2 A comparison of methods for total community DNA preservation and extraction
3 from various thermal environments
6 Kendra R. Mitchell1 and Cristina D. Takacs-Vesbach1
9 Department of Biology, University of New Mexico, Albuquerque, NM
13 Submitted to Journal of Industrial Microbiology & Biotechnology
17 Running Title: Nucleic acid preservation and extraction methods
21 *Corresponding author: firstname.lastname@example.org, 505-277-3418, 505-277-3040 (Fax)
23 1. Abstract
24 The widespread use of molecular techniques in studying microbial communities
25 has greatly enhanced our understanding of microbial diversity and function in the natural
26 environment and contributed to an explosion of novel commercially viable enzymes.
27 Exemplified by the discovery of Taq polymerace, one of the most promising
28 environments for detecting novel processes, enzymes, and microbial diversity is hot
29 springs. This is the first study to specifically examine potential biases introduced by
30 DNA preservation and extraction methods in hot spring microbial samples. We compared
31 the quality, quantity, and diversity of environmental DNA samples preserved and
32 extracted by commonly used methods. We included samples from the whole spectrum of
33 environmental conditions that are found in Yellowstone National Park thermal features
34 and our results are applicable to future studies in the park and other extreme
35 environments. We found that samples preserved in a non-toxic sucrose lysis buffer,
36 along with a variation of a standard DNA extraction method using CTAB resulted in
37 higher quality and quantity DNA than the other preservation and extraction methods
38 tested here. Diversity determined using DGGE revealed that there was some variation
39 within replicates of a sample, but no statistical difference between the methods.
40 However, the sucrose lysis buffer preserved samples extracted by the CTAB method
41 were 15-43% more diverse than the other treatments.
42 Keywords: DGGE, DNA extraction, environmental microbiology, thermophile
43 2. Introduction
44 The impact of molecular studies on our knowledge of the extent of microbial
45 diversity can’t be over stated. This explosion of information has significantly changed
46 the whole field, from basic ecological research into the fundamental controls of microbial
47 communities to bioprospecting for novel and potentially commercially relevant enzymes.
48 Even with recent advances in culturing efforts [10, 22], the majority of microbes in the
49 environment still cannot be cultivated in the laboratory . However, inability to
50 maintain an organism in culture is no longer a major impediment to accessing its genetic
51 diversity. Metagenomic studies similar to those that have been useful in exploring the
52 diversity of uncultivated organisms have also be used to mine for enzymatic diversity
53 . The biotechnology applications that are currently targeting microbial metagenomic
54 studies range from the search for new antibiotics to environmentally sound biocatalysts
55 such as amyases [14, 19].
56 Thermal environments have been a particularly rich source of novel organisms [1,
57 7, 9, 11, 16, 26], processes , novel enzymes [10, 23], and on-going research in the
58 origin and diversity of microbes. Fundamental to any of these studies is maximizing the
59 detectable diversity by maximizing the quality and quantity of DNA examined and
60 minimizing the biases of the methods. There are layers of potential bias in molecular
61 studies because of the sequential nature of the process, including those inherent in PCR
62  which have been well studied and will not be addressed in this paper. In the
63 collection of an environmental sample for molecular study, the initial, and least
64 examined, possible source of biased detection of community diversity is the biases that
65 could be introduced by the preservation method. Another possible source of bias is DNA
66 extraction methods and post-extraction purification. The effect of DNA extraction
67 method on detectable diversity has been examined in soil [5, 6], marine sediments ,
68 and compost [13, 28], but no study has examined the impact that temperature and pH
69 extremes may have on extraction efficiency.
70 We are concerned about the effects of sample preservation and DNA extraction
71 methods on the diversity of the community detected by examining molecular methods.
72 The objective of this study was to find the most effective sample preservation and DNA
73 extraction method that results in high molecular weight DNA, is relatively free from
74 contaminants, and maximizes the amount of diversity that we detect. We compared three
75 preservation methods and six nucleic acid extraction methods. The ideal preservation
76 and extraction method would work well with a variety of samples that included high and
77 low biomass, a wide range of pH (e.g. thermal features range from very acidic, <1, to
78 alkaline), as well as different sample types including biofilms, microbial mats, and
79 sediments. Additionally, this method should be quick to accommodate high throughput
80 of large sample numbers and facilitate sample collection from the remote backcountry
81 where liquid nitrogen, dry ice, and toxic preservatives are not practical. We included
82 samples from the whole spectrum of environmental conditions that are found in YNP
83 thermal features and our results are applicable to future studies in YNP, other extreme
84 environments, and microbial surveys in general.
