- Email: [email protected]
Comparison of efficacy of three extractants to solubilize glomalin on hyphae and in soil
Chemosphere 64 (2006) 1219–1224 www.elsevier.com/locate/chemosphere
Comparison of efficacy of three extractants to solubilize glomalin on hyphae and in soil S.F. Wright
a,*
, K.A. Nichols b, W.F. Schmidt
c
a
Sustainable Agricultural Systems Laboratory, United States Department of Agriculture, Agricultural Research Service, Bldg. 001, Rm 140, BARC-W, Beltsville, MD 20705, USA b Northern Great Plains Research Laboratory, United States Department of Agriculture, Agricultural Research Service, 1701 10th Ave. SW, P.O. Box 459, Mandan, ND 58554, USA c Environmental Quality Laboratory, United States Department of Agriculture, Agricultural Research Service, Bldg. 012, BARC-W, Beltsville, MD 20705, USA Received 30 August 2005; received in revised form 2 November 2005; accepted 8 November 2005 Available online 5 January 2006
Abstract Glomalin, a glycoprotein produced by arbuscular mycorrhizal (AM) fungi, is a major component of the humus fraction of soil organic matter. Glomalin is extracted from soil and hyphae of AM fungi by using sodium citrate at 121 °C in multiple 1-h cycles, but extensive extraction does not solubilize all glomalin in all soils. Efficacies of 100 mM sodium salts of citrate, borate or pyrophosphate (pH 9.0, 121 °C) were tested for two 1-h cycles for hyphae from four AM fungal isolates and four 1-h cycles for seven soils from four US geographic regions. Residual soil glomalin was examined by pyrophosphate extraction of soils previously extracted with citrate or borate followed by extraction of all soils after treatment with NaOH. Hyphal extracts were compared using Bradford-reactive total protein (BRTP) values, and extracts from soils were compared using BRTP, percentage C and C weight. No difference among extractants was detected for AM fungal isolates or across soils. The residual glomalin across soils for extractants contained the following percentages of the total BRTP: pyrophosphate, 14%; borate, 17%; and citrate, 22%. Comparisons among individual soils indicated that pyrophosphate extracted significantly more BRTP (10–53%) than borate or citrate in six soils and borate was equal to pyrophosphate in one soil. Extraction with borate should be compared with pyrophosphate before initiating an experiment. For routine extractions of ca. 85% of the glomalin across a variety of soils, sodium pyrophosphate appears to be equal to or better than borate and better than citrate. Ó 2005 Elsevier Ltd. All rights reserved. Keywords: Humin; Soil organic matter; Arbuscular mycorrhizal fungi
1. Introduction Glomalin, a brown to red-brown colored glycoprotein produced by arbuscular mycorrhizal (AM) fungi, is a major component of soil organic matter that currently is defined operationally by the extraction method (Nichols and Wright, 2005). Solubilization of glomalin on hyphae or in soil is accomplished by using 20–50 mM citrate, pH
*
Corresponding author. Tel.: +1 301 504 8156; fax: +1 301 504 8370. E-mail address: [email protected] (S.F. Wright).
0045-6535/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.chemosphere.2005.11.041
7.0 or 8.0 at 121 °C (Wright and Upadhyaya, 1996; Wright et al., 1996). Extractions are repeated by replenishing the extractant in 1-h cycles until the supernatant is almost colorless. The extract operationally defined as glomalin is relatively free of co-extracted proteins according to gel electrophoresis profiles of citrate extracts of diverse AM fungi or soils which show consistent bands with few, if any, extraneous bands (Wright and Upadhyaya, 1996; Rillig et al., 2001). When glomalin was first identified, sodium citrate (20 mM) at pH 7.0 and 121 °C solubilized glomalin on hyphae (Wright et al., 1996) and was the only solution of many organic and inorganic solutions tested that
1220
S.F. Wright et al. / Chemosphere 64 (2006) 1219–1224
successfully solubilized the compound (Wright and Upadhyaya, 1996; unpublished data). Citrate was used when the extraction of glomalin was extended from hyphae to soils, but higher citrate concentration and higher pH released greater amounts of glomalin from soils than pH 7.0, 20 mM citrate (Wright and Upadhyaya, 1996). Citrate (50 mM), pH 8.0 became the standard solution used for soils. Solubilized glomalin remains in solution at an alkaline pH and flocculates slowly in a mild acid (pH < 5.5) (unpublished data) or rapidly at pH 2.5 (Nichols and Wright, 2005). Therefore, to effectively extract glomalin, alkaline pH of the extractant was thought to be a necessity. Small amounts of glomalin in soil are solubilized by using water at 121 °C (Wright and Upadhyaya, 1996). Rillig and Steinberg (2002) used water at 121 °C to remove glomalin from hyphae produced in root organ cultures. Although an unknown amount of White’s M medium (pH 5.5) was also in the extraction solution, diluted White’s medium contains very low concentrations of salts and no citrate. The effectiveness of citrate for glomalin extraction decreases at 110 °C (Wright and Upadhyaya, 1996; Wright and Jawson, 2001), and only a small amount of glomalin is extracted at room temperature (Wright and Upadhyaya, 1996). These results show the importance of temperature but are confounding about the importance of citrate. We previously proposed that citrate acted as a competitive chelator of iron associated with glomalin (Wright and Upadhyaya, 1998) and thereby assisted in solubilization of the molecule. Glomalin contains iron, but amounts differ for glomalin extracted from pot-cultured hyphae and soil. Iron associated with glomalin from several soils contained 0.8–8.8% iron (Wright and Upadhyaya, 1998). Glomalin freshly produced on hyphae has Fe concentrations of 60.3% (Nichols, 2003). As work on glomalin progressed, it became apparent that citrate is too weak a chelator to compete for iron on glomalin from either hyphae or soil. Some, but not all, iron is released from glomalin by using 8-hydroxyquinolin or Tiron (4,5-dihydroxy-1,3-benzene disulfonic acid) (Nichols, 2003). Chelators such as EDTA (ethylenediaminetetraacetic acid) or EDDHA (ethylenediamine-di(o-hydroxyphenylacecid acid)) unsuccessfully compete with glomalin for iron (unpublished data). Sodium pyrophosphate is used to extracted humic substances from soils (Stevenson, 1994; Clapp and Hayes, 1999). Monovalent sodium disrupts Fe- and Al-(hydr) oxides acting as bridges between organic matter and clay minerals (Clapp and Hayes, 1999). Glomalin may be released from organo-mineral complexes in the soil by a similar reaction. Sodium borate was used to extract glomalin from soils when citrate would interfere with subsequent tests for heavy metals attached to glomalin (Gonza´lezCha´vez et al., 2004). Glomalin is co-extracted with humic acid (HA) in 0.1 M NaOH even after first extracting glomalin with citrate (Nichols, 2003; Nichols and Wright, 2005). This suggests that a residual fraction of glomalin is not released by using citrate even though soil is repeatedly extracted.
