Growth-Dependent Stable Carbon Isotope Fractionation by Basidiomycete Fungi: δ¹³C Pattern and Physiological Process

Abstract

We grew 11 basidiomycetes in axenic culture to characterize their physiological capacities to fractionate stable C isotopes. Generally, δ¹³C values of the fungal biomass were (i) enriched in ¹³C relative to the growth medium, (ii) variable among the isolates, and (iii) dependent on the growth rate and growth stage of the fungi. We found a multiphasic dynamic of fractionation for Cryptoporus volvatus and Marasmius androsaceus during various growth stages. The first phase, P1, corresponded to the exponential growth stage and was characterized by an increasing enrichment in ¹³C content of the fungal biomass relative to the growth medium ranging between 4.6 and 6.9‰. The second phase, P2, exhibited a continual depletion in ¹³C of the fungal biomass, with the δ¹³C values of the fungal biomass asymptotically returning to the δ¹³C value of the growth medium at inoculation. The expression of the various fractionation phases was dependent on the amount of low-concentration micronutrients and growth factors added to the growth medium. The onset of P2 occurred at reduced concentrations of these elements. All of the sugars in the growth medium (sucrose, maltose, and glucose) were utilized for growth, indicating that the observed fractionation was not an artifact derived from the preferential use of ¹³C-rich maltose, which was found at low concentrations in the growth medium. In this study, we establish a framework with which to explore the impact of physiological fractionations by fungal interfaces on natural distributions of stable C isotopes.

Fungi form a ubiquitous interface mediating nutrient movements in terrestrial ecosystems (5, 23, 24, 28). In this role, fungi should be expected to affect the natural distribution of stable C isotopes. Nevertheless, the effect of fungal interfaces on isotopic transformations in terrestrial ecosystems is poorly understood. Recent studies indicate that fractionation of stable isotopes of C by fungi is common in nature (8, 12, 14, 18) and in culture (11, 30, 31). Since this fractionation can result in an alteration of isotope distributions in microbially respired CO₂ and in organic matter that has accumulated in the ecosystem (11), fractionation effects due to fungal processing can be expected to be significant at an ecosystem level.

Research on forest fungi across the Northern Hemisphere (6, 7, 12-15, 18, 29) has consistently suggested an isotopic fractionation pattern in which ectomycorrhizal (EM) and saprotrophic (SAP) sporocarps (mushrooms) display, on average, characteristically different isotopic signatures for both C and N. This “EM/SAP divide” (12) is predominantly determined by substrate effects (7, 12), but there is also evidence of an additional physiological component that appears to be insensitive to these substrate effects or to differences between EM and SAP functional groups (12).

Here we investigate the dynamics and determinants of C isotopic fractionation for 11 basidiomycetes that represent EM and SAP fungi under controlled substrate conditions. For two representative isolates, we provide details on the relationship between the observed isotope fractionation and the growth stage of their mycelium. Our results provide a framework with which to explore the importance of physiological isotopic fractionations that occur in ecosystems during fungal nutrient processing.

MATERIALS AND METHODS

Liquid cultures.

Live mycelium was isolated from fresh sporocarps of 11 basidiomycetes collected in pine-dominated forests (Table 1) and grown on solid modified Melin-Norkrans (MMN) medium (22). Cultures were maintained in the same medium for <10 subcultures until inoculated into experimental media. Following our previous finding that fractionation in these fungi occurred only when grown on C3-C sources (11), we used beet sucrose (98% purity) as the predominant C source for experimental media. Maple sucrose was used for some replicates in the single-time-point experiments (described below), but no difference was observed in results from either C source (P > 0.05; analysis of variance [ANOVA]). Malt extract (ME; Difco) was included in the medium at 0.13% (wt/vol) in the single-time-point experiments and at various concentrations for the dynamic experiments. The amount of sucrose added was adjusted so that, in all cases, the total sugar concentration was 1.3% (wt/vol). The δ¹³C value for the C3 sucrose was −23.6‰ ± 0.1‰, and that for the C4 ME was −11.4‰ ± 0.0‰. The ME contained 24.3% maltose and 33.3% glucose, determined by high-performance liquid chromatography (HPLC) as described below. (Difco reports their ME contains approximately 33% reducing sugars. Various necessary organic and inorganic factors necessary for growth are also contained in the ME.) N was supplied as ammonium phosphate [(NH4)2HPO4] at 0.25 mg/ml. The medium also contained CaCl₂ (0.05 mg/ml), NaCl (0.025 mg/ml), KH₂PO₄ (0.5 mg/ml), MgSO₄ (0.15 mg/ml), FeCl₃ (0.012 mg/ml), and thiamine HCl (0.1 μg/ml). The methods used for preparation of media, harvest of mycelia via filtration through glass fiber filters, and preparation of mycelium and growth medium for isotopic analysis were described previously by Henn and Chapela (11). The controls were flasks not inoculated with any fungus, but incubated in parallel with inoculated flasks.

TABLE 1.

Taxon-specific growth rate, mycelium biomass, percent C, δ¹³C; and Δ_C from single-time-point, plug inoculation experiments

