Tropical Malaria Does Not Mean Hot Environments

FORUM

Tropical Malaria Does Not Mean Hot Environments TAKAYA IKEMOTO1 Department of Microbiology, Teikyo University School of Medicine, Tokyo 173-8605, Japan

J. Med. Entomol. 45(6): 963Ð969 (2008)

ABSTRACT If global warming progresses, many consider that malaria in presently malaria-endemic areas will become more serious, with increasing development rates of the vector mosquito and malaria parasites. However, the correlation coefÞcients between the monthly malaria cases and the monthly mean of daily maximum temperature were negative, showing that the number of malaria cases in tropical areas of Africa decreases during the season when temperature was higher than normal. Moreover, an analysis of temperature and development rate using a thermodynamic model showed that the estimated intrinsic optimum temperatures for the development of the malaria parasites, Plasmodium falciparum and P. vivax, in the adult mosquito stage and that of the vector mosquito Anopheles gambiae s.s. were all ⬇23Ð24⬚C. Here, the intrinsic optimum temperature is deÞned in the thermodynamic model as the temperature at which it is assumed that there are no or negligible adverse effects for development. Therefore, this study indicates that the development of malaria parasites in their mosquito hosts and the development of their vector mosquitoes are inhibited at temperatures higher than 23Ð24⬚C. If global warming progresses further, the present center of malarial endemicity in sub-Saharan Africa will move to an area with an optimum temperature for both the vector and the parasite, migrating to avoid the hot environment. KEY WORDS Anopheles gambiae, Plasmodium falciparum, intrinsic optimum temperature, Africa, malaria

Global warming is a serious problem considering its consequences, such as a rise in the incidence and spread of diseases like malaria. As global warming progresses, there is concern for increased transmission of some tropical diseases and potential for their expansion into temperate regions (Sutherst 1993). Moreover, many consider that malaria in presently malaria-endemic areas will become more serious with increasing developmental rates of vector mosquitoes and malaria parasites (Lindsay and Birley 1996, Massad and Forattini 1998, Craig et al. 1999, Pascual et al. 2006). These predictions are mostly based on the change of global temperature and the present distribution of malaria parasites and their vector mosquitoes. Predictive studies based on biological responses of parasites and vector to warming are rarely performed. Here, I attempt to analyze the correlation between the seasonal number of malaria cases in some districts of tropical Africa and the seasonal ßuctuation of temperature there. Moreover, the intrinsic optimum temperatures (T⌽) for the development of the malaria parasites and the vector mosquito are estimated by using a thermodynamic model, which is able to exhibit the minimum effects on enzyme inactivation in relation to development of ectotherms at low and high temperatures (Ikemoto 2005). 1

Corresponding author, e-mail: [email protected].

The results indicate that the center of malarial endemicity in tropical Africa will move to an area with an optimum temperature for both the vector and the parasite, if global warming progresses further. Although these results may be outside general considerations, this Þnding agrees with previous reports concerned with the relationship between the temperature and sexual events of the malaria parasite in the mosquito gut (OgwanÕg et al. 1993), the relative transcription levels of rRNA involved with parasite sporogony in the mosquito (Fang and McCutchan 2002), and the success of mosquito development from the aquatic stage to adult (Bayoh and Lindsay 2003). Materials and Methods Data Sources. The three sets of monthly malaria cases (incidence data) were analyzed. They are those from Dar es Salaam of Tanzania (WHO 1986, January 1981ÐDecember 1985, low land), Dodowa of Ghana (Dodoo et al. 1999, April 1994 ÐMay1995, low land), and Kericho of Kenya (Malakooti et al. 1998, January 1991ÐDecember 1997, high land). The meteorological data in Dar es Salaam and Kericho were found in the same report of malaria cases and those of Dodowa weretakenfromAccra,located50kmaway(TuTienpo. net 2007). Data for the temperature-dependent development of the human malaria parasites, Plasmodium falcipa-

