
Breeding of Al-tolerant malting barley cultivar
To breed a barley cultivar with acid soil tolerance, we introduced the 1-kb insertion from Murasakimochi to an elite malting barley cultivar, Haruna Nijo, through multiple backcrossing (Supplementary Fig. 1). Haruna Nijo is a two-rowed hulled cultivar and was released in 1981 in Japan. It has high quality for malting and is also often used as a donor of high-quality profiles in Japanese malting barley breeding programs. However, this cultivar is very sensitive to Al toxicity, resulting in low production in acid soil areas. By contrast, Murasakimochi is an old Japanese waxy food barley landrace, which is especially used for making local cakes. The grain shows a purple color and is not suitable for malt production (Supplementary Fig. 2), but it carries 1-kb insertion upstream of HvAACT1 and shows high Al tolerance11.
We first generated recombinant chromosome substitution lines (RCSLs) from the cross between Murasakimochi (Al tolerance donor parent) and Haruna Nijo as a recurrent parent (Supplementary Fig. 1). We selected 83 BC3F3 lines to be genotyped by 1536 SNP markers derived from barley Expressed Sequence Tag (EST) sequences15. Of the 1536 markers, 691 showed polymorphism between Haruna Nijo and Murasakimochi (Supplementary Fig. 3). We used consensus genetic marker position15 to map EST-derived markers, which were closely linked to HvAACT19. Based on these genotyping data, we selected one isogenic line having the Murasakimochi allele on the HvAACT1 locus in the genetic background of Haruna Nijo (Fig. 1).
The gray color indicates the genome of Haruna Nijo, and the black color indicates the substituted segment and a locus of HvAACT1 from Murasakimochi on chromosome 4H.
Characterization of Al-tolerant malting barley cultivar
The BC4F3 line generated was used for evaluation of acid soil tolerance. The root growth of Haruna Nijo was inhibited more compared with Murasakimochi when grown on acid soil (Andosol) (Supplementary Fig. 4), but the BC4F3 line showed similar growth as Murasakimochi, indicating that introgression of 1-kb insertion from Murasakimochi into Haruna Nijo enhanced the acid soil tolerance.
We repeatedly self-pollinated this line and obtained BC4F10 (designated Haruna Nijo-AT (acid soil tolerance)). Haruna Nijo-AT showed similar grain size and color as Haruna Nijo, which are very different from those of Murasakimochi (Supplementary Fig. 2a). When grown in a field, there was no visible difference in spike phenotypes between Haruna Nijo-AT and Haruna Nijo at both grain filling stage and maturity stage (Supplementary Fig. 2b, c), but very different from Murasakimochi. We confirmed the presence of a 1-kb insertion in the genomic sequence of Haruna Nijo-AT (Fig. 2a). Expression analysis showed that the expression level of HvAACT1 was increased 30 times in the roots of Haruna Nijo-AT compared with its original cultivar (Fig. 2b).

a Detection of 1-kb insertion. DNA extracted from both Haruna Nijo and Haruna Nijo-AT was used as a temperate for PCR using specific primers. b Expression level of HvAACT1 in the roots. RNA was extracted from the roots of both Haruna Nijo and Haruna Nijo-AT grown hydroponically, and the expression level was quantified with real-time RT-PCR. Actin was used as an internal standard, and expression relative to the root of Haruna Nijo is shown. Data are shown as the means ± SD (n = 4) for (b). Significant differences are marked with **P < 0.01, by Student’s t-test.
Physiological characterization showed that Haruna Nijo-AT secreted more citrate from the roots than Haruna Nijo (Fig. 3a), being seven times more than the original cultivar. Consistent with this, Haruna Nijo-AT also accumulated less Al in the root tips (Fig. 3b). Al staining using Eriochrome cyanine showed that the root tip of Haruna Nijo was heavily stained (Fig. 3c), whereas that of Haruna Nijo-AT was much less stained. These results indicate that introgression of 1-kb insertion enhanced expression of HvAACT1 in Haruna Nijo-AT, resulting in more secretion of citrate from the root tips. Citrate is able to chelate toxic Al ions in the rhizosphere, resulting in external detoxification of Al. This is also supported by the finding that less Al was bound to the root tips of Haruna Nijo-AT (Fig. 3b, c).