Fungal isolate	Ecology	Petri dish culture specific growth rate (μ)	Liquid culture
			No. of replicatesa	Mycelium accumulated biomass (mg)b	Biomass		ΔCc
					%Cb	δ13C
Suillus granulatus	EM	NAd	15 (0)	NA	NA	−17.6 ± 0.5	5.7 ± 0.5 b,c,d,e
Amanita muscaria	EM	0.15 ± 0.02	16 (13)	8.592 ± 2.770	37.5 ± 0.8	−16.2 ± 0.5	6.5 ± 0.5 a,b,c
Tricholoma sejunctum	EM	0.39 ± 0.04	17 (13)	18.962 ± 4.221	40.6 ± 0.4	−16.2 ± 0.4	7.0 ± 0.4 a,b,c
Suillus pungens	EM	0.30 ± 0.03	15 (12)	25.642 ± 8.236	39.4 ± 0.4	−14.5 ± 0.3	8.8 ± 0.3 a,b
Trichaptum abietinus	SAP	1.11 ± 0.05	6 (3)	37.933 ± 4.384	40.6 ± 0.7	−18.9 ± 0.8	4.0 ± 0.8 c,d,e,f
Cryptoporus volvatus	SAP	1.05 ± 0.02	16 (10)	42.730 ± 4.007	40.6 ± 0.7	−17.8 ± 0.4	5.1 ± 0.4 c,d,e
Ischnoderma resinosum	SAP	0.43 ± 0.01	15 (9)	86.311 ± 12.961	46.0 ± 1.0	−21.8 ± 0.5	2.0 ± 0.5 f
Heterobasidion annosum	SAP	1.59 ± 0.09	7 (3)	117.600 ± 61.149	39.0 ± 0.5	−21.9 ± 0.0	0.7 ± 0.0 f
Stereum sp.	SAP	NA	3 (3)	128.333 ± 23.398	42.1 ± 0.3	−22.2 ± 0.2	NA
Marasmius androsaceus	SAP	0.90 ± 0.07	18 (6)	156.467 ± 20.093	45.2 ± 0.4	−21.8 ± 0.1	1.0 ± 0.2 f
Calbovista subsculpta	SAP	NA	2 (0)	NA	NA	−17.3 ± 0.8	5.5 ± 0.8 b,c,d,e

^aThe numbers of replicates used for the average accumulated biomass values are reported in parentheses.

^bBiomass was not weighed for all replicates.

^cCalculated as described in Materials and Methods by equation 1: Δ_C = δ_F − δ_s,t. Taxa that do not share a letter are significantly different (P = 0.05, Tukey's test).

^dNA, not available.

Two different inoculation strategies were followed. Under the first strategy, a mycelial plug (<1 mm³) containing less than 0.5 μg of C was cut out from the edge of actively growing cultures and placed in the experimental liquid medium (75 ml in a 125-ml Erlenmeyer flask). This plug inoculation was used to study a diverse group of 11 basidiomycete taxa in a single-time-point sampling schedule (i.e., one harvest at 35 days). Plug inoculation yields lower growth rates and larger variation in biomass accumulation as dense mycelial clusters build up during incubation. Increased variation due to pellet growth by the fungi in the single-time-point experiments was encountered, but it was not practical to run these experiments with a diversity of taxa by using the slurry technique (described below). The increased variation in results from plug inoculation was not sufficient to mask detection of the dynamic nature of fractionation elucidated in the dynamic experiments (Fig. 1).

FIG. 1.

Average mycelium δ¹³C value (A) and mycelium percent C (B) as a function of accumulated fungal biomass for A. muscaria (am; n = 13), C. volvatus (cv; n = 10), H. annosum (ha; n = 3), I. resinosum (ir; n = 9), M. androsaceus (ma; n = 6), S. pungens (sp; n = 12), Stereum sp. (s; n = 3), T. abietinus (ta; n = 3), and T. sejunctum (ts; n = 13). The dashed line indicates the average δ¹³C of control medium (−22.5‰ ± 0.3‰).

The second method was used for time sequence (dynamic) experiments with two species, Cryptoporus volvatus and Marasmius androsaceus. We used a slurry inoculation method to reduce variation in biomass accumulation and to increase homogeneity in the culture. Biomass from liquid cultures at the exponential growth stage was blended for 10 s at high speed in a Waring blender to produce small mycelial fragments in a slurry. Experimental flasks (75 ml of medium in 125-ml Erlenmeyer flasks) were inoculated with a 500-μl aliquot from this slurry. The amount of C introduced into the experimental flask by this method was <5 μg.

Single-time-point experiment.

Cultures were grown (plug inoculation method) at room temperature under natural light conditions on an Innova 2100 platform shaker (New Brunswick Scientific) at 180 rpm. The initial pH of the medium was 5.9 to 6.4. The final pHs of the filtered medium were 6.3 for controls (i.e., no fungal inoculum) and 2.9 to 3.6 for medium processed by fungi. We assume that reabsorption of respired CO₂ into the medium is negligible, given the low pH (H₂CO₃ pK₂ = 10.25). At harvest, cultures were checked for contamination by plating and DNA molecular techniques as described by Henn and Chapela (11).

Dynamic experiment 1.

C. volvatus and M. androsaceus were grown (slurry inoculation method, inocula of 97 and 90 CFU, respectively) in complete darkness with an average hourly temperature of 23.6 ± 0.6°C and with constant shaking at 180 rpm (incubator shaker; Lab-Line) in modified liquid MMN medium (22). The ME concentration was 0.13% (wt/vol), and the sucrose concentration was 1.17% (wt/vol). The corresponding ratio of ME to sucrose (mass/mass) was 0.10. The average pH of the control medium flasks (no fungal inoculum) was 6.3 ± 0.2. Mycelia were harvested as described above over 61 days at 2.5, 3.5, 4.4, 6.5, 10.8, 21.3, 32.0, and 60.9 days. For each of the two isolates, three replicate flasks were taken at each harvest date. Plating was used to check cultures for contamination.

Dynamic experiment 2.

In a second experiment, designed to explore the effect of varied concentrations of organic and inorganic growth factors on isotopic fractionation, C. volvatus was inoculated (slurry method, inoculum of 103 ± 10 CFU, mean ± standard deviation [SD]; n = 8) into modified liquid MMN medium (22) with ME concentrations of 0, 0.065, 0.13, and 0.26% (wt/vol) and sucrose concentrations of 1.30, 1.24, 1.17, and 1.04% (wt/vol), respectively. Corresponding ratios of ME to sucrose (mass/mass) were 0, 0.05, 0.10, and 0.20, respectively. Flasks were incubated in complete darkness at a temperature of 22.2 ± 0.9°C (hourly average) and shaken at 180 rpm (incubator shaker; Lab-Line). The pHs of the control medium (from flasks with no fungal inoculum run in parallel with inoculate flasks) were 6.4 ± 0.1, 6.3 ± 0.2, 6.3 ± 0.2, and 6.3 ± 0.2 for the 0, 0.065, 0.13, and 0.26% ME treatments, respectively. Mycelia from three replicates for each harvest date of each ME concentration were harvested at 3.1, 6.1, 9.2, 19.2, and 26.3 days. Harvest times were chosen based on results from dynamic experiment 1 to optimize the sampling scheme for detection of the maximum fractionation point. Additionally, the number of harvesting dates was constrained so that all flasks could be run in the same incubation chamber to avoid variability in environmental conditions. Plating was used to check cultures for contamination. Slopes and y-intercepts for various regressions were compared as described by Zar (32).

