Inversions encounter relaxed genetic constraints and balance birth and death of TPS genes in Curcuma

Abstract

Evolutionary dynamics of inversion and its impact on biochemical traits are a puzzling question. Here, we show abundance of inversions in three Curcuma species (turmeric, hidden ginger and Siam tulip). Genes within inversions display higher long terminal repeat content and lower expression level compared with genomic background, suggesting inversions in Curcuma experience relaxed genetic constraints. It is corroborated by depletion of selected SNPs and enrichment of deleterious mutations in inversions detected among 56 Siam tulip cultivars. Functional verification of tandem duplicated terpene synthase (TPS) genes reveals that genes within inversions become pseudogenes, while genes outside retain catalytic function. Our findings suggest that inversions act as a counteracting force against tandem duplication in balancing birth and death of TPS genes and modulating terpenoid contents in Curcuma. This study provides an empirical example that inversions are likely not adaptive but affect biochemical traits.

Fig. 1. Phenotypic divergence and karyotype evolution among Curcuma species.

a Images of bract, floret, rhizome, and tuberous root of C. alismatifolia, C. petiolata, and C. longa. b Ancestral chromosomal reconstruction of Zingiberaceous species. Phylogeny was constructed based on 1791 single-copy orthologous genes. 11 pseudochromosomes of the common ancestor of Zingiberaceae were marked with different colors. It showed that modern species experienced at least dozens of chromosome fusion and fission events. The red dot represents the species of Zingiberaceae, the blue dot represents the species of Zingiberoideae, and the yellow dot represents the species of Curcuma. c A diagram showed the presence of inversions between haplotypes of C. petiolata and among the three Curcuma species. HapA: haplotype A; HapB: haplotype B. d Hi-C interaction signal validated an inversion between two haplotypes of Chromosome 20 in C. petiolata. e Percentage of the total length of inter-specific inversions between haplotypes from any two of the three Curcuma species.

Fig. 2. Pervasive inversions detected between haplotypes within species.

a Inversions and translocations between haplotypes within each of the three Curcuma species. b Percentages of total length of inversions and translocations among species with different ploidy levels. c Percentages of genes in inversions. INTRA: inversions between haplotypes within species.

Fig. 3. Genomic impacts of inversions in Curcuma species.

a Comparison of gene expression, nucleotide substitution rates, and LTR content between inversions and genome-wide background in the three Curcuma species. b Comparison of gene expression, nucleotide substitution rates, and LTR content between inversions and high LTR regions in C. alismatifolia. The red dashed line indicated the mean LTR content in inversions (window size = 500 Kbp). The red shaded area indicated the non-inverted region where the LTR content exceeded the mean LTR content in the INTRA. The rhombuses represented the mean. The left y axis of the density map corresponded to INTRA, and the right y axis corresponded to HLR. c Linear regression of gene expression levels (R² = 0.91, p = 0.001) among genomic regions when departing away from inversions in C. alismatifolia. d Linear regression of nucleotide substitution rates (R² = 0.84, p = 0.004) among genomic regions when departing away from inversions in C. alismatifolia. e Linear regression of LTR content (R² = 0.45, p = 0.102) among genomic regions when departing away from inversions in C. alismatifolia. a–e The box-plot elements were defined as: center line, median; box limits, upper and lower quartiles; whiskers, 1.5× interquartile range; points, outliers; rhombuses, mean. a, b Two-sided Wilcoxon rank-sum test was conducted for significance evaluation, and multiple comparisons were adjusted with the Bonferroni correction. Asterisks represented significant differences (***p < 0.001, adjusted, Supplementary Data 5). c–e The R² and p values were shown. FPKM fragments per kilobase of transcript per million fragments mapped, Ka nonsynonymous substitution rate, Ks synonymous substitution rate, LTR-gf The content of LTR in the 3 Kbp flanking regions of genes, INTRA inversions between haplotypes within species, INTER inter-specific inversions among the three species, POP inversions among populations of C. alismatifolia. BK genome-wide background. HLR high LTR regions. Source data are provided as a Source Data file.

Fig. 4. Inversions were depleted with selected SNPs in C. alismatifolia.

a Population structure of 56 C. alismatifolia cultivars at K = 2. g1 and g2 represented two groups with large differences in multiple traits. b Comparison of the proportion of selected SNPs (with 95% percentile as threshold for outliers; two-sided χ² test, ***p < 0.001, Supplementary Data 8) between inversions and non-inversions based on XP-CLR (top), nSL analysis (down). c The polymorphism information content (PIC) of multi-allelic SNPs (≥2 alleles per site) in genes. d Comparison of recombination rates among inversions, high LTR regions, and genome-wide background in C. alismatifolia. c, d The box-plot elements are defined as: center line, median; box limits, upper and lower quartiles; whiskers, 1.5× interquartile range; points, outliers; rhombuses, mean. e Comparison of percentage of deleterious mutations between inversions and genome-wide backgrounds in C. alismatifolia. Deleterious mutations of the top five longest inversions were also shown, where Chr08 stood for “CaChr8-TPS-TD-INV” (8-10517073-28214380), the longest inversion on C. alismatifolia chromosome 8. Additive: the additive mode with deleterious alleles in both homozygous and heterozygous genotypes; Recessive: the recessive mode with deleterious alleles only in homozygous genotypes. CDS: coding sequence. Box-plot elements are defined as: center line, median; box limits, upper and lower quartiles; whiskers, 1.5× interquartile range; points, all values, n = 56. INTRA inversions between haplotypes within species, NON-INV non-inversion region, BK genome-wide background, HLR high LTR regions. c–e two-sided Wilcoxon rank-sum test was employed for significance evaluation, and multiple comparisons were adjusted with the Bonferroni correction. Asterisks represented significant differences (***p < 0.001, adjusted, Supplementary Data 5). Source data are provided as a Source Data file.