85 3. Methods and Materials
86 Site description and sampling
87 Samples were collected from thermal features throughout Yellowstone National
88 Park, USA (YNP) during the summer of 2002 as part of our microbial inventory. A
89 subset of fifteen samples was selected for this study that encompassed the full range of
90 pH, temperature, and biomass types found in the park (Yellowstone sample n=15, see
91 Table 1). The pH and temperature of the samples were split into three ranges, low (pH 0
92 to 4, temperature 40 to 60 degrees Celsius), mid (pH 4.01 to 8, temperature 60.1 to 80),
93 and high (pH above 8.01, temperature above 80.1). The samples were also categorized by
94 the type of biomass collected: sediment, microbial mat, or filaments. Additionally, we
95 collected samples from two neutral thermal springs in the Jemez Mountains, New
96 Mexico. Approximately 1-2.5 ml of sample was collected at each site with either sterile
97 forceps or a syringe.
98 Sample Preservation
99 Two replicate YNP samples were collected from each sampling site: one was
100 preserved in an equal volume of sucrose lysis buffer (20 mM EDTA, 200 mM NaCl, 0.75
101 M sucrose, 50 mM Tris-HCl, pH 9.0)  and the other replicate was preserved in an
102 equal volume of GIT (5 M guanidine isothiocyanate, 50 mM Tris pH 7.4, 25 mM EDTA
103 pH 8, 0.8% 2-mercaptoethanol). The YNP samples were held at ambient air
104 temperature (10 to 26 ºC) for up to five days before they were stored at –80 degrees
106 The two Jemez Springs, New Mexico samples were collected from neutral
107 thermal springs one from the Jemez Springs Bath House (BH) and the other from
108 Giggling Star Resort (GS). Three replicate samples were collected. Two of the replicates
109 were preserved in sucrose lysis buffer (SLB), one of these replicates was frozen
110 immediately in liquid nitrogen and the other was held at 20C for 7 days before being
111 frozen at –80C. The third replicate sample was collected and mixed with molten 2%
112 agarose while in the field for extraction with a high molecular weight agarose noodle
113 method (see below).
114 Extraction Methods
115 We used four extraction methods: lysis by pulse boil, a CTAB extraction, agarose
116 noodle which results in high molecular weight DNA , and the Mo Bio Soil DNA
117 Purification kit, which combined with the preservation methods results in six treatments.
118 With the pulse boil method, nucleic acids were extracted from 200 l of the YNP and
119 Jemez samples that were preserved in SLB (referred to here as “Boil”) . Briefly: the
120 samples were boiled at 96C then cooled to 4C three times in a thermocycler, sodium
121 dodecyl sulphate (SDS) (to a final concentration of 2%) and proteinase K (final
122 concentration 250 g/ml) were added and the sample was incubated at 42C for 2.5 hrs,
123 then incubated at 60C for 30 minutes, extracted once with phenol/chloroform then twice
124 with chloroform, finally the DNA was precipitated and washed with ethanol.
125 We extracted nucleic acids from 200 l of the YNP and Jemez SLB preserved
126 samples using a variation of the CTAB  method (referred to as CTAB S). Briefly: 2
127 volumes of 1% CTAB buffer (1% CTAB, 0.75 M NaCl, 50 mM Tris pH 8, 10 mM
128 EDTA) and proteinase K (final concentration 100 g/ml) were added to the SLB
129 preserved samples; incubated for one hour at 60C, SDS (final concentration 2%) was
130 added and incubated one hour at 60C, extracted once with phenol/chloroform then twice
131 with chloroform, finally the DNA was precipitated with ethanol. The YNP samples that
132 were preserved in GIT (referred to as CTAB G) were also extracted using the CTAB
133 method with the modification of washing the sample three times with filter sterilized
134 water before the addition of CTAB to remove the GIT from the sample. GIT is a protein
135 denaturant and inhibits proteinase K activity if not removed.
136 SLB (Mo Bio S) and GIT (Mo Bio G) preserved YNP samples were extracted
137 using the Mo Bio Soil DNA purification kit following manufacture’s suggested protocol
138 (Mo Bio Laboratories, Inc., Solana Beach, CA).
