Root hairs increase rhizosphere extension and carbon input to soil

Abstract

Background and Aims

Although it is commonly accepted that root exudation enhances plant–microbial interactions in the rhizosphere, experimental data on the spatial distribution of exudates are scarce. Our hypothesis was that root hairs exude organic substances to enlarge the rhizosphere farther from the root surface.

Methods

Barley (Hordeum vulgare ‘Pallas’ – wild type) and its root-hairless mutant (brb) were grown in rhizoboxes and labelled with ¹⁴CO₂. A filter paper was placed on the soil surface to capture, image and quantify root exudates.

Key Results

Plants with root hairs allocated more carbon (C) to roots (wild type: 13 %; brb: 8 % of assimilated ¹⁴C) and to rhizosheaths (wild type: 1.2 %; brb: 0.2 %), while hairless plants allocated more C to shoots (wild type: 65 %; brb: 75 %). Root hairs increased the radial rhizosphere extension three-fold, from 0.5 to 1.5 mm. Total exudation on filter paper was three times greater for wild type plants compared to the hairless mutant.

Conclusion

Root hairs increase exudation and spatial rhizosphere extension, which probably enhance rhizosphere interactions and nutrient cycling in larger soil volumes. Root hairs may therefore be beneficial to plants under nutrient-limiting conditions. The greater C allocation below ground in the presence of root hairs may additionally foster C sequestration.

INTRODUCTION

Future agriculture will be limited by drought and nutrient scarcity in many parts of the world due to climate change (Parry et al., 2005; Parry and Hawkesford, 2010; FAO, 2012). It is therefore crucial to improve our understanding of processes that have the potential to increase the ability of plants to extract water and nutrients from the soil. Root traits, such as length and density of root hairs, have been proposed to increase plant productivity especially when soil resources are limited (Jungk, 2001; Bardgett et al., 2014; Pausch et al., 2016).

Root hairs are tubular extensions of epidermal cells (Peterson and Farquhar, 1996), which emerge right behind the zone of root elongation (Jungk, 2001). For cereals, approximately 20–90 root hairs emerge per millimetre of root length (Gahoonia et al., 1997) and they grow 0.2–1.0 mm into the soil (Schweiger et al., 1995; Haling et al., 2010; Brown et al., 2012). Root hairs contribute between 70 and 90 % of the total root surface area and there is evidence that they improve nutrient acquisition (Bates and Lynch, 1996; Gilroy and Jones, 2000; Brown et al., 2012). However, Jungk (2001) calculated that the increase in root surface due to root hair length alone could not explain increased nutrient influx into the root. The author proposed that other processes such as exudation of organic substances by root hairs might additionally increase nutrient availability.

Little is known about the role of hairs in root exudation. Root exudates were defined as materials released from the roots such as simple sugars, amino acids or polysaccharides but did not include dying root cells or root hairs (Leinweber et al., 2008). Although it has been shown that root hairs are covered by mucigels (Dart, 1971; Greaves and Darbyshire, 1972; Sprent, 1975), it is not clear whether the observed materials were released by the root hairs or just transported there simply as a result of root elongation. Moreover, there is nearly no information as to whether root hairs exude other substances apart from the observed mucigels such as low molecular weight exudates. Pausch et al. (2016) studied the impact of root hairs on rhizosphere priming effects (RPEs). They found that the presence of root hairs increased RPEs at least for young plants, hinting to an increase in rhizodeposition. However, the contribution of root hairs to the total root exudation and to the spatial distribution of exudates in the rhizosphere remains unclear.

