Isolation and DNA-Marker Based Genotyping of Individual Pollen Grains

Dry pollen of diploid Solanum clone 9507-04, Cotton chromosomal substitution line and Echinacea purpurea „Magnus‟ were isolated and their cells contents released in individual tubes containing water or pollen germination media, respectively.Six polymorphic markers from individual pollen grains of the diploid potato were amplified using two random amplified polymorphic DNA (RAPD) primers, i.e., UBC291 and UBC504. The procedures for primer extension pre-amplification (PEP) for genome and subsequent amplified fragment length polymorphism (AFLP) analysis with 13 primers pairs were tested on the upland cotton line. The cotton parental AFLP markers were scored for their presence or absences in the 35 pollen samples and majority of the markers were found monomorphic as expected from the gametes of a near isogenic line.PEP and AFLP procedures for linkage mapping were standardized usingpurpleconeflower pollen grains. From 64 possible AFLP primer pairs,37.5% amplified 104 markers for coneflowerthat were linked in 11 linkage groupsranging from 13 to 134 centimorgans (cM). The 409 cM sizedlinkage map constructed forE. purpurea „Magnus‟genome had average distance of 2.4 cM to 22 cMfor the two to 55 polymorphic markers identifiedwithin the 11 groups.The protocols described in this report would serve to enhance breeding efficiencies in plantsimportant for food,fiber, ornamental and medicinal industries.


Introduction
Molecular markers" based plant evaluation for cultivar release is more efficient in terms of time and development costs than conventional breeding approaches [1].
The segregation of parental markers,during anthesis, into its pollen grains can be used by plant breeders for genetic analyses. Plant breeding efficiencies can be enhanced if such pollen based segregation of molecular markers is used in genetic linkage estimations [2].Pollen is a convenient source of DNA as they require only limited freezer storage space and are easily stored for long periods [3,4].DNA typing of individual pollen grains has several uses including assessment of gamete merit. Pollen have been subjected to selection processes since pollen characteristics such as resistance to different forms of stresses have been used to preselect pollenfor fertilization in subsequent genetic hybridizations [5]. In the development of polyploid varieties unreduced (2n) pollen,that can be produced by first or second division restitution [6,7], are used. Unreduced pollen produced by nondisjunction at first meiotic division would retain the heterozygous loci of the parent, whereas nondisjuction at the second meiotic division would produce the 2n pollen segregating for dominant and recessive loci of the parent [8]. Therefore, the segregation analysis of molecular markers in unreduced pollen would help to understand the origin of 2n pollen production.Amplified fragment length polymorphism (AFLP) analysis of rose showed that they produce unreduced male and female gametes [9]. Pollen protoplasts also contain several organelles which contribute to markers" amplification products [4,10], therefore such molecular markers could also be used to study the inheritance of organelles in the pollen populations. AFLP markers have been used to identify male fertility restorer locus [11].Molecular markers" based analysis of individual gametes can be used to compute genetic linkage relationships by determining whether the individual gametes are parental or recombinant in genotype [2,12]. Generation of AFLPs" based data from individual pollen grains for genetic mapping is efficient and cost approach since it can circumvent the need to maintain large plant populations whileconducting controlled pollinations [13,14]. Two approaches have been reported to extract DNA from isolated pollengrains. One approach would be to capture single pollen grain in a PCR tube after drilling a hole in the wall of a pollen grain with a UV-laser microbeam [14]. Alternatively a pollen grain isolated from dry spread was dispersed in liquid to release its cell content [4] .Mature pollen grains contain three haploid nuclei, i.e., one vegetative nucleus and two sperm nuclei [10]. Although individual pollen grains contain very limited amounts of DNA, polymerase chain reaction (PCR) techniques can be used to amplify DNA from genomic (chromosomal) sequences [4]. Since PCR is capable of amplifying a DNA sequence on the order of 10 9 fold [12], PCR-based molecular markers in an individual cell can be analysed. For single gamete genotyping, random amplified polymorphic DNA (RAPDs)and AFLPs are more useful than other PCRbased markers since these can be used without any prior genome information and analysed with less effort [15]. In RAPD and AFLP analyses, the dominant markers are identified by the presence of DNA bands and recessive markers are identified by the absence of the DNA bands [16]. Despite dominance of the markers, RAPD and AFLP can provide the resolution needed with less effort [15]. Use of the gametophytic tissue in RAPD and AFLP analysis can facilitate the detection of co-dominant markers [17] which are otherwise rarely discernible [16]. Since a heterozygous genotype (for an allelic pair) segregates into gametes carrying dominant and recessive markers [17], analysis of pollen via RAPD and AFLP markers would simplify the genetic assessment of the parent since only 1:1 segregation of the markers is expected [2,13]. AFLP based markers are more reliable than RAPDs, since it is based on restriction digest and stringent PCR conditions [1]. However, restriction digest, which is needed for AFLP analysis,is not feasible with the limited amount of DNA available in single cells.Though single cell gametes have sufficient DNA for PCR amplification [18], the small amount of DNA in a pollen grain needs to be increased for subsequent AFLP analysis. The primer extension pre-amplification (PEP) protocol of Zhang [22].Gene Specific-Primer Extension Preamplification (GS-PEP) technique has been developed that amplifies gene of interest after increasing the copy number of genomicDNA [23]. Some PEP based sequence specific methods have been patented by combining randomized primers in the first amplification reaction and specific primers in the second amplification reaction [24]. This study standardizes the individual pollen grain isolation and DNA harvest for subsequent PCR protocols using three unrelated genomes, i.e.,diploid potato clone 9507-04, upland cotton line and purple coneflower plants.RAPD polymorphism from individual pollen grains of the diploid potato was also detected. The protocols for PEP amplification of genome and AFLP analysis were standardized usingupland cotton line (with pima cotton"s chromosome 17) and Echinaceapurpurea "Magnus"pollen grains. The procedures described in this report for isolation of individual pollen grain, release of pollen DNA for subsequent PCR amplification as well as molecular marker and genetic linkage analysis from individual pollen grains would serve as important tools for enhancing breeding efficiencies in plants.

