Microsatellite loci for the white-dotted mosquito (Culex restuans), a principal vector of West Nile virus in North America

Abstract

We characterized the first microsatellite loci in the white-dotted mosquito, Culex restuans, a critical early spring West Nile virus vector. An enrichment protocol yielded 960 positive clones of which we sequenced 300. We designed primers to amplify 29 unique di-, tri- and tetranucleotide microsatellites and chose 17 that amplified consistently across populations and were polymorphic. We developed three multiplex primer combinations for all 17 loci. A survey of 44 individuals revealed two to 20 alleles across loci, and expected heterozygosity ranging from 0.17 to 0.89. These markers will allow examination of the life history of this mysterious early season encephalitis vector.

While over 30 species of mosquitoes have been found positive for the West Nile virus (WNV), Kilpatrick and colleagues (2005) suggested that Culex pipiens L. and Culex restuans Theobald may be responsible for up to 80% of human WNV infections in the northeastern USA. Culex restuans is a spring species, unlike C. pipiens, and it appears to be the critical vector of WNV early in the transmission season (Andreadis et al. 2001). Culex restuans occurs from California to North Carolina, from southern Canada to Honduras (Strickman & Darsie 1988), and locally adapted populations may differ in epidemiologically significant traits. Highly polymorphic molecular markers will allow the examination of life history and ecological traits of this species and more informed exchange of information across epidemiological studies.

We isolated microsatellite loci from C. restuans using a modified version of the enrichment protocols described by Hamilton et al. (1999) and further refined by Glenn & Schable (2005). Approximately, 3 μg of DNA from four mosquitoes from the state of New York were digested with HaeIII, and blunt-end ligated to double-stranded SNX linkers. The ligation was divided into two and hybridized to different groups of biotin-end labelled oligonucleotides using the conditions in the polymerase chain reaction (PCR) program OligoHyb. [Group 1: (AAC)₁₀, (CAT)₁₀, (GT)₁₅, (GA)₁₅; Group 2: (GTC)₁₀, (CAC)₁₀, (GCT)₁₅, (GGT)₁₅]. The hybridization reactions were mixed with iron beads coated with streptavidin (Dynabeads) and incubated for 30 min at 43 °C after which a series of baths of increasing stringency removed most of the unattached DNA that lacks repeats. After denaturation, the single-stranded DNA recovered from the beads was amplified using primers to the linkers using the PCR conditions in Keyghobadi et al. (2004). The amplified DNA was cloned into bacteria using a TOPO TA cloning kit (Stratagene) with a transformation efficiency of 90%. Out of 3750, 960 white colonies were transferred into 96-well plates containing Luria-Bertani liquid broth with 15 μg/mL of ampicillin, and grown for 48 h at 37 °C with moderate shaking. From these we chose 300 colonies that were amplified and sequenced using M13 primers.

We found 220 unique inserts that contained microsatellites, of which 98 had more than four repeats and enough flank to design primers. We refrained from using loci with more than 20 repeats, especially in the case of dinucleotide motifs, because in our experience those loci are hard to score consistently. Primers were designed for 29 loci focusing on an even mix of di- and trinucleotide motifs, using ‘Primer 3′ software (http://frodo.wi.mit.edu/primer3/input.htm); often we designed multiple primers from the same sequence to create multiplexes. We optimized 17 microsatellite loci into only three PCRs (Table 1). In a multiplex (MP) amplification, up to six primer pairs labelled with up to three different colours (Applied Biosystems) are added simultaneously to a single PCR and processed together.

Table 1.

