Genetic evidence for the importance of interrupted feeding by mosquitoes in the transmission of malaria

454 T~N~ACTICJN~OF THE ROYAL SOCIETYOF TROPICALMEDICINEAND HYGIENE

Genetic evidence the transmission

for the importance of malaria

David J. Conway’ and Jana S. McBride Edinburgh,

King’s

Buildings,

West Mains

(1991)

of interrupted

85,

454-456

feeding

Institute of Cell, Animal, and Population Road, Edinburgh, EH3 9JT, UK

Abstract

Plasmodium falciparum isolateswere obtainedfrom 17 Dab of Gambianchildren. eachoair living in the

saniehouseand presentingwith malariaat &e same time. Frequenciesof allelic serotypesof 3 polymorphic blood stageproteins (MSPl, MSPZ, and Exp-1) were previously determinedfrom a large number of isolates from patients in the local area, and the probability of a random pair of isolatescontainingan identical genotypewascalculatedto be lessthan 0’01. However. 3 of 8 householdlairs in onevear. and 6 of 9 in the next year, containid identical p. falciparum genotypes, a much higher frequency than expected randomly (P
In nature, some blood meals are obtained from more than one individual, due to interrupted feeding (BOREHAM& GARRETT-JONES, 1973).Moreover, if a mosquito.fails to locate a blood capillary initially, it may attempt to feed on another individual nearby, and thereby infect more than one individual (ROSS& NOL & MACKAY-R• SSIGNOL. 1988). However. the effectsof repeatedprobing an&or interrupted fe&ling on the natural transmissionof malariahave not been investigated. There are inherent difficulties in correlating inoculation and incidencerates bv conventionalmeansin endemicareas(PULL & GRUB, 1974). Therefore, it would be verv difficult to estimatedirectlv the effects on malariat&smission of repeatedmosq&o probing and/or interrupted blood feeding. A new approachto the problem involves the characterizationof Plasmodium falciparum parasitegenotypesfrom individuals with malaria. A single mosquito infecting 2 people would be expected to inoculate identical parasite genotypesinto both of them. If this occurred frequently, in an area where the malaria vector bites indoors at night, patients from the same house ‘Presentaddress: Department of Biology,ImperialCollege, London, SW72BB, UK.

by mosquitoes Biology,

University

in of

presentingwith malaria at the sametime would be expected to shareidentical P. falciparum genotypes moreoften than thoseliving in different housesin the samearea. Materials and Methods

Practical

methods

The study was conducted in an urban/periurban areaof The Gambia,West Africa, whereP. falciparurn malariais trans&ted by Anojkles gambiae s&w lato. which bites mainlv between 2200 and 0400 h (HOLSTEIN,1954). Male&a transmissionis seasonal and hyperendemic,reachinga peak betweenAugust and November. During July-December 1988and October-December 1989P. falcibarum isolateswere collected from 337 malaria patients presenting to the out-patients departmentsof the Medical ResearchCouncil, Fajara, and the Royal Victoria Hospital, Banjul (220 isolates in 1988, 117in 1989).Among thesepatientswere 17 pairs of siblingspresentingtogether. The parentsof 11 of the pairs were questionedabout where their children sleptin the house.In all cases,both children slept in the sameroom and, in all except onecase,in the samebed. All the patients lived in the same urban/periurban area (CONWAY & MCBRIDE, 1991). Heoarinizedsamolesof parasitizedbloodfrom each patient were culttired from 2u8 h until mature schizonts were obtained. Multi-sDot slidesof schizonts were prepared (MCBRIDE&et al., 1984), for analysis of P. falciparum genotypes by indirect immunofluorescence using a panel of 27 monoclonal antibodies(aslisted by CONWAY&MCBRIDE., 1991), recognizing allelic variants of 3 polymorphic blood stageproteins: (i) the precursorto the major merozoite surface proteins (MSPl, PMMSA, or gp195: HOLDER, 1988), (ii) a second merozoite surface protein (MSPZ, MSA-2, or gp35-56: FENTON et al., 1991),and (iii) an exported protein (Exp-1, CRA, or ~23: SIMMONS et al., 1987). These proteins are encodedby geneson different chromosomes (KEMP er al., 1987),which segregate at meiosisin the mosquito vector (WALLIKER et al., 1987). At least 200 schizontsof each isolatewere typed with each monoclonal antibody by indirect immunofluorescence(CONWAY& MCBRIDE,1991). A method of two-colour differential labelling wasused to resolve the majority parasiteclone in any isolate containing more than one recognizableclone (CONWAY et al., 1991). Only the majority clone in each isolate was included in the statistical analyses. Statistical

