| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Brief Communication |
Laboratory of Quantitative Vegetation Ecology, Institute of Botany, Chinese Academy of Sciences, 20 Nanxincun, Xiangshan, Haidian District, Beijing 100093, P.R. China; School of Engineering and Applied Science, University of Pennsylvania, Philadelphia 19104 USA
Received for publication November 25, 2005. Accepted for publication January 18, 2007.
ABSTRACT
Pollen identification and classification are important not only for palynologists, but also for systematists and ecologists. Because palynological methods for the identification of pollen in surface soil until now could resolve at best to the generic level, we have developed a molecular approach to species-level identification of Chenopodiaceae pollen in surface soils. Surface soil samples were collected in the central area of Junggar Desert Basin, Xinjiang, China. Fresh leaves of 19 Chenopodiaceae species were sampled for DNA sequencing, establishing a database of internal transcribed spacer (ITS) regions of nuclear ribosomal DNA for Chenopodiaceae. Individual chenopod pollen grains in a soil sample were separated from the soil and the ITS1 region of each pollen grain was amplified using nested PCR and sequenced. By comparing the amplified ITS1 sequences to those in the Chenopodiaceous database, we identified the pollen in the soil samples to the level of species. The new method provides a technical reference for species identification of soil surface pollen for other families. This work is necessary for further efforts to interpret the relationship of surface soil pollen to vegetation characteristics. It also has significant potential for enhancing the ability to identify pollen in clinical airborne allergen or criminological studies.
Key Words: Chenopodiaceae • internal transcribed spacer sequence • Junggar Desert Basin • modern pollen • pollen identification
Pollen identification and classification are useful for palynologists, systematists, and ecologists. Species composition of pollen in surface soil is considered reflective of existing vegetation diversity. Information on the relationship of existing vegetation diversity to its pollen diversity is indispensable for studies on global vegetation changes and to deduce past vegetation diversity from the diversity of fossil pollen. A description of pollen diversity depends on the species identification of the pollen. The most common approach for this identification is through morphological characters visible with an optical microscope. However, with this approach, a pollen grain can be identified to the generic level at best, but not to the species level (Mullins and Emberlin, 1997
). For example, Chenopodiaceae pollen in Xinjiang could be classified only into five types, not even into genera (Hao et al., 1989
). Similarly, such pollen in central Jordan could not be separated into genera, only into four types (Davies and Fall, 2001
). While existing vegetation can be quantified at a species level, quantifying soil surface pollen at this level is quite challenging. As vegetation is described for an area, the species of soil surface pollen also needs to be accurately described, which requires a technique able to resolve pollen to that level (Joosten and de Klerk, 2002
). A molecular approach was our first choice for developing such a technique.
Chenopodiaceae, a widespread family in arid regions, contributes abundant pollen to surface soil, and accounts for more than 50% and up to 90% of the Quaternary pollen in Xinjiang (Pan, 1993
), northwestern China, where there are 153 species in 34 genera of this family (Zhu et al., 2003
). This group was chosen due to the great similarity in pollen grain morphology among species within a genus either with an optical or scanning electron microscope (Erdtman, 1978
).
The internal transcribed spacer (ITS1 and ITS2) regions of nuclear ribosomal DNA have been widely used for plant species identification (Shen et al., 1998
). However, this identification was based on fresh pollen grains using PCR and DNA sequencing (Petersen et al., 1996
). Here, we present a method for species identification of individual Chenopodiaceae pollen grains in surface soil based on ITS1 sequences.
MATERIALS AND METHODS
Site and sampling
The study region is located in the Junggar Desert Basin (400–500 m a.s.l.), Xinjiang, China. Winds from the northwest and the south are the most common in the region, determining the directions of pollen dispersal. Accordingly, a 220-km transect parallel to the major northwest wind and a 300-km transect parallel to the major south wind were established for sampling. Along the two transects, 134 plots, each 5 x 80 m2, were systematically sampled for plants and surface soil pollen.
One plant for each Chenopodiaceae species in a plot was collected and independently identified to species by two experienced botanists (K. P. Ma and H. B. Ren, Institute of Botany, Chinese Academy of Sciences) using morphological characteristics. A total of 19 species were identified.
The leaves of each species were molecularly analysed. Because Haloxylon ammodendron is without leaves and uses branches for photosynthesis, new growing shoots were collected. The samples were kept on ice in a cooler in the field and then stored at –20°C.
Each plot was further divided into sixteen 5 x 5 m2 subplots for collecting surface soil pollen. In each subplot, one soil sample was collected to a depth of 0.5 cm using a spatula. All soil samples from individual subplots were homogenized for analyses.
