High-Throughput Isolation and Mapping of C. elegans Mutants Susceptible to Pathogen Infection

We present a novel strategy that uses high-throughput methods of isolating and mapping C. elegans mutants susceptible to pathogen infection. We show that C. elegans mutants that exhibit an enhanced pathogen accumulation (epa) phenotype can be rapidly identified and isolated using a sorting system that allows automation of the analysis, sorting, and dispensing of C. elegans by measuring fluorescent bacteria inside the animals. Furthermore, we validate the use of Amplifluor® as a new single nucleotide polymorphism (SNP) mapping technique in C. elegans. We show that a set of 9 SNPs allows the linkage of C. elegans mutants to a 5–8 megabase sub-chromosomal region.


Introduction
The relatively simple innate immune system of C. elegans and the number of traits that facilitate genetic and genomic analysis using this organism have nurtured rapid advances into the understanding of C. elegans innate immunity during the last few years (reviewed in [1][2][3][4][5]). However, traditional methods of isolating and mapping C. elegans mutants exhibiting aberrant immune responses to pathogen infection are often labor intensive and time consuming.
The conditions that dictate the isolation of C. elegans mutants exhibiting enhanced pathogen susceptibility (eps mutants) or enhanced pathogen resistance (epr mutants) depend on the killing mechanism of a given pathogen. For example, Pseudomonas aeruginosa is a fast killing pathogen of C. elegans which allows the rapid identification of mutants deficient in proper immune response. Indeed, the first screen to identify eps mutants was performed using this pathogen [6]. Because wild-type animals infected with P. aeruginosa typically start to die at approximately 34 hours, mutant dead animals were isolated during a period of 16 to 30 hours. Since C. elegans eggs are not infected by P. aeruginosa, the candidate mutants were recovered by transferring individual dead animals containing their brood to plates seeded with nonpathogenic E. coli. Such an approach would not be possible with any of the various slow-killing pathogens of C. elegans [3]. For example, in the case of infections by Salmonella enterica, hermaphrodite nematodes initially exposed to the pathogen need to be transferred each day to fresh plates to avoid losing track of these initial nematodes in the morass of progeny. This is tedious and time consuming, greatly reducing the number of killing assays and mutant isolations that can be performed.
S. enterica, as well as other Gram negative and Gram positive pathogens, has been described to cause a persistent infection, leading to distension of the C. elegans intestinal lumen [7]. Specifically, a persistent S. enterica infection occurs within 72 hours of initial exposure to the pathogen and it correlates with the premature death of the nematode [7]. Taking advantage of this correlation, we designed a novel approach to avoid the ratelimiting steps of isolating, propagating and screening thousands of individual mutant strains for an eps phenotype. We used a sorting system specifically designed to automate the analysis, sorting, and dispensing of C. elegans by measuring the length of the nematode and the intensity of fluorescent markers. Animals infected with S. enterica expressing GFP were analyzed, sorted according to the profile of bacterial accumulation in the gut, and then dispensed into 96-well plates for subsequent studies and mapping.
Traditionally, C. elegans mutations are identified using a combination of multi-point or deficiency mapping by crossing the mutant of interest to marker strains. More recently, alternative gene mapping techniques such as restriction fragment length polymorphism (RFLP) analysis have been developed [8,9]. These well established techniques, though dependable, can be timeconsuming and limited by the number and location of markers and RFLP single nucleotide polymorphisms (SNPs). The original SNP mapping methods are often laborious and expensive because they require multiple steps including restriction digests and size separation. A fragment length polymorphism (FLP) map was recently established for the automated gene mapping of C. elegans mutants based on small insertions or deletions (InDels) [10,11]. Although this mapping strategy greatly reduces the manual input required for techniques such as RFLP-SNP mapping, this system is limited by the number of polymorphisms that involve InDels. Even though InDels are ubiquitously dispersed across genomes, they constitute only 25% to 28% of the polymorphisms in the most commonly used Hawaiian mapping strain, CB4856 [8,12]. Recently, a new cost-effective, flexible and accurate highthroughput SNP genotyping technique, called AmplifluorH was developed and optimized for use in mouse and human systems [13,14]. We have optimized this technology to provide a fast and reliable method for mapping C. elegans mutants. Taken together, our results indicate that it is possible to use two high-throughput techniques to accelerate the isolation and mapping of C. elegans mutants susceptible to pathogen infection.

