Phylogenomic Analysis of micro-RNA Involved in Juvenile to Flowering-Stage Transition in Photophilic Rice and Its Sister Species

Abstract

Vegetative to reproductive phase transition in phototropic plants is an important developmental process and is sequentially mediated by the expression of micro-RNA MIR172. To obtain insight into the evolution, adaptation, and function of MIR172 in photophilic rice and its wild relatives, we analyzed the genescape of a 100 kb segment harboring MIR172 homologs from 11 genomes. The expression analysis of MIR172 revealed its incremental accumulation from the 2-leaf to 10-leaf stage, with maximum expression coinciding with the flag-leaf stage in rice. Nonetheless, the microsynteny analysis of MIR172s revealed collinearity within the genus Oryza, but a loss of synteny was observed in (i) MIR172A in O. barthii (AA) and O. glaberima (AA); (ii) MIR172B in O. brachyantha (FF); and (iii) MIR172C in O. punctata (BB). Phylogenetic analysis of precursor sequences/region of MIR172 revealed a distinct tri-modal clade of evolution. The genomic information generated in this investigation through comparative analysis of MIRNA, suggests mature MIR172s to have evolved in a disruptive and conservative mode amongst all Oryza species with a common origin of descent. Further, the phylogenomic delineation provided an insight into the adaptation and molecular evolution of MIR172 to changing environmental conditions (biotic and abiotic) of phototropic rice through natural selection and the opportunity to harness untapped genomic regions from rice wild relatives (RWR).

1. Introduction

Micro-RNAs (MIRNAs) function as post-transcriptional regulators of gene expression in eukaryotes [1]. In plants under a given environmental condition, MIRNAs perform a host of regulatory functions, and one important regulation is phase transitions leading to plant morphogenesis and development [2,3,4,5]. Such phase transitions mark cardinal changes in plant development and are mediated by sequentially expressed MIRNAs [6]. Stage transitions are cardinal and necessary changes in the plant developmental processes, and these transitions are mediated by sequentially expressed miRNAs [6]. MIRNA172 is one such family of MIRNA that is involved in phase transition during plant development. Vegetative phase changes in Arabidopsis and maize are controlled by the sequential activity of miR156 and MIR172 [7]. Although miR156 is highly expressed during early developmental stages, MIR172 is highly expressed during later stages of development [7,8,9]. In plants such as Acacia confusa, A. colei, Hedera helix, Eucalyptus globulus, and Quercus acutissima, contrasting expression patterns of miR156 and MIR172 and their target genes were observed [10].

Inter alia, MIR172 is one such family of MIRNA that is sequentially involved in the vegetative to reproductive phase transition in Arabidopsis and maize [7]. MIR172 is highly expressed during later stages of plant development [7,8,9] and, functionally, one of the first plant MIRNA to be identified in Arabidopsis [11]. In most plants, it is 22-nucleotide long and is highly conserved. However, the number of precursor-MIR172, mature MIR172, and the target genes exhibit diversity in different plant species. In rice, two mature MIR172s are encoded by four precursor-MIR172s with several homologs (similar due to descent from a common ancestor) and homeologs (homologs resulting from allopolyploidy).

Functionally, MIR172 is ubiquitous and one of the earliest plant miRNA genes to be identified by small-RNA cloning and sequencing in Arabidopsis [11]. In Arabidopsis, five pri-MIR172s encode three mature MIR172s, and these MIR172s repress the expression of six members of the APETALA 2 (AP2)-like family of transcription factors, three TARGET OF EAT (TOE) proteins, and SCHLAFMUTZE (SMZ) and its paralog SCHNARCHZAPFEN (SNZ) [12]. In maize, five pri-MIR172s encode only one mature MIR172, which represses the expression of six members of the AP2-like family of transcription factors and TS6-GN2230 [13]. In rice, four pri-MIR172s encode two mature MIR172s and repress five members of the AP2-like transcription factors [13,14]. In Populus trichocarpa, nine pri-MIR172s encode four mature MIR172s with six target genes [15].