The role of sodium in the efficiency of extraction has not been resolved. Sodium borate pentahydrate contains 12% sodium, approximately half of the sodium in trisodium citrate (23%) or sodium in sodium pyrophosphate pentahydrate (21%). Therefore, comparisons among sodium salts of borate, pyrophosphate and citrate may resolve the question of the importance of sodium in efficiency of glomalin extraction and yield information on the purity of glomalin. The objectives of this experiment were to (1) compare efficacies of extractants containing different amounts of sodium to extract glomalin from hyphae of four AM fungi using protein yield, (2) compare the same extractants for seven soils using protein yield, percentage carbon, and carbon weight (based on gravimetric weight of freeze-dried glomalin), and (3) explore additional extractions to solubilize a residual glomalin fraction from soils. 2. Materials and methods 2.1. Extractants Glomalin was extracted from hyphae or soil with 100 mM, pH 9.0, sodium citrate (C6H5Na3O7 Æ 2H2O), sodium borate (Na2B4O7 Æ 10H2O), or sodium pyrophosphate (Na4P2O7 Æ 10H2O). The pH was adjusted with HCl. 2.2. Extraction from arbuscular mycorrhizal fungi Four arbuscular mycorrhizal (AM) fungal isolates— Gigaspora (Gi.) rosea (FL224), Acaulospora (A.) morrowiae (CL551), Glomus (G.) etunicatum (BR220), and G. intraradices (WV964) were grown on corn (Zea mays) in sterile, single-species pot cultures. A sterile mixture of glomalinfree coarse sand and coal was placed in 15 cm diameter (1300 cc) pots. Three disinfected seeds plus spores of Gi. rosea and spores and hyphae for the other inocula (>100 infectious units per plant) were placed in 8 cm diameter 38 lm nylon mesh bags in the centers of three replicate pots to contain plant roots while providing a root-free hyphal chamber (Nichols and Wright, 2004). Pots were harvested at 12 weeks of plant growth according to the following protocol. Shoots and the nylon mesh bag containing roots were removed and discarded, and hyphae were removed from the sand outside of the nylon mesh bag by rinsing the sand with water and washing the hyphae into a 53 lm sieve. Hyphae from each isolate was divided into approximately equal amounts and placed in extraction tubes for extraction with citrate, pyrophosphate or borate. Sufficient volumes of extractants were added to cover the hyphae. After 1 h at 121 °C extracts in the supernatant were collected by centrifugation at 10 844g for three min. A second extraction was performed under the conditions described above. Hyphae were rinsed into weigh boats, dried at 70 °C and weighed. Glomalin is reported in mg g 1 hyphae.
S.F. Wright et al. / Chemosphere 64 (2006) 1219–1224
2.3. Extraction from soils Seven soils were used—two soils from each of three geographic areas: Colorado (Sampson and Haxtun soil series), Maryland (Baltimore site A and Baltimore site B soil series), Georgia (Cecil and Pacolet soil series) and one soil from Nebraska (Pawnee soil series). See Nichols and Wright (2005) for soil analysis values. These soils are all slightly acidic to acidic loams with organic carbon contents ranging from 7.0 to 30.3 g kg 1. Three replicate 1-g subsamples of each soil were used for each extraction solution described above. All samples were subjected to four 1-h extraction cycles until the supernatant color was no longer brown. Supernatants were removed for separate analysis after each 1-h cycle. After four 1-h extraction cycles, the remaining soil from citrate and borate extractions was extracted for 1 h at 121 °C with the pyrophosphate solution. Following all of the above extractions, the three subsamples of soil were combined and 2 ml 0.1 M NaOH was added to cover the samples as was done by Nichols and Wright (2005) in the protocol to extract HA after extracting glomalin. Samples were shaken at room temperature overnight. The supernatant was neutralized with HCl and saved for protein analysis. Previous work indicated that separation of glomalin from HA in NaOH is not possible
Location
Georgia Cecil
Maryland Pacolet
Soil Series
Site A
Clay (g kg-1)* 110 12
Bradford-reactive protein (mg g-1)
Baltimore Baltimore
220
250
1221
at this stage of an extraction process (Nichols and Wright, 2005) so the NaOH extract was not processed further to obtain the C content of glomalin. All soil pellets were extracted with 50 mM citrate, pH 8.0 for 1-h cycles at 121 °C until the supernatant was straw-colored. This required two or three cycles. Supernatants from these extractions were pooled. Supernatants from all extractions were collected by centrifugation at 6850g for 10 min. 2.4. Protein and carbon analyses Analyses were performed on samples as described by Wright et al. (1996). Bradford-reactive total protein (BRTP) values for two extractions of hyphae were summed. Values for BRTP were obtained for the separate soil extraction supernatants described above and are reported for the routine extraction sequence for the first cycle and the sum of cycles 2–4, the follow-up pyrophosphate extraction, the NaOH extraction, and the pooled final citrate extracts. BRTP values were corrected for background color contributed by the extractant. Protein in all soil extract supernatants except the treatment with 0.1 M NaOH was precipitated by acidification with HCl, dialyzed against water, freeze-dried, weighed, and analyzed for percentage C by combustion using a Perkin–Elmer Series II C, H, N,
Colorado
Nebraska
Sampson Haxtun
Pawnee
Site B 260
160
110
200
* Data from Nichols and Wright (2005) Routine Follow-up pyrophosphate Residual
10
8
6
4
2 b a a
c b a
b b a
b b a
b c a
b c a
b c a
C B P
C B P
CB P
C B P
CBP
CB P
CBP
0
Fig. 1. Comparison of Bradford-reactive protein (BRTP) in seven soils extracted with 100 mM, pH 9.0 sodium citrate (C), sodium borate (B), or sodium pyrophosphate (P) at 121 °C. Routine extraction series = mean values for the amount extracted from three replicates of each soil in four 1-h extraction cycles. Follow-up pyrophosphate = mean values for the pyrophosphate extracts of three replicates of each soil previously extracted with citrate or borate. Residual = the amount in combined soil replicates treated with 0.1 M NaOH and then extracted with 50 mM citrate, pH 8.0 for two or three 1-h cycles at 121 °C citrate. SD is shown in white within stacked bars. Means separation of values for the routine extraction sequence for each soil are indicated by letters within bars (LSD a = 0.05).
