A Method of Producing STS (Sequence Tagged Site) markers from RAPDs (Random Amplified Polymorphic DNAs) in Ceratopteris richardii

Cora Chaffin (Provine High School, Jackson, Mississippi) and  Robert Hamilton (Mississippi College, Clinton, Mississippi)


Genes are arranged in specific sequences along chromosomes, with each gene occupying its own particular position (locus) on a chromosome. Together the genes on a chromosome form a linkage group in which the relative order of loci corresponds to the arrangement of genes in the chromosomal DNA. Analyses of gene linkage has allowed for the construction of linkage maps indicating the relative locations of genes in the chromosomal DNA. The location of genes along chromosomes are described in terms of their locations relative to identifiable genetic markers.

Traditionally the genetic markers used to construct linkage maps have been mutations that identify loci on the basis of mutant phenotypes. More recently differences in chromosomal DNA that can be detected without a change in phenotype have been used as markers in the construction of linkage maps. Such DNA sequence differences are especially useful markers because they are plentiful and easy to characterize precisely.

Restriction fragment length polymorphisms (RFLPs) are DNA markers that have been widely used to construct linkage maps (Paterson, 1996). However, RFLP analysis is a laborious and time-consuming procedure that requires previous information of the genome. RFLPs would not be not useful for rapidly generating large numbers of DNA markers for species whose genomes are relatively unknown.

Random amplified polymorphic DNAs (RAPDs) is a method of PCR where arbitrarily chosen 10 base primers are used to search for variation in DNA (Williams et al., 1990). RAPDs data can contain artifacts (Ellsworth et al. 1993), and are not fully reproduceable (Paterson 1996). However, RAPDs have been be used to generate large numbers of genetic markers useful for linakge mapping quickly and cheaply (Antolin et al., 1996).

Paran and Michelmore (1993) were able to separate RAPD fragments on an agarose gel, excise the bands from the gel and reamplify individual bands from gel slices using the original RAPDs primer. They were then able to clone and sequence the PCR products and use sequence data to design PCR primers specific to specific RAPDs fragments, and use PCR to produce specific RAPD's fragments from genomic DNA, which then can function as sequence tagged sites (STSs). This method allows for rapid generation of STSs derived from RAPDs fragments, and eliminates the problems associated with reproduceability associated with RAPD analyses. Reamplification from genomic DNA and subsequent sequencing of the PCR products also allows for the identification of any artifacts.

The problem with Paran and Michelmore's (1993) method is poor resolution of bands on an agarose gel. Multiple similar sized DNA fragments may be present in one "band" on a an agarose gel. Purification of DNA fragments from agarose gels is a multi-step process, which increases the time needed to generate each STS, reducing the number of STSs that can be produced per unit time, thus increasing costs per band.

Antolin et al. (1996) solved this problem by separating fragments on an acrylamide gel, improving resolution and making it easier to excise bands (simply cut out the band and resuspend it in 50uL of TE buffer) for PCR amplification. Thus Antolin et al. (1996) provide a superior method of separating RAPDS fragments, which can be used for generating large numbers of STS markers cheaply and quickly.

C. richardii is a pteridophyte, and as such serves as a useful model organism to study gender expression in plants. The independent gametophytic phase allows for the analysis of how the gametophytic genome regulates the events leading to gamete production independent of sporophytic effects.

C. richardii has a rapid life cycle. Spores can germinate to produce mature gametophytes in about 2 weeks, and can complete a generation (spore to spore) in about 2 months. Thus the crosses required by current methods of genomic mapping, such as quantitative trait loci (QTL) mapping, can be accomplished within a reasonable time.

An analysis of the genetic mechanisms regulating gender expression will require detailed genetic maps, if current (and likely future) methods of genetic analysis, such as positional cloning are to be used. There are not any detailed genetic maps for C. richardii, and thus our first step must be to develop a method of generating genetic maps. Genetic maps require markers. As described above, we believe that RAPDs can be used to quickly and cheaply develop large numbers of STS markers for C. richardii., which could then be used in the development genetic maps.