For the 0.13% ME treatment, we determined the concentrations of specific metabolite components in the control medium and medium after C. volvatus growth. The metabolite composition of the growth medium was determined by gel permeation chromatography (GPC)-HPLC in combination with evaporative light scattering (ELS) detection for carbohydrates and UV detection for organic acids. The freeze-dried growth medium remaining after isotopic analysis for each sample (between 131 and 175 mg) or 1 mg of ME was dissolved in 1 ml of ultrapure water (Millipore). The dissolved medium was diluted to a final concentration of 10 mg/ml, and all samples were subsequently filtered (Millex-GV, 0.45-μm pore diameter; Millipore). To separate maltose, sucrose, and glucose, 50 μl of solution was automatically injected into the GPC-HPLC (ASI 100; Dionex). We used a PL Hi-Plex H 8-μm column (300 by 7.7 mm; Polymer Laboratories) for maltose and sucrose separation and a PL Hi-Plex PB 8-μm column (300 by 7.7 mm; Polymer Laboratories) for glucose separation. Both HPLC columns were supplied with a constant flow of water (1 ml/min; P580; Dionex) and maintained at 85°C (STH 585; Dionex). Under the conditions described above, sucrose was completely hydrolyzed, and since the concentration of free glucose in the medium was relatively low, sucrose was quantified as the sum of glucose and fructose. All carbohydrates were detected with an ELS detector (PL-ELS 1000; Polymer Laboratories) with the evaporator temperature set to 120°C and the nebulizer temperature set to 85°C, and the flow of nitrogen was set to 0.4 ml/min. For quantification, the system was calibrated with four concentrations of between 0.1 and 1 mg/liter for maltose and glucose and between 0.5 and 15 mg/liter for sucrose. Relative standard deviations were between 0.9 and 4.7%, 0.3 and 4.3%, and 0.6 and 2.6%, respectively. To quantify organic acids in the control medium and medium after fungal growth, another 50 μl of solution was injected onto a PL Hi-Plex H 8-μm column (300 by 7.7 mm; Polymer Laboratories). In this case, 0.005 M H₂SO₄ was supplied at a flow rate of 0.6 ml/min. Organic acids were quantified with a photo diode array detector (PDA 100; Dionex) at a wavelength of 210 nm. The system was calibrated with five concentrations of between 0.1 and 1 mg of oxalic acid per liter, with a relative standard deviation of between 0.02 and 0.06%.

Petri dish growth rate determination.

Given the growth stage dependency of fractionation, it was important to determine the growth rates of taxa relative to one another. Because determination of specific growth rates is highly questionable under the plug inoculation conditions used in our single-time-point experiments, we performed parallel determinations of relative growth rate for each isolate tested on solid media. Growth rates measured on solid medium cannot be compared directly with those applying to the liquid culture conditions, since the media were different and the geometry and dynamics of growth should be expected to differ under both conditions, but we used these growth rate estimations on a relational basis to make first approximation comparisons between fast- and slow-growing taxa. All fungi grown in liquid culture were also grown on MMN agar plates in replicate for 32 days at 24.1 ± 2.0°C (mean ± SD). Fungal growth was logged every day for the first 5 days and then every 3 days thereafter by marking the edge of the growing mycelium on the reverse of the plate until the extending mycelium reached the edge of the petri dish. We took digital photographs of the plates and used Adobe Photoshop to quantify fungal growth for each growth period. Briefly, growth accrued during each period was converted to an artificial color, and then pixels with that color were quantified by using the “histogram” function in the software program. For each digital image, a standard piece of graph paper was included as an internal standard to transform pixel dimensions to square centimeters. Intrinsic or specific growth rate (μ) was determined for each isolate by using the slope in a graph of the ln (growth) versus time for the exponential growth stage (10). Slopes and y-intercepts of growth regressions for each replicate were compared for significant differences as described by Zar (32). No within-taxon comparisons showed significant differences between regressions (P = 0.05), and therefore the growth rates reported for each taxon were calculated by pooling all replicates (Table 1).

Isotopic analysis.

Isotopic composition and percentage of each element were determined with an online continuous-flow CN analyzer (ANCA NT-System, Europa Scientific) coupled to a stable isotope ratio mass spectrometer (20/20 IRMS; Europa Scientific). For fungal biomass, isotopic signatures represent total C and are not differentiated based on cellular fractions. Values are reported in the standard notation (δ¹³C value [per mille]) relative to the international standard, Pee-Dee Belemnite, by using NIST peach leaves no. 1547 as a reference material, where δ¹³C = [(R_sample/R_standard) − 1] × 1,000, and R is the molar ratio ¹³C/¹²C (19). Unless otherwise noted, isotopic fractionation (the change in δ¹³C from the medium to the fungal biomass) is defined as:

(1)

where δ_F is the per mille isotopic signature of the accumulated fungal biomass and δ_s,t is the per mille isotopic signature of the substrate at the time of harvest. Instrumental precision, estimated with standard ground peach leaves (NIST no. 1547), for solid samples was 0.13‰ (SD for all isotopic runs combined; n = 106). Each sample was run twice, and values were averaged with duplicates always within <0.3‰ and <1.4% of each other for δ¹³C and percent C, respectively. Values were corrected for linearity relative to the beam area of the standard. The values reported are from all experimental runs averaged.

Mass balance.

