associated with the treatment of new cases arising from climate change (CC) [11][12][13][14][15][16] . Most studies evaluating the effects of future CC are based on the Intergovernmental Panel on Climate Change's (IPCC) findings. In this context, we conducted a mini-review to elucidate the important aspects described in the literature on CC regarding the epidemiology and transmission of malaria and elucidated some of the effects of CC on anopheline mosquitoes and on the Plasmodium cycle in the vector. For review purposes, we reviewed the published articles. The identification of the articles was performed through a bibliographic search in the databases PubMed and SciELO Virtual Library, referring to the years 1994 to 2018 in English and Spanish. The search strategy used was as follows: (climate changes AND malaria) OR (temperature increase AND malaria) OR (humidity increases AND malaria) OR (climate change and Anopheles). From the paper evaluated, we made some recommendations for future studies of CC evaluation in the distribution and transmission of malaria worldwide, aiming to predict future impacts of CC on disease distribution. It is important that possible risk management strategies should be developed and surveillance system enhanced, including in the currently malaria-free areas predicted to experience malaria in the future.

Climate change and future climate scenarios and perspective of infection disease distribution
Global emissions of greenhouse gases (GHGs) from human activities have increased to >70% since the preindustrial period, 17,18 and in the face of CC in the late 1980s, the United Nations Environment Program and the World Meteorological Organization established the IPCC for a wide international network of scientists dedicated to assessing relevant scientific, technical, and socioeconomic information to understand the risks posed by CC worldwide and to propose CC scenarios for the future 19 . The Representative Concentration Pathways (RCPs), which are used to make projections based on population size, economic activity, lifestyle, energy use, land use patterns, and technology and climate policies, refer to the number of radiative forcing measured in watts per square meter per year by 2100 20 .
The RCPs include a stringent mitigation scenario (RCP2.6), two intermediate scenarios (RCP4.5 and RCP6.0), and one scenario with very high GHG emissions (RCP8.5) 21 . RCP2.6 represents a scenario that aims to keep global warming likely below 2°C 22 . The RCP8.5 combines, among other factors, high population growth, high energy demand, and high GHG emissions in the absence of CC policies, corresponding to the RCP with the highest GHG emissions. This scenario includes radiative forcing beyond 8.5 W/m2 and carbon dioxide (CO 2 ) concentrations of 1,370 ppm up to the year 2100 23 .
CC or different weather conditions patterns in different climatic scenarios are affecting the human health by increasing morbidity, mortality, and disabilities and through the emergence of diseases in previously non-endemic regions [24][25][26] . CC may impact infectious diseases' epidemiology and transmission, specifically diseases transmitted by insect vectors and contaminated water [27][28][29] , such as dengue, malaria, hantavirus pulmonary syndrome, salmonellosis, cholera, and giardiasis [30][31][32][33][34][35] . CC includes alterations in one or more climate variables (such as temperature, precipitation, wind, and sunshine), and these changes may impact the pathogens, vectors, hosts, and their living environment mainly by altering survival, reproduction, or distribution of disease pathogens and hosts [36][37][38] (Figure 1).
The impacts of CC must be understood in the context of several other forces that drive infectious disease dynamics, such as rapid evolution of drug-and pesticide-resistant pathogens, rapid global dissemination of microbes and vectors through expanding transportation networks, and deterioration of public health programs in some regions 39 .
The IPCC concluded, with high confidence, that CC will cause higher frequency of infectious disease epidemics following floods and storms and anticipated that CC will have a mixed effect on the spread of malaria 40 based on vector biology and vector absence in regions that are currently extremely cold for survival 41,42 . Climate malaria models depict an increase in the geographic distribution of the disease as global environmental temperatures and conditions worsen [43][44][45][46] . Based on the current CC associated with the increase of CO 2 concentrations and increase in atmospheric temperature, the IPCC predicts that in 2050, malaria may threaten some previously unexposed regions of South America, sub-Saharan Africa (SSA), and China, thus causing a 50% higher probability of malaria 47 .

