Near-infrared spectroscopy as a complementary age grading and species identification tool for African malaria vectors
© Sikulu et al; licensee BioMed Central Ltd. 2010
Received: 29 March 2010
Accepted: 4 June 2010
Published: 4 June 2010
Near-infrared spectroscopy (NIRS) was recently applied to age-grade and differentiate laboratory reared Anopheles gambiae sensu strico and Anopheles arabiensis sibling species of Anopheles gambiae sensu lato complex. In this study, we report further on the accuracy of this tool for simultaneously estimating the age class and differentiating the morphologically indistinguishable An. gambiae s.s. and An. arabiensis from semi-field releases and wild populations. Nine different ages (1, 3, 5, 7, 9, 11, 12, 14, 16 d) of An. arabiensis and eight different ages (1, 3, 5, 7, 9, 10, 11, 12 d) of An. gambiae s.s. maintained in 250 × 60 × 40 cm cages within a semi-field large-cage system and 105 wild-caught female An. gambiae s.l., were included in this study. NIRS classified female An. arabiensis and An. gambiae s.s. maintained in semi-field cages as <7 d old or ≥7 d old with 89% (n = 377) and 78% (n = 327) accuracy, respectively, and differentiated them with 89% (n = 704) accuracy. Wild caught An. gambiae s.l. were identified with 90% accuracy (n = 105) whereas their predicted ages were consistent with the expected mean chronological ages of the physiological age categories determined by dissections. These findings have importance for monitoring control programmes where reduction in the proportion of older mosquitoes that have the ability to transmit malaria is an important outcome.
Vector survival is recognised as one of the most imperative determinants of vector-borne pathogen transmission. For example, malaria vectors can only transmit malaria parasites when they are at least 10 days old because of the lengthy period required for Plasmodium parasite development in the mosquito . Traditionally, scientists relied upon observations of morphological changes in the reproductive system of female mosquitoes to estimate their physiological age [2–6] and to assess disease transmission potential [7, 8]. However, these well-established age measurement techniques are labour intensive and they engage highly skilled personnel. These disadvantages render the techniques unsuitable for assessing age distribution at an operational level in large scale, community-randomized trials. New tools are therefore required that can effectively, rapidly, and accurately assess the ages of large numbers of mosquitoes as existing priority intervention technologies are scaled up [9, 10] and new complementary approaches are developed and evaluated.
Furthermore, while accurate polymerase chain reaction methods do exist for differentiating sibling species such as those from the An. gambiae and Anopheles funestus complex in Africa [11–13], these methods are also somewhat laborious and expensive, limiting the numbers of mosquitoes which can be rigorously classified in most field studies. A convenient high throughput technique for simultaneously classifying and estimating the age of large numbers of mosquitoes would therefore enable biodemographic surveys of vector populations, and the impact of specific interventions upon them, on unprecedented scales.
Near-infrared spectroscopy (NIRS) uses the near-infrared region of the electromagnetic spectrum to quantitatively measure organic compounds e.g. O-H, N-H and C-O functional groups in biological samples. The spectrum collected is a result of the near-infra-red energy absorbed by a sample and is proportional to the amount of these functional groups present in samples. It is expected that a unique spectrum would be obtained for different age classes as well as different species of mosquitoes since it has been demonstrated that cuticular hydrocarbons change with age of mosquitoes  and that An. arabiensis have more water content in their body than An. gambiae s.s. . After calibrations have been developed, the technique is very simple, requiring very little training or expertise. Whole insects are placed below a fibre-optic probe, a spectrum collected, and the age and species predicted from stored calibrations. Advantages of this technique are that insects can be scanned non-destructively, no sample preparation is required, and results are obtained in a few seconds.
NIRS was recently applied successfully to age and distinguish laboratory reared An. gambiae s.s. from An. arabiensis. However, this study showed that additional data were needed to further develop calibrations and that additional field validation was needed. Herein we report on evaluation of the accuracy of this NIRS technique to estimate the ages and classify semi-field-maintained members of the An. gambiae s.l. complex, namely An. gambiae s.s. and An. arabiensis. We also further validated the accuracy of NIRS for differentiating wild caught An. gambiae s.s. from An. arabiensis and made a preliminary assessment of whether it might be useful for age grading wild-caught An. gambiae s.l.