Fig. 5. An inversion disabled function of tandemly duplicated TPS genes and impacted terpenoid biosynthesis.

a KEGG enrichment of genes within inversions in three Curcuma species and Z. officinale performed with TBtools. Enrichment of KEGG pathways was calculated with the Hypergeometric test. Statistical tests were one-sided and multiple comparisons were adjusted with the Benjamini-Hochberg correction (Q value). b Schematic diagram showing tandemly duplicated TPSs on chromosome 8 (Chr08) of C. alismatifolia. Genes outside of the inversion maintained the intact structure (dark gray rectangle), and genes within the inversion were pseudogenes (hatching chevron). Genes within inversions on different haplotypes lost different domains. ψ: pseudogene (gray font); CaChr8-TPS-TD-INV: Inversion on C. alismatifolia Chr08 with tandem duplicated TPS genes spanning its inversion breakpoint (8-10517073-28214380), shown as the intersecting dashed lines. Light and dark pink boxes indicated inverted regions on two haplotypes. c Compound 1 content detected by GC-MS and floret phenotype at three developmental stages of florets. The bar indicated 1 cm. Error bars represent the standard deviations (SD), and data are presented as mean values ± SD. Three biological replicates were performed. d GC-MS analysis of the main products formed by prokaryotic expression of germacrene synthases, compound 1 (m/z = 93), compound 3 (m/z = 93), and compound 4 (m/z = 93). Enzymes were incubated with compound 5. Reaction products were identified by standard chemicals and comparison of their mass spectra and retention indices with authentic standards and NIST libraries. Empty vector, pCold-TF. p.Gln64Glu, p.His124Leu, p.Ile228Leu, p.Met303Thr, p.Ile374Val, p.Val435Ala, and p.Val513Ile indicated seven different amino acid sites between Chr08HA727 and Chr08HA736 were mutated one by one accordingly. The direction of mutation is from Chr08HA727 to Chr08HA736. DDXXD deletion: DDXXD domain was deleted. Exon deletion (4): The sequence of exon 4 (containing the DDXXD domain) was deleted. All indicated assays were conducted for four or more repetitions. e Relative catalytic activity of germacrene synthase genes in producing compound 1. Box-plot elements are defined as: center line, median; box limits, upper and lower quartiles; whiskers, 1.5× interquartile range; points, all values. 5–6 biological replicates were performed for each treatment. f GC-MS analysis of the products from prokaryotic expression of Chr08HA727 proteins at different injection temperatures. Reaction products were identified by comparison of their mass spectra and retention indices with authentic standards and NIST libraries. Empty vector, pCold-TF. When the injection temperature was lowered to 150 °C, the content of compound 1 was decreased, and compound 2 was increased significantly (p < 0.01). g Catalytic scheme of germacrene synthase. The triangle represented the cope rearrangement of compound 2 to form compound 1 that easily occurred at high injection temperatures. c, e Statistical tests were two-sided Student’s t-test, and multiple comparisons were adjusted with the Bonferroni correction. Asterisks represented significant differences (*p < 0.05, **p < 0.01, ***p < 0.001, adjusted).

Fig. 6. The evolution of germacrene synthase genes in Zingiberaceous species.

a Annotated functions of TPS genes in Zingiberaceous species in the NR (Non-Redundant Protein Sequence) database revealed that the annotated germacrene synthase genes were expanded in Curcuma. b Collinearity of tandem duplicated genes of germacrene synthase (blue) in C. petiolata and C. longa. c Germacrene synthase tandem duplication genes spanning inversions on C. petiolata chromosome 14 (Chr14). Inversions between haplotypes were marked in red. Germacrene synthase genes in the inversion were labeled blue. Black dots denoted the genes with the complete gene structure. d GC-MS analysis of the products from prokaryotic expression of intact germacrene synthases in Zingiberoideae species.

Fig. 7. A schematic model depicting how inversions underwent relaxed genetic constraints and balanced birth and death of TPS genes in Curcuma species.

Inversions acted as a counteracting evolutionary force of gene tandem duplication to keep a balance between gene birth and death of TPS genes, and genomic characteristics associated with inversions indicated relaxed genetic constraints. Recombination within inversions between haplotypes was suppressed, resulting in inefficient removal of TEs and deleterious mutations. Higher levels of TE content, deleterious mutations, PICs, and base substitutions were observed within inversions. Genes within inversions were barely expressed, and inversion-internal genes on different haplotypes lost different domains, implying that these genes were pseudogenes and the two haplotypes evolved independently. And fewer selected SNPs were detected within inversions. Red and blue arrows indicated an increase and decrease within inversions relative to the genome-wide background, respectively.