139 Nucleic acids were extracted from YNP SLB preserved samples and all Jemez
140 samples using the high molecular weight noodle extraction method (Noodle). Briefly:
141 the sample was mixed with molten 2% agarose and cooled in 1 mL syringes forming
142 agarose noodles, the noodles were incubated for 3 hours at 37C in a lysis buffer (10 mM
143 Tris, 50 mM NaCl, 50 mM EDTA, 0.2% SDS, 1% Sarkosyl, 1mg/ml lysozyme) then
144 incubated at 37C in ESP buffer (1% Sarkosyl, 0.1 M EDTA, 1 mg/ml proteinase K), the
145 ESP buffer was changed once a day for a total of four days, the noodles were then stored
146 at 4C in TE storage buffer (10 mM Tris and 50 mM EDTA). The purified nucleic acids
147 were extracted from the agarose noodles by incubating at 60C to melt the agarose then
148 adding agarase and incubating overnight at 37C. The nucleic acids were purified from
149 the agarose/agarase slurry by extracting once with phenol/chloroform then twice with
150 chloroform. The nucleic acids were precipitated with ethanol. 
151 Measuring extraction success
152 Environmental DNAs were electrophoresed on an ethidium bromide stained 1.2%
153 agarose gel. The size and quality (evidence of shearing etc.) of the DNA was evaluated
154 using a size standard DNA ladder (EZ Load 1Kb, BioRad).
155 The purity of the extracted DNA was quantified by calculating the ratio of the
156 absorbance at 260 nm and 280 nm (A260/A280). Nucleic acids extracted from
157 environmental samples are often contaminated with humic organic carbons, metals, and
158 other compounds that makes using the 260 nm absorbance to calculate DNA
159 concentration unreliable. To circumvent this problem, we also quantified the DNA using
160 PicoGreen dsDNA Quantitation Reagent (Molecular Probes, Inc., Eugene, OR, USA).
161 Extracted DNA (5 l) was mixed with a 1:200 solution of PicoGreen in 1X tris-acetate
162 buffer (TAE) and absorbance was read in a fluorometer.
163 The third parameter that was qualified was the ability to amplify 16S rDNA from
164 the environmental genome of each extraction replicate. Dilutions of the genomic DNA
165 were used as template DNA. The reaction included (1X Promega buffer with 1.5 mM
166 MgCl2, bovine serum albumin (0.04 % final), 2.5 U Taq DNA polymerase (Promega
167 U.S.) 2.5% Igepal CA-630 (Sigma-Aldrich), 10 M each dATP, dGTP, dCTP, dTTP
168 (BioLine USA, Inc.), 20 M Bacterial specific primers 338FGC
170 AGCAG) and 519R (ATTACCGCGGCTGCTGG)). The PCR reaction (50 l) was
171 incubated in a thermocycler (ABI GeneAmp 2700) for 5 minutes at 94.0C then for 30
172 cycles of 30 seconds at 94.0C, 30 seconds at 50.0C and 30 seconds at 72.0C. The
173 reaction was incubated at 72.0C for 7 minutes for final extension.
174 Relative species Richness
175 Differences in relative species richness among the preservation and extraction
176 methods were determined using denaturing gradient gel electrophoresis (DGGE). Ten l
177 of PCR product (approximately 500 ng) was run on a 6% (wt/vol) acrylamide gel with
178 1X TAE (40 mM Tris, 20 mM acetate, and 1mM EDTA) with a denaturing gradient of
179 20% (8% (vol/vol) formamide and 8.4% (wt/vol) urea) to 60% (24% formamide and
180 25.2% urea). The gels were run in a Bio-Rad DCode Universal Mutation Detection
181 System (Cat# 170-9080) at 180V for 3.5 hours. The gels were stained in 1X TAE
182 containing SybrGreen (100 l/L) for 30 minutes then destained in 1X TAE for 15
183 minutes. The gels were photographed under UV light and DGGE bands were identified
184 and analyzed using Kodak 1D software. For our analyses, only bands with a minimum
185 intensity of 72% were recognized (program default). Bands were distinguished based on
186 migration distance within each gel, as determined by the software using the DGGE
187 product from Escherichia coli as a standard. . Although re-amplification and sequencing
188 of individual DGGE bands can be particularly important in resolving heteroduplex
189 fragments, we did not sequence the bands we detected in this study because we were
190 interested in detecting the greatest number of bands possible. A relative species richness
191 index was computed for each case by dividing the number of bands detected within a
192 treatment by the total number of unique bands found within a sample. Relative species
193 indices were computed for samples where at least 2 of the treatments amplified (YNP
194 samples n=12).
