Single nucleotide polymorphisms (SNPs) are the most common source of genetic variation in eukaryotic species and have become an important marker for genetic studies. The mosquito Anopheles funestus is one of the major malaria vectors in Africa and yet, prior to this study, no SNPs have been described for this species. Here we report a genome-wide set of SNP markers for use in genetic studies on this important human disease vector.
Results
DNA fragments from 50 genes were amplified and sequenced from 21 specimens of An. funestus. A third of specimens were field collected in Malawi, a third from a colony of Mozambican origin and a third form a colony of Angolan origin. A total of 494 SNPs including 303 within the coding regions of genes and 5 indels were identified. The physical positions of these SNPs in the genome are known. There were on average 7 SNPs per kilobase similar to that observed in An. gambiae and Drosophila melanogaster. Transitions outnumbered transversions, at a ratio of 2:1. The increased frequency of transition substitutions in coding regions is likely due to the structure of the genetic code and selective constraints. Synonymous sites within coding regions showed a higher polymorphism rate than non-coding introns or 3' and 5'flanking DNA with most of the substitutions in coding regions being observed at the 3rd codon position. A positive correlation in the level of polymorphism was observed between coding and non-coding regions within a gene. By genotyping a subset of 30 SNPs, we confirmed the validity of the SNPs identified during this study.
Conclusion
This set of SNP markers represents a useful tool for genetic studies in An. funestus, and will be useful in identifying candidate genes that affect diverse ranges of phenotypes that impact on vector control, such as resistance insecticide, mosquito behavior and vector competence.
Background
Anopheles funestus and Anopheles gambiae are the major malaria vectors in Africa. Due to the difficulty of laboratory colonization, An. funestus has not received the same attention as An. gambiae and as a consequence there are few molecular markers for this species. However, the recent successful colonization of two strains of An. funestus [1] and the identification of a number of microsatellite markers [2, 3] have facilitated more detailed studies of this species. Microsatellite markers particularly have been used to study population structure and gene flow between An. funestus populations [4–6] and a subset of these microsatellite markers were used to build the first linkage map of this species [7]. However, microsatellite markers are not evenly distributed across the genome, and their low number so far is an obstacle to the development of high resolution linkage maps needed for QTL mapping or association studies in An. funestus. Therefore, this study was initiated to increase the availability of characterized and mapped markers for An. funestus. Physically mapped ESTs were used to identify SNPs. Such ESTs have been used to study the genetic variability in a number of species such as Aedes aegypti, Drosophila melanogaster or Homo sapiens [8–10] and should be a source of DNA polymorphisms for An. funestus as well.
Single nucleotide polymorphisms (SNPs) are by far the most common type of molecular variation in all organisms. They are extremely abundant with an occurrence of about one SNP per kb in human [11] and about one SNP every 125 bp in An. gambiae [10]. Significant progress has been made in the development of tools for detection and genotyping of SNPs and they are now becoming the markers of choice for association studies, high-resolution linkage mapping and population genomics studies [12]. SNPs located in non-coding regions of the genome and synonymous SNPs (sSNPs) in coding regions, which have no impact on the phenotype, may provide useful markers for population genetics studies. Non-synonymous SNPs (nsSNPs) which alter the structure (change of amino acid sequence) and potentially the function of encoded proteins are useful markers for association studies to detect genetic variations linked with phenotypic traits.
Patterns of genetic diversity in An. funestus have not been studied to the same extent as in An. gambiae or Drosophila species. Nucleotide diversity in these species has been used to compare patterns of nucleotide variation, such as the relative occurrence of transitions/tranversions in different regions of the genome [8, 13]. These surveys have established codon usage and usage bias patterns in many species, with bias hypothesized to occur as a result of selection for efficient translation [14, 15].
The sequencing of the 278 million base pairs (Mbp) constituting the An. gambiae genome has revealed more than 400,000 SNPs indicating a high level of polymorphism in mosquito species [16]. We hypothesize that by sequencing DNA fragments of different genes of An. funestus, a similar level of polymorphism should be encountered and will allow the identification of a significant set of SNPs.
Here, we describe the detection and characterization of a set of genome-wide SNP markers from 50 nuclear genes using two laboratory strains and field samples of An. funestus. We also examined patterns of polymorphism and nucleotide diversity in coding and non-coding regions of the genome and define the pattern of codon usage in An. funestus. The utility of the SNPs was assessed by genotyping a subset of these SNPs during a linkage mapping study.
Results and discussion
Gene amplification
In total, 70 primer pairs were tested by PCR, 55 of which gave reliable amplification with PCR products ranging from 194 to 1342 bp. Sequence data from a total of 21 specimens of An. funestus was obtained for 50 of these genes (see Table 1) from laboratory and field samples. Overall, we sequenced a total of 20,547 bp consisting of 14,671 bp of coding region and 5,876 bp of non-coding region. We identified 494 SNPs consisting of 303 coding SNPs (cSNPs) and 191 non-coding SNPs. Each gene contained at least one polymorphism with BU73 having one and BU88 having 29. The distribution of SNPs among the 50 genes is presented in Table 1. All information concerning the location and the nature of each individual SNP have been submitted to dbSNP, the SNP database of GenBank. These SNPs with their respective reference SNP number (rs) are publicly available in dbSNP Build N°127. The NCBI ss (submitted SNP) numbers of these SNPs are ss65917063 to ss65917416.
