The impacts of climate change on worldwide crop production become increasingly severe. Thus, sustainable enhancements of agricultural production are needed. The present study investigated the effects of drought and arbuscular mycorrhizal fungi on wheat plants (Triticum aestivum) and their interaction with aphids. Considering predicted climate change scenarios, wheat plants were exposed to well-watered conditions, continuous (CD) or pulsed (PD) drought and plants were grown without (NM) or with mycorrhizal (AM) fungi. Ear biomass and harvest index were evaluated when grains were produced. Moreover, drought- and mycorrhiza-induced changes in the amino acid composition of leaf phloem exudates were studied and the population growth and survival of Sitobion avenae aphids on those plants measured. Wheat plants responded differently towards the irrigation treatments. Under drought stress, ear biomass was reduced, while AM resulted in an enhanced harvest index. In phloem exudates especially relative concentrations of the osmoprotectant proline were modulated by drought. Aphid population size was influenced by the interaction of drought and mycorrhiza treatment. This study emphasises the pronounced influence of irrigation frequency on plant performance and indicates positive contributions of AM that may be relevant for agriculture.

Plant cultivation and mycorrhiza treatments

The experiment was carried out in a climate chamber at 22.7 ± 1.6 °C (mean ± SD) with a relative humidity (r.h.) of 61 ± 5% and 16 h light:8 h dark. Untreated grains of spring wheat cultivar Tybalt were provided from Borries-Eckendorf (Leopoldshöhe, Germany). Plants were grown in 2 L pots (11.3 ×11.3 × 23 cm) with a 2:1 mixture of steamed (4 h, at 120 °C) sand and soil (Fruhstorfer Pikiererde, HawitaGroup, Vechta, Germany). This mixture was used to be able to harvest and clean the roots at the end of the experiment to determine the total root length colonisation (TRLC). The pots had holes at their base to enable draining of surplus water and were placed on holders to prohibit water loss. To pots of the arbuscular mycorrhizal treatment (AM; n = 30) 150 mL of fungal inoculum (sand spore mixture, pH 7; Inoq GmbH, Schnega, Germany) were added, containing the generalist fungus Rhizoglomus irregulare (Błaszk., Wubet, Renker & Buscot) Sieverd., G. A. Silva & Oehl (Glomerales, Glomeromycota). This AMF species was used because it is commercially available for potential use in agriculture, colonises wheat roots and is known to affect growth and physiological responses of spring wheat (Campos et al., 2019; Zhang et al., 2018). The inoculum was mixed with the upper third of the substrate in each pot. To provide comparable microbial conditions in the non-mycorrhized control treatment (NM; n = 30), 150 mL sterilised (3 h, at 120 °C) inoculum were mixed with the upper third of the substrate of each pot and 45 mL of a microbial solution obtained from the inoculum before sterilisation by filtrating a washing of the inoculum through a 20 µm sieve were added. AM pots received 45 mL water instead. To simulate field conditions with regard to plant density and nutrient competition each pot contained two plants in opposite corners, at a distance of about 10 cm. Sufficient germination was guaranteed by putting three seeds in the pots. Surplus seedlings were removed six days post sowing (dps). Plants were fertilised 18 (1 g / plant) and 35 dps (0.6 g / plant) with a solid long-term, phosphate-free mineral fertiliser (Floranid N-P-K 14-0-19, containing 3% Mg, 11% S and traces of B, Cu, Fe, Mn and Zn; Manna, Düsseldorf, Germany).

Establishment of irrigation treatments

Initially, all pots were kept well-watered near field capacity [~ 18% soil water content (SWC) determined in preliminary experiment] until 24 dps to gain robust plants. The water requirement was determined gravimetrically. Therefore, every other day 15 representative pots were chosen, weighed and weights averaged. Pots then received the calculated amount of water to re-gain the requested soil moisture. Within each mycorrhiza treatment (NM/AM), pots were randomly attributed at 24 dps to one of three irrigation treatments, a control (CTR), continuous drought stress (CD) or pulsed drought stress (PD) (n = 10 per mycorrhiza and irrigation treatment). CTR pots were continued to be weighed as described above and watered to 18% SWC. CD and PD pots were left unwatered until a SWC of ~8% was reached. Then, CD pots received 40% of the water amount that was added to the CTR pots. A reduction of precipitation to 40% may occur in certain hot and dry summers, especially in certain regions under ongoing climate change. Furthermore, this amount of water had been shown to result in a significantly reduced aboveground biomass of wheat in a previous study under similar growth conditions, without killing the plants (Stallmann, Schweiger & Müller, 2018). CTR and CD plants were watered every other day. Pots of the PD treatment were irrigated only every eight days with the cumulated amount the CD pots had received in that time period (Fig. 1). All pots were randomly distributed and their position was changed weekly. The number of replicates was reduced by three pots (2 CD AM, 1 PD AM) due to a local fungal infection and by a further pot (CD NM) due to a failure during phloem exudate sampling. If not stated otherwise, in the following “treatment” is referred to as a combination of mycorrhizal treatment and irrigation treatment, e.g. AM CTR.