Materials and Methods Plant Materials and Growth
Diploid potato (Solanum) clone9507-04 which had several species in its background (50% tuberosum and 50% wild species including S. chacoense and Argentinean wild potato species) were obtained and maintained in greenhouse as described . Four µl of RNaseA stock (100 mg/ml) solution was used to remove RNAs. Manufacturer"s QIA-shredder spin column under centrifugation removed the debris. Precipitated DNA was then collected on DNeasy column membrane and eluted. DNA concentrations were quantified using a NanoDrop 1000 Spectrophotometer (Thermo Fisher Scientific, Wilmington, DE)after verifying DNA isolation via 1% agarose gel (in TAE buffer) electrophoresis and using ethidium bromide (0.5 µgml -1 in TAE buffer) stained gels that were visualized on UV trans-illuminator and photographed (AlphaImager 2000 system, Alpha Innotech, San Leandro, CA).

Gel Electrophoresis
ForRAPD-PCR products"separationhorizontalagarose gel electrophoresissystem (Fisher Scientific, Pittsburgh, PA, USA)on 1.4% agarose gels (3.5Vcm-1 for 3 hrs)stained with ethidium bromide (0.5µgml-1) in 0.5X TBE (0.045M Tris, 0.045M Boric acid, 0.001 M EDTA) was used. Sizes of the ethidium bromide stained RAPD markers were determined by comparing with a DNA Mass™ Ladder (Invitrogen TM Life Technologies, Carlsbad, CA, USA)after being visualized/ photographed over a UV transilluminatorusing an AlphaImager 2000 System (Alpha Innotech, San Leandro, CA, USA). Before analysis on the DNA Sequencer, 3X blue loading dye (Li-Cor) was added to the AFLP samples. As described earlier [2, 27], the AFLP profiles from parent and from each pollen DNA sample (~0.6 µl) were and separated through electrophoresis using denature 6.5% polyacrylamide gel between 25cm 2 Soda-Lime (for single IRD-dye analyses) or Starphire® (for dual IRD-dyes analyses) glass plates that were 0.25mm apart. These DNA bands were scored, by simultaneously electrophoresing and analyzing the selective amplification products for approximately 3.5 hours, on Li-Cor"s automated DNA analyzer (Global IR2 for coneflower or Global 4300 for cotton sample). The Saga TM Generation 2 Version 3.1 (Li-Cor) software was used for maintaining constant electrophoresis conditions of 1500V, 40W and 40mA at 45°C, and for scoring the presence or absence of parental AFLP markers in each pollen sample. UnderWindows XP operating system (Microsoft Corporation, Redmond, WA, USA) the Saga TM software, lane and calibration modes was used to prepare gel for scoring.