Isotopic mass balances were estimated for the dynamic experiments by using a simple mixing model:

where f is the molar fraction of C contained in that component relative to the original amount of C (i.e., C in controls). The controls were flasks that were not inoculated and run in parallel with the inoculated flasks. Respired CO₂ was not collected in these two experiments. was calculated by two methods (Table 2). In the first method, the amount of C respired as CO₂ was calculated assuming a closed reaction in which the amount of CO₂-respired C is equal to the difference in C in the biomass and medium from that in the control. In the second method, the CO₂ respired was calculated by using equation 2 and assuming the δ¹³C of the respired CO₂ is the same as that of the fungal biomass, as illustrated by Henn and Chapela (11). Comparison of calculated by the two methods indicates that the mass balance is closed for almost all conditions (Table 2). By using equation 2 and calculated by method 1, it is possible to calculate the percent ¹³C recovered and test the assumption that the δ¹³C value of the respired CO₂ is the same as the fungal biomass δ¹³C value. Table 2 indicates that this assumption is valid. C excreted by the fungus into the growth medium is accounted for in the mass balance calculation; C contained in the medium is not differentiated based on unprocessed and processed C.

TABLE 2.

Percent C of fungal biomass in dynamic experiments and mass balances for reaction flasks

Expta	Amt of ME (% [wt/vol])	Harvest day	Biomass %C	fbiomass	fmedium	frespired CO2		Fractional recoveryd
						Method 1b	Method 2c
CV-C	0.13	2.5	36.4 ± 1.4	0.00 ± 0.00	0.99 ± 0.13	0.01 ± 0.13	0.01 ± 0.16	1.00 ± 0.19
		3.5	33.5 ± 5.9	0.02 ± 0.01	0.83 ± 0.05	0.16 ± 0.06	0.20 ± 0.07	0.97 ± 0.11
		4.4	34.0 ± 6.3	0.03 ± 0.01	0.78 ± 0.11	0.19 ± 0.12	0.25 ± 0.15	0.95 ± 0.17
		6.5	34.3 ± 6.3	0.04 ± 0.01	0.86 ± 0.06	0.10 ± 0.07	0.12 ± 0.08	0.98 ± 0.12
		10.8	40.4 ± 0.1	0.06 ± 0.02	0.67 ± 0.13	0.27 ± 0.14	0.32 ± 0.17	0.96 ± 0.20
		21.3	41.4 ± 0.4	0.09 ± 0.01	0.86 ± 0.07	0.05 ± 0.07	0.06 ± 0.08	1.00 ± 0.13
		32.0	42.0 ± 0.7	0.08 ± 0.01	0.87 ± 0.07	0.04 ± 0.08	0.04 ± 0.08	1.00 ± 0.14
		60.9	41.2 ± 1.1	0.09 ± 0.01	0.76 ± 0.12	0.15 ± 0.13	0.16 ± 0.13	0.99 ± 0.20
MA-C	0.13	2.5	30.4 ± 6.4	0.01 ± 0.01	0.89 ± 0.12	0.11 ± 0.12	0.13 ± 0.15	0.98 ± 0.18
		3.5	35.9 ± 6.8	0.02 ± 0.01	0.84 ± 0.07	0.14 ± 0.08	0.17 ± 0.09	0.97 ± 0.12
		4.4	34.9 ± 6.5	0.03 ± 0.01	0.90 ± 0.08	0.07 ± 0.08	0.08 ± 0.10	0.99 ± 0.13
		6.5	35.8 ± 6.7	0.04 ± 0.01	0.81 ± 0.13	0.15 ± 0.14	0.21 ± 0.20	0.96 ± 0.19
		10.8	35.0 ± 6.8	0.04 ± 0.01	0.82 ± 0.13	0.14 ± 0.14	0.17 ± 0.18	0.98 ± 0.19
		21.3	39.1 ± 0.1	0.05 ± 0.01	0.82 ± 0.13	0.13 ± 0.13	0.16 ± 0.17	0.98 ± 0.19
		32.0	39.0 ± 0.5	0.03 ± 0.00	1.04 ± 0.11	−0.07 ± 0.10	−0.14 ± 0.16	1.05 ± 0.16
		60.9	38.2 ± 0.8	0.03 ± 0.01	0.91 ± 0.15	0.05 ± 0.16	0.05 ± 0.21	1.00 ± 0.22
CV-V	0.00	3.1	32.7 ± 0.0	0.00 ± 0.00	0.98 ± 0.12	0.01 ± 0.00	0.00 ± 0.18	1.02 ± 0.17
		6.1	29.3 ± 2.7	0.00 ± 0.00	0.97 ± 0.07	0.03 ± 0.00	0.05 ± 0.10	0.99 ± 0.12
		9.2	30.6 ± 0.8	0.01 ± 0.00	0.98 ± 0.07	0.02 ± 0.00	0.06 ± 0.11	0.98 ± 0.12
		19.2	37.3 ± 1.1	0.02 ± 0.00	1.01 ± 0.07	−0.03 ± 0.00	0.03 ± 0.80	0.98 ± 0.13
		26.3	37.7 ± 0.6	0.02 ± 0.00	0.97 ± 0.08	0.01 ± 0.00	−0.95 ± 1.73	0.98 ± 0.13
	0.065	3.1	34.1 ± 0.0	0.00 ± 0.00	0.94 ± 0.10	0.06 ± 0.01	0.07 ± 0.11	0.99 ± 0.17
		6.1	36.8 ± 0.7	0.01 ± 0.00	0.91 ± 0.09	0.08 ± 0.01	0.12 ± 0.15	0.99 ± 0.17
		9.2	38.3 ± 0.8	0.03 ± 0.00	0.90 ± 0.10	0.07 ± 0.01	0.26 ± 0.31	0.99 ± 0.17
		19.2	40.4 ± 0.6	0.04 ± 0.00	0.92 ± 0.10	0.03 ± 0.00	0.76 ± 1.14	1.00 ± 0.17
		26.3	39.9 ± 0.4	0.06 ± 0.01	0.85 ± 0.10	0.09 ± 0.01	−3.97 ± 3.07	1.00 ± 0.18
	0.13	3.1	34.6 ± 0.8	0.00 ± 0.00	1.07 ± 0.07	−0.07 ± 0.00	−0.08 ± 0.08	1.02 ± 0.10
		6.1	38.4 ± 0.6	0.02 ± 0.00	0.89 ± 0.15	0.08 ± 0.01	0.16 ± 0.30	0.99 ± 0.17
		9.2	39.1 ± 1.4	0.04 ± 0.00	0.94 ± 0.05	0.02 ± 0.00	0.16 ± 0.20	0.99 ± 0.09
		19.2	39.1 ± 1.9	0.07 ± 0.01	0.96 ± 0.06	−0.03 ± 0.01	0.70 ± 2.21	1.01 ± 0.10
		26.3	40.8 ± 0.9	0.09 ± 0.01	0.75 ± 0.22	0.16 ± 0.01	−3.03 ± 2.75	0.99 ± 0.24
	0.26	3.1	36.7 ± 0.5	0.00 ± 0.00	0.97 ± 0.12	0.02 ± 0.00	0.02 ± 0.15	1.00 ± 0.22
		6.1	39.6 ± 0.9	0.03 ± 0.01	0.93 ± 0.12	0.04 ± 0.01	0.06 ± 0.29	1.01 ± 0.22
		9.2	40.5 ± 0.6	0.07 ± 0.01	0.91 ± 0.12	0.02 ± 0.00	0.20 ± 0.52	1.00 ± 0.22
		19.2	40.8 ± 0.3	0.11 ± 0.02	0.88 ± 0.12	0.01 ± 0.01	−1.14 ± 2.10	1.03 ± 0.22
		26.3	41.5 ± 0.5	0.12 ± 0.02	0.79 ± 0.12	0.08 ± 0.01	−1.18 ± 0.89	1.03 ± 0.22