The aim of this study was to test how root hairs affect the carbon allocation in the soil–plant system and root exudation. Barley [Hordeum vulgare ‘Pallas’ – wild type (WT)] and its root-hairless mutant (brb) were grown for 4 weeks and labelled with ¹⁴CO₂ to trace the carbon allocation in plant and soil. ¹⁴C imaging was used to quantify the effect of root hairs on exudation and rhizosphere extension. We hypothesized that: (1) the presence of root hairs increases the amount of ¹⁴C allocated to the roots (relative to the total assimilated ¹⁴C) because carbon is needed for root hair production and maintenance; (2) root exudation is greater in the wild type because of the presence of root hairs; and (3) direct exudation from root hairs increases the radial and axial (longitudinal) rhizosphere extension.

MATERIAL AND METHODS

Sample preparation and plant growth

Before germination, barley seeds (wild type: WT, bald root barley: brb) were immersed in a 10 % H₂O₂ solution for 10 min to avoid seed-borne diseases. Seedlings were planted after 3 d of germination.

The growth of plants was staggered over time for all experiments: three to four plants were grown per sowing time. This was done because the exemption limit for radioactive substances allows the handling of only limited amounts of ¹⁴C at a time in a laboratory. Furthermore, the chamber for labelling could not fit more than a maximum of five plants. The sowing dates were between April and June 2016. The two treatments (WT, brb) were randomly distributed within and between the sowing times. The barley plants were grown in rhizoboxes with a size of 12 × 20 × 3.5 cm for 30–31 d (Fig. 1). The soil used was a sandy soil (A-horizon) collected from a field site close to Reinhausen (Göttingen, Germany). C_tot was 2.0 % and N_tot was 0.17 % and pH was 4.9. The soil texture was distributed as follows: clay: 8.6 %; silt: 18.5 %; sand: 73 %. The volumetric soil water content was kept at 23–25 % vol. water content during plant growth. The temperature in the climate chamber was 25 °C during the day and 22 °C during the night, the photoperiod was 14 h and the light intensity was 200 μmol m^–2 s^–1. During the growth period, photographs of the root system were taken at regular intervals to monitor root elongation.

Fig. 1.

(A,B) Close up of the root system of barley plants after 4 weeks of growth: (A) wild type with root hairs (WT); (B) mutant without root hairs (brb). (C,D) Rhizosheaths around barley roots after taking them out of the soil and gently shaking them: (C) wild type with root hairs (WT); (D) mutant without root hairs (brb).

¹⁴CO₂ pulse labelling and CO₂ measurement

After 4 weeks of plant growth each plant was labelled with 0.5 MBq ¹⁴CO₂ (specific activity of 59.6 mCi mmol C⁻¹) for 4.5 h. Labelling was always conducted at noon for plants of all sowing times. The labelling technique has been described in detail elsewhere (Kuzyakov et al., 2006). At the end of labelling, approx. 70 % of the added ¹⁴C had been taken up by the plants. This was tested by collecting 30 mL of chamber air with a syringe four times during labelling (after 5, 30, 60 and 120 min) and injecting it into a scintillation cocktail (C 400; Zinsser Analytics).¹⁴C activity was quantified by a liquid scintillation counter (300 SL; Hidex). The activity of ¹⁴C in the 30 mL of chamber air that had been taken out of the chamber was back-calculated to the volume of the whole chamber. Doing this for all four measuring times, the amount of ¹⁴C in the chamber over time was calculated, which is inversely proportional to the uptake of ¹⁴C.

For the measurement of ¹⁴C allocation in the soil–plant systems (Exp. 1) five replicates of each barley type (WT and brb) were used and four replicates for the measurement of ¹⁴C respiration over time (Exp. 2). For ¹⁴C imaging of roots and exudates three replicates were used (Exp. 3). Immediately after labelling, the rhizoboxes used for the first and second experiments were packed in a plastic bag which was closed with modelling clay at the lower part of the stem of the plants. Inside the bag a 20-mL 1 m NaOH trap was placed to trap the ¹⁴CO₂ released from soil.