0022-2585/08/0963Ð0969$04.00/0 䉷 2008 Entomological Society of America

964

JOURNAL OF MEDICAL ENTOMOLOGY

rum (Welch) (Coccidiina: Plasmodiidae) and P. vivax (Grassi and Feletti), in vector mosquitoes were combined from the literature (Moshkovsky and Rashina 1951, Macdonald 1952, Russell et al. 1963). Data for the temperature-dependent development of the malaria vector mosquito Anopheles gambiae s.s. (Giles) (Diptera: Culicidae) (Bayoh and Lindsay 2003) were used for analysis. Nonlinear Curve Fitting for the Thermodynamic Model. The law of total effective temperature applied to the temperature-dependent development of arthropods and other poikilotherms is well known and is expressed by the equation: 1 t 1 ⫽⫺ ⫹ T D k k

[1]

where D, T, t, and k represent the duration of development (d), environmental (mean/isothermal) temperature (⬚C), the estimated developmental zero (threshold) temperature, and the effective cumulative temperature, respectively. This model is referred to as the linear degree-day (rate) model, which can be applied within an appropriate and middle temperature range (Ikemoto and Takai 2000). However, the nonlinear thermodynamic model (Sharpe and DeMichele 1977, SchoolÞeld et al. 1981, Ikemoto 2005) can be applied for a wider range of temperature. The model equation is expressed as follows:

冋 冉 冋 冉 冋 冉

冊册 冊册 冊册

关T兴 ⌬HA 1 1 exp ⫺ 关T⌽兴 R 关T⌽兴 关T兴 r⫽ ⌬HL 1 1 1 ⫹ exp ⫺ R 关TL兴 关T兴 ⌬HH 1 1 ⫹ exp ⫺ R 关TH兴 关T兴

␳

[2]

where r represents the development rates (the dependent variables) at the absolute temperature [T] (the independent variable). All the other parameters are constants: [TL], [TH] and [T⌽] represent the absolute temperatures, ⌬HA, ⌬HL, and ⌬HH represent enthalpy changes, R is the universal gas constant, and ␳ is the development rate at [T⌽]. Here, [T⌽] is the most important constant with the deÞnitive concept; that is, the intrinsic optimum temperature for development that exhibits the minimum effects on enzyme inactivation related to development at low and high temperatures. In this deÞnition, [T⌽] must have the relationships with the other constants as follows: 关T⌽兴 ⫽

⌬HL ⫺ ⌬HH ⌬HL ⌬HH ⌬HL R ln ⫺ ⫹ ⫺ ⌬HH 关TL兴 关TH兴

冉 冊 冉 冊 冉 冊 (Ikemoto 2005)

[3]

The detailed deÞnitions of all the constants were reported previously (Ikemoto 2005). This multiconstant model is able to derive more information about the temperatures concerned with development than the linear model, when nonlinear line Þtting for data points is completed. However, a set of nonlinear least-

Vol. 45, no. 6

square estimates by general methods such as the Quasi-Newton method is not necessarily unique, and other sets of estimates could also provide acceptable results, according to the initial values. Here, I show the condensed procedure for the curve Þtting of the thermodynamic model developed by Ikemoto (2005) as follows. Step 1. Determination of a tentative value of [T⌽] and replacement of ␳ by 1/D at T⌽. The replacement of ␳ by 1/D at 293.15 K of [T⌽] (⫽20⬚C) was usually set up as the Þrst candidate. Step 2. Replacement of [TL] by [t]. 关TL兴 ⫽ 关t兴

[4]

Step 3. Estimation of ⌬HA. The Arrhenius plot (ln rate versus reciprocal kelvin) gives a line with slope. ⌬HA can be estimated by the following equation: ⌬HA ⬇ ⫺␤ ⫻ R (SchoolÞeld et al. 1981)

[5]

where ␤ is the slope of the line and R is the universal gas constant (1.987 cal/deg/mol). The reduced major axis line-Þtting method (Ikemoto and Takai 2000) was applied to this step. In practice, line-Þtting performed against the limited points of the Arrhenius plot {1/[T], ln r ⫽ ln(1/D)}, namely, the points within the range of appropriate and middle temperature for the linear model. Here,