195 Statistical Analysis
196 Statistical analyses were performed using Minitab software version 13 and SPSS
197 11 (for Mac OS X). The purity of the DNA among treatments was compared using the
198 absolute value that resulted from subtracting the A260/A280 from 1.8, the A260/A280 of pure
199 DNA . Univariate ANOVA was used to detect statistical differences in the quantity
200 and purity of the DNA and relative species richness (dependent variables) among the
201 treatments and by pH, temperature, and biomass type (fixed factors). For analysis of the
202 DGGE bands, only samples where at least 2 of the treatments amplified were included,
203 otherwise a statistical comparison would not be possible. The Bonferroni test (which is
204 more powerful for small sample sizes than Tukey’s test) was performed to identify
205 treatments that were significantly different.
208 Sample Preservation
209 Sucrose lysis buffer preserved the DNA of our extreme environment samples
210 better than the GIT solution. The extracted DNAs were quantified using the PicoGreen
211 reagent. Using our protocol, the PicoGreen assay has a dsDNA detection limit of 500
212 pg/ml. We were able to extract more DNA from the samples preserved in SLB than
213 from the GIT replicates (paired t-test p=0.01). We also compared the amount of DNA
214 recovered from samples that were frozen or mixed with agarose in the field with those
215 that were held at ambient temperature, there was no statistical difference between the two
216 replicates (paired t-test =0.05). Even when samples cannot be frozen immediately, the
217 DNA is adequately preserved in sucrose lysis buffer. Although this study did not assess
218 the relative effectiveness of the different methods in extracting RNA, we have
219 successfully produced cDNA from samples preserved in SLB using reverse transcriptase
220 PCR (unpublished data).
221 There was no significant difference between the preservation methods in the
222 diversity we detected (paired t-test =0.05). However, the 16S rDNA from the samples
223 that were preserved in SLB were amplified by PCR more successfully; 80.5% of SLB
224 preserved samples PCR amplified compared to just 50% of the GIT preserved samples.
225 DNA extraction
226 The CTAB extraction on SLB preserved samples resulted in the greatest yield of
227 DNA. The results of the quantification are displayed in Fig. 1a. ANOVA showed a
228 statistically significant difference in DNA yield between the extraction methods
229 (p<0.001). Post-hoc multiple comparison analysis indicates that the CTAB extraction on
230 SLB preserved samples resulted in the highest quantity of DNA.
231 DNA purity, as measured by A260:A280 ratio, was greatest in samples extracted
232 using the commercial Mo Bio extraction kits, however the kit results in less DNA and
233 smaller fragments (presumably from shearing and the selecting column used in the kit)
234 (Fig. 1b). There was no significant difference between the purity of the DNA extracted
235 with the Boil, CTAB S, or CTAB G methods (ANOVA =0.05). The quantity of DNA
236 extracted by the Noodle method was not sufficient to obtain reliable absorbance data,
237 therefore the purity of the DNA extracted with that method could not be compared to the
238 other methods. Most of the amplification success we observed was in sample extracted
239 with either the CTAB method (89% amplified) or the Mo Bio kit (95% amplified)
240 (ANOVA and Bonferroni test p<0.001).
241 The most important parameter to consider when choosing a preservation and
242 extraction method is the effect a particular method has on the detectable diversity. The
243 number of bands or sequence diversity detected using DGGE varied within the replicates
244 of the samples (Fig. 2). The only statistically significant difference we detected was
245 between the CTAB S and the noodle method (p = 0.014). However, CTAB S resulted in
246 15% to 43% more bands than any of the other treatments (Fig, 3). There were no
247 differences at the fixed effects level (pH, temperature, and biomass type), nor were any
248 interactions (e.g., pH X method) among the factors detected.
250 It is standard practice that investigators informally evaluate methods to determine
251 which are appropriate for their study. However, few of these studies have been published
252  and in many cases, statistical analysis are not used to determine the most appropriate
253 methods . This study presents a framework for the systematic comparison and
254 evaluation of methods by statistical analysis. The current interest in the organisms that
255 inhabit extreme environments makes this study, which used samples from a wide variety
256 of thermal environments, of importance to thermal biology researchers, and applicable to
257 other microbial studies in general.