Table 1
Nucleotide polymorphism in An. funestus genes
Coding region
Non-coding region
Polymorphic sites
Nucleotide diversity
Polymorphic sites
Nucleotide diversity
Transition
Transversion
Gene
nHap
L (bp)
1st
2nd
3rd
Σ
1st
2nd
3rd
Σ
Syn
Rep
Σ
π
πn
Ks
Ka
L(bp)
Ts
Tv
Σ
π
4G17
3
190
0
1
1
2
0
0
0
0
1
1
2
0.0052
0.0036
0.0200
0.0072
0
0
0
0
0.0000
9K1
23
199
3
1
3
7
1
0
1
2
5
4
9
0.0193
0.0107
0.1010
0.0270
0
0
0
0
0.0000
4J10
14
141
1
0
3
4
0
0
1
1
5
0
5
0.0116
0.0000
0.1300
0.0000
73
2
2
4
0.0157
6P3
18
0
0
0
0
0
0
0
0
0
0
0
0
0.0000
0.0000
0.0000
0.0000
179
6
2
8
0.0188
6P4
17
201
0
0
2
2
0
0
0
0
2
0
2
0.0041
0.0000
0.0450
0.0000
189
8
3
11
0.0233
6P5
5
189
0
0
1
1
0
0
0
0
1
0
1
0.0016
0.0000
0.0200
0.0000
112
2
0
2
0.0063
6Z1
14
444
1
0
5
6
0
0
5
5
8
3
11
0.0079
0.0022
0.0750
0.0088
29
1
0
1
0.0159
6Z3
14
453
1
0
13
14
1
0
5
6
18
2
20
0.0169
0.0011
0.1860
0.0057
0
0
0
0
0.0000
9J12
9
156
2
1
0
3
1
0
1
2
0
5
5
0.0117
0.0156
0.0000
0.0510
0
0
0
0
0.0000
9J14
14
174
0
0
3
3
2
0
0
2
3
2
5
0.0140
0.0077
0.0760
0.0150
0
0
0
0
0.0000
BU01
2
162
0
0
1
1
0
0
0
0
1
0
1
0.0011
0.0000
0.0230
0.0000
26
0
0
0
0.0000
BU08
12
394
0
0
2
2
2
1
2
5
4
3
7
0.0042
0.0026
0.0470
0.0098
16
0
0
0
0.0000
Ache
18
249
0
1
7
8
0
1
2
3
8
3
11
0.0149
0.0034
0.1460
0.0160
85
2
2
4
0.0180
BU10
16
294
1
0
3
4
0
0
2
2
5
1
6
0.0084
0.0024
0.0690
0.0045
245
3
1
4
0.0056
BU11
4
177
0
1
0
1
0
0
0
0
0
1
1
0.0029
0.0039
0.0000
0.0074
53
0
1
1
0.0099
BU12
7
504
1
1
1
3
0
1
1
2
3
2
5
0.0032
0.0011
0.0240
0.0053
38
0
0
0
0.0000
BU13
19
363
2
0
1
3
0
0
0
0
1
2
3
0.0026
0.0029
0.0110
0.0073
148
2
5
7
0.0133
BU19
22
526
3
1
4
8
0
1
1
2
6
4
10
0.0072
0.0035
0.0490
0.0100
0
0
0
0
0.0000
BU021
17
267
0
0
2
2
0
0
2
2
4
0
4
0.0044
0.0000
0.0610
0.0000
67
2
2
4
0.0164
BU21
16
417
0
1
2
3
1
0
2
3
4
2
6
0.0073
0.0018
0.0510
0.0077
81
4
1
5
0.0256
BU25
2
234
0
0
0
0
0
0
0
0
0
0
0
0.0000
0.0000
0.0000
0.0000
111
2
0
2
0.0036
BU29
11
258
0
0
2
2
0
0
2
2
3
1
4
0.0038
0.0008
0.0540
0.0049
152
0
2
2
0.0029
BU34
11
231
0
0
2
2
0
2
2
4
4
2
6
0.0101
0.0051
0.0740
0.0113
28
0
0
0
0.0000
BU35
12
264
0
0
7
7
0
0
1
1
8
0
8
0.0071
0.0000
0.1280
0.0000
108
1
1
1
0.0039
BU40
14
255
2
1
1
4
1
0
0
1
1
4
5
0.0074
0.0076
0.0170
0.0200
415
3
2
5
0.0016
BU56
4
156
0
1
2
3
1
0
2
3
3
3
6
0.0214
0.0132
0.0710
0.0263
126
5
2
7
0.0331
BU58
7
261
1
0
2
3
0
1
1
2
3
2
5
0.0053
0.0019
0.0446
0.0103
20
0
0
0
0.0000
BU62
22
213
0
0
1
1
0
0
0
0
1
0
1
0.0013
0.0000
0.0207
0.0000
260
7
5
12
0.0143
BU66
25
282
9
2
0
11
1
2
0
3
3
11
14
0.0174
0.0194
0.0503
0.0494
179
4
4
8
0.0099
BU70
5
228
0
0
2
2
0
0
2
2
2
2
4
0.0066
0.0015
0.0506
0.0059
62
0
1
1
0.0069
BU71
9
573
0
0
6
6
1
1
2
4
8
2
10
0.0044
0.0005
0.0615
0.0045
98
1
0
1
0.0051
BU72
20
354
2
1
2
5
2
1
1
4
3
6
9
0.0082
0.0062
0.0368
0.0220
195
1
1
2
0.0052
BU73
5
277
0
0
0
0
1
2
0
3
0
3
3
0.0039
0.0036
0.0000
0.0135
0
0
0
0
0.0000
BU76
15
330
0
0
7
7
1
1
1
3
9
1
10
0.0087
0.0008
0.1104
0.0041
154
3
3
6
0.0095
BU77
18
261
3
0
1
4
1
2
0
3
6
1
7
0.0125
0.0136
0.0166
0.0298
198
9
7
16
0.0345
BU82
13
228
0
0
2
2
0
2
0
2
2
2
4
0.0059
0.0025
0.0402
0.0112
88
2
2
4
0.0156
BU85
14
299
2
1
8
11
0
0
1
1
11
1
12
0.0108
0.0363
0.1155
0.0033
121
2
0
2
0.0087
BU88
16
441
2
0
12
14
0
0
2
2
16
0
16
0.0135
0.0000
0.1601
0.0000
212
9
4
13
0.0179
BU90
4
0
0
0
0
0
0
0
0
0
0
0
0
0.0000
0.0000
0.0000
0.0000
261
1
2
3
0.0021
BU92
11
345
3
1
3
7
1
1
1
3
5
5
10
0.0075
0.0031
0.0650
0.0186
110
5
3
8
0.0226
BU93
8
510
0
0
5
5
0
0
3
3
7
1
8
0.0035
0.0004
0.0626
0.0025
137
0
2
2
0.0023
BU98
4
369
0
0
5
5
0
0
2
2
7
0
7
0.0101
0.0000
0.0843
0.0000
158
2
6
8
0.0257
BU883
21
354
1
0
5
6
2
1
2
5
7
4
11
0.0120
0.0052
0.0913
0.0145
188
8
3
11
0.0189
BU897
19
303
0
0
2
2
3
1
2
6
3
5
8
0.0080
0.0078
0.0458
0.0210
167
4
3
7
0.0180
BU901
12
231
0
0
3
3
0
0
0
0
3
0
3
0.0035
0.0000
0.0552
0.0000
229
2
2
4
0.0065
BU973
5
471
1
1
0
2
0
0
0
0
2
0
2
0.0015
0.0019
0.0000
0.0054
67
1
0
1
0.0028
BU974
3
459
0
0
2
2
0
0
1
1
3
0
3
0.0036
0.0000
0.0265
0.0000
52
1
0
1
0.0096
BU982
4
240
0
1
1
2
0
0
0
0
1
1
2
0.0030
0.0016
0.0190
0.0053
51
1
0
1
0.0107
BU996
18
543
4
0
2
6
1
1
1
3
6
3
9
0.0073
0.0019
0.0518
0.0070
87
6
6
12
0.0312
Kdr
13
201
0
0
0
0
1
1
0
2
0
2
2
0.0020
0.0026
0.0000
0.0127
501
2
5
7
0.0039
Total
14671
45
17
139
201
25
23
54
102
206
97
303
5876
106
85
191
Average
0.0072
0.0041
0.0537
0.0097
0.0123
nHap, number of haplotypes; L, length of the nucleotide sequence; ∑, total; Syn, synonymous substitutions; Rep, replacement substitutions; π, average number of nucleotide substitution per site; πn, average number of non-synonymous nucleotide substitution per site; Ks, average number of nucleotide substitution per synonymous site; Ka, per non-synonymous site; Ts, transitions; Tv, transversions.