Markers' and Linkage Analyses
Reproducible RAPD-PCR products from the potato parent, used as markers, were visually scored for their presence or absence in the pollen progeny. For each RAPD marker, goodness-of-fit to expected segregation of 1:1 (presence of DNA band: absence of DNA band) was tested by chi-square analysis using SYSTAT version 7.0 (SPSS Inc., Chicago, IL, USA). Segregation ratios that differed from the expected value (significant at P=0.05 or less) were classified as distorted [13]. Desmile lines(Saga TM software) were added to the gel pictures of cotton (IRD 700 and 800 based) and coneflower (only IRD-800 based) samplesusing monomorphic DNA-bands in the gel-lanes. Gels were then scored for the presence or absence of AFLP markers denoted by "+" and "-". Scored data were added to a database (Oracle® 9i) to generate reports consisting of segregation pattern(absence/presencein entire pollen progeny) strings for each loci. Coneflower reports combined in a file were analyzed by genetic mapping software (JoinMap® 3.0, Plant Research International, Wageningen, Netherlands) for genetic linkage analysis and converting recombination fractions to map distances with Kosambi mapping function [2]. Linkage groups were thus constructed by using likelihood of odds (LOD) threshold of 3.0 and software"s default maximal recombination fraction of 0.40 to determine the most likely marker orders inside groups. By lowering the LOD score to 2, unlinked loci were incorporated in linkage group one [2].

Results and Discussion
The segregation of parent plant"s markersinto its pollen grains during anthesis can be used for genetic analysis, which can be facilitated by single gamete genotyping. Using pollen of important food (potato) and nursery/medicinal (coneflower) crops this study standardizes the individual pollen grain isolation and DNA-fingerprinting protocols.
Dry pollen fromSolanumdiploid clone 9507-04, cotton hypoaneuploid lineCS-B17 and Echinacea purpurea "Magnus" were successfully isolated individually.Observing under microscope a micromanipulator was used to move an insect-pin (tip: ~0.1mm thickness) to touch individual pollen (Fig. 1) which would then adhere to the pin.Individual pollen grains were convenient to isolate, by adherence to tip of insect pin, from dry spread. Unlike too flexible hair-point [28], insect pin would move accurately with the help of micro-manipulator over the requisite location on microscope slide. Isolated single cells contents were released in individual tubes containing water or pollen germination media, respectively (Fig.  2).In agreement with Petersen et al. [4] and Krabelet al. [28] the pollen released their contents into water and germination media, respectively. Typically 60% pollen thus isolated would succeed in PCR amplification. A few nonspecific and low intensity PCR products would also be discernible, however, it is unlikely that such extraneous DNA is in quantitative competition [29] with pollen DNA or causes significant errors in genotyping. Six polymorphic markers from individual pollen grains of the diploid potato were amplified using two random amplified polymorphic DNA (RAPD) primers, i.e., UBC291 and UBC504. Some RAPD markers showed expected ratio of 1:1 segregation of 'presence of band: absence of band' while others showed distorted ratio. No linkage analysis was conducted for potato pollen grains since only six polymorphic RAPD markerswere amplified. Since only those pollen samples that were successful for amplification of the PEP products as well as the parental AFLPs were used for subsequentmarker analysis, some distorted segregation was expected. The procedures for primer extension preamplification (PEP) for genome and subsequent amplified fragment length polymorphism (AFLP) analysis with 13 primers pairs (Table 3) were tested on the upland cotton line [30]. Many genomic copies in fragments of varied lengths were produced through PEP procedure by amplifying the tri-nucleate DNAs of coneflower and cotton pollen grains [2,30]. The MasterAmp™ Long PCR kit (EPICENTRE ® , Madison, WI) was used since its enhancers increased the fidelity and accuracy of the resulting PEP products [31]. Thus larger sized (>300 base pairs) andincreased numberof AFLP markers were yielded when pollen PEP amplifications were conducted along withMasterAmp™ Long PCR kit.The parental AFLPsof cotton were markedas present or absentin the 35 pollenPEP products and majority of the markers were found monomorphic [30].No linkage estimations were needed for cotton pollen grains since majority of the AFLP markers amplified from those samples were found monomorphic, as expected from the gametes of a near isogenic chromosomal substitution line [25,26].PEP and AFLP procedures for linkage mapping were standardized usinggerminatedpurple coneflowerpollen. Table 4 lists all the primer pairsthat were identified for AFLP amplifications for E. purpurea "Magnus" genome. The 104 parental coneflower AFLPs were placed in 409 centimorgans(cM) linkage mapafter pollen analyses [2]. The resulting 11 linkage groups ranging from 13 cM to 134 cM in sizes had two to 55 polymorphic markers each with average mapping distance of 2.4 cM to 22 cM.Theprotocols described in this report provide an opportunity to directly observe the meiotic division products. Such innovative genetic analyses tools would serve to enhance breeding efficiencies in plants important for food,fiber, ornamental and medicinal industries. IRDye® 800 only monomorphic bands amplified *Each AFLP (amplified fragment length polymorphism) marker is identified by the primer-pair designed for both EcoRI (E) and MseI (M) adaptors, and the three selective nucleotides of each primer 1 Primers designed for EcoRI adaptors had been labelled by Li-Cor (Inc., Lincoln, NE) IRDye® Phosphoramidite florescent dyes optimized for Li-Cor 700 nm (IRD-700) or 800 nm (IRD-800) infrared channels.