Results and Discussion
Isolation of C. elegans mutants based on the profile of S. enterica colonization of the intestine Typically, C. elegans animals are propagated in the laboratory by feeding them E. coli strain OP50 grown on a relatively low osmolarity medium. E. coli is effectively disrupted by the C. elegans pharyngeal grinder and essentially no intact bacterial cells can be found in the intestinal lumen. In contrast, when C. elegans is fed bacterial pathogens, intact bacteria can be found within the intestine. Interestingly, pathogens such as Enterococcus faecalis [15], Serratia marcescens [16], Yersinia pestis [17], and S. enterica [7] establish persistent infections in the C. elegans intestine which cannot be displaced by transferring the animals from pathogen lawns to E. coli lawns. In the case of S. enterica, specifically, a high titer of S. enterica persists in the intestinal lumen, ultimately killing the nematodes [7].
Taking advantage of S. enterica persistent infection of the C. elegans intestine, we have devised a strategy for identifying putative C. elegans mutants deficient in immune response to S. enterica infection by screening for mutants that exhibit enhanced pathogen accumulation (epa). We used S. enterica/GFP to infect a mutagenized population of nematodes and used the COPAS Biosort system to isolate those mutants which exhibited high GFP intensity at a time point when wild-type nematodes exhibited low GFP intensity. In a trial experiment, F1 mutagenized and nonmutagenized wild-type nematodes were fed S. enterica/GFP strain SMO22 for 24 hours. After extensive washing, these two groups of nematodes were loaded into the COPAS Biosort and the time of flight (TOF; which represents nematode length) and GFP intensity were recorded for each individual nematode. As shown in Figure 1, the non-mutagenized population showed a less dispersed pattern of bacterial accumulation, evaluated in terms of GFP intensity, than the mutagenized population. Thirteen out of a total of 23,191 first generation mutagenized nematodes displayed GFP intensities above the arbitrary baseline of 80 (0.06%), whereas only two nematodes out of a total of 10,087 (0.02%) from the nonmutagenized control group displayed GFP intensity of over 80 (Figure 1a-b). The trial experiment using F1 animals only identifies animals that have acquired a dominant epa mutation. Therefore, we expected to identify many more mutants in a screening using second generation (F2) mutagenized nematodes which should also help identify fixed recessive epa mutations. Indeed, 116 out of a total of 25,590 (0.45%) F2 mutagenized nematodes displayed GFP intensities above the arbitrary baseline of 80 (Figure 1c), whereas only one nematode out of a total of 10,087 (0.01%) from the non-mutagenized control group had a GFP intensity of over 80 ( Figure 1b).
All 116 epa mutants exhibiting high GFP intensity were isolated to individual wells in 96-well plates and allowed to propagate for further studies. Of the 116 epa mutants isolated, 41 mutants successfully propagated. The bacterial accumulation of each isolated mutant was confirmed using the fluorescence stereomicroscope. While most of the epa mutants showed accumulation of S. enterica expressing GFP in the intestinal lumen (Figure 2a), in a limited number of cases, S. enterica appeared to be invading all the tissues of the nematode (not shown). However, additional studies demonstrated that the dissemination of S. enterica throughout the different tissues of the mutants was due to the internal hatching of eggs, which disrupts tissues causing matricide, rather than to specific deficiencies in responses that prevent S. enterica invasion such as those mediated by TOL-1 [18].
S. enterica accumulation correlates with enhanced susceptibility to S. enterica-mediated killing S. enterica persistent colonization of the C. elegans intestine causes a remarkable luminal distension that is similar to that observed in mammalian gastritis and that correlates with C. elegans death. In addition, LPS is required for S. enterica-mediated killing of C. elegans, and is also required for persistent infection of the intestinal lumen [19]. More recently, S. enterica pathogenicity islands-1 and -2 (SPI-1 and SPI-2), and PhoP, which are required for virulence in both mammals and nematodes, were found to be required for persistent colonization in C. elegans [19,20]. These results indicate that S. enterica mutants in genes required for the killing of nematodes exhibit reduced colonization capabilities. This suggests that C. elegans mutants exhibiting enhanced accumulation of S. enterica may also exhibit enhanced susceptibility to S. enterica-mediated killing, To address whether it is indeed possible to identify eps (enhanced pathogen susceptibility) mutants based on increased accumulation of S. enterica, a subset of 25 mutants was infected with S. enterica and their survival monitored throughout the complete course of the infection. Figure 2b shows 19 of 25 mutants (76%) that were killed at significantly faster rates than wild-type animals (P,0.05), indicating that the increased accumulation of S. enterica correlates with an enhanced susceptibility to the pathogen in these mutants.