Type of polymorphism
For all sequenced DNA fragments, transition substitutions were more predominant than transversions (62 % vs 38%). Transitions C↔T and A↔G are over-represented with 35.4 and 27.2 % of the total substitutions respectively while the four transversion classes occurred at similar levels (Figure 1). The higher frequency of C↔T and A↔G SNPs is probably partly related to 5-methylcytosine deamination reactions that occur frequently, particularly at CpG dinucleotides [17]. The preponderance of transitions is more obvious for coding regions where out of the 303 SNPs identified, 201 were transitions (66.3%) and 102 were transversions (33.7%). The ratio of transitions/transversions observed here is close to the 2 : 1 ratio observed for Drosophila and humans [13, 18]. For polymorphism in non-coding regions, transitions accounted for 55% (106) and transversion for 45% (86). The frequency of transitions between coding and non-coding regions were significantly different (66.3% vs 55% respectively; χ2 = 5.86, P < 0.01). This confirms that SNPs occur more frequently as transitions in coding regions than in non-coding regions. There is also a higher frequency of SNPs occurring at the third codon position (63.7%) than at the 1st or 2nd position (Table 1). Similar results have been observed for Aedes aegypti [10] and in three species of Drosophila [13]. The degeneracy of the genetic code and the selective pressure for gene conservation have been suggested as the main reasons for the preponderance of transitions over transversions [13]. Synonymous or silent substitutions are more often transitions than transversions and there is a stronger selection against replacement substitutions than against synonymous, leading to an increase of the relative frequency of transitions [13]. For fourfold degenerate codons, selection should be neutral, since no amino acid change is induced by a nucleotide substitution at the third position, and each of the 4 codons will produce the same amino acid. We tested this hypothesis by comparing the proportions of transversions at fourfold degenerate codon positions and at non-coding positions for all the 50 genes (Table 2). The result shows that there is no significant difference between the frequency of transversions at fourfold degenerate codon positions (36.8%) and at non-coding regions (44.5%) (χ2 = 2.55 P = 0.11), while this difference is significant between coding and non-coding regions (χ2 = 5.33; P = 0.021). The fact that fourfold degenerate sites have a similar ratio of transitions/tranversions to non-coding regions is consistent with an hypothesis that the structure of the genetic code and selection against replacement polymorphisms accounts for the preponderance of transition substitutions in coding regions.
Figure 1
Distribution of transitions and transversions among SNPs.
Table 2
Transition (Ts) and transversion (Tv) polymorphisms for different classes of DNA
Polymorphism
Probability
Ts
Tv
%Tv
Coding Region
3rd coding position
Fourfold
Non coding regions
106
85
44.5
P = 0.021
P = 0.0
P = 0.11
Coding regions (Cd-R)
201
102
33.6
P = 0.204
P = 0.507
Third coding position
139
54
27.9
P = 0.047
Fourfold degenerate sites
60
35
36.8
Five insertion/deletion polymorphisms (indels) were observed in four genes ranging from 1 to 4 bp in coding, intronic and 5'UTR regions (Table 3). Two indels of 2 and 4 bp were observed in the BU10 intron. The frequency of indels (8% for 4/50) is lower than that reported in Ae aegypti of 24% [10] or 25% in An. gambiae [19]. Only one indel, in the BU93 gene, was located in a coding region. This indel was a triplet that did not cause a frame shift. The four indels identified can serve as molecular makers for mapping studies.
Table 3
Indel polymorphism
Non coding region
Gene
Coding region
Introns
5' UTR
3'UTR
6P5
4 bp
BU10
2 bp, 4 bp
BU66
1 bp
BU93
3 bp
Approximately 2/3 (206) of the 303 cSNPs were synonymous substitutions (no modification in amino acid) while around 1/3 (97) were non-synonymous or replacement SNPs leading to a change of amino acid. As approximately two-thirds of random coding substitutions change an amino acid, the fact that only 1/3 of cSNPs are non-synonymous implies strong selection against changes that alter amino acid. This ratio of synonymous and replacement cSNPs is similar to that observed in An. gambiae [19] and Ae. aegypti [10].
Genetic diversity
We estimated the nucleotide diversity for each of the 50 genes in coding and non-coding regions (Table 1). The average nucleotide diversity per gene in coding regions was 7.2 × 10-3 or around 1 SNP every 138 bp similar to that observed in An. gambiae (1 SNPs every 125 bp) [19] but much higher than the frequency of 1 SNP/kb observed in humans [20]. SNPs were observed in non-coding regions at a frequency of 1 SNP per 100 bp, corresponding to π = 10 × 10-3. Figure 2 shows that there is a positive correlation in the level of polymorphism between coding and non coding regions of An. funestus genome within a gene (r = 0.48, P < 0.01). This positive correlation may be the consequence of many factors notably the correlated genealogies existing between coding regions and their surrounding non-coding regions. This correlation may also be strengthened by the presence of indirect selection (hitchhiking or background selection) and probably by variable recombination rate, as it is the case in Drosophila [21]. Mutational effect of recombination or biased gene conversion can also operate, but this needs to be confirmed as even in Drosophila, the effect of biased gene conversion is only suspected but unwarranted [22, 23]. The average nucleotide diversity in non-coding DNA (0.010) was lower than in synonymous sites of the coding regions (0.0207), P < 0.01. This pattern was also observed in An gambiae, Ae aegypti and Drosophila species [10, 13, 19]. This is an indication that non-coding regions are under greater purifying selection than synonymous sites within coding regions. This is not surprising, given that non-coding regions may be involved in gene regulation. The non-coding 5'-flanking sequence of a gene may contain regulatory elements such as the promoter that control the expression of that gene, and single-base mutations can affect essential structures for splicing and processing [24].