Beyond its cardinal role in the transition from the vegetative to the reproductive stage, MIR172 is also involved in a repertoire of developmental processes in plants, such as the determination of stem cell fate [16], developmental timing [12,17,18,19,20], sex determination [21], flowering [22], fruit growth [23,24], spike architecture and grain threshability in wheat [25,26], tuberization in potatoes [27], and nodulation in soybeans [28]. Response to abiotic stress in Arabidopsis [29] and biotic stress resistance in tomatoes [30] are also affected by MIR172 expression. Recently it was observed that auxins also modulate MIR172 [1] activity to affect plant morphogenesis.

Grain production in field crops has doubled in the last decades due to the “Green revolution” [31], along with advances in functional [32,33,34,35] and structural [36,37,38,39,40,41,42] genomics. Further, marker-assisted back-cross breeding (MABB) imparted biotic [43,44,45,46] and abiotic [47,48,49,50] stress tolerance, ultimately leading to higher grain yield. However, we still need to improve food production to maintain the demand-supply balance. In this regard, wild relatives of a crop are more robust genomic sources of traits/genes than their domesticated counterparts in terms of disease tolerance and plant architecture [5]. Agriculturally important crops [47,48,49] are positively photophilic and phototropic and are constantly exposed to changing environmental conditions during their lifetime. Environmental stresses, such as biotic and abiotic stresses, cause adverse effects on growth, development, and yield. However, their wild relatives are reservoirs of untapped genomic resources to be utilized to develop/breed improved cultivars to combat biotic/abiotic threats posed by climate change [51]. Rice is the world’s second most-produced [52] staple food crop, and its evolution from grasses has had an immense impact on human civilization catering to daily nutrition [53] globally. It belongs to the genus Oryza (family poaceae), which comprises 24 species, among which Oryza sativa L. and Oryza glaberrima S. are cultivated, and the remaining species are wild relatives spread all over the world [54]. Wild Oryza species or relatives (RWRs) are adapted to diverse habitats and are more robust compared to cultivated varieties in terms of stress tolerance [55]. RWRs are a rich resource of unexplored genes for breeding improved cultivated rice to combat the threats posed by increasing population and climate change [51]. Previous studies targeting the regulation/expression of genes targeted by MIR172 in poaceae have been accomplished, but studies focusing on evolution, content, genetic diversity, and structure of genes embedded in the MIRNA172 locus per se are lacking. In this endeavor, we employed a comparative genomics approach to identify homologs of MIRNA172 in rice and its wild relatives to gain insight into the expression and evolution of the region in and around MIR172. Our study revealed the conservative and disruptive evolution of the MIR172 gene family in various cultivated/wild grasses and supplanted genomic information to be harnessed from RWR to design a designer crop tolerant to abiotic/biotic stresses.

2. Materials and Methods

2.1. Sequence Retrieval of MIR172 Homologs

MiRBase 22.1 (http://www.mirbase.org/, accessed on 10 March 2021) was used to retrieve the mature and precursor sequences of MIR172 from Oryza sativa, Sorghum bicolor, Zea mays, Triticum aestivum, and Arabidopsis thaliana [56]. Oryza sativa MIRNA precursor sequences were used as the query to execute BLASTN (Local BLAST) against the genome sequences of O. glaberrima, O. glumaepatula, O. rufipogon, O. punctata, O. barthii, and O. brachyantha that are available in the Gramene to extract each plant’s precursor sequence, the MIRNA precursor sequences of sorghum, maize, and Arabidopsis were utilized as the query. Based on the score, e-value (lowest), and percentage identity (highest), high-scoring pairings (HSP) for MIR172 were found. These HSP were then chosen for comparative genomic study in order to comprehend the microsynteny, organizational structure, and evolutionary trend of MIR172 in domesticated and wild grasses. As an outlier for determining the evolutionary trend, the Arabidopsis MIR172 sequence from TAIR (https://www.arabidopsis.org/, accessed on 10 March 2021) was used.

