Understanding the distribution of mosquito species is an important element of surveillance. This is especially true in Florida, where detections of nonnative mosquitoes have been increasing. Collier Mosquito Control District performs routine adult mosquito surveillance for operational purposes throughout the year. Here, we report records for 3 species collected in 2021 that had not been documented previously in Collier County, FL: Aedes tortilis, Culex declarator, and Cx. tarsalis. Specimens were initially identified based on morphology, then each species was confirmed by comparing the cytochrome c oxidase subunit I gene sequences to those of other related mosquito species. Although Ae. tortilis and Cx. declarator were collected at multiple sites, Cx. tarsalis was collected only once, making it unclear whether this species has established a permanent population within the county.
A number of new mosquito species have been reported in Florida during the last decade, including Aedes pertinax Grabham (Shroyer et al. 2015), Culex interrogator Dyar and Knab (Shin et al. 2016), Cx. panocossa Dyar (Blosser and Burkett-Cadena 2017), Ae. japonicus japonicus (Theobald) (Riles et al. 2017), and Aedeomyia squamipennis (Lynch Arribálzaga) (Burkett-Cadena et al. 2017); and the rate at which new species are being detected in the state appears to be increasing (Reeves et al. 2023). Potential factors contributing to this influx include increases in global connectivity due to escalated shipping and air travel, shifts in climate and/or landscape composition that have facilitated mosquito migration into and establishment within previously inhospitable areas, and adaptation of invasive species to urban environments (Lounibos 2002, Shin et al. 2016, Multini et al. 2019, Wilke et al. 2021). At the same time, improvements in mosquito surveillance coverage, trap technology, and molecular tools for species identification have increased researchers' ability to detect mosquitoes that may already have been present within the state, but either had not been collected or had been misidentified (Kröckel et al. 2006, Shroyer et al. 2015, Beebe 2018, Reeves et al. 2021).
Understanding the composition and distribution of various mosquito species is of vital importance for not only ecological researchers but also mosquito control and public health agencies. For invasive species, early detection provides an opportunity for rapid intervention, potentially slowing or preventing the spread of the species into new areas (Reaser et al. 2020). For established but cryptic species, correct identification can have significant impacts on predictive distribution mapping and disease risk assessment. For example, Ae. triseriatus (Say) and Ae. hendersoni Cockerell are found in similar habitats and are difficult to separate morphologically as adults (Reno et al. 2000, Wilson et al. 2014). Although Ae. triseriatus is considered a significant vector of LaCrosse virus, Ae. hendersoni is not (Watts et al. 1975, Grimstad et al. 1985), so misidentification can lead to inaccurate assessments of LaCrosse encephalitis risk potential.
Collier County (Florida, USA) is located in Southwest Florida along the Gulf of Mexico. Its subtropical climate, broad range of habitats, and extensive network of conservation lands make it an ideal environment for a diverse mosquito fauna. It is also a tourist destination with high numbers of seasonal residents that commute from other parts of the country, creating abundant opportunities for the introduction of new, invasive species. As of 2020, 46 mosquito species had been identified within the county (Darsie and Morris 2003, Connelly et al. 2016, Riles and Connelly 2020). Here we report 3 additional species detected in 2021: Ae. tortilis (Theobald), Cx. declarator Dyar and Knab, and Cx. tarsalis Coquillett.
MATERIALS AND METHODS
Surveillance and visual identification
All specimens were initially collected by the Collier Mosquito Control District (the District) during routine surveillance in the city of Marco Island and surrounding preserve lands from January 7, 2021, to December 14, 2021. They were collected from 2 trap types: Centers for Disease Control and Prevention (CDC) Miniature Light Traps (CDC light traps; BioQuip Inc., Rancho Dominguez, CA) and BG-Sentinel 1 Traps (BG-Sentinels; BioGents AG, Regensburg, Germany). The CDC light traps were equipped with a 0.95-W incandescent lightbulb, baited with compressed carbon dioxide that was dispensed at a rate of 500 ml/min, and deployed at 20 designated sites once per week. BG-Sentinels were baited with compressed carbon dioxide and BG Lure Mosquito Attractant (BioGents AG). These traps were deployed at 8 designated sites once per week and at additional temporary sites on an ad hoc basis.