Figure 2
Correlation of nucleotide diversity in coding (πc) and non-coding regions (πnc) πc: nucleotide diversity of coding region, πnc: nucleotide diversity of non-coding region.
Nucleotide diversity varies greatly from one gene to another (Table 1) and this is likely related to individual gene function and potentially to differences in selective constraints. However, non-synonymous diversities need to be compared in order to definitely estimate the influence of differences in selective constraints. Among the most polymorphic genes sequenced were cytochrome P450 genes, lysozyme, translation initiation factor and ubiquitin conjugating genes. The non-synonymous nucleotide diversity of these genes varied from 14 to 36.3 × 10-3. Most of these genes are involved in specific mechanisms that evolve very rapidly, such as detoxification of xenobiotics for cytochrome P450s or defense mechanisms against bacteria like lysozyme. For example, P450s present a high level of redundancy with less genetic constraints and therefore more polymorphism. In contrast some genes showed very low level of variation particularly those involved in transcriptional or translational regulation (BU973 and BU25, BU93) or in signaling processes (BU01, BU08, BU13). Examples of selective constraints have been observed as well in Drosophila spp. where substitution rate between conservative genes and fast evolving genes differ by around 10-fold [25]. Nucleotide diversity was not statistically different between laboratory strains and field collected mosquitoes (7.4 × 10-3 and 6.9 × 10-3; P = 0.21 by Student's t-test), despite an apparent low level of heterozygosity (fewer heterozygote SNPs) observed in the two laboratory strains compared to the field sample. This result could be due to the fact that FUMOZ and FANG (the two laboratory strains used in this study), were only recently colonized in laboratory and therefore still largely retain the polymorphism of natural populations of An. funestus.
The ratio of synonymous to non-synonymous changes (Ka/Ks) gives an indication of the magnitude of the purifying selection against deleterious mutations in a species. The rate of non-synonymous nucleotide substitution per non-synonymous site (Ka) is generally expected to be much lower than the rate of synonymous substitution per synonymous site (Ks), because random amino acid changes are usually deleterious, whereas synonymous changes are likely to be neutral or nearly so [26]. Thus, the expectation is Ka << Ks, except when positive selection is involved favouring particular amino acid replacements, in which case Ka will increase. For An. funestus the Ka/Ks ratio was equal to 0.181 and is similar to the ratio of 0.192 observed in An. gambiae [19] or 0.204 in Ae. aegypti [10] but, higher than the ratio of 0.115 reported in D. melanogaster [13]. This result indicates that the purifying selection against deleterious mutations is acting in An. funestus. Indeed species with large effective population size such Drosophila or Anopheles species are generally more effective at purging deleterious mutations [26].
Clustering pattern of the SNPs
We analyzed the distribution of SNPs identified in this study. We found 16 clusters of two directly neighboring SNPs, one cluster of 3 consecutive SNPs and 13 clusters of two SNPs separated by just 1 bp. For some SNP genotyping methods based on allele-specific amplification, ligation or single base extension principles for which primers need to be designed immediately adjacent to the SNP, it is important that the SNPs are not too close together to prevent primer designing. The presence of a polymorphism within approximately 20 bp will limit the possibilities for designing a robust primer. Most of the 494 SNPs identified in this study do not have a SNP within 20 bp on either or one side thus, and should be easily genotyped by one of these methods.
Genomic position of the SNPs
Among the 50 genes amplified for SNP detection in this study, 45 are already physically mapped to the An. funestus genome by in situ hybridization [27], and the remaining 5 genes were genetically located to their respective chromosome by linkage mapping [7]. Overall, 29 SNPs were located on the X chromosome, 334 on chromosome 2 and 131 on chromosome 3. The higher number of SNPs observed on chromosome 2 is also a consequence of the fact that most of the studied genes are located on that chromosome. Table 4 gives the chromosomal location of the 50 genes across the genome of An. funestus.
Table 4
Characteristics of genes amplified for SNP detection
Genes
Chromosomal Location
Accession no.
