Connect with us


Mycorrhizal C/N ratio determines plant-derived carbon and nitrogen allocation to symbiosis – Communications Biology



Collection of field-grown beech trees

Young, healthy beech (Fagus sylvatica) trees (n = 40) with fully open leaves were excavated from May 10th to 15th in a deciduous forest (Billinghäuser Schlucht, Göttingen, Latitude: 51°35′15.39″, Longitude: 9°58′57.95″, 362 m above sea level, mean annual temperature 8.5 °C, mean sum of annual precipitation 614 mm). The trees were carefully removed and placed with intact soil layers individually in pots (diameter 183 mm, height 255 mm, corresponding to about 6.5 L soil volume). Soil has been classified as Orthic Rendzina on limestone with L-Mull as the humus type; the soil in this forest has a pH value of 4.9 and contains 36 mg C g−1 and 2.6 mg N g−1 dry weight soil61. Herbs were removed. The trees were kept for about 10 weeks under common garden conditions in Göttingen. The plants were watered regularly and were used in summer (end of July to August) for the labeling experiments. The trees were similar in size (±SE): 133 ± 4 cm height, 12.05 ± 0.29 mm width at the stem base, 0.23 ± 0.02 m2 whole-plant leaf surface area, 7.3 ± 0.6 g leaf biomass, 12.5 ± 1.0 g coarse root and 3.9 ± 1.1 g fine root biomass.

Stable isotope labeling of beech trees

For 15N leaf labeling, three adjacent top, three middle, and three bottom leaves were selected, whose surface was slightly abraded with fine grained abrasive paper (Basic Korn 240, LUX, Wermelskirchen, Germany), and immediately placed in a solution of 20 mM 15NH4Cl (99% NH4Cl, Campro Scientific GmbH, Berlin, Germany) in a reagent vessel with the leaves still attached to the tree. The twig with the feeding leaves in the experimental set-up was placed in a plastic bag to avoid overly evaporation. The surface of the pots was covered with plastic foil to avoid soil contamination. After 72 h the feeding device and soil cover were removed and the 15N fed leaves were cut off to finish 15N uptake. The area of the feeding leaves was scanned and corresponded to a mean of 6% of the total tree leaf area. The latter was determined when the trees were harvested. The labeled leaf areas did not vary among different sets of plants (P = 0.921).

The pots with the trees were then placed in plastic bags, which were sealed at the stem bottom to separate the above- and below-ground compartments (Supplementary Fig. S1a, b). To enable irrigation, a tube was introduced into the bag and placed on top of the soil. The prepared trees were transferred into a gas-tight chamber (Weiss Umwelttechnik, Reiskirchen, Germany). The end of the irrigation tube was installed outside of the chamber and used to supply the trees with water. 12CO2 and 13CO2 were supplied by silicon tubing to the air space of the gas-tight chamber and circulated with the help of valves and a pumping system (Supplementary Fig. S1c). In the chambers, the trees were exposed for an adjustment phase of 2 days to 12CO2, then for 3 days to 13CO2 and then for a chase period of 2 days to 12CO2 before they were removed. The duration of the chase period was chosen according to Sommer et al.62 to achieve maximum 13C enrichment in roots and the labeling intensity was chosen to reach a detectable signal in small amounts of biomass (such as ectomycorrhizas) without any known disturbance of metabolism63,64. To produce 13CO2, 106.98 g Na213CO3 (99 atom-% 13C, Sigma-Aldrich, Taufkirchen, Germany) was dissolved in 2 L of deionized H2O and automatically mixed with 5 M lactate (90% C3H6O3 pur Ph Eur. USP, AppliChem, Darmstadt, Germany) by a pumping system. During the adjustment phase, the same device was used but supplied with CO2 produced from Na212CO3 (105.99 g in 2 L deionized H2O, KMF Laborchemie Handes GmbH, Lohmar, Germany) instead. When the 12CO2 concentration (senor: Carbocao, GMM220, Driesen+Kern GmbH, Bad Bramstedt, Germany) in the air space dropped below 180 ppm, the pumps were activated and adjusted 12CO2 to 400 ppm. The 13CO2 concentrations were measured by gas chromatography and reached a maximum of 1800 ppm. 13CO2 was applied only during the light phase. During exposure in the chambers, the trees were kept for 16 h under light (420 µmol photons m−2 s−1 of photosynthetically active radiation) at 70% relative air humidity and 20 °C air temperature. After removal from the chambers, the trees were kept indoors, in ambient air with 16 h/8 h light/dark cycles under the same environmental conditions as in the chambers. The first group of trees was harvested immediately after removal (5 days since the start of the 13CO2 treatment) and the second group 15 days later (20 days since the start of the 13CO2 treatment). The experimental exposures were conducted in two independent runs, each with 10 trees of which half were harvested in the first and half in the second group (5 or 20 days of 13CO2 treatment). Ten non-labeled trees maintained indoors under the same environmental conditions as in the labeling chambers and ambient CO2 (397 ± 4 ppm) were used as controls.