methods

Among the 337isolates,there were 36 distinguishableallelicserotypesof MSPl, 8 types of MSP2, and2

455

types of Exp-1. From observedallelic frequenciesat each locus, the expected frequencies of different three-locusgenotypes(i.e., combinationsof different , Probability I

a

children is given in the Table. The Figure showsthe expected numbers of pairs containing identical genotypesas probability distribution histogramsfor Table.Three-locus gonotypts of P. fkiparum isolated from pairs of cbikhmwhosleptin the samehouse Pair no.

0.6

1988 1

0.6

3 0.2

4

n

D II 0

,

1

2

No.

+ a of identical

4

6

6

gonotyps

7

6

5

palm

6

_ Probability 1,

b 0.60.6-

1989 1

0.4 -

0.2 -

Sex’

AN MN

sJ *J AT ST FC PC BM KM SF MF

I 0

3

2

3

of identical

4

5

genotype

6

7

a

NN SN

9

pairs

Figure. Observed (arrows) versus expected (bars) numbers of identical P. falcipanm three-locus genotype pairs, among pairs of children sleeping in the same house. (a) Eight household pairs in 1988 (expectation: binomial distribution ofP=0.00802 and n=8); (b) nine household pairs in 1989 (expectation: binomial distribution of P=O.O0898 and n=9). Details of statistics are given in the Methods SXtiOtl.

MSPl, MSPZ, and Exp-1 types) were calculatedfor 1988 and 1989 independently, assumingrandom outbreedingin the parasitepopulation. The observed numbers of different three-locusgenotypes (126 of 220 sampledin 1988,and 75 of 117sampledin 1989) were in accordancewith the expectationsof random assortmentamong the loci (CONWAY & MCBRIDE, 1991). The probability (P) that a randomly selectedpair of isolateswould contain an identical three-locusgenotype was calculatedas the sumof the squaresof the individual expectedthree-locusgenotypefrequencies (P=O*OO802 for 1988,and P=O*OO898for 1989:data from CONWAY MCBRIDE, 1991). The expected number of householdpairs of isolatescontaining an identical P. falciparum three-locus genotype was calculatedas the binomial distribution of the above probability valuesfor the total number of household pairs (8 in 1988, 9 in 1989). Results

The three-locusgenotype of the majority P. falciclone within each of the household-paired

5

6 7

SD JD YS MS AC JC

8

KB

9

OB SC AC

3 1 5 2 4
Genotypeb H-8-2 28-S-l 10-g-1, 10-8-l 1542 52-2-2 52-2-2, 52-2-2 28-2-2 32-2-2 20-2-l 11-1-2 11-7-2 12-2-2 12-2-2, 12-2-2

:; MS AS AB MB

BM TM No.

parum

(y*Z)

2

0.4

O-

Initials

M M M M F M F M F F ii M

F M M

F F

20-2-2 11-7-2 U-2-2* 15-2-2 7-l-1, 7-l-l 10-7-l 4-2-l 28-2-2 28-2-2* 52-7-2 41-7-2 8-8-1, 8-8-l 40-2-2, 40-2-2 16-6-2, 16-6-2