2.2. MIRNA Precursor Sequence Analysis

Using the MAFFT version 7.271 program [57] with the L-INS-I approach and output in Phylip format, multiple sequence alignment of MIR172 precursor and mature sequences was carried out in seven Oryza spp., maize, sorghum, and Arabidopsis. The aligned sequences were scored for similarity using ESPRIPT 3.0 [58] (https://espript.ibcp.fr/ESPript/ESPript/, accessed on 12 March 2023) with the default settings.

2.3. Microsynteny Analysis

O. sativa, O. rufipogon, O. glaberrima, O. barthii, O. glumaepatula, O. brachyantha, O. punctata, Zea mays, and Sorghum bicolor genomic sequences were obtained from Gramene database, and Arabidopsis thaliana genomic sequences were obtained from TAIR (https://www.arabidopsis.org/, accessed on 15 March 2021) (Supplementary Table S1). The reference plant for the microsynteny investigation was Oryza sativa. Genes were predicted using the FGENESH tool from Molquest II (http://www.molquest.com/molquest.phtml?group=index&topic=gfind, accessed on 18 March 2021) [59] with the default parameters. For all rice species, Oryza sativa was used as the default template, and the genomes of wheat, maize, and sorghum were employed as templates for their corresponding genomes. Blast2GO software (version 5.2.5) (https://www.blast2go.com/, accessed on 19 March 2021) was used to perform functional annotation and genomics analysis on genes predicted by FGENESH in the 100 kb area [60]. Functionally annotated genes from the 100 kb region of MIRNAs were enlisted from each species and used for microsynteny analysis using BLAST2GO analysis.

The database for microsynteny analysis, as well as the blastp for calculating synteny block input, were created using the NCBI BLAST-2.11.1+ software (makeblastdb and blastp). The query for the makeblastdb script comprised four sets of proteins (one set per MIR172A, B, C, and D) predicted by the MolQuest program in 100 kb genomic regions of 10 genomes under study in the previous stage. The blastp settings were blastp-outfmt 8-e-value 1 × 10⁻¹⁰-max target seqs [5]. Synteny blocks for each MIRNA were computed using the blast output and GFF annotations of seven Oryza, two non-Oryza, and arabidopsis as input. The MCScanX tool [61] was used to determine the interspecies syntenic blocks with the following parameters: Match-score, final score = match score + num gaps × gap penalty (default: 50); gap-penalty, gap penalty (default: 1); match-size, the number of genes required to call a collinear block (default: 5); E-value, alignment significance 1 × 10⁻⁵; max-gaps, maximum gaps allowed (default: 25); and overlap-window, maximum distance 10,000 (number of nucleotides between genes) (default: 5) as well as collinear block patterns: 1 inter-species. The method found two or more species that shared a pairwise synteny block with at least five genes shared and an E-value of 1 × 10⁻¹⁰ in a maximum range of 10,000 nucleotides. The MCScanX software circle plotter was used to generate the figures. After identifying the collection of synteny blocks, in-house scripts were created to sub-set the MCScanX collinearity output file. Circos (Version 0.69-9., http://circos.ca/, accessed on 25 March 2021) [62] was used to generate circular graphs.