Phloem exudate collection and plant harvest

To examine the influence of the different treatments on the phloem sap composition of developing wheat plants, phloem exudates were collected when ears were fully emerged and plants started flowering (52 dps, T1; Fig. 1). At this time point, the aphid bioassays were started as well (see below). Phloem exudates were collected from half of the replicates (group A; n = 4-5 per treatment) from one plant per pot, following a method modified after Kos et al. (2011) and Schweiger, Heise, Persicke & Müller (2014c). To guarantee a sufficient turgor of the phloem sap the phloem exudate collections took place 24 h after watering all plants. The three youngest leaf blades of the main shoot were cut at their base and placed into a 50 mL Falcon tube with 1 mL of an 8 mM ethylenediaminetetraacetic acid solution (EDTA; 99%, AppliChem GmbH, Darmstadt, Germany; pH = 7, adjusted with NaOH) in the dark for 2 h (20 °C, 60% r.h.). After this first incubation, leaves were washed in Millipore water (MicroPure Water Purification System, Thermo Fisher Scientific, Niederelbert, Germany), transferred to a new 50 mL Falcon tube with 1 mL Millipore water and incubated for another 2 h in the dark. These second collections were used for subsequent chemical analyses. Blanks were prepared by keeping Millipore water without plant material in Falcon tubes for the same duration. For amino acid analysis, 300 µL of the exudates and blanks were frozen in liquid nitrogen and stored at -80 °C. Leaf blades used for phloem exudate sampling were dried for 48 h at 40 °C and weighed.

When flowering was completed and grains were watery ripe (68 dps, T2), the total (remaining) aboveground plant biomass of both plants per pot was harvested for groups A and B. Biomass was separated into vegetative and generative (ears) parts and dried for 96 h at 40 °C to determine the dry biomass. The HI was calculated by dividing the dry ear biomass by the dry total aboveground biomass.

Amino acid analysis of phloem exudates

To analyse the amino acid composition, phloem exudates and blanks were lyophilised and redissolved in 80% methanol with norvaline and sarcosine (Agilent Technologies, Waldbronn, Germany) as internal standards for primary and secondary amino acids, respectively. Samples were analysed via high-performance liquid chromatography coupled with fluorescence detection (HPLC-FLD; 1260/1290 Infinity HPLC and FLD, Agilent Technologies, Santa Clara, CA, USA) following the protocol of Jakobs & Müller (2018). Identification of amino acids was performed via comparison of retention times with those of reference standards measured within the same worklist. Amino acids were quantified by integrating the corresponding peaks, using OpenLab ChemStation C.01.07 (Agilent Technologies). Data were normalised by dividing the peak areas by the areas of the corresponding internal standards. Peak areas were related to the dry weight of the leaf blades from which the phloem exudates had been collected. Amino acids were only included into further analysis when they were found in at least three of the five replicates (or two of four replicates) of one treatment and not in the blanks. To compare the relative composition of amino acids in dependence of the plant treatment, for each amino acid the mean percentage (i.e., its mean peak area divided by the mean total peak area of all amino acids) was determined for each treatment group. The amino acid data were only compared visually, because the sample sizes (n = 4-5) were too small for proper statistical analyses.

Root sampling and quantification of AMF colonisation

To determine the TRLC with AMF two subsamples of roots were taken per pot at T2 by punching a cork borer (2.8 cm i.d.) twice vertically into the substrate for about 22 cm. Subsequently roots were washed and representative subsamples were bleached (10% KOH; 20 min, 95 °C), dyed with an acetic solution of ink (royal blue, Ink 4001, Pelikan Group GmbH, Berlin, Germany) (1:1:8 ink:acetic acid:water; 15 min, 90 °C) and conserved (4:2:1 90% lactic acid:89% glycerine:water, 4 °C in the dark). The TRLC was determined using the grid-line intersect method (Giovanetti & Mosse, 1980) by separately counting different AMF structures (i. e., hyphae, vesicles, arbuscules) in 200 intersects per sample.

Bioassays with aphids

Aphids of S. avenae were obtained from Koppert Biological Systems (Suffolk, UK) and kept in tents (58 ´ 58 ´ 58 cm) on wheat plants (cultivar: Tybalt) for several generations at 16 h light:8 h dark. To examine the influences of the plant treatments on aphid performance we investigated the growth of aphid populations as well as the survival of single nymphs on leaves of the experimental plants over 16 days (from T1 to T2). To record the population growth clip cages (2 cm i.d.) with three eight days-old apterous aphids were attached to the second youngest leaf of the main shoot of one intact plant per pot (groups A and B, n = 8-10 / treatment). Under the used conditions, aphids of S. avenae turn into adults at about 9 days and live for about 20 - 60 days. For group A, the plant in each pot was used that had not been used for phloem exudate collections (see above). Numbers of living adult aphids and their offspring were counted after 16 days on the plants (T2). To record the survival of single aphids, a one-day-old nymph was fixed in a clip cage on the flag leaf of the same shoot (n = 8-10 / treatment). Survival of these nymphs was checked every day and offspring counted and removed.