Sleeping -gementc NI NI NI NI Same Same

NI

Different Same Same Sm.2 Same SStlle SZ3tlle

Same NI

*F=female, M=male. bAllelic serotypes of proteins MSPl, MSP2 and Expl (see CONWAY & MCBRIDE, 1991). Asterisks (*) indicate pairs which shared an identical P. fdcipamm genotype. ‘Same= same bed, Different=different beds in same room, NI=no information.

eachyear. In both years, the expectednumber was0 or 1. In 1988, 3 of 8 pairs contained identical genotypes.This washigher than expectedrandomly (binomial distribution P
Approximately 50% of pairs of children with malaria from the same household contained an identical three-locusP. falciparum genotype, compared with approximately 1% of pairs randomly chosenfrom the study area as a whole. This exceptionally high frequency of genetically

456 identical parasite pairs among patients from the same household could be explained in one of 2 ways: (i) single mosquitoes frequently infect 2 or more people on the same occasion, due to interrupted feeding and/or repeated probing, or (ii) extreme clustering of mosquitoes carrying identical P. falciparum genotypes exists within households, due to a lack of dispersal during the whole period of development of the parasite in the mosquito. There is no evidence for the second explanation. An. gambiae s.Z. are not likely to remain clustered in a household during the 10-14 d development period of the parasite, as they lay eggs at outdoor breeding sites approximately every 4 d (HOLSTEIN, 1954). Accordingly, no spatial clustering of P. falciparum genotypes was observed in the study area (CONWAY & IGBRIDE, 1991). Therefore, it is very likely that single mosquitoes are frequently responsible for infecting more than one individual with malaria. Haptoglobin typing of blood meals revealed that 3-10% of human blood meals taken by An. gambiae s.1. in Africa were acquired from 2 or more individuals (BOREHAM et al., 1979; PORT et al., 1980). Here, the: observed frequency of‘ genetic identity between household paired isolates was approximately SO%, suggesting that mosquito probing activity before feeding also contributed significantly to malaria transmission. Field studies__of pre-feeding probing behaviour by malaria -vectors have not been attempted, and until now the effects on malaria transmission have not been demonstrated. There is probably great diversity in both mosquito probing behaviour and interrupted blood feeding, depending on factors such as host irritability (L~NAF~AN & BOREHAM, 1976) as well as salivarv gland oatholoev (ROSSIGNOL et al.. 1984), and &erifore such beKaviour might be difficult & assess precisely. Mosquito pre-feeding probing behaviour will tend to increase malaria transmission, as a larger number of people receive infective bites. The results presented in this paper strongly suggest that the effects of such behaviour are important, and therefore that present models of malaria transmission underestimate the vectorial capacity of mosquito populations. Since a precise estimation of vectorial capacity may not be possible, a more realistic goal of mathematical models is the prediction of relative reduction in vectorial capacity under a given control effort (DYE, 1990). Acknowledgements

We are grateful to workers at the Medical Research Council laboratories in the Gambia, and particularly fo Drs Brian Greenwood, Hilton Whittle, Andrew Wilkins, Dominic Kwiatkowski, Harry Campbell and Adrian Hill, for making the work possible. We thank Dr Peter Keighley and Professor Bill Hill for statistical advice. This investigation received 6nancial support from the Medical Research Council, The Wellcome Trust, and the UNDP/World Bar&WHO Special Programme for Research and Training in Tropical Diseases. References

Aron, J. L. & May, R. M. (1982). The population dynamics of malaria. In: Population Dynamics of Infectious Diseases, Anderson, R. M. (editor). London: Chapman and Hall, pp. 139-179. Boreham, P. F. L. &Garrett-Jones, C. (1973). Prevalence of

mixed blood meals and double feeding in a malaria vector (Anopheles sacharovi Favre). Bulletin of the World Health Organizatidn,

48, 605-614.