Harvest and root morphotyping

The leaves were harvested and scanned to determine plant leaf area. The plant stem was cut off at the soil surface. An ~20 mm long bottom part of the stem was cut and debarked. The bark was dried. The roots with soil were removed from the pots and cautiously shaken. Soil adhering to fine roots was defined as rhizosphere soil. An aliquot was collected with a toothpick (ca. 2 g), and dried. Then the roots were shaken to remove the remaining soil. The total amount of rhizosphere soil (dry) was 54 ± 10 g per root system. The roots were briefly washed and separated into fine (<2 mm diameter) and coarse roots. Subsamples of coarse and fine roots were dried.

Further subsamples of fresh fine roots were viewed under a stereomicroscope (M 205 FA, Leica, Wetzlar, Germany) and about 400 root tips per plant were classified as vital ectomycorrhizal, vital non-mycorrhizal or dry. Vital and dry root tips were distinguished based on their microscopic appearance, categorizing shrunken and dark brown distorted root tips as dried. This technique was previously validated by stable isotope tracing, demonstrating negligible 13C or 15N transfer to the dry root tips of plants supplied via the soil with labeled glucose or ammonium33,65. The number of root tips in each category (vital non-mycorrhizal, vital mycorrhizal, dry) was counted and used to calculate:

$${{{{{\rm{Mycorrhizal}}}}}}\,{{{{{\rm{colonization}}}}}}\,{{{{{\rm{rate}}}}}}\,( \% )= \, {{{{{\rm{Number}}}}}}\,{{{{{\rm{of}}}}}}\,{{{{{\rm{vital}}}}}}\,{{{{{\rm{mycorrhizal}}}}}}\,{{{{{\rm{root}}}}}}\,{{{{{\rm{tips}}}}}}* 100\\ /({{{{{\rm{number}}}}}}\,{{{{{\rm{of}}}}}}\,{{{{{\rm{vital}}}}}}\,{{{{{\rm{mycorrhizal}}}}}}\,{{{{{\rm{root}}}}}}\,{{{{{\rm{tips}}}}}}\\ +{{{{{\rm{number}}}}}}\,{{{{{\rm{vital}}}}}}\,{{{{{\rm{non}}}}}}-{{{{{\rm{mycorrhizal}}}}}}\,{{{{{\rm{root}}}}}}\,{{{{{\rm{tips}}}}}}).$$

$${{{{{\rm{Root}}}}}}\,{{{{{\rm{tip}}}}}}\,{{{{{\rm{vitality}}}}}}\,( \% )= \, ({{{{{\rm{number}}}}}}\,{{{{{\rm{of}}}}}}\,{{{{{\rm{vital}}}}}}\,{{{{{\rm{mycorrhizal}}}}}}\,{{{{{\rm{root}}}}}}\,{{{{{\rm{tips}}}}}}\\ +{{{{{\rm{number}}}}}}\,{{{{{\rm{vital}}}}}}\,{{{{{\rm{non}}}}}}-{{{{{\rm{mycorrhizal}}}}}}\,{{{{{\rm{root}}}}}}\,{{{{{\rm{tips}}}}}})* 100\\ /{{{{{\rm{number}}}}}}\,{{{{{\rm{of}}}}}}\,{{{{{\rm{all}}}}}}\,{{{{{\rm{counted}}}}}}\,{{{{{\rm{root}}}}}}\,{{{{{\rm{tips}}}}}}.$$

Vital ectomycorrhizal root tips were further classified into ectomycorrhizal morphotypes according to fungal characteristics using for instance color, branching, or lengths of emanating hyphae66 and, subsequently counted to determine the relative abundance of the morphotypes in a sample as follows:

Relative abundance of morphotypei = number of root tips with morphotypei/number of vital mycorrhizal root tips

Aliquots of each morphotype consisting of fungal and plant tissues were collected (Supplementary Fig. S2). Because individual trees did not contain all morphotypes in sufficient quantities, the number of replicates per morphotype varied (n = 5 to 7). One subsample was stored at −80 °C for molecular identification, another was fixed in a solution of FAE (formaldehyde 37%:ethanol 70%:acetic acid 97% in proportions of 5:90:5) for Secondary Ion Mass Spectrometry, and the rest sample was dried. The lateral root segment connecting the ectomycorrhizal morphotype with the next higher root branch was also collected27 (Supplementary Fig. S2a) and dried.