For the first experiment the trap was left inside the bag for 24 h. The trap was then removed and a scintillation cocktail (Eco Plus) was added to NaOH at a ratio of 4:1. ¹⁴C activity in NaOH was determined using a liquid scintillation counter (300 SL; Hidex). Shoots were cut and dried at 40 °C. The roots were taken out of the soil and the soil attached to the roots after being gently shaken (rhizosheath) was collected. Roots were dried at 40 °C. Rhizosheath and bulk rhizosphere soil (the soil not adhering to the roots) were freeze dried to avoid microbial degradation of labile carbon compounds. The terminology bulk rhizosphere soil was chosen as the root density in the rhizoboxes was high so that the whole soil volume is assumed to be affected by the activity of the roots. However, it is important to bear in mind that the so defined bulk rhizosphere also contains some small roots and root hairs because it is technically impossible to remove all the roots from the soil after harvesting the plants. Care must therefore be taken when interpreting the results on ¹⁴C in bulk rhizosphere soil as a considerable amount of ¹⁴C might be caused by ¹⁴C in root debris. The biomass of the dried plant samples was determined gravimetrically. To measure ¹⁴C activity, ground shoots, roots, rhizosheath and bulk rhizospere soil samples were combusted in an Oxymat OX500. The released ¹⁴CO₂ was captured in a scintillation cocktail (C400; Zinsser) and quantified using a liquid scintillation analyser (Tricarb, 3180; PerkinElmer). Total CO₂ respiration from soil was measured from a subsample (1 mL) of the NaOH trap: The carbonate in the NaOH solution was precipitated by barium chloride and the trapped CO₂-C was determined by back titration with 0.05 m HCl.

In the second experiment (Exp. 2), ¹⁴CO₂ production in soil was measured over 17 d. The NaOH traps were exchanged after 1, 2, 4, 9 and 17 d and ¹⁴C in the traps was measured as described above. After 17 d, plant biomass and ¹⁴C in plant and soil material was determined as described above.

For the third experiment (Exp. 3), rhizoboxes were opened immediately before labelling and a moist filter paper (Whatman, 1001–917, 11 µm) was attached to the root surface to capture root exudates (Dennis et al., 2010). Note that although root exudates were defined in this study as materials released from the roots such as simple sugars, amino acids or polysaccharides, it cannot be excluded that sloughed off root cells or the contents of damaged root hairs are also contained in the material captured in the filter paper. As the mechanical stress at the rhizobox surface is rather low, it was assumed that those compounds make up only a small portion of ¹⁴C captured at the filter paper and that the ¹⁴C in the filter paper is mainly due to root exudates. The presence of root hairs on the filter paper was checked with a microscope. The filter paper was covered by a thin plastic film to avoid drying of the filter paper. Finally, a thin layer of foam material was placed between the plastic film and the cover of the rhizobox to achieve good contact between soil surface and filter paper. Then, 18 h after labelling, the filter paper was carefully removed from the soil surface and dried in an oven (40 °C) to avoid the decomposition of root exudates captured in the filter paper. ¹⁴C imaging was conducted by placing an imaging plate (Storage phosphor screen, BAS-IP MS 2040 E; VWR) both on the filter paper and on the rooted soil surfaces of each rhizobox. A thin plastic film was placed between the sample (or filter paper) and the imaging screen to protect the imaging screen against the moist soil. The screens were attached to the roots (or filter paper) for 15 h. After this time the screens were scanned (FLA 5100 scanner; Fujifilm) with a spatial resolution of 50 µm.

Quantification of root elongation and image analysis

During the growth period, photographs of the root system were taken at regular intervals to monitor root elongation. For each plant, photographs taken 1 d before labelling, on the day of labelling and 1 d after labelling were analysed. The elongation rate was calculated using the Smart Root plugin (Lobet et al., 2011) in ImageJ (https://fiji.sc/): Roots from each image were segmented and the length of the segment was calculated. Root elongation rate was calculated based on the changes in root length over time.