258 Sample Preservation
259 Sucrose lysis buffer preserved our extreme environment samples better than the
260 GIT solution. This was likely due to more effective lysing, as the SLB initiates the lysis
261 process while the sample is being stored. The alkaline SLB raised the pH within the
262 sample vials, potentially slowing the degradation of the DNA in low pH samples. Even
263 when samples could not be frozen immediately, the DNA is adequately preserved in
264 SLB. This result is especially important because of restrictions on carrying and shipping
265 dry ice and for samples that are collected from remote areas where it is logistically very
266 difficult to freeze samples immediately and maintain them frozen. The non-toxic nature
267 of sucrose lysis buffer may be an additional attraction for field microbiologists, and for
268 those shipping samples either commercially or hand-carrying on airplanes.
269 DNA extraction
270 When considering preservation alone, among the techniques we tested, SLB with
271 or without freezing preserved environmental DNA best. However, we also were
272 interested in the effect of extraction method on the amount and quality of extracted DNA
273 and detectable diversity.
274 CTAB extraction combined with SLB preserved samples resulted in the most
275 DNA. The greater efficiency of the CTAB extraction method has previously been found
276 in several environments, marine sediments  and by one of us in caterpillar gut and
277 filtered water (Takacs-Vesbach, unpublished data) indicating that this method may be
278 effective on a wide range of sample types. An extraction comparison on compost
279 microbial communities found no difference in the amount of DNA recovered by the
280 CTAB method, but it resulted in the highest percentage of cells lysed . It is likely
281 that the combination of proteinase K and hot SDS lyses more cells, including cell types
282 more resistant to lysis (i.e. gram-positive bacteria). There were two exceptions
283 where larger quantities of DNA were not recovered by the CTAB method. The PicoGreen
284 analysis on two samples, 066-MV and 139-LS, indicates that more DNA was extracted
285 using the Mo Bio DNA extraction kit than was extracted with the Boil and CTAB
286 methods. The Boil and CTAB extracted DNA from these two samples was dark brown,
287 which were most likely photosynthetic pigments that were co-extracted with the DNA.
288 The Mo Bio kit, which produced clear DNA extracts, was effective at removing these
289 organics. We suspect that the co-extracted pigments interfered with PicoGreen
290 fluorescence, but both of these samples were successfully amplified with all of the
291 extraction methods, and the co-extracted pigments from the Boil and CTAB methods do
292 not appear to inhibit the PCR reaction.
293 DNA Purity
294 Extracting pure DNA from environmental samples is nearly as important as
295 extraction efficiency and is one of the most difficult technical problems in using
296 molecular techniques on environmental samples. Most DNA extraction procedures co-
297 extract humic organic carbon, pigments, heavy metals, and other contaminants. These
298 contaminants play havoc with PCR reactions and can degrade the DNA during storage.
299 Not surprisingly, the purest DNA was extracted using the commercial Mo Bio extraction
300 kit, which was the only method that incorporated a post-extraction purification, however
301 the kit results in less DNA and smaller fragments. The Mo Bio kit contains a proprietary
302 inhibitor removal solution and a column purification step that were designed to remove
303 humic acid contaminants . There was no significant difference in the purity of the
304 DNA extracted with the Boil, CTAB S, or CTAB G methods (ANOVA =0.05). The
305 concentration of the DNA extracted with the Noodle method was not sufficient to obtain
306 reliable absorbance data, therefore the purity of the DNA extracted with that method
307 could not be compared to the other methods.
309 PCR amplification of environmental DNA is not always a straightforward process.
310 The co-extraction of humics, metals, and other potential inhibitors figure significantly in
311 the analysis of microbial communities. In our experience, all but a very small proportion
312 of the samples that we have examined can be amplified using PCR. Some samples
313 require more effort, troubleshooting, and processing time, but in the end most can be
314 amplified. However, that being said, the samples preserved in SLB and extracted with
315 either the CTAB method or the Mo Bio kit were more successfully amplified relative to
316 the other cases. It should be noted that some of the samples with no detectable DNA, as
317 determined using the PicoGreen fluorometer quantification, still had enough DNA to
318 amplify with PCR (theoretical detection limit for PCR amplification < 1 pg/ml).