␤⫽⫺

冑

冘 n

关lnri ⫺ 共lnr兲兴 2

i⫽1

冘冋 n

i⫽1

1 ⫺ 关T兴 i

冉 冊册 1 关T兴

2

[6]

Step 4. Trial Þtting of ⌬HH and [TH]. Before beginning the trial, a ⌬HL value of ⫺100,000 was tentatively assigned. The combination of ⌬HH and [TH] was determined by manual trials with ␹2 values showing the extent of differences between estimated and observed values. In practice, the determination was carried out by searching for the pair with the minimum ␹2 value from among ⬎4,000 pairs of ␹2 values calculated at 10 cal/mol intervals for ⌬HH and 0.01 K interval for [TH] on the macro function in Excel (Excel 2003; Microsoft, Redmond, WA). Step 5. Correcting of ⌬HL. Equation 3 was used for this purpose. The solution of the ⌬HL value was done by Goal-Seeking software in the Excel tool menu. Step 6. Returning to Step 4. Because the ⌬HL value assigned tentatively in step 4 was changed through step 5, readjustments for ⌬HH and [TH] values were needed by returning to Step 4. The number of iterations was usually several times until the values of these three constants converged. Step 7. Changing the Þrst candidate of [T⌽] value to the next one. To determine the best Þt curve of the thermodynamic nonlinear model, the 293.15 K (⫽20⬚C) of the initial [T⌽] value selected in step 1 was changed to 293.65 (⫽20.5⬚C). Then, steps 1Ð 6 were conducted.

November 2008

IKEMOTO: TROPICAL MALARIA AND TEMPERATURE

965

Fig. 1. Seasonal prevalence of malaria and meteorological elements at three areas in sub-Saharan Africa. Meteorological data of Accra was cited from http://www.tutiempo.net/en/Climate/Accra/654720.htm (TuTienpo.net. 2007), and those of the other two sites were from the data of the same reports of malaria cases and were rearranged (see Table 1).

After this procedure, the replacements of [T⌽] values at 0.5 K intervals were repeated until 298.15 K (⫽25.0⬚C), and then the best Þt set of the constants expressing the model curve was selected by the minimum value of ␹2. Results and Discussion Seasonal Prevalence of Malaria and Temperature in Tropical Africa. The development rate of malaria parasites in their vector mosquitoes and that of ectotherm vector mosquitoes change according to habitat temperature. For example, high temperature increases the rate at which mosquitoes develop into adults, the frequency of their blood feeding, the rate with which parasites are acquired, and the incubation time of the parasite within the mosquito, although these inßuences must be compared with the opposing effects that high temperatures exert in reducing adult mosTable 1.

quito survival (Patz et al. 2000). Generally, the rate of transmission of malaria parasites by vector mosquitoes should be higher at higher temperatures because of their higher vectorial capacities, and the number of malaria cases increases during the hot season (Lindsay and Birley 1996). However, some studies showed a negative correlation between the number of malaria cases and temperature, as determined by an analysis of the monthly mean maximum or minimum temperature and the monthly number of malaria cases in Africa (Fig. 1; Table 1). This negative correlation was remarkable in Dar es Salaam of Tanzania and Dodowa of Ghana, which showed a coefÞcient of correlation between the monthly mean maximum temperature and the number/incidence of malaria cases of ⫺0.7 or higher. This negative correlation was also observed in Dar es Salaam, when the number of malaria cases of the next month with respect to the mean maximum tempera-

Coefficients of correlation between monthly metrological factor and no. of malaria cases per month No month lag

1-mo lag

2-mo lag

Mean/yr

⫺0.640b ⫺0.721a ⫺0.526 ⫺0.255 ⫺0.236 0.356 Dodowa, low land of Ghana (Apr. 1994 ÐMar. 1995)

⫺0.268 0.069 0.532

254 30.8 20.5 98

⫺0.467 ⫺0.700a ⫺0.242 ⫺0.570b ⫺0.183 0.213 Kericho, high land of Kenya (Jan. 1991ÐDec. 1997)