For quantification of ¹⁴C images, the images were converted from a log into a linear system by applying the following equation:

$$\text{PSL}={\left(\frac{\text{Res}}{100}\right)}^2\text{* }\frac{4000}{S}\text{* }{10}^{\left(\frac{QL}{G}-\frac{1}{2}\right)}$$ (1)

where PSL (photostimulated luminescence) is the quantified value of the image in linear scale, Res is the resolution of the image (µm; Res = 50 µm), S is the sensitivity (S = 5000), L is the latitude (L = 5) and G is the gradation (G = 65 535). After conversion of the images, the background noise was removed: the part of the image where the screen was not in contact with the sample was selected and subtracted from the part of the image where the root system was visible. Based on the contrast between roots and soil/filter paper the root tips that showed a high ¹⁴C signal in the images were segmented using the SmartRoot plugin in ImageJ. In this way only the roots that were at the sample surface were selected. Possible artefacts caused by overlapping roots or roots detached from the soil were removed from the segmentation. Two to five roots per sample were segmented depending on how many roots were clearly visible. The signal was averaged as a function of distance from the root centre up to a distance of 4 cm from the root surface using the Euclidean distance mapping functions in MATLAB (The MathWorks). To quantify total exudation, the PSL values around each root tip were summed up and a mean of total exudation per tip and treatment was calculated.

Statistical analysis

The significances of differences between treatments (WT, brb) were tested using R 3.3.1. After testing for normal distribution and homogeneity of variances, a one-way ANOVA was conducted followed by a post hoc test (Tukey test). The level of significance was α = 0.05. To test for significances in radial rhizosphere extension, total exudation and the ratio between axial rhizosphere extension and root elongation between treatments, a mixed effects model (α = 0.05) with treatment as fixed effect and plant as random effect was applied. To account for the differences in numbers of roots sampled per plant, the restricted maximum likelihood (REML) method was applied.

RESULTS

Plant biomass, ¹⁴C recovery and total CO₂ efflux

Shoot biomass measured in the first experiment was similar for plants with and without root hairs. Root biomass was three times greater in plants with root hairs compared to the hairless mutants and the rhizosheath was ten times greater in WT compared to the hairless mutants. The specific ¹⁴C activity of shoots was similar in both plants, while it was 22 % greater in the roots and 80 % greater in the rhizosheath of brb plants compared to WT plants (Table 1). Similarly, total ¹⁴C in shoots did not differ between the genotypes, while total ¹⁴C in root and rhizosheath soil was greater in WT compared to brb. ¹⁴C activity in CO₂ as well as the total CO₂ respiration from soil was similar for the wild type and the hairless mutant.

Table 1.

Shoot, root and rhizosheath dry weight and specific and total ¹⁴C activity for the barley wild type (WT) and the mutant without root hairs (brb) 1 d after labelling

	WT	brb	P
Dry weight shoot (g)	0.37 ± 0.07	0.31 ± 0.06	n.s.
Dry weight root (g)	0.13 ± 0.03	0.05 ± 0.01	*
Dry weight rhizosheath (g)	5.75 ± 0.83	0.49 ± 0.09	*
14C in shoot (kBq g−1)	492 ± 32.3	494 ± 39.2	n.s.
14C in root (kBq g−1)	273 ± 29.9	333 ± 29.2	*
14C in rhizosheath (kBq g−1)	0.55 ± 0.1	0.99 ± 0.16	***
14C in bulk rhizosphere (kBq g−1)	0.03 ± 0.007	0.01 ± 0.002	*
Total 14C in shoot (kBq)	173 ± 27.1	143 ± 16.9	n.s.
Total 14C in root (kBq)	37.5 ± 9.64	16.2 ± 2.38	(*)
Total 14C in rhizosheath (kBq)	3.43 ± 1.00	0.45 ± 0.06	*
Total 14C in bulk rhizosphere (kBq)	30.0 ± 6.19	11.3 ± 2.19	*
14C in CO2 (kBq)	3.68 ± 0.46	2.90 ± 0.39	n.s.
CO2 (mg C kg−1 soil d−1)	41.1 ± 9.74	33.8 ± 5.45	n.s.