2.4. Plant Material, PCR, and Sequencing of Precursor MIR172

Seeds of O. sativa, O. rufipogon, O. glaberrima, O. barthii, O. glumaepatula, O. brachyantha, and O. punctata were procured from AICRP on rice (OUAT), and individual collections of rice breeders, seeds of Zea mays and Sorghum bicolor, were obtained from the division of Genetics, IARI; seeds of Arabidopsis were available in the laboratory. Plants were sown on moistened germination paper for five days, and after initial growth, the seedlings were transferred to 12-inch pots containing a mixture of cocopeat and sand (1:1) for completion of growth. Fresh tissues were harvested from pot-grown plants for amplification of MIRNAs and downstream experiments. Amplification and sequencing of each of the four MIR172 homologs from O. sativa, O. rufipogon, O. glaberrima, O. barthii, O. glumaepatula, O. brachyantha, O. punctata, Zea mays, Sorghum bicolor, and Arabidopsis thaliana were accomplished with Triticum aestivum as an outlier since no homologs of MIR172 were identified in wheat by in-silico analysis. Using the CTAB method, genomic DNA was isolated from ten plant species and used as a template for amplification of MIR172 A, B, C, and D precursor sequences using specific primers (Supplementary Table S2) in a master-cycler using high-fidelity DNA polymerase. The PCR reaction conditions were as follows: 30 s at 98 °C, 30 cycles of 10 s at 98 °C, 15 s at different annealing temperatures, and 10 s at 72 °C. For validation, the amplified PCR products were sequenced and aligned with pre-existing sequences.

2.5. Expression Analysis of Mature MIR172

Total MIRNA was extracted from the stage staggered tissues of seven Oryza spp., such as 2 leaf-shoot, 2 leaf-root, 4 leaf-shoot, 10 leaf-shoot, 10 leaf-shoot apical, 10 leaf-root, flag leaf, booting panicle, panicle (0.5 cm), panicle (0.5–1 cm), panicle (1–2 cm), panicle (2–4 cm), sorghum (3 leaf-shoot, 3 leaf-root, 5 leaf-shoot, 5 leaf-root, growing point, flower that has not yet bloomed, complete flower-blooming and G-blooming), maize (2 leaf-shoot, 2 leaf-root, 4 leaf-shoot, 10 leaf-shoot, 10 leaf-root, flag leaf, tassel, and silk), and Arabidopsis (2 rosette leaf-shoot, 2 rosette leaf-root, 4 leaf rosette-shoot, 10 leaf rosette-shoot, 10 leaf rosette-root, complete rosette growth, inflorescence, floral bud, and open flower) using a commercial MIRNA isolation kit as per manufacturer’s instruction. Using the Mir-XTM MIRNA First-Strand synthesis kit, cDNA was synthesized from the extracted MIRNA. Real-time PCR (qRT-PCR) was used to examine the expression of mature MIR172 using the Mir-X MIRNA qRT-PCR SYBR Kit with the mRQ 3′ universal reverse primer supplied with the kit, the species-specific mature MIR172 forward primer (Supplementary Table S3), and a reference dye (ROX) according to the manufacturer’s protocol. In 96-well qPCR compatible plates, the following cycles were used: initial denaturation for 5 min at 95 °C, 45 cycles of 15 s at 95 °C, 30 s at 60 °C, and 30 s at 72 °C. U6 was used as an endogenous control to normalize the MIRNA expression level. The results were presented as the mean of three biological replicates, with three technical replicates for each biological repeat.

2.6. Phylogenomic Analysis of Precursor Sequences

MIR172 homologs’ precursor sequences, along with 500 bp upstream sequences, were retrieved from the Gramene database. Synteny/evolutionary link between different MIRNAs was deduced by using the Gramene database (Supplementary Table S1). Clustal Omega [63] was used to conduct multiple sequence alignment for the individual MIRNAs, and MEGA10 [64] was used to construct an un-rooted tree using the Maximum-Likelihood (ML) technique. The Nearest Neighbor Interchanges (NNIs) method [65] was used to search for tree topology. For MIR172, the Tamura 3-parameter model with a discrete Gamma distribution (+G) and 5 rate categories substitution was used. The gamma shape parameter was directly determined from the data, and 1000 bootstrap replicates were used in the procedure. The proportion of invariable sites was fixed. The tree was generated in the Newick format. I-TOL (http://itol.embl.de/, accessed on 28 March 2021) produced a graphical representation of the phylogenetic tree.

2.7. Test of Neutrality

The neutrality of MIRNA sequence polymorphisms among poaceae was assessed using DnaSP v5.10 [66] and neutrality tests such as Tajima’s D [67] and Fu and Li’s F [68] on four distinct MIRNA loci, namely MIR172A, MIR172B, MIR172C, and MIR172D.