1222
S.F. Wright et al. / Chemosphere 64 (2006) 1219–1224
S/O 2400 Analyzer (Rillig et al., 2001; Nichols and Wright, 2005). Values are reported in mg g 1 hyphae or soil. 2.5. Soil clay content Values for clay content are from analysis of these soils by Nichols and Wright (2005) (Fig. 1). 2.6. Statistical analysis Mean values for extractants across all soils and assay sequences described above, the sums of mean values for all sequential extracts across soils, values for the four AM isolates as replicates, and values for replicate routine sequence extracts within soils were analyzed. Differences among the extractants were tested by one-way analysis of variance (ANOVA), after Bartlett’s test of equal variances was satisfied, using Statistics (Analytical Software, Tallahassee, FL). Pair-wise comparisons were by LSD (a = 0.05). Pearson product moment correlation coefficients were calculated for soil clay content and BRTP. When necessary, sine or cosine transformations of data were made to meet the assumption of equal variances. 3. Results 3.1. Glomalin from hyphae BRTP yields for AM fungal isolates by extractant are shown in Table 1. Differences were not significant among different extractants across the fungal isolates (p = 0.27), but variation in BRTP was high.
The first extract in the routine extraction sequence contained the following BRTP percentages of the totals across soils: citrate, 44%; borate, 46%; and pyrophosphate, 60% (Table 1). Extracts to define a recalcitrant pool of glomalin showed that sodium pyrophosphate removed BRTP from soils that had previously been extracted in the routine scheme with citrate or borate. When soil samples were extracted in the routine extraction sequence with pyrophosphate, only 14% of the BRTP remained in a residual pool; whereas, when citrate and borate were used in the routine extraction sequence, 22% and 17%, respectively, of the total was in follow-up pyrophosphate extractable and NaOH mediated citrate-extractable glomalin (Table 1). The NaOH extract contained the lowest amount of protein for all extraction stages, but this treatment assisted in the citratemediated extraction of small amounts of residual glomalin. Extractants varied in efficacy in individual soils and between soils from the same geographic region (Fig. 1). In the routine extraction sequence pyrophosphate solubilized from 10% to 53% more BRTP than citrate or borate in six of the seven soils (Fig. 1). In the Georgia soils, pyrophosphate was more effective in Pacolet than in Cecil when compared with the other extractants. In Pawnee, Sampson and Haxtun soils, borate was less effective than pyrophosphate or citrate in extracting glomalin. Clay content of soils was not significantly linearly correlated with BRTP extracted by borate, citrate or pyrophosphate (n = 7) in the routine extraction sequence (r = 0.37, 0.24 and 0.20, respectively), the residual amounts extracted (r = 0.51, 0.33 and 0.55, respectively) or the combined total amounts (r = 0.27, 0.26, and 0.23, respectively). 4. Discussion
3.2. Glomalin from soil No differences were detected for BRTP, percentage C, or C weight across all soils for the different extractants except for BRTP in the NaOH treatment (Table 1). These results indicated that the various extractants did not significantly contaminate the products, and differences in sodium concentration, within the range tested, was not a factor in extractant efficacy.
Table 1 Mean values followed by (SD) of glomalin as measured by Bradfordreactive total protein extracted from hyphae of arbuscular mycorrhizal fungi with 100 mM, pH 9.0, sodium salts of citrate, borate or pyrophosphate Extractant
Mean (SD)
Citrate Borate Pyrophosphate
5.37 (4.09) 7.45 (3.55) 9.84 (3.10) p = 0.27*
Hyphae of the following isolates were used as replicates: Gigaspora rosea, Acaulospora morrowiae, Glomus intraradices and Glomus etunicatum. * One-way ANOVA with four replicates.
Differences in efficacy of extractants across soils or hyphae were not detected, and the sodium ion concentration within the range tested did not influence extraction of glomalin. The trend was that pyrophosphate was more effective than the other extractants (Tables 1 and 2). A small amount of soil glomalin was in a residual pool that could be removed only by treatment with NaOH followed by heat-mediated extraction (Table 2). The three extractants were tested first for the ability to solubilized glomalin on hyphae before making comparisons about the ability to solubilize glomalin in a complex matrix, soil. The extractants tested removed glomalin from hyphae, but high variability in BRTP related to hyphal weight (Table 1) may be due to inclusion of spores, which disproportionately contribute weight and not glomalin to a sample. Spores are very difficult to remove from bulk hyphal samples such as the ones used in this work. Previous work also indicates high variability in the ratio of glomalin to hyphal length (CV = 28%) (Lovelock et al., 2004). No differences were detected among extractants within an isolate (data not shown). Comparison of extractants within soils (Fig. 1) showed that pyrophosphate was superior to citrate and borate in
S.F. Wright et al. / Chemosphere 64 (2006) 1219–1224
1223
Table 2 Mean values of glomalin Bradford-reactive total protein, percentage C and C weight for seven soils from four geographic regions of the US extracted with three different sodium salt extraction solutions examined in the sequence presenteda,b Routine extraction sequence for soils
Extracts to define a residual pool of glomalin
Assay sequence
1
2
3
4
5
Extractant (100 mM, pH 9 at 121 °C for 1 h)
Extract 1
Extracts 2, 3, 4
Follow-up pyrophosphate
NaOH
Residual (extracted with citrate)
1.58 (0.45) 1.19 (0.46) 1.26 (0.39) 0.240
ND 0.70 (0.54) 0.53 (0.21) 0.631
0.09 (0.05)b 0.22 (0.15)a 0.23 (0.11)a 0.046
45.07 (1.50) 43.71 (3.57) 36.13 (7.77) 0.851
38.44 (2.63) 46.61 (2.08) 32.30 (7.46) 0.630
ND 43.98 (2.54) 41.79 (2.85) 0.155
ND ND ND
51.86 (7.46) 48.10 (1.85) 50.56 (5.04) 0.715
3.73 (1.12) 2.98 (1.82) 2.95 (1.13) 0.511
1.54 (0.56) 1.49 (0.68) 1.43 (0.45) 0.938
ND 2.10 (1.64) 1.70 (0.76) 0.573
ND ND ND
0.42 (0.15) 0.31 (0.11) 0.45 (0.08) 0.068
Bradford-reactive total protein (mg g 1 soil) Pyrophosphate 3.61 (1.50) Borate 2.33 (1.49) Citrate 2.36 (0.94) p= 0.150 Carbon (%) Pyrophosphate Borate Citrate p= Carbon weight (mg g Pyrophosphate Borate Citrate p=
1
0.73 (0.35) 0.65 (0.27) 0.96 (0.45) 0.281
Total
6.01 (2.22) 5.09 (2.04) 5.35 (1.94) 0.696 ND ND ND
soil) 5.69 (1.73) 6.88 (2.58) 6.54 (2.04) 0.577
a
Different letters in a column following values and SD in parentheses indicate differences according to LSD (a = 0.05) analyzed by one-way ANOVA. Data was sine or cosine transformed when needed to satisfy Bartlett’s test of equal variances. b ND indicates that the sample was not extracted or that the analysis could not be done.