319 Although the CTAB S treatments consistently had more bands than any of the
320 other methods (15-43%), the only statistical difference detected was between the CTAB
321 S and the noodle extraction method. We also tested for fixed effects such as pH,
322 temperature, and biomass type, but did not detect any statistical differences. However, as
323 stated earlier, there were three samples that we were unable to amplify using two or more
324 of the extraction treatments. It is interesting to note that all three of these samples were
325 sediments from low pH systems (<pH 3.5), indicating that there may be a sample type or
326 geochemistry effect on extraction efficiency or amplification success. This may be a
327 result of low biomass or inhibition by some unknown factor associated with sediments.
328 The widespread use of molecular techniques utilizing 16S rDNA to describe the
329 diversity of a microbial community, such as clone libraries, DGGE, and T-RFLP
330 (terminal restriction fragment length polymorphisms), have greatly increased our
331 understanding of microbial ecology and revealed novel enzymes. We are interested in
332 not only expanding our knowledge of novel species and the environments they inhabit,
333 but also in developing a baseline inventory of microbial diversity. Because of this
334 emphasis on a baseline inventory, it is very important that we maximize our ability to
335 detect all possible diversity. DGGE is an attractive method for microbial ecology studies
336 because of the relatively rapid ease with which a community profile can be generated.
337 Many researchers prefer this method to others, such as T-RFLPS, because individual
338 bands can be sequenced. This can be particularly important in distinguishing truly unique
339 bands. DGGE can resolve minor base differences among samples; therefore, distinct
340 bands may represent sequences that differ by less than 1%. Depending on the nature of
341 the study being performed, 1% may or may not be significant to the research question.
342 Because our objective was to identify the method that resulted in the greatest detectable
343 diversity, we did not sequence the bands we observed in the study. Conversely, we are
344 aware that individual bands can be comprised of several different sequences, however,
345 because our methods were standardized across samples, this effect would not figure in
346 our conclusions. We repeated the PCR amplification and DGGE for selected samples to
347 ensure reproducibility in the number of bands detected.
348 The main goal of this study was to determine the sample preservation and DNA
349 extraction method that would maximize the detectable diversity in molecular analysis of
350 environmental communities. For the purposes of preserving and extracting very high
351 molecular weight DNA, many researchers prefer the noodle method. Based on the results
352 from this study, this preservation method does not seem to work well with the samples
353 we collect which often have low biomass in a dense extracellular matrix. In many
354 samples, no DNA was visible on the ethidium bromide stained gels from the noodle
355 treatments. Additionally, we have found that this method is particularly difficult to use in
356 the remote backcountry. Originally, the noodle method was used to recover large
357 fragments of DNA for metagenomic analysis (25), but the development of new
358 sequencing technologies and metagenomic approaches has made the quality of DNA
359 from routine DNA extraction procedures such as CTAB S acceptable. While the Mo Bio
360 kit resulted in the cleanest DNA of the methods examined, the quantity of DNA obtained
361 with the kit was much lower than the concentration recovered with the Boil, CTAB G,
362 and especially the CTAB S methods. More high molecular weight DNA was extracted
363 using the Boil or CTAB method than was using the Mo Bio kit. The impurities in the
364 DNA extracted with the CTAB S method did not seem to adversely impact our ability to
365 PCR amplify the target sequence. Furthermore, CTAB S resulted in more species
366 richness than of the other treatments. Therefore, the CTAB DNA extraction method on
367 samples preserved in SLB seems to be the most appropriate for our purposes and should
368 be of interest to other projects with similar objectives.
370 We are grateful for the comments and input by Robert Sinsabaugh and Diana
371 Northup who kindly reviewed an earlier version of the manuscript. This work was
372 supported by NSF Biodiversity Surveys and Inventories grant 02-06773 to CTV and NPS
373 funding to Ann Rodman.
376 1. Barns, S.M., et al., Remarkable archaeal diversity in a Yellowstone National Park
377 hot spring environment. Procedures of the National Academy of Science USA,
378 1994. 91(March): p. 1609-1613.
379 2. Braid, M.D., et al. Testing the UltraClean soil DNA purification kit on a diverse
380 range of soils by PCR amplification of 16S rDNA. in ASM General Meeting.
382 3. Bryant, D.A., et al., Candidatus Chloracidobacterium thermophilum: An Aerobic
383 Phototrophic Acidobacterium. Science, 2007. 317: p. 523-526.
384 4. Cary, S.C., et al., Identification and localization of bacterial endosymbionts in
385 hydrothermal vent taxa with symbiont-specific polymerase chain reaction
386 amplification and in situ hybridization techniques. Molecular Marine Biology and
387 Biotechnology, 1993. 2(1): p. 51-62.
388 5. Fromin, N., et al., Statistical analysis of denaturing gel electrophoresis (DGE)
389 fingerprinting patterns. Environmental Microbiology, 2002. 4(11): p. 634-643.