⫺0.164 0.128 0.446

17.6 30.6 23.4 67

Dar es Salaam, low land of Tanzania (Jan. 1981ÐDec. 1985) Number of cases Mean maximum temperature (⬚C) Mean minimum temperature (⬚C) Rainfall (mm) Incidence Mean maximum temperature (⬚C) Mean minimum temperature (⬚C) Rainfall (mm) Number of cases Mean maximum temperature (⬚C) Mean minimum temperature (⬚C) Rainfall (mm)

⫺0.445 0.120 0.183

⫺0.208 0.362 0.544

0.219 0.628b 0.597b

129 26.2 11.2 146

Number of cases/1,000 wk at risk. a SigniÞcance level at P ⬍ 1%. b SigniÞcance level at P ⬍ 5%. No month lag, the no. of malaria cases of the same month as that of the mean max temp.; 1-mo lag, the no. of malaria cases of the next month with respect to the mean max temp of the previous month; 2-mo lag, the no. of malaria cases of the next 2 mo with respect to the mean max temp of 2 mo previously.

966

JOURNAL OF MEDICAL ENTOMOLOGY

ture of the previous month (i.e., 1-mo lag) was analyzed. The signiÞcantly high coefÞcient of negative correlation between the monthly mean minimum temperature and the number/incidence of malaria cases was also found in Dodowa. However, in Kericho, the highland site in Kenya, signiÞcant negative coefÞcients were not detected between temperature and malaria incidence. Shanks et al. (2002) and Teklehaimanot et al. (2004) also did not Þnd a signiÞcant association between temperature and increased malaria in the highland area of Kenya and Ethiopia, respectively. The reason seems to be attributable to the cool temperatures in these highland areas in comparison with the hot temperature in surrounding lowland. Temperature, rainfall, and other environmental conditions must have an effect on the survival and Þtness of the parasites and vectors, in addition to their having an effect on development rate. Several excellent studies, such as those on the relationship between malaria epidemics and rainfall, which is related to El Nin˜ o, and the mapping of malaria-free areas, which are expected to be found in low-temperature areas, have been reported (Craig et al. 1999, Wort et al. 2004). In Uganda, close correlation was found between the peak rainfall and the peak of malaria incidence with a time lag of 2Ð3 mo (Kilian et al. 1999). Unlike temperature, water can persist in the environment and can increase larval habitat as things dry out and pooling occurs. In principle, the amount of larval habitat that leads to abundance of biting vectors and higher humidity caused by rainfall leading to survival of mosquitoes should have an important effect on malaria transmission. However, the causal relationship between rainfall and malaria is far from simple, as already pointed out by Russell et al. (1963). For example, heavy rain may act to ßush out breeding places. However, mild showers have no such effect. Flooding may be beneÞcial or otherwise, depending on the nature of the countryside and malaria vector. Flooding may not be caused by local rain but to rain in the distant hills. In such cases, the local rainfall statistics may have no direct value as an index in malaria epidemiology. Therefore, in this paper, it is not my purpose to simply explain the degree of malaria transmission using only temperature data of restricted areas. I shall only emphasize those examples of cases that show a negative effect of high temperature on malarial transmission. Intrinsic Optimum Temperature for Development of Malaria Parasites. Detailed analysis of how high temperatures affect the development rates of the malaria parasites and vector mosquitoes follows. Macdonald (1952) examined in detail the relationship between temperature and the development rates of P. falciparum and P. vivax on the basis of numerous reported data. Two smoothed curves were drawn for each of these organisms, showing data points for temperature on the y axis and the required number of days for development on the x axis in his original graph. He reprinted the graph without the data points in a subsequent book (Macdonald 1957). To date, the graph