Validation of AmplifluorH SNP mapping for use in C. elegans
Traditional techniques of SNP mapping in C. elegans are often time-consuming and laborious. To automate the mapping of C. elegans mutants, we have optimized the AmplifluorH genotyping technology for SNP mapping in nematodes. This technology, which was originally developed and optimized for use in mouse and human systems [13,14], is both flexible and accurate.
In the AmplifluorH technique, each PCR reaction mixture contains five primers: two fluorescently labeled AmplifluorH primers, two allele specific forward primers, and a single common reverse primer ( Figure 3). The two AmplifluorH primers contain a green or red fluorophore and a quencher that becomes separated when the primer is incorporated in a PCR product allowing the fluorescence to occur. The single common reverse primer and the two allele specific forward primers, which are designed to amplify across the SNP, have a 59 tail that corresponds to one of the two labeled AmplifluorH primers. As allele-specific PCR products are generated, the corresponding AmplifluorH primer recognizes the complement of the tail sequence and is able to prime off the newly generated PCR product. This separates the fluorophore from the quencher and generates a fluorescent signal. Homozygotes produce either a green or red signal while heterozygotes produce a yellow signal combined for both fluorophores. Depending upon which fluorescent label is tagged to which allele-specific primer, recombinants homozygous for wild-type N2 or CB4856 alleles cluster along the red or green axis. Heterozygote animals are represented by a cluster in the middle of the two axes ( Figure 4).
First, we tested this technique by using it to map the mutation in glp-4 nematodes, which fail to develop a germline at 25uC. We chose this mutant because the failure of glp-4 nematodes to produce a germline at the restrictive temperature provides a rapid and simple phenotype to assay. The C. elegans mutant glp-4 was crossed to the Hawaiian mapping strain CB4856 and 93 recombinant progeny exhibiting the glp-4 phenotype were genotyped. In the initial round of gene mapping, a single SNP near the center of each chromosomal linkage group (LG) was analyzed. For LGII, LGIII, LGIV, LGV and LGX (Figure 4), the 93 recombinants were distributed according to normal Mendelian segregation for a recessive mutation, suggesting that the glp-4 mutation is unlinked to any of these chromosomes. However, the SNP analyzed on LGI resulted in a dramatic shift of the recombinants away from the homozygous wild-type allele, suggesting linkage to LGI ( Figure 4). Three additional SNPs flanking the currently predicted location of glp-4 (http://www.wormbase.org) on LGI were used to further genotype the 93 recombinants ( Figure 4). These data linked the glp-4 mutation to a 235 Kb region on the right arm of chromosome I containing 97 predicted genes, which overlaps with the predicted glp-4 location described in the literature (Figure 4). Although the label color varied between the wild-type N2 and mapping strain CB4856 alleles, the called genotype of each SNP in the wild-type and mapping strain controls were always correct, therefore color bias did not appear to be a problem in the Amplifluor analysis. These data indicate that AmplifluorH provides a fast and reliable method for gene mapping mutants and that it is a powerful new method for SNP mapping in C. elegans.

