Nicotine is the predominant alkaloid in tobacco plants, accounting for ~90% of their total alkaloid content. It is the main addictive substance in cigarettes. Reducing nicotine content in tobacco leaves will aid the development of low-nicotine tobacco products. Prior work has shown that the manipulation of genes involved in nicotine biosynthesis can achieve this purpose (Hidalgo Martinez et al., 2020). Here, we focussed on the long-sought major regulator of nicotine biosynthesis, NIC1 (A).

The NIC1 gene, together with a minor locus NIC2 (B), have been identified through genetic analysis of a low-nicotine trait originating from natural mutants of cigar tobacco (Legg and Collins, 1971). Introgression of the low-nicotine trait into Burley 21 (B21) generated near-isogenic lines with different alkaloid levels: high alkaloid (HA, AABB), high intermediate (HI, AAbb), low intermediate (LI, aaBB) and low alkaloid (LA, aabb) (Legg and Collins, 1971). The genes coding for nicotine biosynthetic enzymes, such as the rate-limiting PMT and QPT, are downregulated in LA, suggesting that NIC genes are transcriptional regulators orchestrating nicotine biosynthesis (Saunders and Bush, 1979). Transcriptome-based cloning of NIC2 revealed that this locus is clustered with transcription factors from the ethylene response factor (ERF) subfamily. Of these NIC2-ERFs, ERF189 is the most effective and directly targets the GC-rich P-box element in promoters of nicotine biosynthetic genes (Shoji and Hashimoto, 2012; Shoji et al., 2010). Suppression of NIC2-ERFs reduced nicotine content in tobacco, but a significant amount of nicotine remained due to the major NIC1 locus (Kajikawa et al., 2017).

To isolate the NIC1 gene, we conducted map-based cloning using 600 field-growing F₂s derived from a cross between HI and LA. The segregating population was first genotyped with a custom tobacco 30K Infinium iSelect HD BeadChip. Additional markers were designed through SNP identification based on RNA-seq of the B21 NILs. NIC1 congregated with SNP4 and was flanked by SNP3 and SNP5 on chromosome 7 (Figure 1a). The delimited NIC1 region was bordered by K326 scaffolds Nitab4.5_0003553 and Nitab4.5_0007027. Reciprocal BLAST comparisons were conducted between K326 and TN90 to fill gaps. Gene annotation identified at least seven full-length single-exon ERFs (JRE5L2, ERF199, ERF91, ERF210, ERF29, ERF16 and ERF130) and two truncated ERFs (ERF110 and ERF17L3ΔN) in this region (Figure 1a).

BLAST analysis showed that NIC1 and NIC2 regions were syntenic and originated, respectively, from N. sylvestris (S-genome) and N. tomentosiformis (T-genome). Notably, ERF199 (Nitab4.5_0003090g0030) is homologous to ERF189, sharing an identical binding domain (Figure 1b). Driven by their native promoters, the seven complete NIC1-ERFs were transferred to LA plants using Agrobacterium tumefaciens strain GV3101. Complementation tests showed that transfer of ERF199 (n = 28) significantly increased nicotine levels in potted T0 plants growing in greenhouse. No significant phenotype changes resulted from the remaining six NIC1-ERFs (n > 20) (Figure 1c). The ERF199-mediated nicotine increase was confirmed with T1 plants (n = 35 from 7 T0s) (Figure 1d). We, therefore, concluded that ERF199 is the NIC1 gene.

To determine subcellular localization, we fused ERF199 with a red fluorescent protein (RFP) under the CaMV 35S promoter. Constitutive expression of ERF199-RFP in LA significantly increased nicotine content in both T0 (n = 15) and T1 (n = 28) plants (Figure 1e, f). Co-localization of the fusion protein with 4′, 6-diamidino-2-phenylindole (DAPI)-stained nuclei demonstrated that ERF199 is localized in the nucleus, consistent with its role as a transcriptional regulator (Figure 1g). Direct binding of ERF199 to the P box in the PMT2 promoter was verified by electrophoretic mobility shift assay (EMSA). The resulted mobility shift was eliminated by the addition of excess non-labelled probe in the competition experiment, confirming the specificity of this DNA–protein interaction (Figure 1h). Furthermore, a transient gene expression assay using tobacco protoplasts revealed that ERF199 regulates both PMT2 and QPT, the two key enzymes for nicotine biosynthesis (Figure 1i, j).