Boreham, P. F. L., Lenahan, J. K., Boulzaquet, R., Storey, I.. Ashkar. T. S.. Nambier. R. & Matsushima. T. 71579). St&es on multiple feeding by Anopheles gambiae s.1. in Garki F$yy20ciery

district, northern Nigeria. of Tropical Medicine

Transactions of the and Hygiene, 73,

Boyd, M. F: (1949). Epidemiology of malaria: factors related to the definitive host. In: Malariology, vol. 1, Boyd, M. F. (editor). Philadelphia: W. B. Saunders, pp. 608-697. Conway, D. J. & McBride, J. S. (1991). Population genetics of Plasmodium falciparum within a malaria hyperendemic area. Parasitology, 103, in press. Conwav. D. I.. Greenwood. B. M. & McBride. 1. S. (1991). Th; &pid&iology of m;ltiple-clone Plasm&urn filcipahurn infections in Gambian patients. Parasitology, 103, in press. Dye, C. (1990). Epidemiological significance of vectorparasite interactions. Parasitology, 101, 409-415. Fenton, B., Clark, J. T., Khan, C. M. A., Robinson, V. J., Walliker, D., Ridley, R., Scaife, J. G. & McBride, J. S. (1991). Structural and antigenic polymorphism of the 35 fo 48-kilodalton merozoite surface antigen CMSA-2) of the malaria parasite Plasmodium falcipkm: Molechlar and Cellular Biology, 11, 963-971. Holder, A. A. (1988). The precursor to major merozoite surface antigens: structure and role in immunity. Progress in Allergy, 41, 72-97. Holstein, M. H. (1954). Biology of Anopheles gambiae: research in French West Africa. Geneva: World Health Organization. Kemp, D. J., Thompson, J. K., WaIliker, D. & Corcoran, L. M. (1987X Molecular karvotme of Plasmodium falciparuk conserved linkage iroil& and expendable histidine-rich protein genes. Proceedings of the National Academy of Sciences of the USA, 84, 7672-7676. Lenahan, J. K. & Boreham, P. F. L. (1976). Effect of host movement on multiple feeding by Aedes aegypti (L.) (Diptera: Culicidae) in a laboratory experiment. Bulletin of Entomological Research, 66, 681-684. McBride, J. S., Welsby, P. D. & Walliker, D. (1984). Serotyping of Plasmodium falciparum from acute human infections using monoclonal antibodies. Transactions of the Royal Society of Tropical Medicine and Hygiene, 78, 32-34. Port, G. R., Boreham, P. F. L. & Bryan, J. H. (1980). The relationship of host size to feeding by mosquitoes of the Anopheles gambiae Giles complex (Diptera: Culicidae). Bulletin of Entomological Research, 70, 133-144. Pull, J. H. & Grab, B. (1974). A sunple epidemiological model for evaluating the malaria inoculation rate and the risk of infection in infants. Bulletin of the World Health Organization, 51, 507-516. Rossignol, P. A. & Mackay-Rossignol, A. (1988). Simulations of enhanced malaria transmission and host bias induced by modified vector blood location behaviour. Parasitology, 97, 363-372. Rossignol, P. A., Ribeiro, J. M. C. 81 Spielman, A. (1984). Increased intradermal probing time in sporozoite-infected mosquitoes. AmericanJournal of Tropical Medicine and Hygiene, 33, 17-20. Simmons, D,., Woollett, G., Bergin-Cartwright, M., Kay, D. & Scafe, J. (1987). A malaria protein exported into a new compartment within the host ervthrocvte. _ - EMBO Journal, b, 485-491. Walliker, D., Quakyi, I. A., Wellems, T. E., McCutchan, T. F., Szarfman, A., London, W. T., Corcoran, L. M., Burkot. T. R. & Carter. R. (1987). Genetic analvsis of the human malaria p&as& Plkmodium falciiarum. Science, 236, 1661-1666. Received 25 February March 1991

1991;

accepted

for publication

22