Mapping of C. elegans mutant epa-18
Upon validation of the AmplifluorH SNP mapping technique for use in C. elegans, we chose epa-18 mutant to begin mapping. This mutant was chosen because it appears to affect a pathway required for immunity to bacterial pathogens in general but not for survival in the absence of potentially pathogenic bacteria. As shown in Figure 5, epa-18 animals are susceptible to infection by Gram-negative bacterial pathogens S. enterica and Pseudomonas aeruginosa and by the Gram-positive bacterial pathogen Enterococcus faecalis. However, the life-span of epa-18 animals is not reduced when the animals are grown on aging plates containing E. coli and 5-fluoro-29-deoxyuridine, which prevents the animals from being killed by replicating bacteria. This indicates that epa-18 animals are specifically susceptible to live, pathogenic bacteria and that they are not sickly.
To map the epa-18 mutation, we took advantage of the enhanced pathogen (P. aeruginosa) phenotype of epa-18 mutants that is easier to track than the enhanced pathogen accumulation phenotype when the COPAS Biosort system is not used. Thus, hermaphrodite epa-18 mutants were crossed to male CB4856 and the DNA from 93 recombinant progeny exhibiting enhanced susceptibility to P. aeruginosa were isolated for SNP gene mapping in a 96-well plate including 3 controls; 1) N2 DNA, 2) CB4856 DNA and 3) water template. For LGI, LGII, LGIII, LGIV, and LGX (Figure 6a), the 93 recombinants were distributed according to normal Mendelian segregation for a recessive mutation, suggesting that the epa-18 mutation is unlinked to any of these chromosomes. However, the SNP analyzed on LGV resulted in a dramatic shift of the recombinants away from the homozygous wild-type allele, suggest-  . AmplifluorH technique. The two AmplifluorH primers contain a green or red fluorophore and a quencher that becomes separated when the primer is incorporated in a PCR product allowing the fluorescence to occur. The single common reverse primer and the two allele specific forward primers, which are designed to amplify across the SNP, have a 59 tail that corresponds to one of the two AmplifluorH primers. As allele-specific PCR products are generated, the corresponding AmplifluorH primer recognizes the complement of the tail sequence and is able to prime off the newly generated PCR product. This separates the fluorophore from the quencher and generates a fluorescent signal. Homozygotes produce either a green or red signal while heterozygotes produce a yellow signal combined for both fluorophores. doi:10.1371/journal.pone.0002882.g003 predicted genes (Figure 7a). Thus, by taking advantage of the 93 individual recombinants, we were able to map the epa-18 mutation to a single Mb sub-chromosomal region. Given the high number of polymorphisms present in CB4856, the limiting factor in C. elegans SNP mapping is the number of recombinant animals used for mapping. Genotyping a larger number of recombinants should allow for finer mapping.
In order to validate the AmplifluorH genotyping data, we used the well established RFLP-SNP analysis as an additional mapping approach. The RFLP-SNP system [8,9], which is an accurate and inexpensive mapping methodology that does not require specialized equipment, takes advantage of polymorphisms that alter restriction enzyme recognition sites. Using RFLP-SNP analysis, a region of ,1.5 Mb, which overlapped the AmplifluorH-mapped region, was identified (Figure 7b).

Conclusions
Traditional methods of identifying and mapping unknown C. elegans mutants exhibiting aberrant immune responses to pathogen infection are often labor intensive and time consuming, greatly reducing the number of immune genes that can be identified. Here, we have devised approaches that use high-throughput methods of isolating and mapping C. elegans mutants susceptible to pathogen infection. We show that C. elegans mutants deficient in immune responses can be isolated based on an enhanced accumulation of bacterial pathogens in the intestine that can be rapidly identified using the COPAS Biosort system. In addition, we have applied the AmplifluorH genotyping to the mapping of C. elegans mutants and demonstrated that it is possible to link a C. elegans mutation to 5-8 Mb sub-chromosomal regions using only 9 SNPs.
Taking advantage of S. enterica persistent infection of the C. elegans intestine, we have devised a strategy for identifying putative C. elegans mutants deficient in immune response to S. enterica infection by screening for mutants that exhibit enhanced pathogen accumulation (epa). This strategy was easily automated by using the COPAS Biosort system, allowing for the rapid screening of thousands of epa mutants. Further studies demonstrated that the enhanced accumulation of S. enterica correlates with enhanced susceptibility to S. enterica. In addition, the increased susceptibility to different bacterial pathogens of mutant epa-18 indicates that the approach can be used to genetically dissect pathways involved in immune responses to pathogens in general. By coupling this approach with the AmplifluorH genotyping system for use in C. elegans, we provide a high-throughput method of isolation and mapping mutants deficient in immune responses. Our results also indicate that AmplifluorH genotyping is a powerful approach for the rapid mapping of any C. elegans mutant.

Materials and Methods
Mutagenesis EMS (ethane methyl sulfonate) mutagenesis was performed as previously described [21,22]. Briefly, the wild-type strain Bristol N2 was mutagenized with 50 mM EMS for 4 hours at 20uC as described [21,22]. This is expected to generate ,220 G/CRA/T transition mutations per haploid genome, R50 of which cause amino acid mutations in protein coding genes [21,22]. From the mutagenized worms, progeny were harvested and allowed to selffertilize in order to fix induced mutations. Non-mutagenized control nematodes were treated in an identical manner.