Function
Forward primer
Reverse primer
Product length
No of SNPs
4G21
X
AY648704
Cytochrome P450
GGCGATAGCAAACGTAAAGC
CGCGGTAAACGGAATATAGC
303
2
9K1
X
AY987362
Cytochrome P450
GTACGAGCTGGCCGTTAATC
CCTTTCTGTAGCTGCACCTTG
243
9
4J12
3R
AY648706
Cytochrome P450
CCAACAAATCAGTTCATCAGC
TTGTAAAAGTGCTTAAAATG
270
9
6P9
2R: 9A-12C
AY729661
Cytochrome P450
GCGCCTTAGACAAGAGATCA
AAGGGATGTCGCTTCTTCTC
350
8
6P4
2R: 9A-12C
AY987359
Cytochrome P450
GTACGAGACTGGCAAAGAAT
AAGGAAGACGTATGGATGG
430
13
6P5
2R: 9A-12C
AY987360
Cytochrome P450
CTGGCTTTGAAACTTCCTC
AGATACACGTAGGGATGTCG
550
3
6Z1
2L: 25A-27D
Cytochrome P450
ACGATCCGTTCCGGGTAG
GCTAGCGCAGGATACATTCG
550
12
6Z3
2L: 25A-27D
Cytochrome P450
GACGATCCGTTCCTGAAGAC
ATCGGTAAGCCCGGATATTT
550
20
9J12
3L
AY729663
Cytochrome P450
TACCGGTGTGCAGCTTGA
CTTTGGCGCGAAGGTAAA
194
5
9J14
3L
AY729665
Cytochrome P450
CGGACAACGTATGATCGATTT
TTTGGCTTGCATTAAAAGGTG
214
5
BU01
X:2B
BU039001
type II transforming growth factor-beta receptor
GTGTGTTTGCTTGGGTGTTG
GGCATCGGTAATCAGGATGT
525
1
BU08
2R:7C-10B
BU038908
rhodopsin
CATTTGTGGAACCCCATTTC
GGTCATTGGTTTACCCGAGA
500
7
Ache
2R:9C-12C
DQ534435
Acetylcholinesterase
GGGTACGGGACAACATTCAC
CGTTAACGTACGGGTCGAGT
1050
15
BU10
2L:28A
BU039010
Cyt-c-p-P1
AAGCACAGTTAAACCTTTCG
ACCTAGCCCAATCTCTGTCT
650
10
BU11
3L:43B
BU038911
protein transporter
ATCTGCTTGCGCTAGATCGT
ATCGCCAAATTTCATCTTCG
2
BU12
2R:7B
BU038912
Alpha tubilin
AAGCTCGAGTTCGCCATCTA
CTCCAATCCTTTCCGACGTA
800
5
BU13
2R:15C
BU038913
signal sequence receptor
ACCCTGAGAAATCGTAACAA
CCGATAGTTGAGAGCAATGT
630
10
BU19
2R:12B
BU038919
Chitinase
CTGTTGCTGCTGCTACATAC
CCGGTCACGTACAAATAGTC
670
10
BU021
3L:38C-40B
BU039021
Tubilin beta-3 chain
GAGTTGGTTGATGCCGTGTT
CGTCCGGAAACAAATATCGT
400
8
BU21
X:3A
BU038921
Phosphoribosylaminoimida-zole carboxylase
TTTCAAGGTGAACGGTGTGA
CCATCAAGATGACGACCAGA
475
11
BU25
2R 12B
BU038925
ferritin heavy chain-like protein
GCGTAAAGCTGTCGTCCTTC
ATTCCCCCGTCAGGTAGTCT
1200
2
BU29
2L:27B
BU038929
sensory appendage protein
CACCAAGTACGATGGTGTCG
AGGCACTTGGTTTTGCAGTT
410
6
BU34
X:1C
BU038934
NADH dehydrogenase
GGCAGGTAGCAGCAGTTTTC
CAGTACCAACCGCAACACAC
400
6
BU35
2R 12B
BU038935
CG6846 gene product
TTCAGCAAACACGTTTCGTC
ACTTGCCCTTGTCCTTGTTG
400
9
BU40
2R:14B
BU038940
Glutathion peroxydase
AGGCAAAATCAATTTTTGAA
CGTAACAATTTCTCGACCAT
1150
10
BU56
2R:7B
BU038956
novel An. gambiae salivary protein
AATCTAGAAGCTGCGCCAGA
AATTCTAGGACGGCGATTCC
13
BU58
2R:12D
BU038958
translation initiation factor
ACTTCCACGCCCAGTGTATC
CGTGCAGAGTTCGAAAACAA
650
5
BU62
2L:23A
BU038962
cAMP responsive element binding protein
CAATCGGAGCGTAAGGAAAG
CGTTCTCCCGCAAAAACTAA
475
13
BU66
3R:30A
BU038966
Lysozyme
TAGCTCATAGTGGCGGTTAT
ACTACAACATGTCGTGCAAA
650
22
BU70
2R:7C
BU038970
Ubiquitin fusion 80
GTGGACTCCGTACCTGGTCA
CTGTAGAATTACAGGAGGGCGTA
5
BU71
3L:39A
BU038971
structural protein of peritrophic membrane
GGGAAGTCGGTGTAGGGAAT
ACGTTTGGGTCAGGTAGTCG
750
11
BU72
2R:12B
BU038972
RHO small monomeric GTPase
GATGAAGCTGCCAAAGATCC
TGCCTCGTCGAAAACTTCTT
900
11
BU73
2R:7A
BU038873
actin binding
AGTAAGAAACGAACGCAAAG
CGGAAAAGTTGGAATGTAAC
430
3
BU76
2R:10B
BU038976
translation initiation factor
TGCCTACGAACGACGTAATG
GGCTCGTAGCTGGTCACTTC
500
16
BU77
2R:10C
BU038877
ubiquitin conjugating enzyme
CAACACACTAGCCAGCAAGG
TTTGGTTCGGCCAACATACT
408
23
BU82
2R:14D
BU038882
Unknown
AGGGCGGTACAACAAAATCT
GCATCGGAGCGTTTCCTA
400
8
BU85
2R:12E
BU038885
phosphoglycerate mutase
AAAAAGAATGGCCGGAAAGT
CTCATCGCCCAGAATTTCAT
800
14
BU88
2R:11B
BU038988
translation initiation factor
GTGGCCTCCCACTTTGTTAG
TACCGGATACGGTTGACGAT
800
29
BU90
3R:35B
BU038990
gustatory receptor
GGGACATCATCATCATCGAC
TTTCGCTTCTCGCGTTAAAT
300
3
BU92
3L:39A
BU038892
Microtubule binding
CATGCGACCGAAGAGAAGTT
ATCCTGATTCTGGCTCATGG
550
18
BU93
2R:7C-10C
BU038893
prefoldin subunit 2
CACCGGAAACTCGGCTATTA
TATCGGTTCCATCCGAAAAG
550
10
BU98
3L:46B
BU038898
CG7630 gene product
TGCGTCACCCGTTACAAATA
ACGTGTACGCTTTCCACCTC
550
15
BU883
3R:32B
BU038883
peritrophin
TTCGTGACACAGTTATACGC
GCACACTTCAGACTTCCTGT
650
22
BU897
3R:36C
BU038897
NADH dehydrogenase (ubiquinone)
GGGAATTCCGTGATTTTT
GGCAGAAATATCCATAATCG
700
15
BU901
2L:20C
BU038901
CG18397 gene product
AAAGACACTCCCGCATTACG
CTCGTGTCTGTTTGGCTTGA
480
7
BU973
3R:36F
BU038973
polyA-binding protein II
AGTAAGAAACGAACGCAAAG
CGGAAAAGTTGGAATGTAAC
630
3
BU974
3L:40A
BU038974
serine-type peptidase
ACTGGCGGAGAACGTACAAC
TGCTGCACATTAATCAAAGGTT
4
BU982
2R:12B
BU038982
ferritin 2 light chain homologue
CTAGTTTCCTGTCGCGTTCC
CATCGTCTCCTCCATTACCG
400
3
BU996
2R:8D
BU038996
vacuolar hydrogen-transporting ATPase
GTTCGCCTACATGTGCTTCA
ACAAAGGGTGTGCAAAAAGG
800
21
Kdr
3R: 36A-37E
DQ534436
Sodium channel gene
TGCAAAATAGAGTCATTGGTGAA
ATCATCTTCATCTTTGC
1342
9
Polymorphism reliability
To assess the validity of the SNPs identified in this study, 30 SNPs were tested for segregation in isofemale lines. These SNPs were tested using different methods (pyrosequencing, HOLA, SBE and AS-PCR) [7, 28]. The Mendelian segregation ratio of each of these SNP loci at F0, F1 and F2 generations was examined in four families from reciprocal crosses between a pyrethroid resistant strain (FUMOZ-R) and a susceptible strain (FANG). Homozygous and heterozygous genotypes for each of these SNPs were observed. Importantly, the expected Mendelian ratio of 1:2:1 was respected in 27 of these 30 SNPs [7], confirming the polymorphism observed at these different positions. We can conclude from this result that the SNPs described in this study are then likely to be true polymorphisms rather than sequence artifacts and our scoring results indicate that they are suitable for use as genetic markers.