ITS sequence
The total DNA from 0.1 g fresh leaves of each species was extracted following the method of Doyle and Dolye (1987)
. The ITS region was amplified using a primer combination of N-nc18s10 and C26A (Wen and Zimmer, 1996
). Genomic DNA of 10–20 ng was added to 24 µL PCR mixture containing 1.5 mM MgCl2, 0.2 mM dNTP mixture (TaKaRa, Shiga, Japan), 0.4 µM primers, 1.5 U Taq DNA polymerase (Promega, Madison, Wisconsin, USA), and 2.5-µL 10x buffer (Promega). The PCR procedure was 94°C for 2 min; 30 cycles of 94°C for 30 s, 50°C for 1 min, and 72°C for 2 min; followed by a final extension of 72°C for 7 min. The PCR products were electrophoresed in a 1.0% agarose gel, purified with the Agrose Gel DNA Fragment Recovery Kit (TaKaRa), and sequenced with the ABI Prism Dye Terminator Cycle Sequencing Reaction Kit on an ABI3700 DNA Sequencer (Perkin-Elmer, Foster City, California, USA). The amplification primers were used as the sequencing primers. The sequencing was done by the GeneCore Company (Shanghai, China).
All ITS sequences of complete ITS1, 5.8S, and partial ITS2 were submitted to the GenBank Database (http://www.ncbi.nlm.nih.gov) under accession numbers AY556425–AY556443 (Appendix). Nineteen different complete ITS1 sequences, ranging from 236–242 bp, were aligned using ClustalX 1.81 (Thompson et al., 1997
).
Determination of similarity threshold
The similarity between ITS sequences is defined as a ratio of the same bases in a match (m) after alignment to all bases (a), given by S = 100m/a in percent, measuring the DNA molecular matches of two biological bodies (Altschul et al., 1997
). Compared to other species, a plant has the greatest similarity in DNA (or ITS) to another plant of the same species; therefore, we used this similarity to identify species for this molecular approach. The similarity within a species is not the same for different species (Baldwin et al., 1995
). The minimum similarity within a species for the 19 species is defined as the similarity threshold for our species identification. If the similarity between an unknown pollen and a species in our database was greater than or equal to the threshold, the unknown pollen was identified as this known species.
Due to budget limitations, we were not able to replicate the sequencing of all 19 species for our measurements of the similarity within a species, but we were able to infer a reasonable value of the threshold for our conservative (more confident) identification. Of the 19 species, Haloxylon ammondendron had the greatest variability in genotype (Y. Sheng, Institute of Botany, Chinese Academy of Sciences, unpublished manuscript); therefore, we used the minimum similarity in ITS within this species as the threshold, which was determined to be 98% after sequencing five individuals of this species. Thus, our defined threshold value must be lower than 98.0%. If we used 98.0% as the threshold, our identification must be conservatively accurate.
Among the 19 species, there were 171 pairwise similarities in the ITS1 sequences. The highest interspecific similarity in ITS was 93.4%, between Horaninowia ulicina and Haloxylon ammodendron. The similarity between the species was much smaller than 98%. Therefore, we were confident in using 98.0% as the threshold.
DNA extraction
A 10 g surface soil sample was washed through a 50 µm sieve (300 mesh) with 0.25 L sterilized water. The sieved mixture of soil and pollen was centrifuged at 5000 rpm for 10 min. The sediments were collected, resuspended, and diluted five times with sterilized deionized water. This diluted sample was placed on a sterilized slide under an optical microscope of 100–400x, and individual Chenopodiaceae pollen grains were picked up using a sterilized glass capillary tube. The pollen grains ranged from 15 to 35 µm in diameter. Each pollen grain was transferred to another sterilized slide and rinsed with sterilized deionized water four times. A new sterilized slide was placed on the slide with the pollen grain. A mark was made on the top slide to note the position of pollen grain. Then the slides were pressed together by hitting the top slide with a columned crabstick 5 mm in diameter to rupture the pollen grain wall. As a result, the cytoplasm was released onto the slide. The wall fragments and cytoplasm were transferred to a 0.5-µL Eppendorf tube containing 4 µL sterilized, deionized water. Then 20 µL InstaGene matrix (Bio-Rad, Hercules, California, USA) was added to the tube. Genomic DNA from the single pollen grain was extracted following the standard InstaGene matrix protocol. Then 2 µL supernatant and the genomic DNA of the pollen grain were added to a 23 µL PCR reaction mixture. As a blank control, 4µL deionized water was added to the InstaGene matrix and was treated in the same way as the extracted cytoplasm and wall fragments.