a Citrate secretion from the roots. The root exudates were collected after exposure to 10 μM Al in 1.0 mM CaCl2 solution (pH 5.0) for 6 h. The concentration of citrate was enzymatically determined. b Al accumulation in the root tips. c Staining of Al in the root tips. Scale bar = 1 mm. The roots were exposed to a 10 μM Al in 1.0 mM CaCl2 solution (pH 5.0) for 24 h, and the root tips (0–1 cm) were excised for Al determination by ICP-MS (b) or subjected to staining by 0.1% eriochrome cyanine (c). Data are shown as the means ± SD (n = 3) for (a) and (n = 4) (b). Significant differences are marked with *P < 0.05; **P < 0.01, by Student’s t-test.
Field test in acid soil
We performed a field test in acidic (pH 4.9–5.0) and neutral soils (pH 6.5) in different years (2022–2023 and 2023–2024). In the field test during 2022–2023, there was no difference in the plant height, spike number, seed yield, and straw weight between Haruna Nijo-AT and Haruna Nijo when grown in neutral soil (Fig. 4). However, when grown in acidic soil, significant difference was observed; Haruna Nijo-AT showed higher plant height and more spike numbers than Haruna Nijo (Fig. 4b, c). Importantly, the seed yield of Haruna Nijo-AT was 3.3 times more than that of Haruna Nijo (Fig. 4d). Haruna Nijo-AT also produced more than 3 times the biomass of the straw compared with Haruna Nijo (Fig. 4e). To confirm these results, we performed the field test on acid soil again during 2023–2024. Similar to the results during 2022–2023, Haruna Nijo-AT produced two times more seed yield than Haruna Nijo (Supplementary Fig. 5). The plant height, spike number, and straw weight were also higher or more in Haruna Nijo-AT compared with the original cultivar. These results consistently support that introgression of the 1-kb is very effective in enhancing acid soil tolerance. The difference in seed yield and other traits between different years could be attributed to different climates and slight pH increases.

a, b Pictures of Haruna Nijo and Haruna Nijo-AT grown in acidic soil (a) and neutral soil (b) at harvest. Scale bar = 10 cm. c–f Comparison of plant height (c), spike number (d), seed yield (e), and straw weight (f) of Haruna Nijo and Haruna Nijo-AT grown in acidic and neutral soil at harvest. Both Haruna Nijo and Haruna Nijo-AT were grown in acidic (pH 4.9) and neutral (pH 6.5) soils for 6 months from November 2022 to May 2023 and harvested for analysis of growth parameters. Data are shown as the means ± SD (n = 15–18 from three plots) for (c–f). Different letters indicate significant differences (P < 0.05), by Tukey’s test.
Analysis of the mineral element profile showed that there was no difference in mineral element accumulation of seeds between Haruna Nijo-AT and Haruna Nijo grown on both acidic soil and neutral soil (Supplementary Table 1), although higher accumulation of micronutrients including Fe, Cu, Mn, and Zn was found in grain harvested from acidic soil than those from neutral soil. This could be attributed to increased bioavailability of these micronutrients in acidic soil.
Due to the high sensitivity of barley to acid soil, its expansion as a crop into many agricultural areas of the world has been limited3. Although liming can improve barley growth and productivity on acid soils, this practice is often not economically feasible16. Furthermore, surface application of lime cannot alleviate toxic subsoil Al, which presents a barrier to deep rooting and the uptake of water and nutrients. So far, some attempts have been made to breed tolerant barley cultivars by transgenic strategies and conventional breeding. For example, overexpression of the wheat TaALMT1 gene enhanced the Al tolerance of barley due to increased Al-activated malate efflux from the roots17. Overexpression of HvAACT1 in barley under the control of maize ubiquitin promotor also enhanced Al tolerance at the seedling stage18. A new barley variety, “Litmus” with higher Al tolerance, but blue aleurone, was developed based on a cross to the acid-tolerant line WB22919. However, no available cultivars have been released due to public acceptance of transgenic crops and the aleurone color problem, which affects the grain quality. By contrast, the cultivar (Haruna Nijo-AT) bred in this study showed similar grain color and mineral accumulation (Supplementary Fig. 2, Table S1). In the future, other barley cultivars with enhanced acid soil tolerance could also be generated by introgression of the 1-kb insertion through multiple backcrossing, although it takes several years to generate the near isogeneic lines. This approach will potentially provide a sustainable and economic way to boost barley productivity in areas with acid soil.