We compared the genomic sequences of EFR199 alleles between HI and LA (~5kb), including 2.3kb upstream of the start codon and 2kb downstream of the stop codon, but no SNPs were detected. Expression analysis indicated that Nic1 was root-specific and significantly downregulated in LA (Figure 1k), suggesting the recessive allele was epigenetically silenced. However, targeted bisulphite sequencing of the same 5kb did not reveal significantly different DNA methylation in CpG, CHG and CHH sites. A thorough epigenome sequencing may provide insight into the underlying epimutations in the NIC1 locus.

We predicted that lower nicotine than LA could be attained by eliminating ERF199. Using CRISPR technology, we generated a mutated allele caused by an ‘A’ insertion in HI (Figure 1l) and evaluated its phenotypic effect at T1. The genotypes of wild-type (WT-T1), heterozygous (Het-T1) and homozygous mutant (Mut-T1) plants were determined by DNA sequencing. Total alkaloid levels in WT-T1 plants (n = 22) were comparable with HI plants, and Het-T1s (n = 29) had intermediate levels. However, alkaloid content was barely discernible in Mut-T1 plants (n = 35); approximately 1/10 of that in LA plants (Figure 1m). Thus, manipulation of the NIC1 gene provides a new strategy for nicotine control. Furthermore, the ultra-low-nicotine levels in Mut-T1 confirmed that ERF199 is the only causal gene for nicotine biosynthesis within the NIC1 locus.

Transcriptional regulation of secondary metabolite production can be controlled by a single ERF. Indeed, GAME9 locates within an ERF cluster and is the only functional regulator of steroidal glycoalkaloid biosynthesis in tomatoes (Cardenas et al., 2016). Although both ERF189 and ERF199 share the same DNA-binding domain and directly bind to P-box elements within the promoters, ERF199 is more effective. A plausible explanation may be the presence of cofactor/coactivator-recruiting activation domains. Further investigation is required to determine whether a unique activation domain or additional transcriptional cofactors play critical roles in the ERF199-regulatory network.

Significant endeavours have been made to attenuate tobacco nicotine content (Hidalgo Martinez et al., 2020). Our study genetically and functionally validated ERF199 as the NIC1 gene. The NIC1 locus, originating from the S-genome, is homologous to the T-genome-donated NIC2 locus. Constitutive expression of ERF199 caused increased nicotine levels, and disruption of ERF199 function in the absence of NIC2 dramatically reduced nicotine accumulation in leaves. Thus, genetic regulation of nicotine levels in tobacco plants can be achieved by manipulating the NIC1 gene.

尼古丁是烟草植物中的主要生物碱，约占其总生物碱含量的 90%。它是香烟中的主要成瘾物质。降低烟叶中的尼古丁含量将有助于开发低尼古丁烟草产品。先前的工作表明，对参与尼古丁生物合成的基因进行操作可以实现这一目的（Hidalgo Martinez等人，2020 年）。在这里，我们专注于长期寻找的尼古丁生物合成主要调节剂NIC1 ( A )。

通过对源自雪茄烟草天然突变体的低尼古丁性状的遗传分析，已经鉴定出NIC1基因和一个次要基因座NIC2 ( B )（Legg 和 Collins， 1971 年）。将低尼古丁性状渗入白肋烟 21 (B21) 产生了具有不同生物碱水平的近等基因品系：高生物碱 (HA, AABB )、高中间 (HI, AAbb )、低中间 (LI, aaBB ) 和低生物碱 ( LA, aabb）（Legg 和 Collins，1971 年）。编码尼古丁生物合成酶的基因，例如限速 PMT 和 QPT，在洛杉矶被下调，这表明NIC基因是协调尼古丁生物合成的转录调节因子（Saunders 和 Bush，1979 年）。基于转录组的NIC2克隆显示该基因座与乙烯反应因子 (ERF) 亚家族的转录因子聚集在一起。在这些NIC2- ERF 中，ERF189是最有效且直接靶向尼古丁生物合成基因启动子中富含 GC 的 P-box 元件（Shoji 和 Hashimoto，2012；Shoji等，2010）。抑制NIC2 - ERF降低了烟草中的尼古丁含量，但由于主要的NIC1 ，大量尼古丁仍然存在位点（Kajikawa等人，2017 年）。

为了分离NIC1基因，我们使用来自 HI 和 LA 杂交的600 个田间生长的 F _{2 s 进行了基于图谱的克隆。}首先使用定制的烟草 30K Infinium iSelect HD BeadChip 对隔离群体进行基因分型。其他标记是通过基于 B21 NIL 的 RNA-seq 的 SNP 鉴定设计的。NIC1与 SNP4 聚集，并在 7 号染色体上侧接 SNP3 和 SNP5（图 1a）。分隔的NIC1区域与 K326 支架 Nitab4.5_0003553 和 Nitab4.5_0007027 接壤。在 K326 和 TN90 之间进行了相互 BLAST 比较以填补空白。基因注释确定了至少 7 个全长单外显子 ERF（JRE5L2、ERF199、ERF91、ERF210、ERF29、ERF16和ERF130）以及该区域中的两个截短的 ERF（ERF110和ERF17L3ΔN）（图 1a）。