Spores of C. richardii (Hn-n) were sprinkled onto 10mm diameter sterile nutrient agar plates. Plates were placed in an enclosure made of plastic wrap located on a light bench with continuous fluorescent and incandescent light, with temperatures ranging between 280C to 310C. Fans were placed at each end of the light bench to reduce variation in temperature and humidity. Gametophytes were harvested 14 days following the date of inoculation and gametophyte tissue was removed from the agar, being careful not to include any of the agar with the tissue. Sporophytes were grown from gametophytes in pots placed inside an aquarium, with the top of the aquarium covered with plastic wrap. The aquarium was placed on the light bench described previously.

DNA was extracted from either a pool of 10 gametophytes or 100mg of sporophyte tissue using the Qiagen DNeasy DNA extraction kit. Tissue was ground in buffer AP1, and debris was pelleted in a microcentrifuge at 13,000RPM, with 400ul of the supernatant used to complete the protocol, starting at step 1, included with the DNAeasy kit. DNA was eluted in 2 separate elutions using 50ul of the elution buffer for each elution. The elution with the highest absorbance at 260nm was used for subsequent analysis.

The "hot start" protocol, as described in the information accompanying the Molecular BioProducts hot start tubes was used to generate random amplified polymorphic DNAs (RAPDs). Hot start tubes were purchased from Fisher Scientific. Hot start tubes are sterile DNA free microcentrifuge tubes with a bead of wax on the inside. The hot start protocol requires the preparation of an "upper" and a "lower" solution. The upper solution contains the DNA polymerase (Perkin Elmer Taq polymerase, Stoeffel fragment) and the template DNA. The lower solution contains the dNTPs, primers and magnesium chloride. The hot start protocol involves placing the lower solution in the hot start tube, heating the tubes to 95oC for 30 seconds, placing the tubes on ice, adding the upper solution and then completing the PCR.

A previous investigation indicated that 12ul of genomic gametophyte DNA sample extracted as described above would yield consistent results, and thus 12ul of our genomic gametophyte DNA sample was used for RAPDs reactions. Most RAPDs protocols call for 25ng of template DNA (eg. Williams et al. 1990). Since we did not have a high quality fluorimeter available to accurately quantitate DNA samples, the separate investigation to empirically determine the amount of template DNA needed for each reaction was necessary.

The upper solution consisted of 2.5 ul of the buffer included with the Stoeffel fragment of Taq polymerase (final concentration of 10mM tris-HCl, pH 8.3, 50mM KCl), 0.5 ul (5 units) of the Stoeffel fragment, 12.0 ul template (extracted DNA from C. richardii), and 10.0 ul deionized water for a total of 25ul.

The lower solution consisted of 0.5 ul of dNTPs obtained from Perkin Elmer (10mM solutions), 2.5 ul of the buffer included with the Stoeffel fragment Taq polymerase, 13.5 ul deionized water, 5.0 ul MgCl2 solution included with the Stoeffel fragment Taq DNA polymerase (25mM), and 2.0 ul of Operon A0-10 primer (5'GTGATCGCAG3') for a total of 25ul.

The Operon AO-10 primer was rehydrated by mixing 1.0 ml of 10mM TE buffer with the dehydrated primer. We had previously determined empirically that rehydrating Operon primers in 1ul ot 10mM TE and using 2ul per RAPDs reaction yielded consistent results with Stoeffel fragment DNA polymerase. This is about 12.5 pmoles per reaction, which is greater than the 5 pmoles per reaction suggested in the protocols accompanying the Operon primers.

The final reagent concentrations of each reaction was 100uM of each dNTP, 2.5 units DNA polymerase, 0.25uM primer and 2.5mM MgCl2.