390 6. Gabor, E.M., E.J. de Vries, and D.B. Janssen, Effecient recovery of
391 environmental DNA for expression cloning by indirect extraction methods. FEMS
392 Microbiology Ecology, 2003. 44: p. 153-163.
393 7. Ghosh, D., et al., Molecular phylogenetic exploration of bacterial diversity in a
394 Bakreshwar (India) hot spring and culture of Shewanella-related thermophiles.
395 Applied and Environmental Microbiology, 2003. 69(7): p. 4332-4336.
396 8. Giovannoni, S.J., et al., Tangential flow filtration and preliminary phylogenetic
397 analysis of marine picoplankton. Applied and Environmental Microbiology,
398 1990. 56: p. 2572-2575.
399 9. Harris, K.J., S.T. Kelley, and N.R. Pace, New Perspective on Uncultured
400 Bacterial Phylogenetic Division OP11. Applied and Environmental
401 Microbiology, 2004. 70(2): p. 845-849.
402 10. Hobel, C.F.V., et al., Use of low nutrient enrichments to access novel amylase
403 genes in silent diversity of thermophiles. World Journal of Microbiology &
404 Biotechnology, 2004. 20: p. 801-809.
405 11. Hugenholtz, P., B.M. Goebel, and N.R. Pace, Impact of culture-independent
406 studies on the emerging phylogenetic view of bacterial diversity. Journal of
407 Bacteriology, 1998. 180(18): p. 4765-4774.
408 12. Kowalchuk, G.A., et al., Finding the Needles in the Metagenome Haystack.
409 Microbial Ecology, 2007. 53: p. 475-485.
410 13. LaMontagne, M.G., et al., Evaluation of extraction and purification methods for
411 obtaining PCR-amplifiable DNA from compost for microbial community
412 analysis. Journal of Microbiological Methods, 2002. 49: p. 255-264.
413 14. Lorenz, P. and J. Eck, Metagenomics and industrial applications. Nature Reviews
414 Microbiology, 2005. 3: p. 510-516.
415 15. Luna, G.M., A. Dell'Anno, and R. Danovaro, DNA extraction procedure: a critical
416 issue for bacterial diversity assessment in marine sediments. Environmental
417 Microbiology, 2006. 8(2): p. 308-320.
418 16. Meyer-Dombard, D.R., E.L. Shock, and J.P. Amend, Archaeal and bacterial
419 communities in geochemically diverse hot springs of Yellowstone National Park,
420 USA. Geobiology, 2005. 3: p. 211-227.
421 17. More, M.I., et al., Quantitative cell lysis of indigenous microorganisms and rapid
422 extraction of microbial DNA from sediment. Applied and Environmental
423 Microbiology, 1994. 60(5): p. 1572-1580.
424 18. Pace, N.R., A molecular view of microbial diversity and the biosphere. Science,
425 1997. 276: p. 734-740.
426 19. Pontes, D.S., et al., Molecular approaches: advantages and artifacts in assessing
427 bacterial diversity. Journal of Industrial Microbiology and Biotechnology, 2007.
428 34: p. 463-473.
429 20. Reysenbach, A.L., M. Ehringer, and K. Hershberger, Microbial diversity at 83
430 degrees C in Calcite Springs, Yellowstone National Park: another environment
431 where the Aquificales and "Korarchaeota" coexist. Extremophiles, 2000. 4(1): p.
433 21. Sambrook, J., E.F. Fritsch, and T. Maniatis, Molecular Cloning, A Laboratory
434 Manual. 2nd ed. 1989, Plainview, New York: Cold Spring Harbor Laboratory
436 22. Schoenborn, L., et al., Liquid Serial Dilution is Inferior to Solid Media for
437 Isolation of Cultures Representative of the Phylum-Level Diversity of Soil
438 Bacteria. Applied and Environmental Microbiology, 2004. 70(7): p. 4363-4366.
439 23. Schoenfeld, T., et al., Viral Diversity and Improved DNA Polymerases, in YNP
440 RCN/TBI Workshop. 2008: Mammoth Hot Springs, WY, USA.
441 24. Steffan, R.J. and R.M. Atlas, DNA amplification to enhance detection of
442 genetically engineered bacteria in environmental samples. Applied and
443 Environmental Microbiology, 1998. 54(9): p. 2185-2191.
444 25. Stein, J.L., et al., Characterization of uncultivated prokaryotes: isolation and
445 analysis of a 40-kilobase-pair genome fragment from a planktonic marine
446 archaeon. Journal of Bacteriology, 1996. 178(3): p. 591-599.