^aCV-C, C. volvatus constant [ME] experiment (dynamic experiment 1); MA-C, M. androsaceous constant [ME] experiment (dynamic experiment 1); CV-V, C. volvatus varied [ME] experiment (dynamic experiment 2).

^bAmount of CO₂ respired = total C − (biomass C + medium C).

^cAmount of CO₂ respired was calculated with the isotopic mixing model (see Materials and Methods).

^dCalculated with a simple mixing model. The mass of CO₂ respired was calculated by method 1, and the δ¹³C value of the respired CO₂ was assumed to be the same as the fungal biomass (see Materials and Methods). A value of 1.00 indicates complete recovery.

RESULTS

Single-time-point experiment.

All 11 basidiomycete taxa discriminated between the two C isotopic species. Fractionations were affected by the growth characteristics of the fungi (Table 1 and Fig. 1). The observed δ¹³C values for fungal biomass had a range of −24.3 to −13.3‰, and the measured values were comparable to values obtained by others under laboratory and field conditions (11, 12, 14, 18, 30, 31). The average magnitude of Δ_C varied significantly among taxa (P < 0.0001, ANOVA) (Table 1), and is negatively correlated with both the average accumulated biomass (Pearson correlation = −0.90) (Table 1) and the specific growth rate (μ) for each isolate (Pearson correlation = −0.61) (Table 1). The relationship of fungal biomass δ¹³C values to the amount of accumulated biomass appears bimodal, with a transition between phases occurring at a critical point (Fig. 1).

The transition in biomass δ¹³C values corresponds with a shift in the mycelium percent C. In general, fungal isolates that grew faster and achieved greater biomasses contained proportionally more C than fungi with smaller biomasses (Table 1 and Fig. 1). This is not clearly the case for Heterobasidion annosum, but replicates for this fungus greatly varied in the amount of biomass achieved at 35 days of growth. When all replicate flasks are treated as independent experimental units, there is a significant difference in mycelial percent C between all fungal replicates that achieved a biomass greater than 50 mg versus those that did not (44.4% ± 0.6% and 39.5% ± 0.4%; P < 0.001, Wilcoxon rank-sum test).

Dynamic experiment 1 (single ME concentration).

C. volvatus and M. androsaceus exhibited growth curves with a short lag stage followed by exponential growth (Fig. 2). After this exponential growth stage, C. volvatus and M. androsaceus differed in their behavior, with the former maintaining a stationary stage of relatively constant biomass, while the latter followed an initial stationary stage by a lytic stage for approximately 10 days before maintaining a final stationary stage (Fig. 2). Both fungi slowed growth by 6.5 days (Fig. 2). The difference in slope for the linearized log stage of C. volvatus (0.58 ± 0.05) and M. androsaceus (0.47 ± 0.10) was not significant (P_slope = 0.05, slope of regressions using first three points of growth curves) (Fig. 2). There was no significant difference between the y-intercepts of the two regressions (P_y-intercept = 0.05), confirming that initial inoculum sizes were comparable (10) (Fig. 2).

FIG. 2.

Biomass accumulation for C. volvatus (A1) and M. androsaceus (B1). (A2 and B2) δ¹³C value of accumulated biomass for C. volvatus (A2) and M. androsaceus (B2). Circles represent the δ¹³C value of the respective medium at harvest, and dashed lines represent the δ¹³C of control medium (A2 and B2). Averages and errors represent values for three independent replicates at each harvest date (see Material and Methods). Points with no visible errors have errors smaller than the symbol.