Identification of morphotypes by internal transcribed spacer (ITS) sequencing

The DNA of individual morphotypes was extracted in 100 µl lysis buffer of the innuPrep Plant DNA kit (Analytik Jena, Jena, Germany) with a pellet mixer (VWR Pellet Mixer, VWR International, Darmstadt, Germany). Further steps were conducted as described by Lang et al.3 using the internal transcribed spacer (ITS) region of the fungal rRNA (primers ITS1f (5′CTTGGTCATTTAGAGGAAGTAA-3′) and ITS4 (5′TCCTCCGCTTATTGATATGC-3′)) and the Fermentas protocol for polymerase EP 0402 (Fermentas, Waltham, Massachusetts, USA). Sequencing was conducted by the company Seqlab (Sequence Laboratories, Göttingen, Germany). The sequences were analyzed with Staden package ( and blasted in NCBI GenBank ( and UNITE ( databases. Fungal sequences have been deposited in NCBI GenBank under the accession numbers MK430999 to MK431014.

Image analyses and secondary ion mass spectrometry (SIMS)

The morphotype samples were dehydrated by incubation in ethanol solutions of increasing concentrations, 80%, 96%, and 100% for 90 min each. Subsequently, the morphotypes were embedded in Technovit 7100 resin (Heraeus Kulzer GmbH & Co. KG, Hanau, Germany) according to the manufacturer´s instructions. Technovit is a well-established acryl plastic embedding system commonly used in NanoSIMS analyses67,68 because it has negligible N contents. Cross-sections with a thickness of 1.0 µm were cut with an autocut microtome (Ultracut E, Reichert-Jung, Vienna, Austria), stained with toluidine blue36, and were viewed and photographed under a microscope (Axioplan Observer Z1, Carl Zeiss GmbH, Göttingen, Germany). Digital images of the cross sections were used to measure surface areas of root tissue and fungal tissue with ImageJ ( Based on these measurements, we calculated the percentage occupied by fungal structures within ectomycorrhizas as Fungal tissue (%) = area of fungal tissue × 100/total cross-sectional area. Details are shown in Supplementary Table S1. For NanoSIMS imaging, 200 nm thick cross sections were cut using an EM UC6 ultramicrotome (Leica Microsystems, Wetzlar, Germany). The sections were analyzed with a NanoSIMS 50 L instrument (Cameca, Gennevilliers, France) equipped with an 8 kV Cs+ primary ion source69. To reach a steady state of ionization, samples were first implanted applying a primary current of ~100 pA for 1.5 minutes. During the analysis, a current of ~1 pA was employed. The detectors were set to collect 12C14N and 12C15N secondary ions. The mass resolving power was adjusted to differentiate 12C15N from potential interferences such as 13C14N. Ion images of 512 × 512 pixels were acquired setting a raster size of 30 ×30 μm or 80 ×80 µm, and a dwell time of 5.07 ms/pixel from three consecutive layers. NanoSIMS data processing, including image exportation, drift correction, image stacking, and ratio measurements was performed using the OpenMIMS plugin of Fiji70.

N and C analyses

For analyses of 14N, 15N, 12C, and 13C all dried samples (except morphotypes and lateral roots) were milled and weighed (0.5 to 1.5 mg on a supermicro balance, type S4, Sartorius, Göttingen, Germany) into tin cartouches (5 × 9 mm, HEKAtech, Wegberg, Germany). Each collected morphotype (Supplementary Fig. S2) and lateral root sample was weighed and used as whole sample in the analysis. Since the whole sample was used, no milling was necessary to obtain representative subsamples. The samples of the labeled and the non-labeled controls were analyzed in an isotope ratio mass spectrometer (IRMS; Delta C, Finnigan MAT, Bremen, Germany) in the Center for Stable Isotopes (KOSI, University of Göttingen, Göttingen, Germany). An acetanilide standard (C = 71.09%, N = 10.36%, δ13C = −29.6‰, δ15N = −1.6‰) was measured after each 10th sample and used for point calibration. A second in-house standard (wild boar liver δ13C = −17.3‰), δ15N = 7.3) was used periodically to check for scaling. The standard materials were calibrated against international standards (International Atomic Energy Agency: IAEA-N1 and IAEA-N2 for 15N and NBS18 and IAEA 600 for 13C, Atom-percent excess (APE) and the 13C enrichment (which is the concentration of newly acquired 13C) were determined as follows:

$${{{{{\rm{APE}}}}}}^{13}{{{{{\rm{C}}}}}}( \% )=\left({\left(\frac{{\,}^{13}{{{{{\rm{C}}}}}}}{{\,}^{13}{{{{{\rm{C}}}}}}+{\,}^{12}{{{{{\rm{C}}}}}}}\right)}_{{{{{{{\rm{labelled}}}}}}}}-{\left(\frac{{\,}^{13}{{{{{\rm{C}}}}}}}{{\,}^{13}{{{{{\rm{C}}}}}}+{\,}^{12}{{{{{\rm{C}}}}}}}\right)}_{{{{{{{\rm{non}}}}}}}-{{{{{{\rm{labelled}}}}}}}}\right)* 100$$