Relevance of the SNPs
The set of SNPs identified in this study provide a very useful tool for future genetic studies in An. funestus. These markers are of immediate use for association and QTL mapping studies. Some of these SNPs have been used for linkage mapping and identification of QTL involved in pyrethroid resistance in An. funestus [7]. This set of SNPs can be used as tools for population genetic studies in An. funestus. Genotyping large number of SNP markers will facilitate the study of genetic structure of natural populations and provide independent estimates of gene flow. It may provide additional markers to study the speciation process observed between the Folonzo and Kiribina chromosomal forms of An. funestus [29]. These markers may also be invaluable in monitoring insecticide resistance genes or genes involved in vector competence.
Conclusion
Through the sequencing of DNA fragments from 50 genes of An. funestus, we identified a set of 494 SNP markers and studied the pattern of genetic variability in this species. The distribution of SNPs in An. funestus was not neutral but under the influence of regional factors such as recombination, the degeneracy of the genetic code and selective constraints for gene conservation. The SNP markers described constitutes an important resource for more genetic studies in this important malaria vector.
Methods
Mosquito samples used for polymorphism discovery
We used adult female specimens of An. funestus from two laboratory strains, FANG and FUMOZ-R (seven specimens for each strain) as well as seven field specimens. FANG is a pyrethroid susceptible strain from Calueque, southern Angola and FUMOZ-R is a pyrethroid resistant strain from southern Mozambique [1]. Field specimens of An. funestus were collected from Kela village in Chikwawa district in southern Malawi.
Selection of gene sequences for SNP identification
Target genes were selected among cytochrome P450 genes for their putative involvement in insecticide resistance [30] or among genes of a broad range of functions that had been physically mapped to An. funestus polytene chromosomes [27] (see Table 4; Figure 3). They were also chosen to be distributed across the genome of An. funestus. The sequences of the physically mapped cDNAs were retrieved from Genbank. Determination of coding sequence, UTRs and intronic regions were done using the BLAST procedure through NCBI.
Figure 3
Relative location of studied genes on the An. funestus genome. For definitions of genes, see Table 4. This figure was adapted from [37].
Gene amplification and sequencing
Genomic DNA was extracted using the LIVAK method as described previously [31]. Primers were designed using Primer3 software [32] to flank putative intron sites to maximize the chance of SNP identification. Genomic DNA from 21 individuals (7 from FUMOZ-R, 7 from FANG and 7 from Kela) was amplified for each gene. PCR was performed with 10 ng of genomic DNA in a final volume of 25 μl containing, 2.5 μl Taq buffer, 0.2 mM of dNTPs, 10 pmoles of each primer, 2.5 mM of MgCl2, 0.2 unit of Taq polymerase (Qiagen). Amplification was performed with the following conditions: 1 cycle at 94°C for 3 min; 35 cycles of 94°C for 30 s, 57°C for 30 s and elongation at 72°C for 30 s; followed by 1 cycle at 72°C for 10 min. The annealing temperature was optimized for each primer pair and varied between 53°C to 62°C.
PCR products were purified using the QIaquick PCR purification kit (Qiagen) and directly sequenced on both strands using a Beckman CEQ 8000 automatic sequencer.
Sequence analysis and SNP detection
SNPs were detected as sequence differences in multiple alignments using Clustalw [33]. Electrophoregrams were visually inspected using BioEdit and heterozygotes were identified [12]. SNPs were identified as transitions or transversions in coding and non-coding regions. SNPs located within coding regions were classified as synonymous or non-synonymous and their codon position determined. Nucleotide diversity analyses were performed using DnaSP 4.0 [34]. The average number of nucleotide substitutions per site between two sequences, π was calculated for each gene as well as the haplotype diversity. The average number of synonymous substitutions per synonymous site (Ks) and non-synonymous substitutions per non-synonymous site (Ka) was computed according to [35].
SNP validation
Many of the SNPs discovered in this study were validated by different methods. As a part of an effort to construct a genetic map and to identify QTL involved in pyrethroid resistance, 30 SNP loci were genotyped in several families generated from a cross between FANG and FUMOZ-R strains of An. funestus. These SNPs were scored using a HOLA (Hot Oligonucleotides Ligation Assay) method [36], single base extension (SBE) using Beckman CEQ8000 and a pyrosequencing method [7].
Declarations
Acknowledgements
This work was supported by the National Institute of Health (NIH) grant IU01 A1058271-01. The authors acknowledge Keith Steen for technical help, John Morgan and Maureen Coetzee for providing the An. funestus specimens used in this study. We are grateful to three anonymous reviewers for their helpful comments on this manuscript.