Variation is given as standard error. n = 5. n.s.: P ≥ 0.1, (*)P < 0.1, *P < 0.05, **P < 0.01, ***P < 0.001.

¹⁴C recovery was calculated as percentage of total ¹⁴C measured in all pools 1 d after labelling. On average, 70 % of the ¹⁴C was recovered in shoots, 10 % in roots, 1 % in rhizosheath, 9 % in bulk rhizosphere soil and 10 % in CO₂. Plants without root hairs allocated relatively more ¹⁴C to shoots compared to plants with root hairs. In contrast, WT allocated twice as much ¹⁴C into roots compared to brb and allocated five times more ¹⁴C into rhizosheath (Fig. 2).

Fig. 2.

¹⁴C recovered in shoots, roots, rhizosphere, bulk soil and CO₂ of the barley wild type (WT) and the mutant without root hairs (brb) 1 d after labelling. Variation is given as standard error. The number of replicates was 5. *P < 0.05.

For both plant genotypes, similar amounts of ¹⁴C were found in the respired ¹⁴CO₂. The cumulative ¹⁴C respiration over the sampling period 17 d after labelling was measured in Exp. 2 and was similar for WT and brb (Fig. 3). After 17 d, 21 % of ¹⁴CO₂ was recovered for plants with root hairs and 19 % for plants without root hairs.

Fig. 3.

¹⁴C recovery (% ± s.e.) in CO₂ from soil and root respiration over a period of 17 d. Differences between treatments were not significant (P < 0.05).

¹⁴C imaging and rhizodeposition

Root hairs were clearly visible along all the roots of WT plants grown in rhizoboxes (Fig. 1A, B). Root hairs reached a length of up to approx. 1 mm and were in general longer around older root parts (visual, qualitative impression). Root hairs favoured rhizosheath formation while nearly no soil was attached to the roots for the mutant without root hairs after their removal from soil (Fig. 1C, D). ¹⁴C was allocated to growing root tips of main and lateral roots (Fig. 4A, B). The region of roots where ¹⁴C was allocated correlated well with the region where ¹⁴C was found on the filter paper (Fig. 4C, D). The radial profiles of ¹⁴C activity around roots imaged directly on the soil samples were similar for plants with and without root hairs (Fig. 5A). Activity decreased to zero at a distance of 1 mm from the root centre. ¹⁴C activity on filter paper (i.e. root exudates) at the location of the root centre was approximately three times lower compared to that of roots in soil for WT plants and was approximately ten times lower for the hairless mutant (Fig. 5B). This is because only a small portion of the ¹⁴C taken up by roots is exuded as exudates. Plants with root hairs showed broader profiles of exudates compared to plants without root hairs. ¹⁴C activity at the location of the root centre was on average around 2 PSL for WT plants and 0.5 PSL for plants without root hairs. For plants with root hairs, ¹⁴C activity decreased to zero at a distance of 1.5 mm from the root centre, while for the mutant without hairs it decreased to zero at a distance of 0.5 mm from the root centre (Fig. 5B).

Fig. 4.

¹⁴C phosphor images of the root system of the barley plants (A,B) and of the root exudates that diffused into the filter paper (C,D) which was attached to the plant during labelling. The intensity of the dark colour corresponds to ¹⁴C activity.

Fig. 5.

Radial profiles around the roots in the soil (A) and of ¹⁴C activity (i.e. root exudates) on filter paper (B). The data sets were fitted using a linear model (solid lines). Confidence intervals are shown as dashed lines. Differences between the profiles were significant for the profiles on filter paper (B) but not for those of the roots in soil (A). Two to five roots were analysed from each of three replicate plants.