six of the seven soils. Stability constant rankings of Fe(III) with the ligands used are: pyrophosphate > citrate > borate (www.coldcure.com/html/stability_constants.html; Elrod and Kester, 1980). Stability of ligand-Fe did not appear to explain differences in extractants because borate extracted more glomalin than citrate in the routine sequence for four of the seven soils tested (Fig. 1). There were no significant correlations between BRTP yields for different extractants and Fe concentration in soils using values from Nichols and Wright (2005) (data not shown). Borate interacts with glycoproteins (Kennedy and How, 1973) and pyrophosphate may bind to Fe in glomalin. Also, nuclear magnetic resonance (NMR) spectroscopy indicates that citrate is present in citrate-extracted glomalin, potentially adding weight and C to the molecule (Nichols, 2003). However, percentage C and C weight of glomalin did not indicate that weight was different for glomalin extracted by any of the extractants (Table 1). For each extraction cycle 8 ml of a 100 mM solution g 1 soil was used (Wright and Upadhyaya, 1996). Therefore, for each extraction cycle, solutions contained from 232 to 356.9 mg of the salts of extractants tested (citrate the lowest and pyrophosphate the highest amounts). If extractants bind irreversibly to glomalin in large amounts during the extraction procedure, gravimetric analyses, such as percentage C and C weight would be affected. Clay content was not correlated with BRTP, however, clay type may influence efficiency of an extractant in a particular soil. Pacolet and Pawnee have approximately the same amount of clay (Fig. 1), but Pacolet clay is kaolinitic and Pawnee clay is smectitic (http://ortho.ftw.nrcs.usda.
gov). Clays in the remaining soils are of mixed types. The influence of smectitic clay on borate extraction needs to be explored further. Release of residual glomalin facilitated by 0.1 M NaOH treatment of soils may be due to dispersion of clays, but is not recommended as a pre-treatment of soils before extraction and quantification of glomalin. Nichols and Wright (2005) show that 0.1 M NaOH co-extracts HA and a fraction of the total glomalin, and HA and glomalin are not easily separated. Therefore, for routine extractions of ca. 85% of the glomalin across a variety of soils, sodium pyrophosphate appears to be equal to or better than borate and better than citrate. Citrate-extracted glomalin 1H NMR spectra show broad peaks that are not adequately definitive for structural interpretation and indicate that citrate binds to glomalin (Nichols, 2003). Although pyrophosphate and borate 1H NMR spectra show no apparent binding, the spectra are not greatly improved over citrate. Further work will be necessary to find an extractant that can be used to solubilize glomalin for 1H NMR structural elucidation. In conclusion, pyrophosphate was superior to citrate or borate in extracting BRTP within the same soil. Glomalin on hyphae was extracted by all three extractants. Differences in percentage C and C weight due to the extractant used were not evident. References Clapp, C.E., Hayes, M.H.B., 1999. Characterization of humic substances isolated from clay- and silt-sized fractions of a corn residue-amended agricultural soil. Soil Sci. 164, 899–913.
1224
S.F. Wright et al. / Chemosphere 64 (2006) 1219–1224
Elrod, J.A., Kester, D.R., 1980. Stability constants of iron(III) borate complexes. J. Solution Chem. 9, 885–894. Gonza´lez-Cha´vez, M.C., Carrillo-Gonza´lez, R., Wright, S.F., Nichols, K.A., 2004. The role of glomalin, a protein produced by arbuscular mycorrhizal fungi, in sequestering potentially toxic elements. Environ. Pollut. 130, 317–323. Kennedy, G.R., How, M.J., 1973. Interaction of sugars with borate, an NMR spectroscopic study. Carbohydr. Res. 28, 13–19. Lovelock, C.A., Wright, S.F., Nichols, K.A., 2004. Using glomalin as an indicator for arbuscular mycorrhizal hyphal growth: an example from a tropical rain forest soil. Soil Biol. Biochem. 36, 1009–1012. Nichols, K.A., 2003. Characterization of glomalin—a glycoprotein produced by arbuscular mycorrhizal fungi. Ph.D. Thesis, University of Maryland, College Park, MD. Nichols, K.A., Wright, S.F., 2004. Contributions of soil fungi to organic matter in agricultural soils. In: Magdoff, F., Weil, R. (Eds.), Functions and Management of Soil Organic Matter in Agroecosystems. CRC Press, Boca Raton, FL, pp. 179–198. Nichols, K.A., Wright, S.F., 2005. Comparison of glomalin and humic acid in eight native US soils. Soil Sci. 170, in press.
Rillig, M.C., Steinberg, P.D., 2002. Glomalin production by an arbuscular mycorrhizal fungus: a mechanism of habitat modification?. Soil Biol. Biochem. 34 1371–1374. Rillig, M.C., Wright, S.F., Nichols, K.A., Schmidt, W.F., Torn, M.S., 2001. Large contribution of arbuscular mycorrhizal fungi to soil carbon pools in tropical forest soils. Plant Soil 233, 167–177. Stevenson, F.J., 1994. Humus chemistry, Genesis, composition, reactions, second ed. John Wiley and Sons Inc., New York. Wright, S.F., Jawson, L., 2001. A pressure cooker method to extract glomalin from soils. Soil Sci. Soc. Am. J. 65 (6), 1734–1735. Wright, S.F., Upadhyaya, A., 1996. Extraction of an abundant and unusual protein from soil and comparison with hyphal protein of arbuscular mycorrhizal fungi. Soil Sci. 161, 575–586. Wright, S.F., Upadhyaya, A., 1998. A survey of soils for aggregate stability and glomalin, a glycoprotein produced by hyphae of arbuscular mycorrhizal fungi. Plant Soil 198, 97–107. Wright, S.F., Franke-Snyder, M., Morton, J.B., Upadhyaya, A., 1996. Time-course study and partial characterization of a protein on hyphae of arbuscular mycorrhizal fungi during active colonization of roots. Plant Soil 181, 193–203.