Amplification of the ITS1 sequence
The ITS1 region of a single pollen grain was amplified by a nested-PCR reaction with two pairs of primers: the outer primer pair (N-nc18s10 and C26A) and the inner primer pair [ITSD1 (5'-CCTGCGGAAGGATCATTGTC-3') and ITSD2 (5'-GCTAGCTTCTTCATCGATGC-3')].
The PCR reaction mixture contained 1.5 mM MgCl2, 0.2 mM dNTP mixture, 1.5U Pyrobest DNA polymerase (TaKaRa), and 2.5-µL 10x Pyrobest buffer (TaKaRa). The first round of PCR was processed the same way as for fresh leaves. It was performed in a volume of 25 µL by adding 2 µL pollen genomic DNA and 0.8 µM outer primer pair (0.4 µM from each) to the reaction mixture while three blanks were used as negative controls. The second round of amplification was completed in a volume of 25 µL with 1 µL product from the first-round PCR, 1 µL sterilized water, and 0.8 µM inner primer pair (0.4 µM from each) in the reaction mixture. The same PCR procedure was used in the second, except that the annealing temperature was increased to 59°C. The amplified products of the second-round PCR were electrophoresed on a 1.0% agarose gel and then purified with the Agrose Gel DNA Fragment Recovery Kit (TaKaRa). The negative control did not produce any band in their PCR (Fig. 1).
|
(DGBIO, Beijing, China) by the heat-shock method (Sambrook and Russell, 2001
). For plasmid isolation, the positive clones were screened by the PCR screening method (Sambrook and Russell, 2001
) using the inner primer pair under the same reaction conditions as in the second round of the PCR procedure. Plasmids were then isolated using the rapid alkaline lysis method of Ish-Horowitz and Burke (1981)
. The sequencing method was the same as that used for fresh leaves except that the primers T7 and SP6 (Promega) were used. As identified, there are 19 Chenopodiaceae species in the study area (see Appendix). The sequences were aligned with the sequences of the 19 known species in ITS1 database using the ClustalX program. A neighbor-joining dendrogram (Saitou and Nei, 1987
) was produced with the program MEGA version 3.1 (Kumar et al., 2004
) and 500 random addition sequence replicates. Ranunculus sceleratus (Ranunculaceae) was used as the outgroup. RESULTS AND DISCUSSION
Identification results
Three surface soil samples in the middle section of each of the two transects were randomly selected, and the six samples were mixed into one working sample. Twenty pollen grains were separated from this working sample, and their genomic DNA was extracted. The ITS1 regions from eight of 20 pollen grains were successfully amplified and sequenced. The similarities in the ITS1 sequence of each pollen grain to the 19 known species were separately calculated. The greatest similarity in the ITS1 sequence of each pollen to a known species (98.1%-99.2%) in the database is shown in the NJ dendrogram (Fig. 2). The lowest similarity of 98.1% is greater than the threshold value of 98.0%. The eight pollen grains belonged to five species (W-C2 to Ceratocarpus arenarius, Sp1–1 and C5–17 to Bassia dasyphylla, C7 to Halostachys caspica, 47–3 to Suaeda physophora, and C6–7, C9–13, and C5–9 to Haloxylon ammodendron). We believe these identification results are satisfactory for this preliminary test.
|
), the PCR and the sequencing are both highly efficient in terms of labor and time.
Crushed grains without subsequent DNA extraction were also used as templates in the PCR reaction. The PCR products were weak but consistent with Petersen et al. (1996)
who demonstrated that ITS sequences could be amplified from Hordeum vulgare (Gramineae) or Secale strictum (Gramineae) without genomic DNA extraction.
Contamination is a major concern in genomic DNA extraction and PCR procedures; therefore, a negative DNA control must be used in all procedures. Primers ITSD1 and ITSD2 were designed for flowering plants as determined by sequence comparison to GenBank. Compared with soil samples, airborne pollen has less contamination (Menking et al., 1999
). Thus, our method is likely to be applicable to identify species of airborne pollen.
The successful rate of DNA amplification
Matsunaga et al. (1999)
amplified DNA of a single fresh pollen with relative ease. However, our efforts to obtain and amplify the DNA from a pollen grain in surface soil were more difficult because pollen contents decompose and leak into the soil. However, our PCR-based method to obtain the ITS1 fragments from individual Chenopodiaceae pollen in surface soil is still a relatively simple technique, which can be tried for other plant families.