Characterization of 17 microsatellite loci in Culex restuans

Locus	Primer sequences (5′-3′)	M	A	Allele range (bp)	HO	HE	P	Motif	GenBank Accession no.
Crest1T1.2d	F: ATTAGCTTCCCAGGCTCTCC R: CGGAAAGCGGAATCGATAG(H)	3	13	142–166	0.83	0.89	0.34	CA10	FJ373274
Crest1T1.17	F: ACCACACCGATCATTATCCTC R: GAAACATGGAAACGGCAAAC(N)	3	8	131–153	0.72	0.73	0.58	GTT6	FJ373273
Crest1T1.30	F: TCTTCGCGCAGAAACGTC(F) R: ACTGCTGCTGGACAGTACCC	3	7	118–130	0.68	0.77	0.38	CA8	FJ373275
Crest1T2.21 L	F: GCTGATCTTGGCTTGTGGTC(F) R: AACTAAACAATGTCACGAGATACAAGG	1	2	181–189*	0.14	0.17	0.29	GAGT5	FJ373276
Crest1T2.23	F: CAAGCCTTGGTCATGTGAG(H) R: CGCGCCATTTTGTCTTTAC	2	9	101–117	0.69	0.79	0.72	CA7	FJ373277
Crest1T2.33	F: AGGGCACATGTTAAGCCTTG R: TCGGAAATCGAATGATGTTG(F)	2	6	88–100	0.81	0.75	0.34	GA7	FJ373278
Crest1T2.54	F: TCTCCTGTACGCTTGTGTGC(H) R: CCTGGCTGATCTTTGGACTG	1	4	103–112	0.4	0.43	0.48	CAA6	FJ373279
Crest1T2.83a	F: CCCATTCAAAGTTAAAAGATTGC(F) R: GCCAAGACTGATTGGAAACG	1	9	118–164	0.6	0.79	< 0.01	CAA5	FJ373280
Crest1T3.7	F: CAATCATCGAGGAGCATCAG(H) R: TTGCGATGATCCTGTCGTAG	1	6	132–152	0.42	0.43	0.62	CAA4	FJ373281
Crest1T4.76	F: CGATATGGGAACGCGTTAGG(F) R: GCAGTTTGAGAGGCTCAACG	3	3	174–178	0.16	0.33	< 0.01	TG6	FJ373282
Crest1T4.82	F: GAAGTGTGAGCTAGGGCAAG(N) R: CAATTCAATGCCAAATACCG	1	10	156–168	0.67	0.63	0.49	GT7	FJ373283
Crest1T5.52	F: GCGAGATTTCGTTGACAAAC R: ACGAAGAGCCGAGGTACAAC(N)	2	7	99–128	0.46	0.44	0.74	CAA6	FJ373284
Crest1T5.87	F: TGTTTATCCTGGAGGAAATGAG R: CCAAAGCCTTTCCTGTCATC(F)	2	20	159–204	0.83	0.87	0.28	GT9	FJ373285
Crest2T2.25	F: GACACGGAAGCGAAATTAAAC(N) R: ACTCCAACCCGACAAGAAAC	1	7	104–117	0.58	0.61	0.66	GTC5	FJ373286
Crest2T2.56	F: AAATTGCGAAAGGACGATTC R: TCGATGCTTGAAGTTTATCACG(N)	2	13	182–218	0.71	0.86	0.4	TGC9	FJ373287
Crest2T4.30	F: CCGGAAGTGAACCGCAAC(H) R: TGTTGAAACCCACTGTTCTAAACC	2	14	164–185	0.7	0.74	0.45	TC9	FJ373288
Crest2T4.71	F: TCTCGTTTCGTCCATTCTCC R: AACCACGCACGATGAGATTC(H)	3	11	71–97	0.75	0.82	0.75	GTC5	FJ373289

F and R indicate forward and reverse primer. The letter after the primer sequence (F, H, or N) indicates that primer is labelled with blue, green, or yellow fluorescence, respectively, and M indicates which multiplex it is part of. Number of alleles (A), their size range, observed (H_O) and expected (H_E) heterozygosities, and the P value of the comparison (P) were estimated from a sample of 44 individuals. We list the microsatellite motif we focused on, but other microsatellite types were often present in the sequences in GenBank.

^*This locus amplifies in Culex pipiens. We have found a single allele (186) not found in Culex restuans.

The variability of each locus was assessed in 44 individuals of Cx. pipiens from Albany, New York (Table 1). PCRs were in 20-μL final volume and contained 1× buffer (10 mm Tris-HCl pH 8.3, 50 mm KCl), 2.0 mm MgCl₂ (MP1 and MP2, 2.5 mm for MP3), 150 μg/mL BSA, 200 μm each dNTP, 0.2 μm each primer (except for loci Crest2T4_30 and Crest1T2.83a that need 0.3 μm), 0.5 U Taq polymerase (Applied Biosystems), and 1 μL DNA template (~5 ng). Thermal cycling was performed on an MJ Research Peltier machine: 5 min at 96 °C; 30 cycles of 30 s at 96 °C, 30 s at annealing temperature (MP1 and MP2 = 52 °C; MP3 = 55 °C), 30 s at 72 °C; 5 min at 72 °C. PCR products were electro-phoresed and detected on a 3100 Model capillary sequencer (Applied Biosystems).

Although we routinely examine the inheritance of the loci by testing them in known family groups (Keyghobadi et al. 2004), Cx. restuans refuses to mate in captivity. Instead, we tested 44 specimens all collected in a restricted area of Albany county, New York, and examined departures from Hardy–Weinberg equilibrium. We observed two to 20 alleles per locus, and expected heterozygosities ranged from 0.17 to 0.89 (Table 1). Statistical tests for Hardy–Weinberg and linkage equilibrium were conducted in GenePop (Raymond & Rousset 1995). A significant heterozygote deficiency was observed for two loci, Crest1T2.83a and Crest1T4.76, suggesting null alleles. All pairwise tests of linkage disequilibrium between loci were nonsignificant after sequential Bonferroni correction. These microsatellite loci are highly polymorphic and suitable for population studies in Cx. restuans.