Authors’ Affiliations
(1)
Liverpool School of Tropical Medicine, Pembroke Place
References
Hunt RH, Brooke BD, Pillay C, Koekemoer LL, Coetzee M: Laboratory selection for and characteristics of pyrethroid resistance in the malaria vector Anopheles funestus. Med Vet Entomol. 2005, 19 (3): 271-275. 10.1111/j.1365-2915.2005.00574.x.PubMedView ArticleGoogle Scholar
Sinkins SP, Hackett BJ, Costantini C, Vulule J, Ling YY, Collins FH, Besansky NJ: Isolation of polymorphic microsatellite loci from the malaria vector Anopheles funestus. Mol Ecol. 2000, 9 (4): 490-492. 10.1046/j.1365-294x.2000.00871-2.x.PubMedView ArticleGoogle Scholar
Cohuet A, Simard F, Berthomieu A, Raymond M, Fontenille D, Weill M: Isolation and characterization of microsatellite DNA markers in the malaria vector Anopheles funestus. Molecular Ecology Notes. 2002, 2 (4): 498-500. 10.1046/j.1471-8286.2002.00290.x.View ArticleGoogle Scholar
Cohuet A, Dia I, Simard F, Raymond M, Fontenille D: Population structure of the malaria vector Anopheles funestus in Senegal based on microsatellite and cytogenetic data. Insect Mol Biol. 2004, 13 (3): 251-258. 10.1111/j.0962-1075.2004.00482.x.PubMedView ArticleGoogle Scholar
Cohuet A, Dia I, Simard F, Raymond M, Rousset F, Antonio-Nkondjio C, Awono-Ambene PH, Wondji CS, Fontenille D: Gene flow between chromosomal forms of the malaria vector Anopheles funestus in Cameroon, Central Africa, and its relevance in malaria fighting. Genetics. 2005, 169 (1): 301-311. 10.1534/genetics.103.025031.PubMed CentralPubMedView ArticleGoogle Scholar
Michel AP, Ingrasci MJ, Schemerhorn BJ, Kern M, Le Goff G, Coetzee M, Elissa N, Fontenille D, Vulule J, Lehmann T, Sagnon N, Costantini C, Besansky NJ: Rangewide population genetic structure of the African malaria vector Anopheles funestus. Mol Ecol. 2005, 14 (14): 4235-4248.PubMedView ArticleGoogle Scholar
Wondji CS, Hunt RH, Pignatelli P, Steen K, Coetzee M, Besansky N, Lobo N, Collins FH, Hemingway J, Ranson H: An integrated genetic and physical map for the malaria vector Anopheles funestus. Genetics. 2005, 171 (4): 1779-1787. 10.1534/genetics.105.044800.PubMed CentralPubMedView ArticleGoogle Scholar
Cargill M, Altshuler D, Ireland J, Sklar P, Ardlie K, Patil N, Shaw N, Lane CR, Lim EP, Kalyanaraman N, Nemesh J, Ziaugra L, Friedland L, Rolfe A, Warrington J, Lipshutz R, Daley GQ, Lander ES: Characterization of single-nucleotide polymorphisms in coding regions of human genes. Nat Genet. 1999, 22 (3): 231-238. 10.1038/10290.PubMedView ArticleGoogle Scholar
Hoskins RA, Phan AC, Naeemuddin M, Mapa FA, Ruddy DA, Ryan JJ, Young LM, Wells T, Kopczynski C, Ellis MC: Single nucleotide polymorphism markers for genetic mapping in Drosophila melanogaster. Genome Res. 2001, 11 (6): 1100-1113. 10.1101/gr.GR-1780R.PubMed CentralPubMedView ArticleGoogle Scholar
Morlais I, Severson DW: Intraspecific DNA variation in nuclear genes of the mosquito Aedes aegypti. Insect Mol Biol. 2003, 12 (6): 631-639. 10.1046/j.1365-2583.2003.00449.x.PubMedView ArticleGoogle Scholar
Wang DG, Fan JB, Siao CJ, Berno A, Young P, Sapolsky R, Ghandour G, Perkins N, Winchester E, Spencer J, Kruglyak L, Stein L, Hsie L, Topaloglou T, Hubbell E, Robinson E, Mittmann M, Morris MS, Shen N, Kilburn D, Rioux J, Nusbaum C, Rozen S, Hudson TJ, Lipshutz R, Chee M, Lander ES: Large-scale identification, mapping, and genotyping of single-nucleotide polymorphisms in the human genome. Science. 1998, 280 (5366): 1077-1082. 10.1126/science.280.5366.1077.PubMedView ArticleGoogle Scholar
Black WC, Baer CF, Antolin MF, DuTeau NM: Population genomics: genome-wide sampling of insect populations. Annu Rev Entomol. 2001, 46: 441-469. 10.1146/annurev.ento.46.1.441.PubMedView ArticleGoogle Scholar
Moriyama EN, Powell JR: Intraspecific nuclear DNA variation in Drosophila. Mol Biol Evol. 1996, 13 (1): 261-277.PubMedView ArticleGoogle Scholar
Powell JR, Moriyama EN: Evolution of codon usage bias in Drosophila. Proc Natl Acad Sci U S A. 1997, 94 (15): 7784-7790. 10.1073/pnas.94.15.7784.PubMed CentralPubMedView ArticleGoogle Scholar
Holt RA, Subramanian GM, Halpern A, Sutton GG, Charlab R, Nusskern DR, Wincker P, Clark AG, Ribeiro JM, Wides R, Salzberg SL, Loftus B, Yandell M, Majoros WH, Rusch DB, Lai Z, Kraft CL, Abril JF, Anthouard V, Arensburger P, Atkinson PW, Baden H, de Berardinis V, Baldwin D, Benes V, Biedler J, Blass C, Bolanos R, Boscus D, Barnstead M, Cai S, Center A, Chaturverdi K, Christophides GK, Chrystal MA, Clamp M, Cravchik A, Curwen V, Dana A, Delcher A, Dew I, Evans CA, Flanigan M, Grundschober-Freimoser A, Friedli L, Gu Z, Guan P, Guigo R, Hillenmeyer ME, Hladun SL, Hogan JR, Hong YS, Hoover J, Jaillon O, Ke Z, Kodira C, Kokoza E, Koutsos A, Letunic I, Levitsky A, Liang Y, Lin JJ, Lobo NF, Lopez JR, Malek JA, McIntosh TC, Meister S, Miller J, Mobarry C, Mongin E, Murphy SD, O'Brochta DA, Pfannkoch C, Qi R, Regier MA, Remington K, Shao H, Sharakhova MV, Sitter CD, Shetty J, Smith TJ, Strong R, Sun J, Thomasova D, Ton LQ, Topalis P, Tu Z, Unger MF, Walenz B, Wang A, Wang J, Wang M, Wang X, Woodford KJ, Wortman JR, Wu M, Yao A, Zdobnov EM, Zhang H, Zhao Q, Zhao S, Zhu SC, Zhimulev I, Coluzzi M, della Torre A, Roth CW, Louis C, Kalush F, Mural RJ, Myers EW, Adams MD, Smith HO, Broder S, Gardner MJ, Fraser CM, Birney E, Bork P, Brey PT, Venter JC, Weissenbach J, Kafatos FC, Collins FH, Hoffman SL: The genome sequence of the malaria mosquito Anopheles gambiae. Science. 2002, 298 (5591): 129-149. 10.1126/science.1076181.PubMedView ArticleGoogle Scholar