An. gambiae s.s. (colony established in 1996 from Njage village, Kilombero, Tanzania) and An. arabiensis (colony established in 2007 from Sagamaganga village, Kilombero, Tanzania) were reared in the semi-field system established in the Ifakara Health Institute . To avoid variance in adult emergence rates and development characteristics arising from environmental differences, larvae and pupae of An. gambiae s.s. and An. arabiensis were reared under their usual rearing conditions in their respective laboratory and semi-field colonies. Pupae of both siblings were transferred into small cages measuring 40 × 40 × 30 cm for emergence. Adult mosquitoes were moved to larger netted cages measuring 250 × 60 × 40 cm at day 0 (within a day of emergence). Adult cages for both sibling species were maintained within the semi-field system. A total of four cages for each species, representing four different age cohorts were reared. In addition to paper cups lined with moist filter papers as oviposition sites, two clay pots lined with a black cloth were positioned inside each cage to provide cool resting sites for these mosquitoes. Adult females were blood-fed twice a week by inserting a human (volunteer) arm inside the cage each time for 15 minutes (Ethical clearance No. IHRDC/EC4/CL.N96/2004) and provided with fresh 10% glucose solution daily. Wild mosquitoes were collected from Njage village, Kilombero, Tanzania using CDC-Light traps over two consecutive nights.
Prior to scanning, mosquitoes were anesthetised using chloroform. At least 40 females of each sibling species in each age cohort were scanned using an ASD Lab Spec 5000 (Boulder, Colorado) NIR spectrometer. Nine ages (1, 3, 5, 7, 9, 11, 12, 14, 16 d) of An. arabiensis and eight ages (1, 3, 5, 7, 9, 10, 11, 12 d) of An. gambiae s.s. from the semi field system were scanned. 105 wild caught An. gambiae s.l. were scanned for age and the sibling species identified as soon as they were collected from the field.
The scanning protocol has previously been described elsewhere . Soon after the NIRS scans, 27% (n = 28) of the wild An. gambiae s.l. were dissected to determine their parity and ovarian development status . Polymerase Chain Reaction was used to validate the accuracy of NIRS for differentiating wild An. arabiensis from An. gambiae s.s. .
A calibration model developed from partial least squares regression cross-validation  was used to predict the age and differentiate semi-field reared and wild caught An. gambiae s.l. sibling species. This was achieved by using the semi-field reared data in the calibration models. The improved species identification model included the following ages: 1, 5, 7, 11, 12 d for An. arabiensis and. 1, 3, 5, 9, 10 d for An. gambiae s.s. For differentiating species, An. arabiensis was assigned a value of "1", and An. gambiae s.s. assigned a value of "2". Samples were then classed depending on whether they were predicted above or below 1.5 classification cut off point.
The cross-validation technique distinguished the two sibling species reared in the semi-field system with 90% accuracy (Figure 1C) and the rest of the species categories that were not included in the initial calibration model (An. arabiensis 3, 9, 14 d and An. gambiae s.s. 7, 11, 12 d) were classified with 88% accuracy.
The accuracy of NIRS for predicting the age and species of semi-field and wild An. arabiensis and An. gambiae s.s.
An. gambiae s.s.
Studies involving vector age assessments are crucial for determining the success or failure of any malaria control strategy that targets mosquito life span [19–22]. This study represents an extension of previous work  on laboratory and field specimens which indicated that additional data were needed to further develop calibrations and that further field validation was essential to verify the results. We have established and confirmed that NIRS can be applied to rapidly distinguish young from old An. gambiae s.s. and An. arabiensis and separate these morphologically similar species in areas where they do not occur sympatrically with other sibling species. Although NIRS was applied to age grade a relatively small sample of wild-caught specimens, insights were gained into the stability of the mosquito population age structure. Large-scale studies could potentially enable far more ambitious studies of malaria vector ecology, as well as the impact of control interventions upon vector bio-demography and transmission potential.