Vol. 45, no. 6

has been reprinted in many textbooks and papers on malaria (Pampana 1969, Bruce-Chwatt 1985, Wernsdorfer and McGregor 1988, Patz and Olson 2006). Although the graph is excellent in principle, the values that the graph curves show are not reliable, even if the error estimates are not large. Mcdonald (1952) wrote in his original paper, “Smoothed curves have been drawn by eye to represent the mean of the observations for each of the parasites.” Therefore, here, the thermodynamic mathematical model that was Þrst developed by Sharpe and DeMichele (1977) and further developed by SchoolÞeld et al. (1981) and Ikemoto (2005), was applied to analyze the relationship between temperature and development rate. In this paper, I focus my discussion on the most important constant, that is [T⌽], which is the temperature at which it is assumed that there is negligible or no enzyme inactivation. On the basis of this assumption, I previously suggested (Ikemoto 2005) that [T⌽] be conferred a new deÞnition, that is, it is the intrinsic optimum temperature for development that tends to be Þxed for a taxon (Ikemoto 2003, 2005) and that it exerts a minimal effect of inhibiting development at low and high temperatures. The T⌽ values of P. falciparum and P. vivax were estimated to be 23.5 (⫽296.65 K) and 24.0⬚C (⫽297.15 K), respectively (Fig. 2). These results indicate that 23Ð24⬚C is the approximate range of T⌽values for the development of both malaria parasites in their vector mosquitoes. The negative effects on development gradually become large at higher temperatures, thereby decreasing the development rates of P. falciparum and P. vivax at 31.0 and 29.8⬚C, respectively. Schematically, the seasons in which the daily mean temperature is much higher than 24⬚C would have deleterious effects on the development of the malarial parasites. At temperatures lower than T⌽, the negative effects on development also gradually become large, thereby decreasing development rate. Although no critical points are found in the lower-temperature range of the thermodynamic nonlinear model, the standard linear model can well express the critical point, that is, the developmental zero temperature. The present estimated values were ⬇16⬚C for both parasites, whereas previously reported data (Detinova 1962) showed that temperatures ⬍18 and 16⬚C caused the cessation of the development of P. falciparum and P. vivax in the vector mosquitoes. A negative effect of high or low temperature was shown as early as 1940 by Stratman-Thomas (1940) for P. vivax in An. quadrimaculatus Say held ⬎30 or ⬍15Ð 17⬚C, respectively. Noden et al. (1995) found that P. falciparum development to ookinete was almost prevented in An. stephensi Liston held at 30 Ð32⬚C, although at 21Ð27⬚C, the development was normal. More recently, Okech et al. (2004) found that the same development in An. gambiae was normal at 24 Ð 28⬚C. All of these results support this analysis, although the viewpoints of the optimum temperature range and the intrinsic optimum temperature were different from each other.

November 2008

IKEMOTO: TROPICAL MALARIA AND TEMPERATURE

967

Fig. 2. Temperature-dependent sporogony of the two malaria parasites in vector mosquitoes is expressed by the thermodynamic model and the linear model. Intrinsic optimum temperatures were estimated to be in the 23.5Ð24.0⬚C range for both parasites. Data are from three reports (Macdonald 1952, Moshkovsky and Rashina 1951, Russell et al. 1963). Circle, observed data point with a number of replications; open data point, out of range for linear model; x, omitted data points for nonlinear Þtting; square (middle), intrinsic optimum temperature T⌽ for development of parasite; square (lower), TL; square (upper), TH of constants.

Intrinsic Optimum Temperature for Development of the Vector Mosquito, Anopheles gambiae. Generally, the duration of immature stages of mosquitoes is shorter at a water temperature in the upper limit of an optimum temperature range. Therefore, the possibility that the transmission rate of malaria reaches its peak during the hot season with increasing vector mosquito supply rate may be reasonable. However, there is not necessarily a direct correlation between the supply of adult mosquitoes and the development rate of immature mosquitoes. That is, although development is rapidly completed at high temperatures, mortality in the aquatic period might become high, and the adult body size might be small, resulting in a decrease in the number of eggs laid (Gillooly et al. 2002). At low temperatures, however, the risk of water drying up in a breeding puddle might increase during a prolonged development phase, and the mosquito might be vulnerable to attacks by its predators and pathogens. Therefore, the vector An. gambiae might have also acquired its own intrinsic optimum temperature for development during the course of its evolution. The T⌽ of An. gambiae s.s. was estimated to be 22.5⬚C (⫽295.65 K) and the temperature at which development rate reaches the peak was 30.1⬚C, as determined from reported data (Bayoh and Lindsay 2003) using the same method as that used for malaria