To separate the effect of different root elongation rates between genotypes on the axial rhizosphere extension, the ratio between axial rhizosphere extension and root elongation was calculated. This ratio was around 2 for plants with root hairs and around 1 for plants without root hairs (Fig. 6A). This means that the presence of root hairs extended the zone of exudation to slightly older root segments. Root hairs caused a three-fold increase in total exudation on filter paper compared to plants without root hairs (Fig. 6B).

Fig. 6.

(A) Ratio between axial rhizosphere extension and root elongation for the barley wild type (WT) and the mutant without root hairs (brb). Variation is given as standard error (n = 3). *P < 0.05. (B) Total exudation of barley plants with root hairs (WT) and without root hairs (brb) calculated based on the ¹⁴C activity on the filter paper, which was attached to the roots and soil in the rhizobox during labelling. Variation is given as standard error (n = 3). **P < 0.01.

DISCUSSION

Root exudates can increase nutrient availability in soil. On the one hand, they directly improve nutrient acquisition by mobilization of nutrients such as phosphorus, iron and micronutrients (Hinsinger, 2001; Dakora and Phillips, 2002; Lynch, 2007; Marschner et al., 2011), particularly in nutrient-poor soils. On the other hand, they strongly affect soil microbial activity and turnover of microbial biomass (Bertin et al., 2003; Gunina and Kuzyakov, 2015). Several studies have shown that nutrient availability increases due to higher microbial activity in the rhizosphere compared to the bulk soil (Hamilton and Frank, 2001; Herman et al., 2006; Landi et al., 2006). It is likely that increased rhizodeposition by plants with root hairs increases nutrient availability under nutrient-limiting conditions. Apart from the increase in total exudation for WT plants, the increase in radial rhizosphere extension is of particular importance because it expands the volume of soil where root exudates can interact with the soil matrix and with microorganisms.

Root hairs also increased the axial extension of the rhizosphere as indicated in Fig. 7. Commonly, root exudation is highest at the root tip and immediately behind the tip (Jones et al., 2009). Conceptual models of plant–microbe interactions along the root axis assume that root exudation and microbial activity are high in the immediate vicinity of the root tip where nutrients are immobilized by microorganisms (Kuzyakov and Xu, 2013). Behind the root tip, in the root hair zone, exudation is assumed to be low and microbes start to starve so that nutrients are mobilized and can be taken up by the plants (Marschner et al., 2011; Kuzyakov and Xu, 2013). We found that plants with root hairs increase axial rhizosphere extension compared to hairless plants. This indicates that the root hairs themselves exude organic substances. We conclude that root exudation is high over a longer distance from the root tip compared to what is commonly expected. For models of nutrient uptake this would imply that nutrients are mobilized farther behind the root tip than so far assumed.

Fig. 7.

Distribution of root exudates in the radial and axial direction for plants with root hairs (right) and the mutant without root hairs (left).

¹⁴C imaging was used to estimate rhizosphere extension and total exudation on filter paper. Based on the distribution of ¹⁴C on filter paper, root exudation was approximately three times greater for the wild type compared to the hairless mutant. These differences are smaller than the eight times difference in the total carbon found in the rhizosheaths (Table 1). This great difference can be explained by the ten times larger rhizosheath mass for WT plants, which confirms former experiments with the same genotypes (Haling et al., 2010). Indeed, a good part of the organic material might move farther away from the root surface and still contribute to rhizosphere processes, which is likely to occur for plants without root hairs. ¹⁴C allocation in the rhizosheath should therefore not be confounded with ¹⁴C allocation in the rhizosphere. A more extensive discussion on the differences between rhizosheath and rhizosphere terminology is given by York et al. (2016).

