Recipient Organization
UNIV OF MINNESOTA
(N/A)
ST PAUL,MN 55108
Performing Department
Veterinary Biomedical Sciences
Non Technical Summary
Mosquito-borne arboviruses are a significant threat to global health. The Amazon rainforest contains the greatest number of mosquitoes and arboviruses on the planet.Due to extensive deforestation, Amazonia now consists of either: a) highly fragmented forests or b) large, mostly undisturbed Indigenous reserves. Our current knowledge of sylvatic mosquitoes comes from fragmented forests and therefore there is a major gap in knowledge on what mosquitoes and arboviruses are circulating in large, intact, rainforest environments.Our research will advance our knowledge of mosquito-borne disease in an area of the world that is highly understudied and is recognized as an arboviral hotspot.
Animal Health Component
(N/A)
Research Effort Categories
Basic
90%
Applied
(N/A)
Developmental
10%
Goals / Objectives
1) Evaluate the diversity and distribution of mosquitoes at edge habitats across an anthropogenically-induced disturbance gradient (intact rainforest, small-scale horticultural farm plots, Waiwai village, mining camp). 2)Characterize the presence and diversity of arboviruses along edge habitats. Our working hypothesis is that the presence and diversity of arboviruses will be altered by anthropogenic disturbance.
Project Methods
Experimental methods and designStudy site - The Waiwai are a group of Indigenous forager-horticulturalists who own and manage the KCOCA, a 625,000 ha Indigenous reserve in the far south of Guyana, South America (1°37'43.8"N 58°38'07.2"W). The KCOCA was granted to the Waiwai by the Guyanese government in 2004 and is the only formally protected reserve in the country under Indigenous control. This extremely remote region of Amazonia is one of the largest areas of contiguous rainforests in the world, exhibits high biodiversity and contains large populations of threatened, endemic species. The Waiwai have occupied the region that is now the KCOCA since at least the mid 19th century. Today, approximately 250 Waiwai live within the KCOCA, concentrated in Masakenari village.Characterize levels of disturbance in the KCOCA - For the purpose of this research, we will collect mosquitoes in four of the previously characterized land-use classes (see Preliminary Data section): 1) Masakenari village: the only permanent settlement within the KCOCA; 2) active agricultural plots: those cleared within the last 3 years and currently used by the Waiwai for agriculture; 3) mining camp: a semi-permanent gold-mining camp with relatively continuous occupation but a maximum of 30 workers; 4) undisturbed forest: forest with no evidence of past human modification.To quantify disturbance associated with each land-use class where mosquito sampling will be conducted, we will calculate Normalized Difference Vegetation Index (NDVI), mean canopy cover (%), and number of mature trees (greater than 30cm diameter at breast height). More intact vegetation has low red-light reflectance and high near-infrared reflectance, producing higher NDVI values. We will then classify NDVI values using an equal frequency classification as either low, medium, or high. We will use a spherical densiometer to measure canopy cover directly above where mosquitoes are collected at each site. We will record the number of mature trees (greater than 30cm diameter at breast height) present within a 5m radius of the collection sites.Mosquito collection and morphological identification - We will focus our mosquito sampling at the edges of each land-use class. The "edge" will be defined as the buffer region between a land-use class (the anthropic matrix) and natural forest, as distinguished on Landsat 8 imagery. Using ArcGIS 10.6.1, we will create a 100m buffer surrounding each land-use class (except for continuous, undisturbed forest). For mosquito collection, we will randomly select four sites within each habitat class and repeat sampling for four consecutive nights (for a total of 16 sampling nights). To generate an optimal assessment of the mosquito fauna present across the landscape of the study site, mosquitoes will be collected using a combination of trapping approaches. Due to their role in the maintenance and transmission of arboviruses, all collection efforts will preferentially target host-seeking female mosquitoes, though limited numbers of male mosquitoes and blood-engorged or gravid female mosquitoes may also be collected and retained for identification and analysis. At each sampling event, we will set up two CDC light traps and two BG traps: 