Comparison of efficacy of three extractants to solubilize glomalin on hyphae and in soil S.F. Wright
a,*
, K.A. Nichols b, W.F. Schmidt
c
a
Sustainable Agricultural Systems Laboratory, United States Department of Agriculture, Agricultural Research Service, Bldg. 001, Rm 140, BARC-W, Beltsville, MD 20705, USA b Northern Great Plains Research Laboratory, United States Department of Agriculture, Agricultural Research Service, 1701 10th Ave. SW, P.O. Box 459, Mandan, ND 58554, USA c Environmental Quality Laboratory, United States Department of Agriculture, Agricultural Research Service, Bldg. 012, BARC-W, Beltsville, MD 20705, USA Received 30 August 2005; received in revised form 2 November 2005; accepted 8 November 2005 Available online 5 January 2006
Abstract Glomalin, a glycoprotein produced by arbuscular mycorrhizal (AM) fungi, is a major component of the humus fraction of soil organic matter. Glomalin is extracted from soil and hyphae of AM fungi by using sodium citrate at 121 °C in multiple 1-h cycles, but extensive extraction does not solubilize all glomalin in all soils. Efficacies of 100 mM sodium salts of citrate, borate or pyrophosphate (pH 9.0, 121 °C) were tested for two 1-h cycles for hyphae from four AM fungal isolates and four 1-h cycles for seven soils from four US geographic regions. Residual soil glomalin was examined by pyrophosphate extraction of soils previously extracted with citrate or borate followed by extraction of all soils after treatment with NaOH. Hyphal extracts were compared using Bradford-reactive total protein (BRTP) values, and extracts from soils were compared using BRTP, percentage C and C weight. No difference among extractants was detected for AM fungal isolates or across soils. The residual glomalin across soils for extractants contained the following percentages of the total BRTP: pyrophosphate, 14%; borate, 17%; and citrate, 22%. Comparisons among individual soils indicated that pyrophosphate extracted significantly more BRTP (10–53%) than borate or citrate in six soils and borate was equal to pyrophosphate in one soil. Extraction with borate should be compared with pyrophosphate before initiating an experiment. For routine extractions of ca. 85% of the glomalin across a variety of soils, sodium pyrophosphate appears to be equal to or better than borate and better than citrate. Ó 2005 Elsevier Ltd. All rights reserved. Keywords: Humin; Soil organic matter; Arbuscular mycorrhizal fungi
1. Introduction Glomalin, a brown to red-brown colored glycoprotein produced by arbuscular mycorrhizal (AM) fungi, is a major component of soil organic matter that currently is defined operationally by the extraction method (Nichols and Wright, 2005). Solubilization of glomalin on hyphae or in soil is accomplished by using 20–50 mM citrate, pH
*
Corresponding author. Tel.: +1 301 504 8156; fax: +1 301 504 8370. E-mail address: [email protected] (S.F. Wright).
0045-6535/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.chemosphere.2005.11.041
7.0 or 8.0 at 121 °C (Wright and Upadhyaya, 1996; Wright et al., 1996). Extractions are repeated by replenishing the extractant in 1-h cycles until the supernatant is almost colorless. The extract operationally defined as glomalin is relatively free of co-extracted proteins according to gel electrophoresis profiles of citrate extracts of diverse AM fungi or soils which show consistent bands with few, if any, extraneous bands (Wright and Upadhyaya, 1996; Rillig et al., 2001). When glomalin was first identified, sodium citrate (20 mM) at pH 7.0 and 121 °C solubilized glomalin on hyphae (Wright et al., 1996) and was the only solution of many organic and inorganic solutions tested that
1220
S.F. Wright et al. / Chemosphere 64 (2006) 1219–1224
successfully solubilized the compound (Wright and Upadhyaya, 1996; unpublished data). Citrate was used when the extraction of glomalin was extended from hyphae to soils, but higher citrate concentration and higher pH released greater amounts of glomalin from soils than pH 7.0, 20 mM citrate (Wright and Upadhyaya, 1996). Citrate (50 mM), pH 8.0 became the standard solution used for soils. Solubilized glomalin remains in solution at an alkaline pH and flocculates slowly in a mild acid (pH < 5.5) (unpublished data) or rapidly at pH 2.5 (Nichols and Wright, 2005). Therefore, to effectively extract glomalin, alkaline pH of the extractant was thought to be a necessity. Small amounts of glomalin in soil are solubilized by using water at 121 °C (Wright and Upadhyaya, 1996). Rillig and Steinberg (2002) used water at 121 °C to remove glomalin from hyphae produced in root organ cultures. Although an unknown amount of White’s M medium (pH 5.5) was also in the extraction solution, diluted White’s medium contains very low concentrations of salts and no citrate. The effectiveness of citrate for glomalin extraction decreases at 110 °C (Wright and Upadhyaya, 1996; Wright and Jawson, 2001), and only a small amount of glomalin is extracted at room temperature (Wright and Upadhyaya, 1996). These results show the importance of temperature but are confounding about the importance of citrate. We previously proposed that citrate acted as a competitive chelator of iron associated with glomalin (Wright and Upadhyaya, 1998) and thereby assisted in solubilization of the molecule. Glomalin contains iron, but amounts differ for glomalin extracted from pot-cultured hyphae and soil. Iron associated with glomalin from several soils contained 0.8–8.8% iron (Wright and Upadhyaya, 1998). Glomalin freshly produced on hyphae has Fe concentrations of 60.3% (Nichols, 2003). As work on glomalin progressed, it became apparent that citrate is too weak a chelator to compete for iron on glomalin from either hyphae or soil. Some, but not all, iron is released from glomalin by using 8-hydroxyquinolin or Tiron (4,5-dihydroxy-1,3-benzene disulfonic acid) (Nichols, 2003). Chelators such as EDTA (ethylenediaminetetraacetic acid) or EDDHA (ethylenediamine-di(o-hydroxyphenylacecid acid)) unsuccessfully compete with glomalin for iron (unpublished data). Sodium pyrophosphate is used to extracted humic substances from soils (Stevenson, 1994; Clapp and Hayes, 1999). Monovalent sodium disrupts Fe- and Al-(hydr) oxides acting as bridges between organic matter and clay minerals (Clapp and Hayes, 1999). Glomalin may be released from organo-mineral complexes in the soil by a similar reaction. Sodium borate was used to extract glomalin from soils when citrate would interfere with subsequent tests for heavy metals attached to glomalin (Gonza´lezCha´vez et al., 2004). Glomalin is co-extracted with humic acid (HA) in 0.1 M NaOH even after first extracting glomalin with citrate (Nichols, 2003; Nichols and Wright, 2005). This suggests that a residual fraction of glomalin is not released by using citrate even though soil is repeatedly extracted.