The lower solution was placed in a hot start tube designed for 50ul reaction volumes. The hot start tube was placed into a thermal cycler and heated to 950C for 30 seconds. The tube was then removed from the thermal cycler and placed on ice. When the wax had hardened over the lower reaction, 25ul of upper solution was added to the tube and it was returned to the thermal cycler. The thermal cycler was programmed for 2 min at 950C and then 45 cycles of (1 minute at 940C, 1 minute at 370C, 2 minutes at 720C); followed by 7 minutes at 720C and then held at 40C until samples were removed.

After removing the tube from the thermal cycler, a small pin hole was made in the bottom of the tube and it was placed into another 500ul microcentrifuge tube to spin the reactions from the hot start tube into the other micorcentrifuge tube. 10 ul of sample buffer was added to the reaction and it was stored in a -100C freezer.

To assay for the quality of RAPDs reactions, we prepared a 1.5% agarose gel (agarose LE) to separate RAPDs fragments. Fragments from reactions of sufficient quality (the presence of distinct bands on the agarose gel) were separated on a polyacrylamide gel using the SSCP protocol described by Antolin et al. (1996). Gels were silver stained using the Promega silver stain kit. The gel prepared was similar to those used in SSCP analysis, however we were analyzing double stranded RAPDs products, not single stranded DNAs.

Two bands were excised from the lane containing fragments derived from RAPDs reactions using primer AO-10, one approximately 450bp and one approximately 150 bp. Fragments were resuspended in 50 ul TE, and reamplified with AO-10 primer using the RAPDs-PCR protocol described earlier, except that 1 ul of resuspended RAPD fragment was used as template with the amount water adjusted accordingly. Four separate PCR reactions were performed for each resuspended RAPD fragment. PCR products were screened, and the products with a single, sharp band and no evidence of contamination were used for cloning.

Reamplified RAPD's fragments were cloned using the Invitrogen TA cloning kit. Two recombinant (white) colonies were picked for each fragment (~150bp and ~450bp fragment) and cultured. Plasmids were extracted using a Qiagen midi-prep kit. Plasmids were sequenced using LiCor 4000L automated DNA sequencer and Amersham Thermosequence Kit with a 4% FMC Long Ranger polyacrylamide gel. The thermosequenase kit we used was the 7-deaza GTP kit, catalog number RPN2438. Sequence data were analyzed using DNASIS sequence analysis software.

Oligo primer selection software was used to design PCR primers specific to the DNA sequences derived from the two recombinant plasmids resulting from the cloning of the ~450bp RAPD fragment.

The primers specific to the two DNA sequences derived from the ~450bp RAPDs product were then used for PCR reactions using C. richardii gametophyte and sporophyte genomic DNA described above as a template, and the PCR protocols described in the documentation accompanying the Stoeffel fragment Taq DNA polymerase, using either 12ul of gametophyte DNA and 1 ul of sporophyte DNA containing 85ng/ul DNA).

Addresses and telephone numbers of suppliers are listed in appendix 2.


Fragments generated using primer AO-10 and cut from the polyacrylamide gel were successfully reamplified using primer AO-10, as shown in figure 1. Of the two sequences generated from plasmids with the insert of ~150bp, one had an insert of approximately 232bp (appendix a), and one had an insert greater than 1000bp. Of the two sequences generated from plasmids with the insert of ~450 bp, one had an insert of 414bp and one had an insert of 444bp (appendix a). A BLAST search indicated that neither the 414bp nor the 444bp RAPD product were highly similar to any other sequences in the Genebank database.

The ~1000bp insert lacked the AO-10 primer. The AO-10 primer was present in all other inserts. None of the inserts matched the cloning vector.

Figure 1. 450bp RAPD product and 150bp RAPD product generated from fragments cut from polyacrylamide gel and reamplified using primer AO-10. Hind III and EcoR I are Lambda DNA cut with the named restriction endonucleases. AO-10 total indicates lane with all the RAPDs fragments generated from genomic DNA using primer AO-10.

The PCR primers selected for the 414bp insert were 5'GTGATCGCAGATTTAATTGGTC3', 5'GTGATCGCAGAGAACGCAACCATA3'. The PCR primers selected for the 444bp insert were 5'TCGCAGATGGAGTGAATTT3', 5'GTTCCTATGATATCTACCACA3'.