Morlais I, Poncon N, Simard F, Cohuet A, Fontenille D: Intraspecific nucleotide variation in Anopheles gambiae: new insights into the biology of malaria vectors. Am J Trop Med Hyg. 2004, 71 (6): 795-802.PubMedGoogle Scholar
Aquadro CF, Bauer DuMont V, Reed FA: Genome-wide variation in the human and fruitfly: a comparison. Curr Opin Genet Dev. 2001, 11 (6): 627-634. 10.1016/S0959-437X(00)00245-8.PubMedView ArticleGoogle Scholar
Begun DJ, Aquadro CF: Levels of naturally occurring DNA polymorphism correlate with recombination rates in D. melanogaster. Nature. 1992, 356 (6369): 519-520. 10.1038/356519a0.PubMedView ArticleGoogle Scholar
Kern AD, Begun DJ: Patterns of polymorphism and divergence from noncoding sequences of Drosophila melanogaster and D. simulans: evidence for nonequilibrium processes. Mol Biol Evol. 2005, 22 (1): 51-62. 10.1093/molbev/msh269.PubMedView ArticleGoogle Scholar
Galtier N, Bazin E, Bierne N: GC-biased segregation of noncoding polymorphisms in Drosophila. Genetics. 2006, 172 (1): 221-228. 10.1534/genetics.105.046524.PubMed CentralPubMedView ArticleGoogle Scholar
Shen LX, Basilion JP, Stanton VP: Single-nucleotide polymorphisms can cause different structural folds of mRNA. Proc Natl Acad Sci U S A. 1999, 96 (14): 7871-7876. 10.1073/pnas.96.14.7871.PubMed CentralPubMedView ArticleGoogle Scholar
Schmid KJ, Tautz D: A screen for fast evolving genes from Drosophila. Proc Natl Acad Sci U S A. 1997, 94 (18): 9746-9750. 10.1073/pnas.94.18.9746.PubMed CentralPubMedView ArticleGoogle Scholar
Kimura M, Takahata N: Selective constraint in protein polymorphism: study of the effectively neutral mutation model by using an improved pseudosampling method. Proc Natl Acad Sci U S A. 1983, 80 (4): 1048-1052. 10.1073/pnas.80.4.1048.PubMed CentralPubMedView ArticleGoogle Scholar
Sharakhov IV, Serazin AC, Grushko OG, Dana A, Lobo N, Hillenmeyer ME, Westerman R, Romero-Severson J, Costantini C, Sagnon N, Collins FH, Besansky NJ: Inversions and gene order shuffling in Anopheles gambiae and A. funestus. Science. 2002, 298 (5591): 182-185. 10.1126/science.1076803.PubMedView ArticleGoogle Scholar
Lynd A, Ranson H, McCall PJ, Randle NP, Black WC, Walker ED, Donnelly MJ: A simplified high-throughput method for pyrethroid knock-down resistance (kdr) detection in Anopheles gambiae. Malar J. 2005, 4 (1): 16-10.1186/1475-2875-4-16.PubMed CentralPubMedView ArticleGoogle Scholar
Costantini C, Sagnon N, Ilboudo-Sanogo E, Coluzzi M, Boccolini D: Chromosomal and bionomic heterogeneities suggest incipient speciation in Anopheles funestus from Burkina Faso. Parassitologia. 1999, 41 (4): 595-611.PubMedGoogle Scholar
Amenya DA, Koekemoer LL, Vaughan A, Morgan JC, Brooke BD, Hunt RH, Ranson H, Hemingway J, Coetzee M: Isolation and sequence analysis of P450 genes from a pyrethroid resistant colony of the major malaria vector Anopheles funestus. DNA Seq. 2005, 16 (6): 437-445.PubMedView ArticleGoogle Scholar
Collins FH, Mendez MA, Rasmussen MO, Mehaffey PC, Besansky NJ, Finnerty V: A ribosomal RNA gene probe differentiates member species of the Anopheles gambiae complex. Am J Trop Med Hyg. 1987, 37 (1): 37-41.PubMedGoogle Scholar
Rozen S, Skaletsky H: Primer3 on the WWW for general users and for biologist programmers. Methods Mol Biol. 2000, 132: 365-386.PubMedGoogle Scholar
Thompson JD, Higgins DG, Gibson TJ: CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res. 1994, 22 (22): 4673-4680. 10.1093/nar/22.22.4673.PubMed CentralPubMedView ArticleGoogle Scholar
Rozas J, Sanchez-DelBarrio JC, Messeguer X, Rozas R: DnaSP, DNA polymorphism analyses by the coalescent and other methods. Bioinformatics. 2003, 19 (18): 2496-2497. 10.1093/bioinformatics/btg359.PubMedView ArticleGoogle Scholar
Nei M, Gojobori T: Simple methods for estimating the numbers of synonymous and nonsynonymous nucleotide substitutions. Mol Biol Evol. 1986, 3 (5): 418-426.PubMedGoogle Scholar
Black WC, Gorrochotegui-Escalante N, Duteau NM: Heated oligonucleotide ligation assay (HOLA): an affordable single nucleotide polymorphism assay. J Med Entomol. 2006, 43 (2): 238-247. 10.1603/0022-2585(2006)043[0238:HOLAHA]2.0.CO;2.PubMedView ArticleGoogle Scholar
Sharakhov I, Braginets O, Grushko O, Cohuet A, Guelbeogo WM, Boccolini D, Weill M, Costantini C, Sagnon N, Fontenille D, Yan G, Besansky NJ: A microsatellite map of the African human malaria vector Anopheles funestus. J Hered. 2004, 95 (1): 29-34. 10.1093/jhered/esh011.PubMedView ArticleGoogle Scholar
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.