NIRS is non-destructive, rapid, and is associated with minimal sample processing costs after an initial outlay (approximately $40,000) for a NIR spectrometer. On the basis that NIRS can rapidly handle a large set of data with minimal labour and resources, the overall cost of using it on a large scale is dramatically reduced in the long run. Comparatively, the cost of cuticular hydrocarbon analysis depends largely on the accessibility of GC/FID or GC/MS instrumentation and, if the analysis is outsourced, can reach over US$50 per sample while the cost of transcriptional mosquito age grading has been estimated to be between $US7.5 to 10 per sample [23, 24]. Additionally, current age assessment tools are not conducive to the rapid assessment of mosquito population age structure on a large scale while standard PCR  and multiplex PCR [12, 13] for differentiating morphologically indistinguishable species is costly and time consuming. In terms of speed, NIRS was more than 10 times faster than parity dissections to determine physiological age. It required less than 15 seconds to scan one mosquito using NIRS and the spectrum obtained was analysed for both age and sibling species identification. Additionally, only one out of four technicians involved in this study could perform dissections to determine parity and these required approximately 3-5 minutes per mosquito. PCR reactions to determine sibling species required 2 people who took on average 5 hours working on 50 samples. Comparatively, it took less than 30 minutes to train 2 people to operate NIRS.
Although this study provides grounds for optimism, further work is clearly needed. Refinements to the method, such as the capacity to perform NIRS on preserved samples will practically facilitate large studies particularly in areas where field specimens must be preserved for future analysis. While there is a need for further age grading methods to determine population age structure on a finer scale, the value of NIRS is in the unique capacity to rapidly identify changes to mosquito population demography. Further studies are required to determine the capacity of this tool to differentiate and age grade other morphologically indistinguishable species in the An. gambiae complex and mosquitoes in other genera.
We thank the insectary staff at IHI; Paulina Kasanga, and Ally Daraja for rearing mosquitoes and Hassan Mtambala, Peter Pazia, Daniel Lugiko, Nuru Nchimbi and Japheth Kihonda for their technical assistance. We thank residents of Njage village for allowing us to trap mosquitoes in their houses. We acknowledge the Griffith Medical Research College and the Science, Engineering, Environment and Technology group, Griffith University, for providing a PhD scholarship to MS. This study was supported by the Bill & Melinda Gates Foundation (awards 51431 and 45114) as well as a Research Career Development Fellowship (076806) provided to GFK by the Wellcome Trust and IAEA fellowship funding provided to FED.
- Beier CJ: Malaria parasite development in mosquitoes. Annu Rev Entomol. 1998, 43: 519-543. 10.1146/annurev.ento.43.1.519.View ArticlePubMed
- Detinova T: Age-grouping methods in Diptera of medical importance, with special reference to some vectors of malaria. Monogr Ser World Health Organ. 1962, 47: 13-191.PubMed
- Polovodova VP: Age changes in ovaries of Anopheles and methods of determination of age composition in mosquito population. Med Parazit (Mosk). 1941, 10: 387-
- Gillies M, Wilkes T: A study of the age composition of Anopheles gambiae Giles and A. funestus Giles in north-eastern Tanzania. Bull Entomol Res. 1965, 56: 237-262. 10.1017/S0007485300056339.View ArticlePubMed
- Gillies M: A modified technique for the age-grading of populations of Anopheles gambiae. Ann Trop Med Parasit. 1958, 52: 261-273.PubMed
- Gillies M: Studies on the dispersion and survival of Anopheles gambiae in East Africa, by means of marking and release experiments. Bull Entomol Res. 1961, 52: 99-127. 10.1017/S0007485300055309.View Article
- Garret-Jones C: Prognosis for interuption of malaria transmission through assessment of the mosquito's vectorial capacity. Nature. 1964, 204: 1173-1175. 10.1038/2041173a0.View Article