The role of sodium in the efficiency of extraction has not been resolved. Sodium borate pentahydrate contains 12% sodium, approximately half of the sodium in trisodium citrate (23%) or sodium in sodium pyrophosphate pentahydrate (21%). Therefore, comparisons among sodium salts of borate, pyrophosphate and citrate may resolve the question of the importance of sodium in efficiency of glomalin extraction and yield information on the purity of glomalin. The objectives of this experiment were to (1) compare efficacies of extractants containing different amounts of sodium to extract glomalin from hyphae of four AM fungi using protein yield, (2) compare the same extractants for seven soils using protein yield, percentage carbon, and carbon weight (based on gravimetric weight of freeze-dried glomalin), and (3) explore additional extractions to solubilize a residual glomalin fraction from soils. 2. Materials and methods 2.1. Extractants Glomalin was extracted from hyphae or soil with 100 mM, pH 9.0, sodium citrate (C6H5Na3O7 Æ 2H2O), sodium borate (Na2B4O7 Æ 10H2O), or sodium pyrophosphate (Na4P2O7 Æ 10H2O). The pH was adjusted with HCl. 2.2. Extraction from arbuscular mycorrhizal fungi Four arbuscular mycorrhizal (AM) fungal isolates— Gigaspora (Gi.) rosea (FL224), Acaulospora (A.) morrowiae (CL551), Glomus (G.) etunicatum (BR220), and G. intraradices (WV964) were grown on corn (Zea mays) in sterile, single-species pot cultures. A sterile mixture of glomalinfree coarse sand and coal was placed in 15 cm diameter (1300 cc) pots. Three disinfected seeds plus spores of Gi. rosea and spores and hyphae for the other inocula (>100 infectious units per plant) were placed in 8 cm diameter 38 lm nylon mesh bags in the centers of three replicate pots to contain plant roots while providing a root-free hyphal chamber (Nichols and Wright, 2004). Pots were harvested at 12 weeks of plant growth according to the following protocol. Shoots and the nylon mesh bag containing roots were removed and discarded, and hyphae were removed from the sand outside of the nylon mesh bag by rinsing the sand with water and washing the hyphae into a 53 lm sieve. Hyphae from each isolate was divided into approximately equal amounts and placed in extraction tubes for extraction with citrate, pyrophosphate or borate. Sufficient volumes of extractants were added to cover the hyphae. After 1 h at 121 °C extracts in the supernatant were collected by centrifugation at 10 844g for three min. A second extraction was performed under the conditions described above. Hyphae were rinsed into weigh boats, dried at 70 °C and weighed. Glomalin is reported in mg g 1 hyphae.
S.F. Wright et al. / Chemosphere 64 (2006) 1219–1224
2.3. Extraction from soils Seven soils were used—two soils from each of three geographic areas: Colorado (Sampson and Haxtun soil series), Maryland (Baltimore site A and Baltimore site B soil series), Georgia (Cecil and Pacolet soil series) and one soil from Nebraska (Pawnee soil series). See Nichols and Wright (2005) for soil analysis values. These soils are all slightly acidic to acidic loams with organic carbon contents ranging from 7.0 to 30.3 g kg 1. Three replicate 1-g subsamples of each soil were used for each extraction solution described above. All samples were subjected to four 1-h extraction cycles until the supernatant color was no longer brown. Supernatants were removed for separate analysis after each 1-h cycle. After four 1-h extraction cycles, the remaining soil from citrate and borate extractions was extracted for 1 h at 121 °C with the pyrophosphate solution. Following all of the above extractions, the three subsamples of soil were combined and 2 ml 0.1 M NaOH was added to cover the samples as was done by Nichols and Wright (2005) in the protocol to extract HA after extracting glomalin. Samples were shaken at room temperature overnight. The supernatant was neutralized with HCl and saved for protein analysis. Previous work indicated that separation of glomalin from HA in NaOH is not possible
Location
Georgia Cecil
Maryland Pacolet
Soil Series
Site A
Clay (g kg-1)* 110 12
Bradford-reactive protein (mg g-1)
Baltimore Baltimore
220
250
1221
at this stage of an extraction process (Nichols and Wright, 2005) so the NaOH extract was not processed further to obtain the C content of glomalin. All soil pellets were extracted with 50 mM citrate, pH 8.0 for 1-h cycles at 121 °C until the supernatant was straw-colored. This required two or three cycles. Supernatants from these extractions were pooled. Supernatants from all extractions were collected by centrifugation at 6850g for 10 min. 2.4. Protein and carbon analyses Analyses were performed on samples as described by Wright et al. (1996). Bradford-reactive total protein (BRTP) values for two extractions of hyphae were summed. Values for BRTP were obtained for the separate soil extraction supernatants described above and are reported for the routine extraction sequence for the first cycle and the sum of cycles 2–4, the follow-up pyrophosphate extraction, the NaOH extraction, and the pooled final citrate extracts. BRTP values were corrected for background color contributed by the extractant. Protein in all soil extract supernatants except the treatment with 0.1 M NaOH was precipitated by acidification with HCl, dialyzed against water, freeze-dried, weighed, and analyzed for percentage C by combustion using a Perkin–Elmer Series II C, H, N,
Colorado
Nebraska
Sampson Haxtun
Pawnee
Site B 260
160
110
200
* Data from Nichols and Wright (2005) Routine Follow-up pyrophosphate Residual
10
8
6
4
2 b a a
c b a
b b a
b b a
b c a
b c a
b c a
C B P
C B P
CB P
C B P
CBP
CB P
CBP
0
Fig. 1. Comparison of Bradford-reactive protein (BRTP) in seven soils extracted with 100 mM, pH 9.0 sodium citrate (C), sodium borate (B), or sodium pyrophosphate (P) at 121 °C. Routine extraction series = mean values for the amount extracted from three replicates of each soil in four 1-h extraction cycles. Follow-up pyrophosphate = mean values for the pyrophosphate extracts of three replicates of each soil previously extracted with citrate or borate. Residual = the amount in combined soil replicates treated with 0.1 M NaOH and then extracted with 50 mM citrate, pH 8.0 for two or three 1-h cycles at 121 °C citrate. SD is shown in white within stacked bars. Means separation of values for the routine extraction sequence for each soil are indicated by letters within bars (LSD a = 0.05).