C isotope fractionation was strongly growth dependent and was characterized by a multiphasic function of time and biomass (Fig. 2). The first phase (P1), coinciding with early growth stages and higher growth rates, exhibited large fractionations (Δ_C; see equation 1) that increased with time (Fig. 2). The average maximum fractionation during P1 was 6.8‰ ± 1.1‰ (mean ± standard error [SE]) (Fig. 2) for the two taxa tested. A second phase (P2) was observed in C. volvatus marked by a sustained depletion in ¹³C from the fungal biomass, which caused its δ¹³C value to asymptotically approach the δ¹³C value of the growth medium at inoculation (Fig. 2). The return in δ¹³C value towards original medium values in this second phase was only briefly observed in M. androsaceus (Fig. 2). M. androsaceus did exhibit an initial decline in observed δ¹³C values, but did not monotonically approach the δ¹³C value of the original medium (Fig. 2). The marked difference between C. volvatus and M. androsaceus in δ¹³C values during late stages of their growth coincides with the lytic stage displayed by M. androsaceus, which was not observed for C. volvatus. For both fungi, a transition point marking the end of P1 and the onset of P2 is characterized by (i) the first observed decline in fractionation (Δ_C) (Fig. 2) and (ii) an increase in percent C in the mycelial biomass (Table 2). There was a significant (t test; P < 0.001) shift in the percent C of the fungal biomass after the first four sampling dates for C. volvatus and first five sampling dates for M. androsaceus: from 34.4% ± 0.5% (n = 9) to 40.2% ± 0.6% (n = 7) (Table 2). Notably, percent C measurements of earlier sampling dates showed a much larger variation among replicate experimental flasks (i.e., large SE) than did later sampling dates associated with P2 (Table 2). The transition point between phases occurred at similar biomasses for both fungi (Fig. 2).

Using the first four sampling dates for C. volvatus and the first five sampling dates for M. androsaceus (harvests corresponding to the shift in percent C), the relationship between accumulated biomass and its δ¹³C value for the first phase, P1, can be approximated for both fungi by the linear relationship:

(3)

(R² = 0.33, P_slope = 0.08, P_y-intercept < 0.001; Fig. 2). Removal of the fourth and fifth sampling points for C. volvatus and M. androsaceus, respectively, from the P1 regression model results in a more robust regression (R² = 0.66; P_slope = 0.01; P_y-intercept < 0.001), suggesting that the transition from P1 may occur slightly before the detectable shift in biomass percent C. It is important to note that our sampling scheme is not sufficient to determine the detailed nature of P1 dynamics, and a linear relationship is used only as a first approximation.

The second fractionation phase (P2), applying to C. volvatus and M. androsaceus prior to the lytic stage of this fungus, is characterized following Mariotti et al. (21) by the equation:

(4)

where δ¹³C_F is the isotopic signature of the accumulated fungal biomass, δ¹³C_s,0 is the isotopic signature of the C available at time zero, ɛ_p/s is the “per mille enrichment factor,” and f is the unused fraction of the substrate at time t (R² = 0.85, P_slope = 0.003, P_y-intercept = 0.001) (Fig. 2). It is important to note that this equation can only approximate our results, since our measure of f is estimated: respired CO₂ was not collected in these experiments and could only be estimated (Table 2). Additionally, some of the C in the medium is derived from fungal metabolism (described below). Infrequent sampling did not allow for the characterization of the M. androsaceus dynamics once a lytic growth stage was initiated.

Dynamic experiment 2 (varied ME concentration).

Growth curves obtained for C. volvatus with various amounts of ME showed the expected differences in growth rates and maximum biomass achieved (Fig. 3). Both the rate of biomass accumulation and maximum biomass accumulated increased with increasing ME concentration (Fig. 3). In all treatments containing ME, there was initially a rapid drop in the medium pH of approximately 2.8 ± 0.1 pH units followed by a slower decline to a final pH of 3.0 ± 0.0 (Fig. 3). A similar pattern was seen in the 0% ME treatment, but here the pH remained above 3.0 (Fig. 3). The rate of decrease in medium pH over time was similar for all treatments when the change in pH relative to the amount of biomass produced is considered.

FIG. 3.

Time-dependent biomass accumulation (A), medium pH at harvest (B), and δ¹³C value of fungal biomass (various symbols) and respective medium at harvest (circles in panels C to F) for C. volvatus grown on modified MMN medium at various concentrations of ME. Dashed lines represent the δ¹³C value of control medium (C to F). Triangles, 0% [ME] (wt/vol) (C); squares, 0.065% [ME] (D); diamonds, 0.13% [ME] (E); X, 0.26% [ME] (F). Means and errors represent the values for three independent replicates at each harvest date (see Materials and Methods). Points with no visible errors have errors smaller than the symbol.

Growth-dependent isotopic fractionation was again observed under various ME concentrations, but fractionation patterns were sensitive to these concentrations (Fig. 3 and 4). In the 0% ME treatment, where biomass remained low, we observed a depletion in biomass ¹³C (Fig. 3). For all other treatments, we observed the biphasic fractionation pattern as described for C. volvatus in the single-time-point experiments (Fig. 2 and 3). The maximum fractionation observed during P1 was higher as the concentration of ME increased (Fig. 4). The first observed decline in observed fractionation (Δ_C) occurred at a larger biomass at higher ME concentrations (Fig. 4). No significant differences were observed in the rate of change in Δ_C as a function of accumulated biomass between the ME treatments during P1 (Fig. 4; regressions from initial harvest to harvest with maximum fractionation; 0% ME treatment not included; P_slope > 0.10, P_y-intercept > 0.25). This was also the case for P2 with the three ME treatments (P_slope > 0.50, regression through first harvest point after maximum Δ_C to end of experiment) (Fig. 4). Similarly, there was not a significant difference in the slope of the 0% ME treatment (all harvest points included) and that of the three ME treatments during P2 (P_slope > 0.25) (Fig. 4). As in the previous dynamic fractionation experiment, there was a significant (P = 0.002) shift in the fungal biomass percent C in each treatment when initial samplings were compared to those that occurred after the maximum fractionation (Table 2), with percent C = 37.0% ± 0.6% and 40.1% ± 0.6% for P1 and P2, respectively. Results obtained for C. volvatus in both dynamic experiments with similar ME concentrations (i.e., 0.13% [wt/vol]) are comparable (Fig. 1 and A2 and Fig. 3A and E).

FIG. 4.

Stable C isotopic fractionation (Δ_C; see equation 1) as a function of accumulated biomass for C. volvatus grown on modified MMN medium with various concentrations of ME (triangles, 0% [wt/vol] [ME]; squares, 0.065% [ME]; diamonds, 0.13% [ME]; X, 0.26% [ME]). Means and errors represent the values for three independent replicates at each harvest date (see Materials and Methods). Points with no visible errors have errors smaller than the symbol. Dotted line represents 1/2 · Δ_Cmax (see text), and arrows represent the approximate biomass at 1/2 · Δ_Cmax.