$${\,}^{13}{{\mbox{C}}}\,{{\mbox{enrichment}}}\left({{{{{\rm{mg}}}}}}\; {{{{{\rm{g}}}}}}^{-1}{{{{{\rm{dry}}}}}}\; {{{{{\rm{mass}}}}}}\right)=\frac{{{\mbox{APE}}}^{13}{{\mbox{C}}}* 1000}{100}* \frac{({\,}^{12}{{\mbox{C}}}+{\,}^{13}{{\mbox{C}}})({{\mbox{g}}})}{{{\mbox{Dry}}}\,{{\mbox{mass}}}({{\mbox{g}}})}$$

The amount of newly acquired 13C in ectomycorrhizas (per plant) was calculated as

$$ {{{{{{\rm{Ectomycorrhizal}}}}}}}{\,}^{13}{{{{{\rm{C}}}}}}\,{{{{{{\rm{amount}}}}}}}\,({{{{{{\rm{mg}}}}}}})\\ =\mathop{\sum }\limits_{i=1}^{n}({{\,}^{13}{{{{{\rm{C}}}}}}}_{{{{{{{\rm{EM}}}}}}}\,{{{{{{\rm{species}}}}}}}\,i}\times {{{{{{{\rm{root}}}}}}}\,{{{{{{\rm{tip}}}}}}}\,{{{{{{\rm{biomass}}}}}}}}_{{{{{{{\rm{EM}}}}}}}\,{{{{{{\rm{species}}}}}}}\,i}\times {{{{{{{\rm{number}}}}}}}\,{{{{{{\rm{of}}}}}}}\,{{{{{{\rm{root}}}}}}}\,{{{{{{\rm{tips}}}}}}}}_{{{{{{{\rm{EM}}}}}}}\,{{{{{{\rm{species}}}}}}}\,i})$$

The number of root tips(EM species i) was calculated as relative abundance(EM species i) × total number of root tips per plant. The mean total number of EM tips per plant was 29215 ± 4361 (n = 25 plants). The root tip biomass(EM species i) for different EM species is reported in Supplementary Table S1.

The amount of newly acquired 13C in roots (per plant) was calculated as

$${{{{{{\rm{Amount}}}}}}}\,{{{{{{\rm{of}}}}}}}{\,}^{13}{{{{{\rm{C}}}}}}\,{{{{{{\rm{in}}}}}}}\,{{{{{{\rm{roots}}}}}}}\,({{{{{{\rm{mg}}}}}}})= {{\,}^{13}{{{{{\rm{C}}}}}}}_{{{{{{{\rm{fine}}}}}}}{{{{{{\rm{root}}}}}}}}\times {{{{{{{\rm{biomass}}}}}}}}_{{{{{{{\rm{fine}}}}}}}\,{{{{{{\rm{root}}}}}}}}\\ +{{\,}^{13}{{{{{\rm{C}}}}}}}_{{{{{{{\rm{coarse}}}}}}}\,{{{{{{\rm{root}}}}}}}}\times {{{{{{{\rm{biomass}}}}}}}}_{{{{{{{\rm{coarse}}}}}}}\,{{{{{{\rm{root}}}}}}}}$$

APE 15N, the concentrations of newly acquired 15N and the 15N amounts were calculated accordingly.

Statistics and reproducibility

All source data used for statistical analyses and to produce the graphs and tables have been deposited in Dryad71. Statistical analysis was performed using Statgraphics Centurion XVI (StatPoint, Inc., St Louis, MO, USA). To test whether the data were normal distributed, the plots of the residuals were visually inspected. If the data deviated from normal distribution, they were log-transformed and rechecked. The variance was plotted and checked visually for homogeneity prior to one-way or multivariate analysis of variance. Main factors in the ANOVA were time and tissue. When P ≤ 0.05, a post hoc test was conducted (HSD) to compare means. When time (day 5 and day 20) showed no significant differences at P ≤ 0.05, the sets of both time points were analyzed together applying general linear models and time as random factor. Data are shown as means (±SE, n = 5 or 10) when not indicated otherwise. Linear regression curves between 13C enrichment in lateral roots and associated EM fungal species were compared using the function “comparison of regression lines”. Comparison of the ectomycorrhizal community composition present at different sampling dates was conducted using the Bray Curtis Similarity Index by ANOSIM in PAST software72.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Continue Reading
Click to comment

Leave a Reply

Your email address will not be published. Required fields are marked *