1222
S.F. Wright et al. / Chemosphere 64 (2006) 1219–1224
S/O 2400 Analyzer (Rillig et al., 2001; Nichols and Wright, 2005). Values are reported in mg g 1 hyphae or soil. 2.5. Soil clay content Values for clay content are from analysis of these soils by Nichols and Wright (2005) (Fig. 1). 2.6. Statistical analysis Mean values for extractants across all soils and assay sequences described above, the sums of mean values for all sequential extracts across soils, values for the four AM isolates as replicates, and values for replicate routine sequence extracts within soils were analyzed. Differences among the extractants were tested by one-way analysis of variance (ANOVA), after Bartlett’s test of equal variances was satisfied, using Statistics (Analytical Software, Tallahassee, FL). Pair-wise comparisons were by LSD (a = 0.05). Pearson product moment correlation coefficients were calculated for soil clay content and BRTP. When necessary, sine or cosine transformations of data were made to meet the assumption of equal variances. 3. Results 3.1. Glomalin from hyphae BRTP yields for AM fungal isolates by extractant are shown in Table 1. Differences were not significant among different extractants across the fungal isolates (p = 0.27), but variation in BRTP was high.
The first extract in the routine extraction sequence contained the following BRTP percentages of the totals across soils: citrate, 44%; borate, 46%; and pyrophosphate, 60% (Table 1). Extracts to define a recalcitrant pool of glomalin showed that sodium pyrophosphate removed BRTP from soils that had previously been extracted in the routine scheme with citrate or borate. When soil samples were extracted in the routine extraction sequence with pyrophosphate, only 14% of the BRTP remained in a residual pool; whereas, when citrate and borate were used in the routine extraction sequence, 22% and 17%, respectively, of the total was in follow-up pyrophosphate extractable and NaOH mediated citrate-extractable glomalin (Table 1). The NaOH extract contained the lowest amount of protein for all extraction stages, but this treatment assisted in the citratemediated extraction of small amounts of residual glomalin. Extractants varied in efficacy in individual soils and between soils from the same geographic region (Fig. 1). In the routine extraction sequence pyrophosphate solubilized from 10% to 53% more BRTP than citrate or borate in six of the seven soils (Fig. 1). In the Georgia soils, pyrophosphate was more effective in Pacolet than in Cecil when compared with the other extractants. In Pawnee, Sampson and Haxtun soils, borate was less effective than pyrophosphate or citrate in extracting glomalin. Clay content of soils was not significantly linearly correlated with BRTP extracted by borate, citrate or pyrophosphate (n = 7) in the routine extraction sequence (r = 0.37, 0.24 and 0.20, respectively), the residual amounts extracted (r = 0.51, 0.33 and 0.55, respectively) or the combined total amounts (r = 0.27, 0.26, and 0.23, respectively). 4. Discussion
3.2. Glomalin from soil No differences were detected for BRTP, percentage C, or C weight across all soils for the different extractants except for BRTP in the NaOH treatment (Table 1). These results indicated that the various extractants did not significantly contaminate the products, and differences in sodium concentration, within the range tested, was not a factor in extractant efficacy.
Table 1 Mean values followed by (SD) of glomalin as measured by Bradfordreactive total protein extracted from hyphae of arbuscular mycorrhizal fungi with 100 mM, pH 9.0, sodium salts of citrate, borate or pyrophosphate Extractant
Mean (SD)
Citrate Borate Pyrophosphate
5.37 (4.09) 7.45 (3.55) 9.84 (3.10) p = 0.27*
Hyphae of the following isolates were used as replicates: Gigaspora rosea, Acaulospora morrowiae, Glomus intraradices and Glomus etunicatum. * One-way ANOVA with four replicates.
Differences in efficacy of extractants across soils or hyphae were not detected, and the sodium ion concentration within the range tested did not influence extraction of glomalin. The trend was that pyrophosphate was more effective than the other extractants (Tables 1 and 2). A small amount of soil glomalin was in a residual pool that could be removed only by treatment with NaOH followed by heat-mediated extraction (Table 2). The three extractants were tested first for the ability to solubilized glomalin on hyphae before making comparisons about the ability to solubilize glomalin in a complex matrix, soil. The extractants tested removed glomalin from hyphae, but high variability in BRTP related to hyphal weight (Table 1) may be due to inclusion of spores, which disproportionately contribute weight and not glomalin to a sample. Spores are very difficult to remove from bulk hyphal samples such as the ones used in this work. Previous work also indicates high variability in the ratio of glomalin to hyphal length (CV = 28%) (Lovelock et al., 2004). No differences were detected among extractants within an isolate (data not shown). Comparison of extractants within soils (Fig. 1) showed that pyrophosphate was superior to citrate and borate in
S.F. Wright et al. / Chemosphere 64 (2006) 1219–1224
1223
Table 2 Mean values of glomalin Bradford-reactive total protein, percentage C and C weight for seven soils from four geographic regions of the US extracted with three different sodium salt extraction solutions examined in the sequence presenteda,b Routine extraction sequence for soils
Extracts to define a residual pool of glomalin
Assay sequence
1
2
3
4
5
Extractant (100 mM, pH 9 at 121 °C for 1 h)
Extract 1
Extracts 2, 3, 4
Follow-up pyrophosphate
NaOH
Residual (extracted with citrate)
1.58 (0.45) 1.19 (0.46) 1.26 (0.39) 0.240
ND 0.70 (0.54) 0.53 (0.21) 0.631
0.09 (0.05)b 0.22 (0.15)a 0.23 (0.11)a 0.046
45.07 (1.50) 43.71 (3.57) 36.13 (7.77) 0.851
38.44 (2.63) 46.61 (2.08) 32.30 (7.46) 0.630
ND 43.98 (2.54) 41.79 (2.85) 0.155
ND ND ND
51.86 (7.46) 48.10 (1.85) 50.56 (5.04) 0.715
3.73 (1.12) 2.98 (1.82) 2.95 (1.13) 0.511
1.54 (0.56) 1.49 (0.68) 1.43 (0.45) 0.938
ND 2.10 (1.64) 1.70 (0.76) 0.573
ND ND ND
0.42 (0.15) 0.31 (0.11) 0.45 (0.08) 0.068
Bradford-reactive total protein (mg g 1 soil) Pyrophosphate 3.61 (1.50) Borate 2.33 (1.49) Citrate 2.36 (0.94) p= 0.150 Carbon (%) Pyrophosphate Borate Citrate p= Carbon weight (mg g Pyrophosphate Borate Citrate p=
1
0.73 (0.35) 0.65 (0.27) 0.96 (0.45) 0.281
Total
6.01 (2.22) 5.09 (2.04) 5.35 (1.94) 0.696 ND ND ND
soil) 5.69 (1.73) 6.88 (2.58) 6.54 (2.04) 0.577
a
Different letters in a column following values and SD in parentheses indicate differences according to LSD (a = 0.05) analyzed by one-way ANOVA. Data was sine or cosine transformed when needed to satisfy Bartlett’s test of equal variances. b ND indicates that the sample was not extracted or that the analysis could not be done.