We were able to amplify an appropriate sized fragment using the primer derived from the sequence of the 414bp insert using both sporophyte and gametophyte genomic DNA, however we were not able to amplify an appropriate sized fragmnent using the primer derived from sequence of the the 444bp insert from either sporophytic or gametophytic DNA (figure 2). Figure 2 indicates products from sporophytic DNA. We have not yet attempted to select primers specific to the 232bp insert and amplify it from genomic DNA.

Figure 2. Silver stained polyacrylamide gel with products of PCR with primers specific to 414 and 444 bp RAPDs fragments from sporophyte genomic DNA. 100bp ladder markers (lanes 1 and 4) flank lanes with PCR products (lanes 2 and 3). Product of primers to 414bp product is in lane 2 and product of primers specific to 444bp product is in lane 3.


We were able to use the methods described by Antolin et al. (1996) to separate RAPDs fragments on a polyacrylamide gel, clone a fragment, sequence that fragment, and design PCR primers that resulted in the amplification of that specific RAPDs fragment from genomic DNA, resulting in the production of an STS.

We generated at least two artifacts, one in the cloning of the ~150bp RAPDs product (the >1000bp insert) and one in the cloning of the ~450bp product (the 444bp insert that could not be reamplified from genomic DNA). The 444bp insert had the AO-10 primers on each and and must have been produced during the PCR using the template DNA derived from the band excised from the polyacrylamide gel. The 30bp separating a 414 and a 444bp is near enough for the possibility that fragments from both bands were included when the ~450bp band was excised from the polyacrylamide gel. The 444bp band could be from a contaminant in the PCRs of the ~450bp band that did not appear on the screeing gel. Our inability to re-amplify the 444bp fragment from genomic DNA of C. richardii suggests that is is a contaminant, however our primer design may have been flawed. Since the 1000bp insert lacks the AO-10 primer and any similarity to the cloning vector, we believe that the cloning of this fragment was the result of contamination of the cloning vector solution included in the TA cloning kit.

While one can investigate the causes of artifacts, the focus of this investigation was the production of STS's. The fact that the primers specific to the ~450bp band produced an appropriate sized fragment from both sporophytic and gametophyte genomic DNA indicates that we did in fact produce an STS for C. richardii.

RAPDs is a very sensitive PCR method, and moreso than most methods of PCR, will result in the amplification of contaminants in any reaction. Our experience has been that if clean, high quality template is used, template DNA will outcompete contaminants in PCR reactions, as suggested by Williams et al (1990). We did have one serious problem with DNA polymerase that was contaminated with E. coli DNA. We tried many methods of eliminating this contamination (Alford, 1997), however none were effective. The solution to this problem was to use another lot of DNA polymerase. We have also been informed anecdotally of equally serious problems with bacterial DNA in water due to the growth of bacterial cultures in water purification systems. We have never experienced this problem, however we have our water purification system serviced yearly, including a changing of cartridges.

The method of generating STS’s described above is an easy, inexpensive means of generating large numbers of genetic markers useable for analyses of genetic diversity and linkage mapping. The method described above can be completed using a manual sequencer as easily as with an automated sequencer as the fragments cloned are usually less than 500bp, and thus sequence data generated using a manual sequencer can be used to design appropriate primers for each fragment cloned.

To be used effectively, STS markers must be mapped onto chromosomes. STS polymorphisms must be discovered, as well as STS polymorphisms linked to genes of interest. We have generated a single STS. Thousands of such markers are needed if genes of interest are to be located and mapped. Our intital goal is to generate as many STS markers as possible. Future goals include the idenfication of markers unique to known gender expression mutants and the identification of the genes associated with the mutant phenotypes. Since there are very few markers published for C. richardii, methods such as ours are needed if the number of markers needed for the genetic dissection of any C. richardii pathway, including the gender expression pathway, is to be completed in a reasonable amount of time.