In our study, 40% (eight of 20) of the pollen had successfully amplified DNA. This percentage seems low, but is acceptable for pollen identification because a large number of pollen grains is generally available. The lack of amplification in some cases might be caused by pollen DNA degradation in the dry, hot environment of the sampled pollen, especially if the pollen had been in the soil for over a year. This degradation hypothesis was tested using five fresh single pollen of Atriplex cana as positive controls. The ITS1 sequences of the four fresh Atriplex cana pollen grains were amplified and matched to the ITS1 sequence of Atriplex cana in the database with similarities of 98.8–99.6%, verifying the hypothesis.
Nevertheless, the high percentage (4 out of 5) of amplified DNA from fresh pollen only signifies that different tissues can be amplified with differing efficiencies. There must be other causes for the PCR failure. For example, sequence mismatching during the PCR may be one cause, and the conditions for the PCR procedure may be another. Further efforts are needed to address the causes in order to refine the technique.
Threshold of similarity
This methodology is feasible if reference sequences are available for all species in the families in a database such as GenBank or are established by extracting and sequencing DNA from fresh leaf or pollen samples. Moreover, the threshold of similarity was variable for different families or different genera. The threshold of 98.0% in this study was higher than the 92.0% used by Peterson et al. (1996) for Secale strictum. If overlap does not occur between species with high intraspecific divergence and between species with low or lacking intraspecific divergence, the threshold of similarity should be the minimum similarity among individuals of species with the lowest intraspecific divergence. Nevertheless, our determined threshold may not be universal for the Chenopodiaceae family because taxon sampling in this study was limited. If taxon sampling were expanded, then the maximum pairwise similarity in ITS1 sequences between species might be greater and the minimum similarity within species lower. From our results for 19 species, we conclude that the maximum similarity between species may be higher than 93.4% and the minimum similarity within a species may be lower than 98%. The threshold for more species, however, must be within 93.4–98.0%.
Potential problems
Chenopodiaceous pollen are not dispersed in the air as far as those of Picea, which have airbags; thus their composition in the soil surface can be used to describe the vegetation with reliable accuracy (Li et al., 2000
). Perhaps the pollen can be carried kilometers by the wind, but the study region is large enough that pollen are not likely to have been carried from outside our study area to the center of site for collecting surface soil. Nevertheless, it is possible that we were not able to sample all species of Chenopodiaceae in the region. Thus, the sequence database established through sampling plants by us might lack some species. If this was the case the ITS1 region from a pollen grain of an unsampled species that had an ITS1 sequence with higher similarity to one of our 19 species, then the ITS1 sequences would be incorrectly identified. Such a scenario can be avoided if more species from extended regions were included or if a species-specific DNA fragment is used as a marker to identify each species.
Summary
We have developed a molecular method for identifying the species of surface soil pollen. The method was developed using 19 species in the Chenopodiaceae family, providing a reference for other species. This method will be help improve the pollen–vegetation models that are used to interpret the composition of fossil pollen in Quaternary research. This method may also be used in allergy, clinical, and forensic research for the diagnostics of pollen in the air, dust and soil.
APPENDIX
Taxon—Voucher; GenBank ITS sequence accession number.
Agriophyllum squarrosum (L.) Moq.—PE 00120509; AY556431.
Atriplex cana C. A. Mey.—PE 00121008; AY556428.
Bassia dasyphylla (Fisch. et Mey.) O. Kuntze.—PE 00146119; AY556434.
Ceratocarpus arenarius L.—PE 00418311; AY556430.
Ceratoides latens (J. F. Gmel.) Reveal et Holmgren.—PE 00146575; AY556427.
Corispermum declinatum Steph.ex Stev—PE 00542306; AY556432.
Halostachys caspica (Bieb.) C. A. Mey.—PE 00541081; AY556429.
Haloxylon ammodendron (C. A. Mey.) Bunge.—PE 00146302; AY556438.
Horaninowia ulicina Fisch. et Mey.—PE 00165575; AY556437.
Kalidium caspicum (L.) Ung. Sternb.—PE 00541521; AY556426.
Kalidium foliatum (Pall.) Moq.—PE 00509737; AY556443.
Kochia prostrata (L.) Schrad.—PE 00526039; AY556433.
Petrosimonia sibirica (Pall.) Bunge.—PE 00526794; AY556524.
Salsola korshinskyi Drob.—PE 00515850; AY556441.
Salsola nitraria Pall.—PE 00528019; AY556439.
Salsola praecox Litv.—PE 00528312; AY556442.