Isolation of epa C. elegans mutants
Both mutagenized and non-mutagenized nematodes were grown to one day old gravid adults on E. coli strain OP50 [23] before transfer to S. enterica/GFP strain SMO22 [24] for 24 hours. Nematodes were harvested in M9 buffer, washed in M9 three times for ten minutes and resuspended in 30 mL S-basal buffer for 30 minutes to remove any external bacterial cells not persistently attached to the C. elegans intestine. To select nematodes which contained high numbers of S. enterica/GFP proliferating in the gut, fluorescence was measured using the COPAS Biosort (Union Biometrica) system which reports time of flight (TOF; nematode length, measured in arbitrary COPAS Biosort units) and green fluorescence (GFP-induced fluorescence, measured in arbitrary units) for each individual animal.
C. elegans killing assays C. elegans wild-type Bristol N2 animals and mutants were maintained as hermaphrodites at 20uC, grown on modified nematode growth medium (NGM) agar plates, and fed with E. coli strain OP50 as described [23]. S. enterica strain SL1344 [25], Psuedomonas aeruginosa strain PA14 [26], and Enterococcus faecalis strain OG1RF [27] cultures were grown in Luria-Bertani (LB) broth at 37uC. S. enterica and P. aeruginosa bacterial lawns used for C. elegans killing assays were prepared by spreading 20 ml of an overnight culture of bacteria on modified NGM agar (0.35% instead of 0.25% peptone) in plates 3.5 cm in diameter. E. faecalis bacterial lawns were prepared by spreading 20 ml of an overnight culture on brain-heart infusion (BHI) agar in plates 3.5 cm in diameter. The S. enterica and E. faecalis killing assays were performed at 25uC, and animals were scored and transferred once a day to fresh plates. P. aeruginosa killing assays were performed at 20uC, and animals were scored and transferred once a day to fresh plates. Animals were considered dead when they failed to respond to touch.
For the life-span assays, bacterial lawns were prepared by spreading 50 ml of a 106 concentrated overnight E. coli culture on modified NGM agar (0.35% instead of 0.25% peptone) containing 100 mg/ml 59-fluoro-29-deoxyuridine in plates 3.5 cm in diameter. Life-span assays were performed at 25uC, and animals were scored once a day. Animals were considered dead when they failed to respond to touch.

Statistical Analyses
Animal survival was plotted as a nonlinear regression curve with the PRISM 4.00 computer program. Survival curves are considered significantly different from the control when P,0.05. Prism uses the product limit or Kaplan-Meier method to calculate survival fractions and the logrank test, which is equivalent to the Mantel-Heanszel test, to compare survival curves.

Generation of Hawaiian recombinant mutant strains
For glp-46CB4856 recombinant analysis, five glp-4 hermaphrodites were placed with 10 Hawaiian CB4856 males for mating at 20uC overnight. After 24 hours, males were removed and hermaphrodite glp-4 nematodes were isolated to separate plates and transferred to 15uC, which is the permissive temperature for glp-4. Thirty F1 progeny were isolated from a single successful mating and allowed to self-fertilize. Twelve F2 progeny were collected from each F1 progeny (a total of 360) and transferred to separate plates for self-fertilization. Adult F3 nematodes were phenotyped for lack of germline when grown at 25uC.
After all 360 recombinant F2 lines had begun laying eggs, each F2 animal was removed from the plate and transferred to a second plate to continue egg laying. The first set of plates was kept at 15uC to keep the lines fertile, and the second set of plates was transferred to 25uC, where the F3 progeny could be screened for loss of germline.
For epa-186CB4856 recombinant analysis, five epa-18 hermaphrodites were placed with 10 Hawaiian CB4856 males for mating at 20uC overnight. After 24 hours, males were removed and hermaphrodite epa-18 nematodes were isolated to separate plates. Thirty F1 progeny were isolated from a single successful mating and allowed to self-fertilize. Twelve F2 progeny were collected from each F1 progeny (a total of 360) and transferred to separate plates for self-fertilization. Adult F3 nematodes were phenotyped for an increased susceptibility to P. aeruginosa (as described in the Methods section for C. elegans killing assays) and positive recombinants were used for genotyping.