All traps were placed in the field and retrieved approximately 24 h later. Captured mosquitoes were preserved at −80°C, then visually identified to species level using a dissecting microscope and regional identification keys (Darsie and Morris 2003, Darsie and Ward 2005, Burkett-Cadena 2013). A leg from at least one specimen of each species was preserved in 70% ethanol and shipped to the Florida Medical Entomology Laboratory for secondary confirmation via DNA barcoding analysis of the cytochrome c oxidase subunit I (COI) gene (Hebert et al. 2003).
Molecular procedures followed those described in Reeves et al. (2021). Briefly, for each specimen, DNA was extracted from a single leg using Zymo Quick-DNA™ Miniprep Plus Kits (Zymo Research, Irvine, CA). Extraction buffer was added to each tube, and the leg was macerated with a sterile plastic pestle for approximately 3 min. Subsequent steps followed the manufacturer protocol with an overnight incubation at 56°C. Extracted DNA was used as a template in polymerase chain reactions (PCR) to amplify a 658-bp fragment of the barcoding region of the COI gene. The final volume of each reaction was 20 μl and consisted of 10 μl 2× Apex Taq RED Master Mix (Genesee Scientific Corp., San Diego, CA), 2 μl of 10 μM forward primer LepF1 (5′-ATT CAA CCA ATC ATA AAG ATA T-3′; Hebert et al. 2004), 2 μl of 10 μM reverse primer LepR1 (5′-TAA ACT TCT GGA TGT CCA AAA A-3′; Hebert et al. 2004), 1 μL extracted DNA, and 5 μL nuclease-free water. Reactions were performed in a Bio-Rad Thermocycler (Bio-Rad Laboratories, Hercules, CA) programmed to 94°C for 1 min; 5 cycles of 94°C for 30 sec, 45°C for 40 sec, and 72°C for 1 min; 35 cycles of 94°C for 30 sec, 51°C for 40 sec, and 72°C for 1 min; and a final extension step of 72°C for 10 min. After PCR, 7 μL of each PCR product were stained and electrophoresed on a 1.5% agarose gel for 45 min at 50 V and examined under a transilluminator to ensure an amplicon of the expected size was produced.
The remaining volume of each resulting PCR product was sent to Eurofins Genomics (Louisville, KY) for amplicon sequencing in one direction via chain-termination sequencing (Sanger et al. 1977). Sequence files were examined and edited for quality in the bioinformatic software Geneious Prime Version 11.0.6 (Biomatters, Inc.; Boston, MA) to remove ambiguous base calls. Edited sequences were submitted to the Barcode of Life Data System (BOLD) v. 4 Identification Engine (Ratnasingham and Hebert 2007), but because of issues with misidentified reference material within publicly accessible sequence databases (BOLD and the National Center for Biotechnology Information GenBank), we constructed a neighbor-joining tree to visualize the similarity of the sequences derived from morphologically identified Ae. tortilis, Cx. declarator, and Cx. tarsalis to those of other congeneric Florida species collected and identified by Lawrence Reeves (LER). Pinned voucher specimens of all species have been preserved as part of the District's reference collection.
On April 21, 2021, 4 female Ae. tortilis specimens were collected from a permanent CDC light trap located at the southern end of the District (lat 25.934, long −81.671; Fig. 1). The site abutted a small residential development, nestled within healthy mangrove swamps and damaged mangrove habitat that is undergoing restoration in the Rookery Bay National Estuarine Research Reserve (Radabaugh et al. 2017). Along with Ae. tortilis, the trap also contained specimens of Ae. taeniorhynchus (Wiedemann), Ae. infirmatus Dyar and Knab, Anopheles atropos Dyar and Knab, and Cx. nigripalpus Theobald. The DNA barcoding confirmed the identity of the specimen, though in the BOLD Identification Engine and in comparisons to other Florida Aedes COI sequences (Fig. 2), this sequence could not be distinguished from those of specimens identified as Ae. condolescens Dyar and Knab.