- Dye C: The analysis of parasite transmission by bloodsucking Insects. Annu Rev Entomol. 1992, 37: 1-19. 10.1146/annurev.en.37.010192.000245.View ArticlePubMed
- Sharp BL, Kleinschmidt I, Streat E, Maharaj R, Barnes KI, Durrheim DN, Ridl FC, Morris N, Seocharan I, Kunene S, La grange JJP, Mthembu JD, Maartens F, Martin CL, Barreto A: Seven years of regional malaria control collaboration - Mozambique, South Africa, and Swaziland. Am J Trop Med Hyg. 2007, 76: 42-47.PubMed CentralPubMed
- Noor AM, Mutheu JJ, Tatem AJ, Hay SI, Snow RW: Insecticide-treated net coverage in Africa: mapping progress in 2000-07. The Lancet. 2009, 373: 58-67. 10.1016/S0140-6736(08)61596-2.View Article
- Scott J, Brogdon W, Collins F: Identification of single specimens of the Anopheles gambiae complex by the polymerase chain reaction. Am J Trop Med Hyg. 1993, 49: 520-529.PubMed
- Vezenegho S, Bass C, Puinean M, Williamson M, Field L, Coetzee M, Koekemoer L: Development of multiplex real-time PCR assays for identification of members of the Anopheles funestus species group. Malar J. 2009, 8: 282-10.1186/1475-2875-8-282.PubMed CentralView ArticlePubMed
- Bass C, Williamson M, Wilding C, Donnelly M, Field L: Identification of the main malaria vectors in the Anopheles gambiae species complex using a TaqMan real-time PCR assay. Malar J. 2007, 6: 155-10.1186/1475-2875-6-155.PubMed CentralView ArticlePubMed
- Caputo B, Dani RF, Horne LG, Petrarca V, Turillazzi S, Coluzzi M, Priestman AA, Torre A: Identification and composition of cuticular hydrocarbons of the major Afrotropical malaria vector Anopheles gambiae s.s. (Diptera: Culicidae): analysis of sexual dimorphism and age-related changes. J Mass Spectrom. 2005, 40: 1595-1604. 10.1002/jms.961.View ArticlePubMed
- Gray EM, Bradley T: Physiology of dessication resistance in Anopheles gambiae and Anopheles arabiensis. Am J Trop Med Hyg. 2005, 73: 553-559.PubMed
- Mayagaya VS, Michel K, Benedict MQ, Killeen GF, Wirtz RA, Ferguson HM, Dowell FE: Non-destructive determination of age and species of Anopheles gambiae s.l. using Near-infrared spectroscopy. Am J Trop Med Hyg. 2009, 81: 622-630. 10.4269/ajtmh.2009.09-0192.View ArticlePubMed
- Ferguson H, Ng'habi K, Walder T, Kadungula D, Moore S, Lyimo I, Russell T, Urassa H, Mshinda H, Killeen G, Knols B: Establishment of a large semi-field system for experimental study of African malaria vector ecology and control in Tanzania. Malar J. 2008, 7: 158-10.1186/1475-2875-7-158.PubMed CentralView ArticlePubMed
- Scott JA, Brogdon WG, Collins FH: Identification of single specimens of the Anopheles gambiae complex by the polymerase chain reaction. Am J Trop Med Hyg. 1993, 49: 520-529.PubMed
- Scholte EJ, Knols BGJ, Samson RA, Takken W: Entomopathogenic fungi for mosquito control: A review. J Insect Sci. 2004, 4: 19-PubMed CentralView ArticlePubMed
- Scholte EJ, Knols BGJ, Takken W: Infection of the malaria mosquito Anopheles gambiae with the entomopathogenic fungus Metarhizium anisopliae reduces blood feeding and fecundity. J Invertebr Pathol. 2005, 91: 43-49. 10.1016/j.jip.2005.10.006.View ArticlePubMed
- Blanford S, Chan BHK, Jenkins N, Sim D, Turner RJ, Read AF, Thomas MB: Fungal pathogen reduces potential for malaria transmission. Science. 2005, 308: 1638-1641. 10.1126/science.1108423.View ArticlePubMed
- Brownstein JS, Hett E, O'Neill SL: The potential of virulent Wolbachia to modulate disease transmission by insects. J Invertebr Pathol. 2003, 84: 24-29. 10.1016/S0022-2011(03)00082-X.View ArticlePubMed
- Hugo EL, Peter EC, Petrina HJ, Luke PR, Brian HK, Peter AR, Scott AR, Scott LON: Field Validation of a Transcriptional Assay for the Prediction of Age of Uncaged Aedes aegypti Mosquitoes in Northern Australia. PLoS Negl Trop Dis. 2010, 4: e608-10.1371/journal.pntd.0000608.PubMed CentralView ArticlePubMed
- Cook PE, Hugo LE, Iturbe-Ormaetxe I, Williams CR, Chenoweth SF, Ritchie SA, Ryan PA, Kay BH, Blows MW, O'Neill SL: Predicting the age of mosquitoes using transcriptional profiles. Nat Protoc. 2007, 2: 2796-2806. 10.1038/nprot.2007.396.View ArticlePubMed
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.