¹⁴C imaging allows estimation of the spatial distribution of C in the soil–plant system. However, care must be taken when interpreting the images in terms of root exudation. When calculating the profiles of ¹⁴C as a function of distance from the root centre for images taken from the soil–root surface (i.e. intact samples with roots in the soil), no differences were found between the treatments (Fig. 5A). In contrast, significant differences between treatments where found when analysing the spatial distribution of ¹⁴C on filter paper. Possibly, the ¹⁴C signal originating from the roots in the intact samples is so strong that it overshadows the signal from root exudates. This is particularly problematic if the soil–root surface is not perfectly flat (which normally is the case). In this case small gaps are present between the soil–root surface and the imaging screen and the β^– electrons from ¹⁴C decay from the roots travel in any direction through air, where they are barely attenuated. Consequently, the β^– signal hits the screen at a position farther from the root, from where it does not originate. Using filter paper those two sources of errors were avoided because: (1) the root signal is excluded from the image; and (2) the surface of the paper is perfectly flat and air gaps between the surface and the screen can be excluded.

So far, rhizodeposition from root hairs has been observed for specific compounds such as mucigels (Dart, 1971; Greaves and Darbyshire, 1972; Sprent, 1975), acid compounds (Yan et al., 2004) and sorgoleone (Czarnota et al., 2003; Dayan et al., 2009). Sorgoleone secretion is specific for Sorghum plants and was seldom observed at root hairs of other graminaceous plants (Weston et al., 2012). There is less information on the effect of root hairs on total rhizodeposition. Pausch et al. (2016) quantified the effect of roots hairs on RPEs. Priming was increased for plants with root hairs at tillering stage compared to plants without root hairs which showed negative RPEs. Because plants had similar root biomass the authors concluded that the presence of root hairs may explain differences in RPEs. Possibly priming was increased in WT plants because the extension of the rhizosphere by root hairs accelerated soil organic matter decomposition (Pausch et al., 2016). In the present study, barley plants with root hairs exuded significantly more C than hairless mutants, which suggests that exudates are not only released from the tips of main and lateral roots but also from root hairs. The increase in RPEs found by Pausch et al. (2016) may therefore be explained not only by a shift in microbial utilization of exudates but also by an increase in total exudation for plants with root hairs.

The images showing increased root exudation for WT plants fitted well with the data on ¹⁴C recovery: plants with root hairs allocated more carbon to roots and rhizosheath soil while plants without root hairs allocated relatively more C to shoots. For both plant types, ¹⁴CO₂ efflux from soil was similar although root exudation was increased for WT plants. It follows that more C derived from roots was retained in soil for WT plants compared to plants without root hairs. There are two possible explanations for this observation: (1) roots hairs decrease the local soil water content in the rhizosphere, as shown by Segal et al. (2008), which might result in a slower decomposition of ¹⁴C-labelled compounds in the rhizosphere (Sanaullah et al., 2012), thus reducing ¹⁴CO₂ efflux from soil; and (2) alternatively, or additionally, the metabolic quotient (qCO₂) of microorganisms in the rhizosphere of WT plants decreased, i.e. microorganisms used C more efficiently for the buildup of biomass and respired relatively less C compared to the brb plants. As plants with root hairs allocated more C below ground, while ¹⁴C efflux as well as total CO₂ efflux remained unchanged, more C derived from roots remained in the soil for WT plants. Considering that the mean residence time of root-derived C in soils is 2.4 times greater than that of shoot-derived C (Rasse et al., 2005), root hairs may play a significant role in soil C sequestration.

CONCLUSIONS

Plants with root hairs exuded significantly more carbon into the soil than plants without hairs and extended the rhizosphere in radial and axial directions (Fig. 7). The higher exudation and the increased rhizosphere extension might be an advantage for plants with root hairs because both favour plant–microbial interactions and therefore nutrient mobilization in the rhizosphere. Barley with root hairs allocated more C below ground compared to plants without hairs, but this did not increase CO₂ efflux. As root carbon has a longer mean residence time in soil compared to shoot carbon, it is likely that plants with root hairs foster C sequestration. Breeding for long root hairs and a high root hair density may be a suitable strategy for future agriculture, where nutrients are expected to become scarce and where C sequestration is a major issue due to climate change and resource depletion.