1) BG-Sentinel 2 traps (Biogents, Regensburg, Germany) and 2) CDC light traps (John W. Hock Company, Gainesville, FL, USA). Both of these trap types are widely used for entomological surveillance in dipteran vectors. BG-Sentinel 2 traps will be baited with a combination of CO2 and human-scented 'BG-lure', and will be used to target anthropophilic mosquitoes. Traps will be set to run for approximately 24 hours before mosquitoes are collected. We will collect microclimate data at each collection site by recording temperature (°C) and relative humidity (%) at 30-minute intervals for the 24-hour duration of mosquito collection using Hygrochron iButton data loggers.Mosquitoes will be anesthetized using an ethyl acetate soaked swab and morphologically examined under a stereomicroscope equipped with a digital camera. Taxonomic field identifications will be based on standard morphological features and will be performed by an in-country entomologist from the University of Guyana. Voucher photos of relevant taxonomic features will be collected for mosquitoes of individual morphotypes. Mosquitoes will be grouped to the species or genus (if unable to speciate) level.As previous studies have shown the importance of calculating both number of species and their relative contribution to the community to adequately describe diversity, we will calculate species specific richness (S) and the Shannon Wiener diversity Index (H') for each habitat. We will assess the similarity between trapping locations habitats using the Morisita-Horn (IM-H) index for quantitative data and qualitative Jaccard (IJ). We will use Pielou's index (J') to determine evenness, the relative abundance of each species in the community. We will use logistic regression to assess the influence of NDVI, distance from the edge, canopy cover, mature tree density, and microclimate variables on the occurrence, and abundance and diversity of important vector species for important mosquito vector species.For arboviral screening (Aim 2), mosquitoes will be pooled into taxonomic groups of up to 20 and stored in Zymo DNA/RNA shield (Zymo Research, USA) until transported back to the Larsen/Wolf BSL- 2+ Laboratory for molecular analysis. DNA/RNA shield effectively renders infectious agents associated with biological samples non-infectious upon preservation, including viruses. Appropriate in-country permits to conduct the proposed research and export samples (i.e., Ministry of Amerindian Affairs, Guyana Environmental Protection Association, Guyana Wildlife Commission) and US import permits (USDA, USFWS) will be secured prior to the start of the study.RNA extraction and nanopore sequencing - Following methods developed by the Schroeder's Laboratory (Co-I) for screening of RNA viruses, pooled mosquito samples will be liquified using a gentleMACS dissociator (Miltenyi Biotec, Germany), and filtered to remove host and bacterial cellular debris. Total RNA will be extracted from filtrates using a Zymo Quick-DNA/RNA Pathogen MagBead Kit following manufacturer's instructions. RNA extractions will be quantified using a Qubit 4 Fluorometer (Invitrogen, USA). First strand cDNA synthesis will be performed using a template switching reverse transcriptase (New England Biolabs, United States) with N6 template switching random primers (Integrated DNA Technologies, USA). Amplification and second strand cDNA synthesis will be performed using a long-range, high fidelity PrimeSTAR HS polymerase (Takara, USA). Resulting double-stranded cDNA libraries will be sequenced using Oxford Nanopore Technologies (ONT) MinION sequencers and R10 flow cells, constructed with the rapid barcoding sequencing kit (SQK-RBK004). Basecalling will be performed in real-time using a GPU-configured version of the ONT basecalling software package guppy, using a super-accuracy base-calling model (r10.4.1_450bps_sup.cfg).Bioinformatic pipeline and phylogenetic analysis -ONT sequence data will be quality filtered using NanoFilt software and a minimum Q score of 8. Surviving sequences will be reference assembled using Epi2ME WIMP, used for rapid species identification and quantification of nanopore metagenomic samples. Consensus genomes will be aligned using available genome references from the NCBI genome database using Muscle (3.8) and visualized using FastTree (2.1) as implemented in Geneious Prime software. Nodal statistical support will be examined using both bootstrapping and bayesian methods. These data will then be incorporated into the four land use classes for preliminary characterization of abundance, diversity and distribution of arboviruses in the KCOCA.?