parasites (Fig. 3). From the data obtained at 2⬚C intervals, the percentage of adult eclosion from the egg reaches the maximum (⬇80%) at ⬇24.0⬚C (Bayoh and Lindsay 2003), which is about the same value as the estimated T⌽. Coevolution of Malaria Parasites and Their Vector Hosts. It was noteworthy that the estimated T⌽values for the development of two human malaria parasites and their important vector mosquito were determined to be nearly the same (23Ð24⬚C) and the developmental limit for high temperatures was at ⬇30⬚C for both the parasites and their vector. These suggest the coevolution of temperature adaptation for the malaria parasites and their vector hosts. Sexual Events of Malaria Parasites in Mosquito Gut and Temperature. Because malaria parasites complete their life cycle by alternating between the human and mosquito bodies, they are exposed to a large difference in temperature between humans and mosquitoes. However, the malaria parasites seem to be able to use this large difference for the regulation of their bionomics. Malaria parasites are exposed to high temperature in humans ranging from 37 to 41⬚C during febrile episodes in the patient. Pavithra et al. (2004) showed that the repeated exposure to elevated temperatures promoted the development of P. falciparum in the erythrocytes. The heat shock protein 90 (PfHsp90) multi-chaperon complex produced by the ma-

968

JOURNAL OF MEDICAL ENTOMOLOGY

Vol. 45, no. 6

Fig. 3. Temperature-dependent development of the malaria vector mosquito An. gambiae s.s. is expressed by the thermodynamic model and the linear model. The estimated intrinsic optimum temperature T⌽ is 22.5⬚C. Data were from Bayoh and Lindsay (2003). Circle, observed data point; open data point, out of range for linear model; square (middle), intrinsic optimum temperature T⌽ for development of mosquito; square (lower), TL; square (upper), TH of constants.

laria parasite plays a role in physiological adaptation to higher temperatures. Inhibition of PfHsp90 function using geldanamycin attenuated the temperature-dependent progression from the ring to the trophozoite stage. Although these results might suggest the positive effect of high temperature on the whole life cycle of the parasite, the lack of heat shock proteins in the sporozoite, which is the transmission form from the mosquito to the human, indicates that this crucial stage of the parasite life cycle is susceptible to deleterious effects of high temperature (Pavithra et al. 2007). When a mosquito ingests human blood with malaria parasites, the sexual forms (gametocytes) of malaria parasites undergo further development in the mosquito midgut. In a male gametocyte, the nucleus divides into four to eight nuclei, each of which forms a long ßagellum 20 Ð25 ␮m in length. The ßagella shoot out from the original cell, lash about for a while, and then break free. This process, called exßagellation, takes only 2Ð3 min at the appropriate temperature. A female gametocyte undergoes maturation into a female gamete. In the midgut of the mosquito, a male gametocyte is attracted to a female gamete and then completes fertilization (Bruce-Chwatt 1985). The exßagellation of P. falciparum gametocytes occurs optimally at 23⬚C compared with that at 37⬚C (OgwanÕg et al. 1993). Although OgwanÕg et al. (1993) did not determine the optimum temperature for exßagellation, they observed that gametocytes maintained at temperatures ⬍30⬚C exßagellated during 1 h of incubation. Thermoregulation, in which the transcription of selected RNAs is up-regulated at temperatures lower than the optimal, is crucial for the developmental transition that occurs during the transmission of P. falciparum from humans to mosquitoes (Fang and McCutchan 2002). Namely, A1-type and A2-type rRNAs predominate during schizogony in humans, whereas S1-type rRNA predominates during gametogony and S2-type rRNA during the human-infective sporogony in mosquitoes. S gene transcription is sen-

sitive to temperature: at 42⬚C, neither the S1 nor S2 gene is transcribed. The S2 transcription levels at 31 and 26⬚C are, respectively, 4.4- and 15-fold to that at 37⬚C, suggesting that the development of a malaria parasite in its host mosquito is strongly inhibited in a high-temperature environment. All of the results analyzed here and the reported data indicate that the development of the malaria parasites in their mosquito host and that of their vector mosquitoes are gradually inhibited in temperatures far from the intrinsic optimum temperature, particularly at temperatures ⬎23Ð24⬚C. If the present temperature in the sub-Saharan region further increases with global warming, malarial endemicity in this region should decrease because of the negative effect of high temperature.