C. elegans recombinant analysis using AmplifluorH
AmplifluorH assays were designed and performed as recommended by the manufacturer (Serologicals, Georgia, USA). DNA lysates were generated by suspending 50 nematodes in 100 ml lysis buffer (50 mM NaCl, 10 mM Tris-Cl pH 7.5, 2.5 mM MgCl2, 0.45% Tween 20, 0.01% gelatin, 0.2 mg/ml proteinase K) and lysed at 65uC for one hour. 1 ml of this crude lysate was used as the PCR template in the PCR reaction. Each PCR reaction mixture contained five primers: two allele-specific primers, each one containing one of two different sequences at its 59-end, and a 39part corresponding to the target sequence with alternating nucleotides at the 39-end, a common reverse primer, and two fluorescently labeled AmplifluorH primers matching the 59-portion of the allele-specific primers. The two common primers include either a green or a red fluorophore and a quencher that forms a loop in the free primer but becomes separated when the primer is incorporated in a double-stranded PCR product. A fluorescent signal is emitted only when the fluorophore and the quencher are beyond a set distance from each other. In the course of the PCR reaction, only the primers that anneal perfectly to the target DNA amplify a DNA fragment. The physical position of each SNP used in AmplifluorH mapping is shown in (Figure S1).
The DNA from each recombinant progeny exhibiting the mutant phenotype is assigned to a specific well in a 96-well plate LGV within the previously mapped region were used to genotype each of the 93 epa-18 recombinant nematodes. Only data from the two most informative SNPs are shown. The relative location of each genotyped SNP is shown along the length of LGV. The region linked to epa-18 is marked in red. (b) RFLP-SNP analysis data confirms the region defined by AmplifluorH genotyping. The percent of wild-type homozygote recombinants for each SNP is shown. The mapped region for epa-18 using RFLP-SNP analysis overlaps with that from AmplifluorH genotyping, confirming the validity of the AmplifluorH system. doi:10.1371/journal.pone.0002882.g007 and remains the same for each individual reaction. This makes it possible to record the genotype of each SNP analyzed for each individual recombinant progeny. Although the mutated chromosome is identified by a significant shift in allele distribution, the final candidate region containing the mutation is determined by identifying the position of the crossover event in each recombinant progeny. In other organisms, double recombination events could interfere with this type of analysis. However, in C. elegans, the frequency of double recombinants is essentially zero due to the nematode's highly efficient crossover interference of 1 [28]. Therefore, the chance of double recombinants is negligible and would not compromise the interpretation of the data.

Primer design for AmplifluorH analysis
The allele-specific PCR primers and the common reverse primers were designed from each surrounding SNP sequence using AmplifluorH AssayArchitect software (http://www.assayarchitect. com/, Chemicon International, a division of Serologicals Corporation, Norcross, GA, USA). Oligonucleotides were synthesized and HPLC purified by MWG Biotech (Ebersberg, Germany). One unique 21-nucleotide tail that is complementary to the Universal AmplifluorH primer [29] was added to the 59 end of each allelespecific primer [13]. Fluorescent measurements and data analysis for AmplifluorH were performed as previously described [29]. The sequences of the PCR primers (two allele-specific primers, and the reverse primer per SNP) used to amplify each of the SNPs are shown in (Table S1).

Primer design for RFLP analysis
RFLP-SNPs and surrounding sequences were obtained from the C. elegans SNP database (http://genome.wustl.edu/genome/celegans/celegans_snp.cgi). Primers were designed using the Primer 3 program (http://frodo.wi.mit.edu/cgi-bin/primer3/primer3_ www.cgi). Oligonucleotides were synthesized and HPLC purified by MWG Biotech (Ebersberg, Germany). The sequences of the primers used to amplify the surrounding sequences for each SNP are shown in (Table S2).

PCR conditions for RFLP analysis
The same DNA lysates that were used for the PCR step during AmplifluorH analysis were used as the DNA template during RFLP-SNP analysis. 2 ml of this crude lysate was used as the PCR template. The PCR reactions contained 100 mM of each the forward and reverse primer, 125 mM dNTPs, 0.5 U Choice-Taq TM DNA polymerase and 16 Choice-Taq TM PCR buffer (Denville Scientific Inc., Metuchen, NJ, USA), in a 25 ml total volume. All reactions were performed in 96-well PCR plates sealed with an adhesive cover as follows: initial denaturation at 95uC for 2 minutes, followed by 30 cycles of 95uC denaturation for 30 seconds, 55uC annealing for 30 seconds, 72uC extension for 30 seconds, and ending with a 7 minute final extension at 72uC.

Restriction digest conditions
All restriction digests were performed using New England BioLabs, Inc (NEB) enzymes and the recommended corresponding NEB buffer in a total reaction volume of 20 ml. A 26digestion master mix was made as follows: 26 NEB buffer, 26 BSA, 3 U NEB enzyme to a total volume of 1 mL. 10 ml of each PCR reaction was added to 10 ml of 26 digestion master mix in a new 96-well plate and incubated at the appropriate temperature for 5 hours. Digested PCR fragments were resolved on 2% agarose gels and visualized with ethidium bromide staining.