Both the sucrose and the maltose contained in the growth medium were used by C. volvatus. Sucrose was the dominant sugar used by the fungus (Fig. 5), and there was no clear preference for either of the two disaccharides during P1 (Fig. 5). There was an increase in the preference for maltose during P2 (Fig. 5). Oxalate produced during the exponential stage of growth was the only organic acid detected in the medium (Fig. 3A and 6); the amount of oxalate excreted by C. volvatus can account for the drop in medium pH observed (Fig. 3B and 6). The average concentration of oxalate during the incubation was 0.07 ± 0.02 mg/ml (Fig. 6), which would correspond to a pH of 2.2 in an unbuffered solution (H₂C₂O₄ pK₁ = 1.23).

FIG. 5.

The ratio of maltose-C to sucrose-C in medium originally containing ME at a concentration of 0.13% (wt/vol) after growth by C. volvatus. Results are from dynamic experiment 2. Means and errors represent values for three independent replicates at each harvest date (see Material and Methods). Points with no visible errors have errors smaller than the symbol.

FIG. 6.

Concentration of oxalate in medium as a function of time after growth by C. volvatus at an ME concentration of 0.13% (wt/vol). Results are from dynamic experiment 2. Means and errors represent values for three independent replicates at each harvest date (see Material and Methods). Points with no visible errors have errors smaller than the symbol.

DISCUSSION

The transformation of C compounds through decomposition and other microbial processes comprises a multitude of pathways, each of which is a potential point for isotopic fractionations. Inclusion of the impact of microbial interfaces on the incorporation of C isotopic species into ecosystem components must account for two sources of isotopic differentiation: intrinsic physiological fractionations and substrate selection by various microbes. This is in contrast to ecological studies considering only photosynthesis. Differential patterns of C isotope distributions in plants are attributed to physiological differences in C fixation and reprocessing within plants (4, 8, 25). The isotopic composition of the sole substrate of photosynthesis, CO₂, is relatively constant and homogeneous (4).

Substrate preferences can result in differences in δ¹³C values of fungi collected in the field (12), but physiological fractionations are apparent even after variation in δ¹³C values resulting from substrate preferences is taken into account (12). Here we focused on the physiological component of fungal fractionation in a sample of forest basidiomycetes (Table 1). It is clear that at least two fractionation phases occur during growth (Fig. 2). The first phase, P1, was characterized by an increasing enrichment in ¹³C content of the fungal biomass relative to the growth medium, while the second phase, P2, resulted in a continual depletion in ¹³C of the fungal biomass. During P2, the δ¹³C values of the fungal biomass asymptotically approached the δ¹³C value of the growth medium at inoculation. These results are robust and are independent of environmental conditions, such as inoculation strategy, temperature, and light, suggesting a mechanism that is biologically inherent to the various fungi tested.

Based on previous studies (11) and the new results presented here, we suggest that the observed P1 fractionation occurs during C uptake. Because the δ¹³C values obtained here are derived for total biomass C and are not differentiated among cellular components, fractionation must result from either the preferential uptake of the ¹³C isotopic species or discriminating excretion of ¹²C in the form of either CO₂ or other metabolites. Results from C. volvatus indicate that the production of potentially ¹³C-depleted metabolites is minimal (Fig. 6). Furthermore, previous studies indicate that the respired CO₂ is not enriched in ¹²C relative to the fungal biomass under comparable conditions (i.e., growth conditions and biomass) for the same fungi as those used here (11) (see Table 2 for confirmation of this assumption). Additionally, the consistency of our results would be unlikely if fractionation did not occur early during incorporation of C into fungal biomass, since small error values (Fig. 2 to 3), even across experimental runs (Fig. 2 and 3E), would not be expected once C isotopic species were scrambled during various enzymatic reactions and intracellular recycling of metabolites (16).

The measured enrichment in ¹³C in the fungal biomass is a physiologically based fractionation effect and is not due to the selective uptake of ¹³C-enriched sugars from the medium. Although small amounts of C4-derived maltose were available in the medium, sucrose was the dominant sugar used, and the ratio of maltose-C to sucrose-C is relatively constant during P1 (Fig. 5). Given the ratio of the two sugars used, a δ¹³C value of the fungal biomass would be −23.4‰ ± 2.3‰ (simple mixing model) if no fractionation occurred during C processing. Furthermore, assuming that the affinity for the maltose transporter is constant, fractionations due to a preference for maltose would result in an observed decrease in Δ_C with time, not an increase (Fig. 2 and 3), as maltose is consumed. Therefore, a hypothetical preferential uptake of C derived from ME is not the cause of the observed enrichment in ¹³C in the fungal biomass.

The exact means by which ¹³C is selectively taken into the fungal cell from the medium is unclear. Sugar molecules are known to have an asymmetrical distribution of C isotopic species (26), and we previously hypothesized that this asymmetry might be at the root of observed enrichments of ¹³C in fungal biomass (“dual-uptake hypothesis”) (11). Such a mechanism is dependent on the existence of at least two C uptake pathways with different kinetics (11). Fungi have separate transport mechanisms for sucrose and its carbohydrate monomers, glucose and fructose (20), providing a minimum molecular infrastructure for fractionation to take place following the dual-uptake hypothesis. Alternatively, differential retention times between enriched and depleted molecules at a single transport-enzyme complex could also result in the observed fractionation. This would be analogous to the mechanism documented for fractionation during the fructose-1,6-bisphosphate aldolase reaction in plants that concentrates ¹³C in chloroplast starches versus in cytosol sugars (2, 8, 9). Sucrose transport in fungi can happen directly or via a hexose transporter after exocellular degradation of the disaccharide (20). Nevertheless, detailed biochemical experiments would be necessary to confirm the mechanism leading to fractionations documented here.