The usefullness of RAPDs in quickly generating markers linked to specific mutants is described well by Antolin et al. (1996), who were able to indentify 94 markers in five weeks for the mosquito Aedes aegypti. No other method allows for the generation of such a large number of markers in such a short period of time. The lack of such technically demanding and expensive steps such as autoradiography and Southern blotting makes the use of RAPDs as a means of generating STS markers relatively easy to incorporate into any laboratory, particularly those in which there is limited technical support, and a relatively rapid turnover of student researchers who vary greatly in ability.

Acknowledgments We would like to acknowledge financial support from the Howard Hughes Medical Institute Undergraduate Biological Sciences Education Program Grant #71195-538901.


Alford, Mac H. 1997. Methods of dealing with contamination in random amplified polymorphic DNA polymerase chain reactions (RAPD-PCR). Honors Thesis, Mississippi College.

Antolin, M., C. Bosio, J. Cotton, W. Sweeney, M. Strand and W. Black IV. 1996. Intensive linkage mapping in a wasp (Bracon hebetor) and a mosquito (Aedes aegypti) with single strand conformation polymorphism analysis of random amplified polymorphic DNA markers. Genetics 143: 1727-1738.

Ellsworth, D., K. Rittenhouse, and R. Honeycutt. 1993. Artifactual variation in randomly amplified polymorphic DNA banding patterns. BioTechniques 14: 214-217.

Paran, I. and R. Michelmore. 1993. Development of reliable PCR-based markers linked to downy mildew resistance genes in lettuce. Theoretical and Applied Genetics 85: 985-993.

Paterson, A. 1996. Genome Mapping in Plants. R.G. Landes, Austin.

Williams, J., M. Hanafey, J. Rafalski, and S. Tingey. 1990. DNA polymorphisms by arbitrary primers are useful as genetic markers. Nucleic Acids Research 18: 6531-6535.


Sequences of of cloned fragments.

Sequence of 414 base pair fragment, including operon AO-10 primer (GTGATCGCAG).


Sequence of 444 base pair fragment, including operon AO-10 primer (GTGATCGCAG).


Sequence of 232 base pair fragment, including operon AO-10 primer (GTGATCGCAG).



List of suppliers

Amersham Life Sciences Inc. 2636 S. Clearbrook Drive, Arlington Heights IL 60005 Ph. (800) 323-9750, FAX (800) 228-8735

Fisher Scientific, P.O. Box 4829, 2775 Pacific Avenue, Norcross, GA 30091 Ph. (800) 766-7000, FAX (800) 926-1166

FMC BioProducts, 191 Thomastown Street, Rockland, Maine 04841, Ph. (800) 341-1574, FAX (800) 362-5552

Hitachi Software Engineering America, Ltd. 601 Gateway Boulevard, Suite 500, South San Francisco, CA 94080-7025 Ph. (800) 624-6176, FAX (650) 615-9600

Invitrogen Corporation, 1600 Faraday Ave. Carlsbad, CA 92008 Ph. (800) 955-6288, FAX (760) 603-7201

LiCor Biotechnology Division, 4421 Superior Street, Lincoln, NE 68504 Ph. (800) 645-4267, FAX (402) 467-0819,

Molecular Biology Insights, Inc. (Oligo), 4515 Ranchview Lane, Plymouth, MN 55446-2128 Ph. (612) 559-0447, FAX (612) 559-3359

Operon Technologies Inc. 1000 Atlantic Avenue, Suite 108, Alameda, CA 94501 Ph. (510) 865-8644, FAX (510) 865-5255

PE Applied Biosystems (Perkin-Elmer), 850 Lincoln Center Drive, Foster City CA 94404 Ph. (800) 327-3002, FAX (650) 638-5998

Promega Corporation, 2800 Woods Hollow Road, Madison, WI 53711-5399 Ph. (800) 356-9526, FAX (800) 356-1970

Qiagen Inc. 28159 Avenue Stanford, Valencia, CA 91355-1106 Ph. (800) 426-8157, FAX (800) 718-2056