Jodi A. Irwin, Department of Cell and Molecular Biology, Northwestern University Medical School, 303 E. Chicago Ave., Chicago, Il 60611-3008 email@example.com; Cristin Orrego, Conservation Genetics Laboratory, Department of Biology, 1600 Holloway Avenue, San Francisco State University, San Francisco, CA 94132 firstname.lastname@example.org
The applications of denaturing gradient gel electrophoresis (DGGE) to the investigation and diagnosis of human genetic diseases are well-known (Fodde and Losekoot 1994). More recently, however, DGGE has also become a valuable tool for addressing questions pertaining to natural history. It has been applied to studies of population genetics (Lessa and Applebaum 1993) and phylogeography (Lessa 1992), and it has been particularly useful in analyses of microbial communities (Muyzer et al. 1993, Murray et al. 1996).
We demonstrate the use of DGGE as a molecular technique that can be employed in the absence of behavioral data for examining the diet of free-ranging howler monkeys (Alouatta pigra) (Figure 1). The difficulties associated with the collection of behavioral data for diet studies are well known. These include problems related to observing monkeys through thick foliage, the large distances separating observer from subject, and the difficulty of identifying tree species with very similar leaf morphologies. In order to circumvent these problems, we investigated the potential of a polymerase chain reaction* (PCR)/DGGE method for augmenting knowledge of the howler diet. We hope that the development of a non-invasive technique for examining diet will facilitate studies of wild populations, particularly those for which behavioral data are not available .
Materials and Methods
Genomic DNA was extracted from fecal samples and reference plants, and the chloroplast rbcL gene amplified via the PCR with one GC-clamped primer. PCR mixtures contained 1x PC2 PCR buffer (Ab Peptides), 0.1 units/l Klentaq DNA polymerase (Ab Peptides), MgCl2 (5.0 mM), 0.25 M each primer, and 1.5% dimethyl sulfoxide. Thermal cycling conditions were: an initial denaturation at 94 C for four minutes, 3 cycles of 94 C, 1 minute; 62 C, 1 minute; 80 C, 45 seconds; 4 cycles of 94 C, 1 minute; 64 C, 1 minute; 80 C, 45 seconds; 32 cycles of 94 C, 1 minute; 68 C, 1 minute; 80 C, 45 seconds; and a 5 minutes final extension at 80 C. PCR products were subsequently subjected to DGGE, using the DCode universal mutation detection system from Bio-Rad, to resolve the various plant sequences produced by PCR amplification.
To determine the melting temperature of PCR products, they were first analyzed on a perpendicular denaturing gradient gel poured with Bio-Rads Model 475 Gradient Former. The gel was 6.5% 37.5:1 acrylamide/bis-acrylamide, and contained a gradient of 1100% denaturant. Approximately 100 l of PCR product were loaded on a gel, and electrophoresis performed at 130 V and 60 C for 3 hours (Figure 2).
Optimal resolution of fragments on subsequent parallel gels was obtained with gradients ranging from 3040% denaturant, with electrophoresis at 130 volts and 60 C for approximately 4 hours (Figure 3). Gels were stained with ethidium bromide, and visualized by UV-fluorescence.
Results and Discussion
Our results indicate that DGGE is an effective technique for resolvi ng unique sequences from mixed template PCR amplifications. Analyses of eighteen plant species consumed by captive animals showed that DGGE permitted resolution of 15 of the 18 species, or over 80% (Figure 3). The three unresolved species possessed identical sequences (unpublished) in the melting domain under examination, and thus differentiation could not be expected.
Additional results, from samples obtained from captive animals whose diets were known, indicate that, on average, only about 30% of all plants ingested by the monkeys were detected by PCR/DGGE. Further experiments on constructed plant mixtures confirmed that a moderate amount of amplification bias occurs in PCR amplifications of mixed templates (Figure 4). Experiments to assess the preferential amplification observed in both the control fecal samples and the constructed plant mixtures suggest multiple factors at play, including template concentration and differential primer fidelity to various templates.
These data suggest that the PCR/DGGE approach tested in this investigation should be useful in situations where a universal detection system is not required, or where nothing is known about the diet of an animal. Further optimization of the technique should permit greater fragment resolution and enhanced sensitivity. Yet, any future investigations utilizing these methods should carefully consider the effects of PCR bias (Reysenbach et al. 1992, Ferris et al. 1996, Suzuki and Giovannoni 1996).
1. Ferris, M. J., Muyzer, G. and Ward, D. M., Applied and Environmental Microbiology, 62, 340346 (1996).
2. Reysenbach, A., et al. Applied and Environmental Microbiology, 58, 34173418 (1992).
3. Suzuki, M., Giovannoni, S., Applied and Environmental Microbiology, 62, 625630 (1996).
4. Fodde, R. and Losekoot, M., Human Mutation, 3, 8394 (1994).
5. Murray, A. E., Hollibaugh, J. T. and Orrego, C., Applied and Environmental Microbiology, 62, 26762680 (1996).
6. Lessa, E., Methods in Enzymology, 244, 419428 (1993).
7. Lessa, E. and Applebaum, G., Molecular Ecology, 2, 119129
We are grateful to Hal Markowitz (SFSU) for the opportunity to conduct this work. This project was supported by a U.S. Department of Education GAANN Pre-doctoral Fellowship, a San Francisco State University Mini-grant, and Oceanic Society Expeditions. This work was in partial fulfillment of a Masters in science degree by J.I.
* The Polymerase Chain Reaction (PCR) process is covered by patents owned by Hoffmann-LaRoche. Use of the PCR process requires a license.
back to top