447 26. Takacs-Vesbach, C.D., et al., Volcanic calderas delineate biogeographic
448 provinces among Yellowstone thermophiles. Environmental Microbiology, in
450 27. von Wintzingerode, F., U.B. Gobel, and E. Stackebrandt, Determination of
451 microbial diversity in environmental samples: pitfalls of PCR-based rRNA
452 analysis. FEMS Microbiology Reviews, 1997. 21(3): p. 213-229.
453 28. Yang, Z.H., et al., Comparison of methods for total community DNA extraction
454 and purification from compost. Applied Microbiol Biotechnology, 2007. 74: p.
456 29. Zhou, J., M.A. Bruns, and J.M. Tiedje, DNA recovery from soils of diverse
457 composition. Applied and Environmental Microbiology, 1996. 62(2): p. 316-322.
462 Figure Legend
463 Fig. 1 a) Mean DNA concentration as measured using PicoGreen assay for each
464 preservation and extraction method with 95% confidence intervals, ourtliers are filled
465 circles. b) Mean DNA purity with 95% confidence interval and outliers. Pure DNA has
466 a A260/A280 ratio of 1.8, showen as the gray bar. The concentrations of the DNAs from
467 the Noodle extractions were not high enough concentration to be detected by the
470 Fig. 2 Example DGGE of PCR products, primers 338FGC and 519R, on 6% acrylamide
471 gel with 20%-60% urea/formamide denaturing gradient, stained with SybrGreen
472 [Treatments: a) Boil; b) CTAB S; c) CTAB G; d) Mo Bio S; e) Mo Bio G; f) Noodle].
474 Fig. 3 a) Rarefaction curves of DGGE bands detected with each extraction method. The
475 95% confidance interval for the diversity of all samples and extraction methods
476 combinded is in gray to demonstrate that there was no significant difference between
477 diversity detected by each method. b) Mean diversity detected as number of DGGE
478 bands with 95% confidence intervals, outliers are filled circles.
479 Table 1 Sample site, description and environmental parameters.
Sample ID Sample Area1 pH T (C) Sample description Northing2 Easting2
007-L YNP Lower GB 7.54 79.9 black powder 4932385.831 517103.977
010-L YNP Lower GB 3.55 90.6 gray clay-like sediment 4933099.227 515316.123
022-L YNP Lower GB 6.89 85.9 black powder 4933820.851 513269.383
045-L YNP Lower GB 2.68 42.4 brown foam and water 4934500.042 513403.550
048-L YNP Lower GB 3.39 48.8 grey mud sediment 4934417.810 513990.150
058-L YNP Lower GB 2.96 61.4 yellow and tan powder 4953191.966 522871.118
orange mat and black
066-MV YNP Mud Volcano 6.41 67.4 filaments 4939784.796 544533.185
072-CH YNP Crater Hills 5.47 55.6 yellow powder 4952741.659 523490.278
088-L YNP Lower GB 8.46 52.8 orange and green mat 4935485.818 515973.327
126-MM YNP Mary Mountain 6.58 79.8 grey powder 4940597.962 532861.547
131-LS YNP Lone Star GB 4.24 43.9 yellow filaments 4916381.813 514106.247
green mat and gray
139-LS YNP Lone Star GB 2.49 54.8 sediment 4919322.475 515169.854
171-S YNP Shoshone GB 8.63 68.7 orange and green mat 4911109.135 515791.782
184-S YNP Shoshone GB 8.92 77.3 tan filament 4911522.596 515932.872
190-S YNP Shoshone GB 9.08 44.5 layered orange mat 4911267.100 515925.080
BH Jemez Bath House 7.15 76.0 yellow filaments 3959977.588 347240.017
Jemez Giggling Star
GS Resort 6.45 53.4 green mat 3959468.000 347235.000
481 Geyser Basin is abbreviated GB. 2Northing and Easting are given in UTM, grid 12N for
482 YNP, 13N for Jemez Springs, NM, datum NAD83
487 Mitchell et. al. Fig. 1
493 Mitchell et. al. Fig. 2
498 Mitchell et. al. Fig. 3