Effects of climate change on malaria cases and their distribution: evaluating different theories and papers
The epidemiology of malaria is significantly complex and multifactorial 48 , and understanding the association between climate and malaria is essential 41 . The association between climate warming and a malaria epidemic has been strongly debated, and recent evidence suggests that the incidence of malaria is increasing in colder regions of the world due to global warming 43,49-51 . However, the CC effects are potentially complex, and the implications for malaria risk in optimal transmission settings remain poorly defined [49][50][51][52] . It most likely depends on the ecology of the prevalent main vector species, the biology of the causative organism (Plasmodium), the resistance and immunity of the host, and climatic factors such as temperature, precipitation, and humidity 38 (Figure 1).
Babaie et al. 53 suggested that temperature, precipitation, relative humidity, and wind intensity and direction are the most important climatic factors affecting the growth and proliferation of Anopheles, the Plasmodium life cycle, and the prevalence of malaria. A series of studies associating temperature variations to incidences of malaria has been debated by several researchers, with particular emphasis on changes in the malaria distribution scenario, specifically in the African continent 43,54-61 , although other studies haves been conducted in Asia 62,63 , South America, and Latin America 64,65 (Figure 2).
Egbendewe-Mondzozo et al. 66 studied the association between malaria cases and climatic factors in 25 African countries that were observed over a period of 11 years. They performed an econometric analysis and found that the number of malaria cases per 1,000 people is significantly influenced by climatic factors, with the incidence of malaria generally increasing under CC. According to the authors, socioeconomic factors also significantly affect the number of malaria cases.
According to Jaenisch and Patz 67 , Halimi et al., 68 and Salahi-Moghaddam et al., 69 the incidence of malaria is sensitive to short-term fluctuations in temperature and rainfall patterns, and with extreme precipitation, washing anopheline mosquito larvae from the breeding sites may lead to decreased incidence of malaria. Hay et al. 70 demonstrated that the mean temperature and precipitation did not change significantly over the past century in highland East Africa, where malaria incidence has increased. However, Pascual et al. 43 reanalyzed the temperature trend data in similar locations in East Africa between 1950 and 2002 and found that where there was an increase in the number of cases of malaria, the temperature increased by approximately 0.5ºC in the last half of the twentieth century. They also evaluated the biological significance of this trend and conducted an active model on the population dynamics of mosquito vectors with time series of temperature.
Halimi et al. 68 showed that in Iran, precipitation has been the most important climatic factor associated with annual malaria changes; hence, the outbreaks of the disease had a high correlation with precipitation and humidity index and lower correlation with the annual temperature mean. A study  MMs using equations with multiple climate variables and the biology of the Plasmodium in the mosquito vector that interferes with the epidemiology of malaria have been created.

Climatic changes
Over the last few decades, a number of MMs, typically statistical (using data and statistical approaches to correlate some climate variables with malaria incidence) or mechanistic (accounting for the detailed dynamic nonlinear processes involved in disease transmission, also sometimes referred to as "process based"), have been used to assess the likely impact of anthropogenic CC on malaria transmission dynamics and control 72 .
Omumbo et al. 57 have indicated through MM that the increase in temperatures observed in the regions of East Africa has an effect on the epidemiology of malaria since the transmission 5/9 of this condition depends on vector abundance, mosquito bite rate, vector survival, development rate, and environmental changes. Alonso et al. 54 developed a model that incorporated the population dynamics of the mosquito vector with the time series of temperature and found that a small increase in environmental temperature has an important but nonlinear effect on malaria transmission. Interestingly, temperature and precipitation exhibited nonlinear and synergistic effects on the incidence of malaria.
Parham and Michael 46 have developed a simple model of climate-related malaria transmission that provides insights into the sensitivity of disease transmission to changes in precipitation and temperature. These authors have shown that the development and analysis of such dynamic climate-based transmission models are crucial in understanding the rate at which P. falciparum and P. vivax can infect, expand, or disappear in the populations as local environmental conditions change. At fixed vector densities, P. falciparum and P. vivax differ by a few weeks in their rate of spread, with P. vivax spreading more rapidly than P. falciparum 73 .