At the end of the exponential stage of growth, there is a transition from the large, increasing ¹³C enrichment of P1 to a new dynamic, P2, which leads to an asymptotic depletion in ¹³C of the fungal biomass (Fig. 2 and 3). Several signals suggest a shift in metabolic state at the P1-P2 transition. There is a slowing of growth (Fig. 2 and 3) with a simultaneous significant shift in the biomass percent C (Table 2). This change in percent C, albeit small in magnitude, is so consistent in our experiments that we must surmise that it derives from an important change in the molecular composition of the fungal biomass between P1 and P2. The specifics of this percent C transition remain to be studied, but the difference is similar in magnitude to the difference in percent C between molecules more abundantly found in young cells, such as amino acids, proteins, and nucleic acids, versus structural and storage compounds typical of older cells, such as fatty acids and chitin.

We cannot conclude whether the asymptotic depletion in ¹³C during P2 fractionation results from a switch in how the fungus immobilizes C, or whether it results from the excretion of ¹³C-enriched C. A hypothesis involving the excretion of ¹³C-enriched C compounds is less plausible, since (i) the respired CO₂ would need to be enriched in ¹³C, and we know from previous studies that it is likely the same isotopic signature as the fungal biomass (11), and (ii) the concentration of oxalate, the most prominent metabolite excreted, was extremely low and relatively constant over the course of P2 (Fig. 3E and 6). It is possible that the transition from P1 to P2 is a result of a mechanism similar to that hypothesized by Henn and Chapela (11) in that there are multiple C transport pathways. In this case, the operation of P1, which fractionates C isotopes during uptake, decreases, and an uptake mechanism that selectively incorporates ¹²C becomes kinetically favored at the onset of P2. Importantly, P2 fractionation does not result from the simple arrest of a fractionation mechanism that then leads to the accumulation of additional ¹²C-enriched biomass, because the overall amount of biomass remains relatively constant throughout P2 dynamics (Fig. 2 and 3). Instead, there must be an active replacement of biomass enriched in ¹³C by corresponding biomass that is relatively depleted in the heavy isotope. Notably, M. androsaceus was not able to maintain P2 dynamics during its lytic stage (Fig. 2), further highlighting the active nature of the mechanism mediating this phase. P2 fractionation is observed again for this fungus when it “recovers” and is able to maintain a stationary phase (Fig. 2). That stationary growth stages are maintained through significant metabolic activity has been described for the production of secondary metabolites in other fungi (10, 27), and we suggest that the P2 dynamics are equally demanding of important metabolic activity, which nevertheless does not result in significant biomass accumulation.

Activation of the biochemical mechanisms responsible for P1 and P2 dynamics is dependent on the availability of micronutrients provided in the ME. P1 dynamics are only observed when ME is a constituent of the growth medium (Fig. 3C to F and 4). P2 dynamics are observed in all ME treatments, including the 0% treatment (Fig. 3C to F and 4). The relative constancy of the P1 fractionation across treatments with ME added followed by variation in the onset of P2 dynamics at various ME concentrations indicates that the switch between P1 and P2 is dependent on the presence of a growth factor in catalytic amounts (Fig. 4). Extrapolation of the biomass values accrued by the fungus at one-half the average maximum Δ_C (1/2Δ_Cmax) during P2 (arrows in Fig. 4; Δ_Cmax determined from the average of the maximum Δ_C values achieved for three ME treatments) approximates the stoichiometric relationship of 1:2:4 for the amount of ME added to the medium. This dose-dependent onset of P2 suggests that maintenance of the fractionating P1 requires the sequestering of critical nutrients from the ME in a nonreversible reaction. Although there is no evidence to suggest what such a reaction might be, a plausible example of the type of mechanism involved could be the sequestration of P from the medium during phosphorylation of sugars during uptake.

It is possible that the medium pH also influences the dominance of P1 versus P2. H⁺ protons are known to be involved in the cotransport of sugars (1, 17). In the treatments containing ME, the onset of P2 does appear related to a significant drop in pH (Fig. 3B and D to F). However, a P2-like decrease in the δ¹³C value of the fungal biomass in the 0% ME treatment is evident from the start of the experiment irrespective of the medium pH (Fig. 3B and C). This suggests that pH is not the sole factor regulating the onset of P2 and that some nutrient limitation is likely the dominant factor controlling the transition between phases. The acidification of the growth medium appears to be a result of the excretion of oxalate into the medium.

Our finding of growth-dependent isotopic fractionation in fungi provides a solid framework to investigate the impact of residual physiological fractionations at fungal interfaces on the natural distribution of stable C isotopes in ecosystems. Under natural conditions, the rates of C input and various nutrients can vary and determine which fractionation phase dominates. There are conditions in nature under which fungi are restricted to bulky natural C sources (e.g., a stump or a given patch of litter), while under other conditions, they are maintained in a state of continuous substrate replenishment. These differences, which could be analogized to the difference between batch- and continuous-culture fermentation, would result, over time and on average, in a relatively more enriched δ¹³C value, characteristic of P1, for the former (bulk substrate decomposition) than for the latter (replenished substrate processing). For example, we may expect a P1 type of fractionation to dominate in EM fungi, since their symbiotic association with a host plant allows for a sustained supply of nutrients. Similarly, some fungi exhibit rapid turnover of their mycelium and may more frequently display exponential growth, while others form more permanent mycelial structures, such as rhizomorphs and mats. Again, the former situation may result over time and on average in a relatively more enriched δ¹³C value versus the latter.

Implementation of a multiphase, dynamic model of fungal C isotopic fractionation over the growth cycle of various fungi coupled with specifics of their life history can be used to establish expected fractionation values in various ecological situations under changing environmental conditions. We find that determination of the physiological status of fungi involved in ecological processing should be a requirement before we can interpret bulk isotopic measurements of ecosystem components. Our findings suggest that variability in the magnitude of fungal fractionations due to varied ecophysiological conditions should be considered the norm, unless evidence to the contrary is available. Instead of assuming that no fractionation occurs in ecosystems at fungal interfaces, careful selection of samples (3), parallel collections of substrates (12), and determination of the physiological status of mycelia should result in a more reliable utilization of stable C isotopes in ecological analysis.