On July 20, 2021, 56 female Cx. declarator specimens were collected from a CDC light trap located at the same site where Ae. tortilis was initially detected (Fig. 1). The majority of the other mosquitoes in the collection were Cx. nigripalpus and Ae. taeniorhynchus, but there were also small numbers of Ae. infirmatus, An. atropos, Cx. coronator Dyar and Knab, and various Culex (Melanoconion) species. The barcoding region of the COI gene was sequenced for 3 specimens captured from CDC light traps elsewhere in the District later that year: 2 specimens collected July 27, 2021 (lat 26.046, long −81.708), and 1 specimen collected November 8, 2021 (lat 25.991, long −81.591). For this species, morphological identifications could not be resolved using the BOLD Identification Engine as the sequences were >99% similar to reference sequences labeled as Cx. nigripalpus, Cx. erraticus (Dyar and Knab), Cx. pipiens Linnaeus, Cx. taeniopus Dyar and Knab, Cx. declarator, and several others. To resolve the question of species identity, the Collier County Cx. declarator sequences were compared to COI sequences for all other Florida Culex (Culex) species collected and identified by LER. The Collier County sequences were similar (approximately 98%) to Cx. nigripalpus and Cx. interrogator sequences, but were most similar (99.8–99.82%) to Cx. declarator sequences collected in Indian River and Charlotte counties. In the neighbor-joining tree (Fig. 3), specimens identified as Cx. declarator formed a clade sister to Cx. nigripalpus.
On November 3, 2021, a single adult female Cx. tarsalis was collected from a BG-Sentinel trap alongside small numbers of Ae. aegypti (L.), Cx. nigripalpus, and Cx. quinquefasciatus Say. Unlike the CDC trap site described above, this was a temporary site used for arbovirus surveillance. It was located in a suburban residential area at the southeastern tip of Marco Island (lat 25.913, long −81.726; Fig. 1). Mosquitoes were collected on a daily basis rather than once per week, and the trap was only active from October 28 to November 5, 2021. Morphological identification of the Cx. tarsalis specimen was confirmed by DNA barcoding. The COI sequence was reliably identified by the BOLD Identification Engine, exhibiting >99% similarity to reference specimens labeled as Cx. tarsalis. When compared with sequences derived from collections by LER of all other US species except Cx. chidesteri Dyar, the sequence was 98.91–99.27% similar to Cx. tarsalis specimens collected in Arizona and Texas. In the neighbor-joining tree (Fig. 3), the sequence from the Collier County specimen formed a clade along with Arizona and Texas Cx. tarsalis, sister to Cx. restuans Theobald + Cx. stigmatosoma Dyar + Cx. thriambus Dyar.
Aedes tortilis was initially identified based on its unusual combination of basal pale bands on the abdominal terga and dark, unbanded hind tarsomeres (Fig. 4A), which distinguish it from other Aedes species in Collier County (Darsie and Morris 2003). Aedes tortilis is a Caribbean species that is found in coastal regions of southern Florida, especially along the Atlantic Coast (Arnell 1976, Darsie and Ward 2005). This species was originally recorded in the USA in 1945 in Key West (Staebler and Buren 1946), and by 2003, it had been identified in Indian River County, St. Lucie County, and much of South Florida (Darsie and Morris 2003). Between April and October of 2021, a total of 92 specimens were collected from 8 separate locations across the District (Fig. 1), the majority (62) of which came from the original trap location.
Phylogenetic analysis placed the Collier Ae. tortilis specimen in a mixed clade with other Ae. tortilis and Ae. condolescens from Broward and Miami-Dade counties. Reeves et al. (2021) found similar results when comparing the COI and ribosomal internal transcribed spacer 2 sequences from 7 Ae. tortilis and Ae. condolescens specimens, going so far as to suggest that they may in fact represent a single species. The main morphological difference between the two is scutal patterning: Ae. condolescens has a broad, pale patch of scales on the anterior scutum, while Ae. tortilis does not (Reeves et al. 2021). Unfortunately, the specimens retrieved from the District's traps were usually missing most if not all of their scutal scales, making it difficult to assess this feature. The Collier specimen described here was identified as Ae. tortilis because it did not appear to have any remnants of pale scales on the anterior of its scutum, but further investigation will be required to determine whether mosquitoes with the Ae. condolescens phenotype also occur in Collier County.
The next species, Cx. declarator, has pale-scaled abdominal sterna, a dark proboscis with a pale patch on the underside, and pale patches along the basal border of its abdominal terga (Darsie and Ward 2005; Fig. 4B). Although Darsie and Ward (2005) include the presence of distinct basal and apical bands on the hind tarsomeres in an early couplet, Darsie and Shroyer (2004) observed that these bands were frequently missing from Florida specimens, leading them to conclude that it may not be a reliable feature for identification purposes. If leg bands are not present, specimens may be misidentified as Cx. nigripalpus, which lacks the scale patches found on the thoracic pleura of Cx. declarator, or Cx. salinarius Coquillett, which has pale rather than dark scales on its abdominal terga VII (Darsie and Ward 2005).