References Cited Bayoh, M. N., and S. W. Lindsay. 2003. Effect of temperature on the development of the aquatic stages of Anopheles gambiae sensu stricto (Diptera: Culicidae). Bull. Entomol. Res. 93: 375Ð381. Bruce-Chwatt, L. J. 1985. Essential malariology, 2nd ed. William Heinemann Medical Books, London, United Kingdom. Craig, M. H., R. W. Snow, and D. Le Sueur. 1999. A climatebased distribution model of malaria transmission in SubSaharan Africa. Parasitol. Today 15: 105Ð111. Detinova, T. S. 1962. Age-grouping methods in Diptera of medical importance with special reference to some vectors of malaria. WHO, Geneva, Switzerland. Dodoo, D., T. G. Theander, J.A.L. Kurtzhals, K. Koram, E. Riley, B. D. Akanmori, F. K. Nkrumah, and L. Hviid. 1999. Levels of antibody to conserved parts of Plasmodium falciparum merozoite surface protein 1 in Ghanaian children are not associated with protection from clinical malaria. Infect. Immun. 67: 2131Ð2137. Fang, J., and T. F. McCutchan. 2002. Malaria: thermoregulation in a parasiteÕs life cycle. Nature (Lond.) 418: 742. Gillooly, J. F., E. L. Charnov, G. B. West, V. N. Savage, and J. H. Brown. 2002. Effects of size and temperature on developmental time. Nature (Lond.) 417: 70 Ð73.

November 2008

IKEMOTO: TROPICAL MALARIA AND TEMPERATURE

Ikemoto, T. 2003. Possible existence of a common temperature and a common duration of development among members of a taxonomic group of arthropods that underwent speciational adaptation to temperature. Appl. Entomol. Zool. 38: 487Ð 492. Ikemoto, T. 2005. Intrinsic optimum temperature for development of insects and mites. Environ. Entomol. 34: 1377Ð 1387. Ikemoto, T., and K. Takai. 2000. A new linearrized formula for the law of total effective temperature and the evaluation of line-Þtting methods with both variables subject to error. Environ. Entomol. 29: 671Ð 682. Kilian, A.H.D., P. Langi, A. Talisuna, and G. Kabagambe. 1999. Rainfall pattern, El Nino and malaria in Uganda. Trans. Roy. Soc. Trop. Med. Hyg. 93: 22Ð23. Lindsay, S. W., and M. H. Birley. 1996. Climate change and malaria transmission. Ann. Trop. Med. Parasitol. 90: 573Ð 588. Macdonald, G. 1952. The analysis of the sporozoite rate. Trop. Dis. Bull. 49: 569 Ð586. Macdonald, G. 1957. The epidemiology and control of malaria. Oxford University Press, London, United Kingdom. Malakooti, M. A., K. Biomndo, and G. D. Shanks. 1998. Reemergence of epidemic malaria in the highlands of western Kenya. Emerg. Infect. Dis. 4: 671Ð 676. Massad, E., and O. P. Forattini. 1998. Modelling the temperature sensitivity of some physiological parameters of epidemiologic signiÞcance. Ecosystem Health 4: 119 Ð129. Moshkovsky, S. D., and M. G. Rashina. 1951. Epidemiology and medical parasitology for entomologists. Moscow. In Russian, unknown publisher, after Detinova, T.S. 1962. Age-Grouping Methods in Diptera of Medical Importance with Special Reference to Some Vectors of Malaria. WHO, Geneva, Switzerland. Noden, B. H., M. D. Kent, and J. C. Beier. 1995. The impact of variations in temperature on early Plasmodium falciparum development in Anopheles stephensi. Parasitology 111: 539 Ð545. Ogwan’g, R. A., J. K. Mwangi, J.B.O. Were, C. R. Roberts, and K. Martin. 1993. Factors affecting exßagelation of in vitro-cultivated Plasmodium falciparum gametocytes. Am. J. Trop. Med. Hyg. 49: 25Ð29. Okech, B. A., L. C. Gouagna, E. Walczak, E. W. Kabiru, J. C. Beier, G. Yan, and J. I. Githure. 2004. The development of Plasmodium falciparum in experimentally infected Anopheles gambiae (Diptera: Culicidae) under ambient microhabitat temperature in western Kenya. Acta Trop. 92: 99 Ð108. Pampana, E. 1969. A textbook of malaria eradication, 2nd ed. Oxford University Press, London, United Kingdom. Pascual, M., J. A. Ahumada, L. F. Chaves, X. Rodo, and M. Bouma. 2006. Malaria resurgence in the East African highlands: temperature trends revisited. Proc. Natl. Acad. Sci. U.S.A. 103: 5829 Ð5834. Patz, J. A., and S. H. Olson. 2006. Malaria risk and temperature: Inßuences from global climate change and local