Salsola subcrassa M. Pop.—PE00528607; AY556440.
Suaeda physophora Pall.—PE 00527387; AY556435.
Suaeda salsa (L.) Pall.—PE 00527572; AY556436.
FOOTNOTES
1 The authors thank H.B. Ren for his field work. This study was funded by the National Natural Science Foundation of China (NSFC 30670403) and the State Key Basic Research and Development Plan (G2000046803). 
2 Author for correspondence (e-mail: makp{at}brim.ac.cn
), fax: +86-10-82591781
LITERATURE CITED
Altschul S. F. Madden T. L. Schäffer A. A. Zhang J. H. Zhang Zh. Miller W. Lipman D. J.. 1997. Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Research 25: 3389-3402.
Baldwin B. G. Sanderson M. J. Porter J. M. Wojciechowski M. F. Campbell C. S. Donoghue M. J.. 1995. The ITS region of nuclear ribosomal DNA: a valuable source of evidence on angiosperm phylogeny. Annals of the Missouri Botanical Garden 82: 247-277.
Davies P. C. Fall P. L.. 2001. Modern pollen precipitation from an elevational transect in central Jordan and its relationship to vegetation. Journal of Biogeography 28: 1195-1210.
Doyle J. J. Dolye J. L.. 1987. A rapid DNA isolation procedure for small quantities of fresh leaf tissue. Phytochemistry Bulletin 19: 11-15.
Erdtman G.. 1978. The handbook of pollen analysis Science Press, Beijing, China (translated into Chinese).
Hao H. P. Zhang J. T. Yan Sh.. 1989. Scanning electron microscope observation on the pollen grains of Chenopodiaceae. Acta Botanica Sinica 31: 650-652.
Ish-Horowicz D. Burke J. F.. 1981. Rapid and efficient cosmid cloning. Nucleic Acids Research 9: 2989-2998.
Joosten H. de Klerk P.. 2002. What's in a name? Some thoughts on pollen classification, identification, and nomenclature in Quaternary palynology. Review of Palaeobotany and Palynology 122: 29-45.[CrossRef][ISI]
Kumar S. Tamura K. Nei M.. 2004. MEGA3: integrated software for molecular evolutionary genetics analysis and sequence alignment. Briefings in Bioinformatics 5: 150-163.
Li Y. Y. Zhang X. Sh. Zhou G. Sh. Ni J.. 2000. The quantity relationship between surface pollen and vegetation in north of China. Chinese Science Bulletin 45: 761-765.
Matsunaga S. Schutze K. Donnison I. S. Grant S. R. Kuroiwa T. Kawano S.. 1999. Single pollen typing combined with laser-mediated manipulation. Plant Journal 20: 371-378.[CrossRef][ISI][Medline]
Menking D. E. Emanuel P. A. Valdes J. J. Kracke S. K.. 1999. Rapid cleanup of bacterial DNA from field samples. Resources, Conservation and Recycling 27: 179-186.
Mullins J. Emberlin J.. 1997. Sampling pollens. Journal of Aerosol Science 28: 365-370.
Pan A. D.. 1993. The study on morphology pollen of Chenopodiaceae, Xinjiang. Arid Land Geography 3: 22-27 (in Chinese).
Petersen G. Johansen B. Seberg O.. 1996. PCR and sequencing from a single pollen grain. Plant Molecular Biology 31: 189-191.[CrossRef][ISI][Medline]
Saitou N. Nei M.. 1987. The neighbor-joining method: a new method for reconstructing phylogenetic trees. Molecular Biology and Evolution 4: 406-425.[Abstract]
Sambrook J. Russell D. W.. 2001. Molecular cloning: a laboratory manual Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York, USA.
Shen Y. L. Newbury H. J. Ford-Lloyd B. V.. 1998. Identification of taxa in the genus Beta using ITS1 sequence information. Plant Molecular Biology Reporter 16: 147-155.
Thompson J. D. Gibson T. J. Plewniak F. Jeanmougin F. Higgins D. G.. 1997. The ClustalX windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Research 24: 4876-4882.
Wen J. Zimmer E. A.. 1996. Phylogeny of Panax L. (the ginseng genus, Araliaceae): inferences from ITS sequences of nuclear ribosomal DNA. Molecular Phylogenetics and Evolution 6: 167-179.[CrossRef][ISI][Medline]
Zhu G. L. Mosyakin S. L. Clemants S. E.. 2003. Chenopodiaceae. In Flora of China Editorial Committee [eds.], Flora of China, vol. 5 351-414 Science Press, Beijing, China.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