Beck-Johnson et al. 74 developed a temperature-dependent differential equations model structured in stages to better understand how climate determines the risk of malaria transmission. By including the complete life cycle of the mosquito in the model, it is revealed that the abundance of the mosquito population is more temperature sensitive than previously thought as it is strongly influenced by the dynamics of the life cycle of the mosquito at a juvenile stage whose vital rates are also temperature dependent. Additionally, the model they developed predicts a peak in the abundance of mosquitoes old enough to transmit malaria at high temperatures.
Studies evaluating the effect of future CC, based on the IPCC findings, show the role of different scenarios in the epidemiology and distribution of malaria. In a multimodel set, using variables predicted by the IPCC and based on five different global climate models, each run under four emissions and a single population projection. Carmine et al. 75 investigated the modeling uncertainty associated with future projections of populations at risk for malaria due to CC. Their findings show the global net increase in climate suitability and the net increase in the population at risk, but with large uncertainties. The model outputs indicate a net increase in the annual person-months at risk when comparing from RCP2.6 to RCP8.5 from the 2050s to the 2080s. Hundessa 76 projected the long-term future distribution of P. vivax and P. falciparum malaria in China using 10 years of malaria surveillance data and 26 global climate models under two emission pathways (RCP8.5 and RCP4.5). The results indicate that P. vivax and P. falciparum malaria will increase in China, but by a larger amount in the RCP8.5 scenario. Moreover, a principal component regression model, constructed by Kwak et al. 77 considering multicollinearity, shows an increase in the occurrence of malaria and the shortening of annual time of occurrence in the future.
Future climate data, generated from RCP4.5 CC scenario climate model, were applied to the constructed regression model to simulate future malaria occurrence and analyze the trend of occurrence and showed an increase in the occurrence of malaria and the shortening of annual time of occurrence in the future. According to Kibret 78 , in SSA, the number of dams located in malaria areas is projected to increase in both RCPs. However, the number of cases will always be higher in RCP 8.5 than in RCP 2.6.

Effects of climate change on anopheline mosquitoes and on the Plasmodium cycle in the vector
Changes in vector body temperature due to variation in ambient temperature are able to significantly change the metabolic rates of insect vectors and the parasites they host 76,79 . Anopheline mosquitoes in environments with climatic warming effects have a higher metabolite rate, which in turn directly interferes with the development of larvae that take less time to mature 78 . Additionally, temperature increases promote the rapid digestion of blood supply, which in turn promotes a significant increase in fecundity, with the development of better reproductive fitness and a greater ability to produce more offspring 45 . Additionally, an increase in the ambient temperature can intensely increase the feeding frequency in the hosts and increase the annual temporal patterns of mosquito activity (particularly biting rates) and promotes a reduction in the time of sporogonic development of Plasmodium from an average of 14 days to 12.6 days 45 .
Noden et al. 80 demonstrated that temperature affects the sporogonic development of P. falciparum in A. stephensi mainly by altering the kinetics of ookinete maturation. High temperatures (30ºC and 32ºC) are considered to significantly affect parasite densities and infection rates, interfering in the developmental processes that occur between fertilization of the parasite and ookinete formation, specifically during zygote formation and early maturation of ookinete. Okech et al. 81 demonstrated that in the experimental model for P. falciparum and A. gambiae, the prevalence of ookinete infection decreased with an increase in temperatures, while the mean oocyst intensities for infected mosquitoes increased with temperatures. Furthermore, they demonstrated that the success of infections reduced by 30°C and 32°C and resulted in greater losses during consecutive periods of development of the interstage parasite. These results show that the most significant impact of high temperatures occurs mainly at the transition between macrogametocytes and ookinetes, while the transition between oocysts and oocysts was apparently unaffected.