six of the seven soils. Stability constant rankings of Fe(III) with the ligands used are: pyrophosphate > citrate > borate (www.coldcure.com/html/stability_constants.html; Elrod and Kester, 1980). Stability of ligand-Fe did not appear to explain differences in extractants because borate extracted more glomalin than citrate in the routine sequence for four of the seven soils tested (Fig. 1). There were no significant correlations between BRTP yields for different extractants and Fe concentration in soils using values from Nichols and Wright (2005) (data not shown). Borate interacts with glycoproteins (Kennedy and How, 1973) and pyrophosphate may bind to Fe in glomalin. Also, nuclear magnetic resonance (NMR) spectroscopy indicates that citrate is present in citrate-extracted glomalin, potentially adding weight and C to the molecule (Nichols, 2003). However, percentage C and C weight of glomalin did not indicate that weight was different for glomalin extracted by any of the extractants (Table 1). For each extraction cycle 8 ml of a 100 mM solution g 1 soil was used (Wright and Upadhyaya, 1996). Therefore, for each extraction cycle, solutions contained from 232 to 356.9 mg of the salts of extractants tested (citrate the lowest and pyrophosphate the highest amounts). If extractants bind irreversibly to glomalin in large amounts during the extraction procedure, gravimetric analyses, such as percentage C and C weight would be affected. Clay content was not correlated with BRTP, however, clay type may influence efficiency of an extractant in a particular soil. Pacolet and Pawnee have approximately the same amount of clay (Fig. 1), but Pacolet clay is kaolinitic and Pawnee clay is smectitic (http://ortho.ftw.nrcs.usda.
gov). Clays in the remaining soils are of mixed types. The influence of smectitic clay on borate extraction needs to be explored further. Release of residual glomalin facilitated by 0.1 M NaOH treatment of soils may be due to dispersion of clays, but is not recommended as a pre-treatment of soils before extraction and quantification of glomalin. Nichols and Wright (2005) show that 0.1 M NaOH co-extracts HA and a fraction of the total glomalin, and HA and glomalin are not easily separated. Therefore, for routine extractions of ca. 85% of the glomalin across a variety of soils, sodium pyrophosphate appears to be equal to or better than borate and better than citrate. Citrate-extracted glomalin 1H NMR spectra show broad peaks that are not adequately definitive for structural interpretation and indicate that citrate binds to glomalin (Nichols, 2003). Although pyrophosphate and borate 1H NMR spectra show no apparent binding, the spectra are not greatly improved over citrate. Further work will be necessary to find an extractant that can be used to solubilize glomalin for 1H NMR structural elucidation. In conclusion, pyrophosphate was superior to citrate or borate in extracting BRTP within the same soil. Glomalin on hyphae was extracted by all three extractants. Differences in percentage C and C weight due to the extractant used were not evident. References Clapp, C.E., Hayes, M.H.B., 1999. Characterization of humic substances isolated from clay- and silt-sized fractions of a corn residue-amended agricultural soil. Soil Sci. 164, 899–913.
1224
S.F. Wright et al. / Chemosphere 64 (2006) 1219–1224
Elrod, J.A., Kester, D.R., 1980. Stability constants of iron(III) borate complexes. J. Solution Chem. 9, 885–894. Gonza´lez-Cha´vez, M.C., Carrillo-Gonza´lez, R., Wright, S.F., Nichols, K.A., 2004. The role of glomalin, a protein produced by arbuscular mycorrhizal fungi, in sequestering potentially toxic elements. Environ. Pollut. 130, 317–323. Kennedy, G.R., How, M.J., 1973. Interaction of sugars with borate, an NMR spectroscopic study. Carbohydr. Res. 28, 13–19. Lovelock, C.A., Wright, S.F., Nichols, K.A., 2004. Using glomalin as an indicator for arbuscular mycorrhizal hyphal growth: an example from a tropical rain forest soil. Soil Biol. Biochem. 36, 1009–1012. Nichols, K.A., 2003. Characterization of glomalin—a glycoprotein produced by arbuscular mycorrhizal fungi. Ph.D. Thesis, University of Maryland, College Park, MD. Nichols, K.A., Wright, S.F., 2004. Contributions of soil fungi to organic matter in agricultural soils. In: Magdoff, F., Weil, R. (Eds.), Functions and Management of Soil Organic Matter in Agroecosystems. CRC Press, Boca Raton, FL, pp. 179–198. Nichols, K.A., Wright, S.F., 2005. Comparison of glomalin and humic acid in eight native US soils. Soil Sci. 170, in press.
Rillig, M.C., Steinberg, P.D., 2002. Glomalin production by an arbuscular mycorrhizal fungus: a mechanism of habitat modification?. Soil Biol. Biochem. 34 1371–1374. Rillig, M.C., Wright, S.F., Nichols, K.A., Schmidt, W.F., Torn, M.S., 2001. Large contribution of arbuscular mycorrhizal fungi to soil carbon pools in tropical forest soils. Plant Soil 233, 167–177. Stevenson, F.J., 1994. Humus chemistry, Genesis, composition, reactions, second ed. John Wiley and Sons Inc., New York. Wright, S.F., Jawson, L., 2001. A pressure cooker method to extract glomalin from soils. Soil Sci. Soc. Am. J. 65 (6), 1734–1735. Wright, S.F., Upadhyaya, A., 1996. Extraction of an abundant and unusual protein from soil and comparison with hyphal protein of arbuscular mycorrhizal fungi. Soil Sci. 161, 575–586. Wright, S.F., Upadhyaya, A., 1998. A survey of soils for aggregate stability and glomalin, a glycoprotein produced by hyphae of arbuscular mycorrhizal fungi. Plant Soil 198, 97–107. Wright, S.F., Franke-Snyder, M., Morton, J.B., Upadhyaya, A., 1996. Time-course study and partial characterization of a protein on hyphae of arbuscular mycorrhizal fungi during active colonization of roots. Plant Soil 181, 193–203.