BLAST 分析表明NIC1和NIC2区域是同线的，并且分别来自于樟子松（S-基因组）和毛细棉（T-基因组）。值得注意的是，ERF199 (Nitab4.5_0003090g0030) 与 ERF189 同源，具有相同的结合域（图 1b）。在其天然启动子的驱动下，使用根癌农杆菌菌株 GV3101将七个完整的NIC1-ERF转移到洛杉矶植物中。互补试验表明，ERF199 ( n  = 28) 的转移显着增加了温室中生长的 T0 盆栽植物的尼古丁水平。其余六种没有引起显着的表型变化NIC1-ERF ( n  > 20) (图 1c)。T1 植物证实了ERF199介导的尼古丁增加（n  = 35，来自 7 T0）（图 1d）。因此，我们得出结论，ERF199是NIC1基因。

为了确定亚细胞定位，我们在 CaMV 35S 启动子下将 ERF199 与红色荧光蛋白 (RFP) 融合。LA中ERF199-RFP的组成型表达显着增加了T0（n  = 15）和T1（n  = 28）植物中的尼古丁含量（图1e，f）。融合蛋白与4'，6-二脒基-2-苯基吲哚（DAPI）染色的细胞核共定位证明ERF199定位于细胞核中，与其作为转录调节剂的作用一致（图1g）。ERF199 与PMT2中的 P 盒直接结合通过电泳迁移率变动分析（EMSA）验证启动子。通过在竞争实验中添加过量的非标记探针消除了由此产生的迁移率变化，证实了这种 DNA-蛋白质相互作用的特异性（图 1h）。此外，使用烟草原生质体进行的瞬时基因表达测定表明，ERF199 调节 PMT2 和 QPT，这两种关键酶用于尼古丁生物合成（图 1i，j）。

我们比较了HI 和 LA (~5kb) 之间EFR199等位基因的基因组序列，包括起始密码子上游 2.3kb 和终止密码子下游 2kb，但未检测到 SNP。表达分析表明Nic1是根特异性的并且在 LA 中显着下调（图 1k），表明隐性等位基因在表观遗传上被沉默。然而，相同 5kb 的亚硫酸氢盐靶向测序并未发现 CpG、CHG 和 CHH 位点的 DNA 甲基化存在显着差异。彻底的表观基因组测序可以深入了解NIC1基因座中的潜在表观突变。

我们预测通过消除ERF199 可以获得比 LA 更低的尼古丁。使用 CRISPR 技术，我们生成了由 HI 中的“A”插入引起的突变等位基因（图 1l），并评估了其在 T1 的表型效应。通过DNA测序确定野生型（WT-T1），杂合（Het-T1）和纯合突变体（Mut-T1）植物的基因型。WT-T1 植物（ n = 22）中的总生物碱水平 与 HI 植物相当，而 Het-T1s（n  = 29）具有中等水平。然而，在 Mut-T1 植物中几乎看不到生物碱含量（n  = 35）；大约是 LA 植物的 1/10（图 1m）。因此，对NIC1的操作基因为尼古丁控制提供了新的策略。此外，Mut-T1 中的超低尼古丁水平证实ERF199是NIC1基因座内尼古丁生物合成的唯一因果基因。

次生代谢产物的转录调控可以由单个ERF控制。事实上，GAME9位于ERF簇内，是番茄中甾体糖苷生物碱生物合成的唯一功能调节剂（Cardenas等人，2016 年）。尽管 ERF189 和 ERF199 共享相同的 DNA 结合域并直接与启动子内的 P-box 元件结合，但 ERF199 更有效。一个合理的解释可能是辅因子/辅激活因子招募激活域的存在。需要进一步调查以确定独特的激活域或额外的转录辅因子是否在 ERF199 调节网络中发挥关键作用。

为减少烟草尼古丁含量做出了重大努力（Hidalgo Martinez等人，2020 年）。我们的研究在基因和功能上验证了 ERF199是NIC1基因。源自 S 基因组的NIC1基因座与 T 基因组捐赠的NIC2基因座同源。ERF199的组成型表达导致尼古丁水平升高，而在没有NIC2的情况下，ERF199 功能的破坏显着降低了叶片中尼古丁的积累。因此，可以通过操纵NIC1基因来实现烟草植物中尼古丁水平的遗传调控。