969

land use practices. Proc. Natl. Acad. Sci. U.S.A. 103: 5635Ð 5636. Patz, J. A., T. K. Graczyk, N. Geller, and A. Y. Vittor. 2000. Effects of environmental change on emerging parasitic diseases. Int. J. Parasitol. 30: 1395Ð1405. Pavithra, S. R., G. Banumathy, O. Joy, V. Singh, and U. Tatu. 2004. Recurrent fever promotes Plasmodium falciparum development in human erythrocytes. J. Biol. Chem. 279: 46692Ð 46699. Pavithra, S. R., R. Kumar, and U. Tatu. 2007. Systems analysis of chaperone networks in the malarial parasite Plasmodium falciparum. PLoS Comput. Biol. 3: e168. Russell, P. F., L. S. West, R. D. Manwell, and G. Macdonald. 1963. Practical malariology, 2nd ed. Oxford University Press, London, United Kingdom. Schoolfield, R. M., P.J.H. Sharpe, and C. E. Magnuson. 1981. Non-linear regression of biological temperature-dependent rate models based on absolute reaction-rate theory. J. Theor. Biol. 88: 719 Ð731. Shanks, G. D., S. I. Hay, D. I. Stern, and K. Biomndo. 2002. Meteorologic inßuences on Plasmodium falciparum malaria in the highland tea estates of Kericho, Western Kenya. Emerg. Infect. Dis. 8: 1404 Ð1408. Sharpe, P.J.H., and D. W. DeMichele. 1977. Reaction kinetics poikilotherm development. J. Theor. Biol. 64: 649 Ð 670. Stratman-Thomas, W. K. 1940. The inßuence of temperature on Plasmodium vivax. Am. J. Trop. Med. 20: 703Ð715. Sutherst, R. W. 1993. Arthropods as disease vectors in a changing environment, pp. 124 Ð145. In J. V. Lake, G. R. Bock, and K. Ackrill [eds.], CIBA foundation symposium. Environmental Change and Human Health, Wiley, New York. Teklehaimanot, H. D., J. Schwartz, A. Teklehaimanot, and M. Lipsitch. 2004. Weather-based prediction of Plasmodium falciparum malaria in epidemic-prone regions of Ethiopia II. Weather-based prediction systems perform comparably to early detection systems in identifying times for interventions. Malaria J. 3: 44. TuTienpo.net. 2007. Historical weather. (http://www. tutiempo.net/en/Climate/Accra/654720.htm). Wernsdorfer, W. H., and S. I. McGregor (eds.) 1988. Malaria: principles and practice of malariology. Churchhill Livingstone, Edinburgh, United Kingdom. World Health Organization. 1986. Malaria control programme in Dar es Salaam and Tanga, Tanzania: WHO/ STC feasibility study to justify proposal for external assistance. Ministry of Health of Tanzania, Dar es Salaam, Tanzania. Wort, U. U., I. M. Hastings, A. Carlstedt, T. K. Mutabingwa, and B. J. Brabin. 2004. Impact of El Nin˜ o and malaria on birthweight in two areas of Tanzania with different malaria transmission patterns. Int. J. Epidemiol. 33: 1311Ð 1319. Received 4 March 2008; accepted 1 August 2008.