Culex declarator first appeared in Florida in 1998 in Indian River County (Darsie and Shroyer 2004). Since then, it has been collected from Charlotte, Pinellas, Monroe, and South Walton counties (Hribar 2018, 2019; Riles and Connelly 2020; Amos et al. 2022; Kovach et al. 2022). It has also been reported to occur in southern Texas, Central and South America, and the West Indies (Bram 1967, Darsie and Ward 2005). Between July and December of 2021, a total of 1,838 specimens were collected from 14 locations throughout the District, with the majority (1,807) coming from CDC light traps located in the southern portion of the District along the border of or within the Rookery Bay National Estuarine Research Reserve and Collier-Seminole State Park (Fig. 1), suggesting that this species may have established populations in Collier County's preserve lands before it was initially detected by the District.
The final species identified, Cx. tarsalis, has a pale band on its proboscis, light-colored tarsal bands, and broad, pale basal bands on its abdominal terga (Darsie and Morris 2003; Fig. 4C), a combination that distinguishes it from most other Florida Culex species. In Florida, Cx. tarsalis is most similar to Cx. coronator as both species have apparent bands of pale scales on the hind tarsi. The 2 species can be distinguished from each other by the presence of a complete band of pale scales on the proboscis of Cx. tarsalis (absent or incomplete in Cx. coronator), or the presence of a V-shaped pattern of dark scales on the abdominal sterna in Cx. tarsalis (absent in Cx. coronator).
Culex tarsalis is common throughout the western USA and southern Canada (Darsie and Ward 2005), with specimens having been collected as far north as Alaska (NEON 2022). Within Florida, this species has been reported from much of the Florida Panhandle; along the west coast from Levy, Pinellas, Manatee, and Sarasota counties; and from the Florida Keys (Darsie and Morris 2003, Pruszynski and Hribar 2012, Smith 2020).
Of the 3 species described here, only Cx. tarsalis is considered a significant vector of human disease, serving as a competent vector of western equine encephalomylitis, Rift Valley fever, St. Louis encephalitis, and West Nile virus (Turell et al. 2005, 2010; Clements 2012). West Nile virus is endemic in South Florida, and introduction of a new vector for this disease would present additional challenges to arbovirus control in the area. However, although the location reported here is no longer being trapped, no additional Cx. tarsalis have been captured in surveillance traps nearby, suggesting that this species may not yet have established a permanent population within the county. This may be due in part to Collier County's historically warm winter temperatures, which recent species distribution modeling suggests are higher than is optimal for this species (Rhodes et al. 2023).
Between 2019 and 2023, including the species described above, the District has reported 7 new mosquito species (Riles and Connelly 2020, Reeves et al. 2023), bringing the total number of species to 50. During this period, the District has increased the number of trap sites included in its weekly surveillance program from 22 to 46. At the same time, the population of Collier County has grown dramatically (20% increase from 321,520 in 2010 to an estimated 384,902 in 2019; USCB 2022), and the corresponding demand for housing has led to new construction and development. In addition, the county is involved in a number of wetlands restoration projects that have had significant impacts on the quantity and quality of local mosquito habitat (Radabaugh et al. 2017, Lucas et al. 2021, SFWMD 2022). This confluence of factors makes it impossible to say whether these new species represent new introductions to the county or improved resolution on existing mosquito populations. What is clear is that mosquito control districts serve a vital role in detecting and assessing the distribution of nonnative mosquito species, particularly those that are not currently considered primary vectors of human disease. Collier County is home to a variety of poorly studied arboviruses (Fish et al. 2021); and the combination of human immigration, residential expansion, and wetlands restoration is likely to increase the encounter rate between human hosts and novel viruses and mosquito species. A thorough understanding of where and when neglected species occur will allow researchers, mosquito control agencies, and public health professionals to better prepare for future arboviral disease threats.
The authors thank the Collier Mosquito Control District Board of Commissioners and Executive Director Patrick Linn; as well as Mathew Minio, Noe Pineda, and all the employees at the District who participated in sample collection and provided technical assistance. The authors also thank the Florida Park Service for granting access to Collier-Seminole State Park under Scientific Research/Collection Permit No. 05032214.
Collier Mosquito Control District, Naples, FL 34104.
University of Florida, Institute of Food and Agricultural Sciences, Florida Medical Entomology Laboratory, Vero Beach, FL 32962.