Temperature can shape the parasite's resistance and growth phenotype through indirect effects, which are still poorly studied, but they may interfere with the innate immune system of the mosquito and influence the development of the parasite cycle 82 . Suwanchaichinda et al. 83 investigated how larval nutrition and environmental temperature changes affect the melanization process in adults of A. gambiae after challenging with Sephadex beads and have shown that melanization progressively decreases as the temperature increases by 24ºC to 30°C 81 . Murdock et al. 82 emphasized that with A. stephensi, temperature change (ranging from 12°C to 34°C) is responsible for shaping variable cellular 6/9 and humoral immune responses. The temperature was shown to significantly affect all measured immune responses, but in different ways. Unexpectedly, melanization, phagocytosis, and expression of defensin (an antimicrobial peptide) were all higher at 18°C (8°C below the standard breeding temperature for this mosquito). The expression of nitric oxide synthase reached its peak at 30°C, while the expression of the antimicrobial peptide cecropin was not directly affected by temperature. Based on these findings, it is possible to conclude that the immunological profile described during a Plasmodium infection is potentially different if the same experiment is conducted at different temperatures.
Although most studies available in the literature describe the association between high temperatures and the increase of malaria cases in certain regions of the globe, specifically in the African continent, Murdock et al. 84 demonstrated in the laboratory that temperature increases are capable of affecting the ability of A. gambiae and A. stephensi to transmit; increase in temperature reduces the rate and intensity of infection and increasing mosquito mortality, thus reducing the vector capacity of both species of mosquitoes 84 . These authors demonstrated that increases of 3°C from 27°C reduced vector capacity by 51%-89%, depending on the species, and at 33°C, the transmission potential was even lower for A. stephensi and completely blocked in A. gambiae. Additionally, these authors suggest that instead of increasing the risk, current and future warming may reduce transmission potential in existing hightransmission environments.

Recommendations for future studies
Assessing the potential change in the risk of malaria caused by CC and climate variability remains an important topic. The effects of CC on malaria appear to be multifactorial, and these distinct factors should be evaluated individually and in conjunction with modeling CC and malaria transmission to determine the factors involved in the incidence of malaria currently and in future climatic conditions as predicted by the IPCC. This will help to effectively control malaria; additionally, for greater impact, interventions could be synchronized with the most important climatic predictors of the disease. Furthermore, a better understanding of the association between precipitation patterns, temperature and humidity changes, and malaria cases and associated mortality is necessary to establish effective strategies for CC, involving the planning and implementation of appropriate control interventions for diseases. The distinct search allows decision-makers to assess changes in malaria in the different territories.
Furthermore, we would like to emphasize some important topics for elucidation in future research: i) Development and validation of climate and ecosystem based on distinct early malaria epidemic prediction models.
ii) Malaria modeling at the regional or national scales and integrating regional environmental programs and socioeconomic variation 25 , mainly in Latin America, where studies regarding malaria are insufficient.
iii) Assessment through multidisciplinary perspectives (life science, human science, and biophysics) of the qualitative factors that affect the transmission of malaria at different spatial and temporal scales, evaluating the interactions, synergism, and nonlinearity between the different factors involved in the transmission. iv) Development of malaria transmission models to assess the risks posed by climatic and ecological changes. Observational, experimental, and modeling activities are all highly interdependent and must progress in a coordinated fashion 51 .
v) Construction of the malaria transmission risk maps according to the IPCC forecast of future climatic conditions to provide an analysis guide for the planning and implementation of appropriate control interventions based on the research agenda for the elimination of malaria.
vi) Understanding the behavior of the malaria vector and the etiological agent in the face of CCs, predicted by the climatic scenarios provided by the IPCC, with special emphasis on the tropical countries of America.
vii) Association between temperature-induced variations in gene expression and functional resistance or malaria vector competence. Thus, the effects of temperature on the production of antimicrobial peptides and on the activity of the nitric oxide enzyme and phagocytosis of pathogens should be investigated under which mosquitoes are subject to Plasmodium.

Final remarks
Different studies were conducted to estimate the impact of future CC and population scenarios on malaria transmission at a global scale and to provide recommendations for future interventions and research. The research's results indicate that future climate might become more suitable for malaria transmission in the tropical regions. However, other factors such as land use change, population growth and urbanization, migration changes, economic development, and associated costs will need to be factored into future risk assessments for the treatment of the disease or for the determination of alternative treatments to block vector transmission through the use of mass treatment of the population.