3. Results

3.1. Identification of MIR172 Homologs

MIR172 homologs were identified in all seven Oryza spp., sorghum, maize, and Arabidopsis except wheat, as no precursor of MIR172 could be identified in wheat using its miRBase database. In rice and its wild relatives, four homologs of MIR172, i.e., MIR172 (A–D), were identified. These homologs are located on different chromosomes throughout the rice genome (

Table 1. Chromosomal locations and coordinates of each MIR172 homologs in selected grasses. MIR172 homolog were identified in all seven rice species, sorghum, maize, and Arabidopsis, except wheat.

Species	MIRNA	Location	Precursor Coordinates
			Start	End
Oryza sativa	MIR172A	Chr 9	21,688,003	21,688,111
	MIR172B	Chr 1	46,953,572	46,953,809
	MIR172C	Chr 7	12,288,035	12,288,145
	MIR172D	Chr 2	37,298,091	37,298,220
		Chr 2	37,328,665	37,328,794
Oryza barthii	MIR172A	Chr 2	30,983,600	30,983,623
	MIR172B	Chr 1	36,558,471	36,558,708
	MIR172C	Chr 7	10,711,579	10,711,689
	MIR172D	Chr 2	30,983,582	30,983,711
Oryza glaberrima	MIR172A	Chr 2	28,473,275	28,473,298
	MIR172B	Chr 1	32,331,490	32,331,727
	MIR172C	Chr 7	10,632,540	10,632,650
	MIR172D	Chr 2	28,473,257	28,473,386
Oryza glumaepatula	MIR172A	Chr 9	23,407,188	23,407,296
	MIR172B	Chr 1	46,207,906	46,208,139
	MIR172C	Chr 7	12,334,350	12,334,460
	MIR172D	Chr 2	37,285,412	37,285,541
Oryza punctata	MIR172A	Ch9	26,230,640	26,230,686
	MIR172B	Chr1	45,791,747	45,791,791
	MIR172C	Chr2	38,821,836	38,821,861
	MIR172D	Chr2	38,821,883	38,821,953
Oryza rufipogon	MIR172A	Chr 9	20,417,006	20,417,114
	MIR172B	Chr 1	39,578,018	39,578,255
	MIR172C	Chr 7	11,443,165	11,443,275
	MIR172D	Chr 2	33,182,612	33,182,741
Oryza brachyantha	MIR172A	Chr9	14,032,760	14,032,800
	MIR172B	Chr1	33638435	33638477
	MIR172C	Chr7	7,379,288	7,379,325
	MIR172D	Chr2	26,379,138	26,429,195
Sorghum bicolor	MIR172A	Chr 9	58,774,558	58,774,659
	MIR172B	Chr 3	74,188,339	74,188,508
	MIR172C	Chr 4	67,645,991	67,646,109
	MIR172D	Chr 2	22,201,215	22,201,302
	MIR172E	Chr 2	14,122,957	14,123,071
	MIR172F	Chr 5	19,295,951	19,296,068
Zea mays	MIR172A	Chr 7	54,276,707	54,276,789
	MIR172B	Chr 5	221,721,349	221,721,474
	MIR172C	Chr 4	174,154,928	174,155,050
	MIR172D	Chr 6	172,029,786	172,029,859
	MIR172E	Chr 3	145,801,592	145,801,765
Arabidopsis thaliana	MIR172A	Chr2	11,942,914	11,943,015
	MIR172B	Chr5	1,188,207	1,188,301
	MIR172C	Chr3	3,599,776	3,599,908
	MIR172D	Chr3	20,587,904	20,588,027
	MIR172E	Chr5	23,988,472	23,988,596

). In sorghum, a total of six homologs of MIR172, i.e., MIR172 (A–F), were identified, whereas, in maize and Arabidopsis, five homologs of MIR172, i.e., MIR172 (A–E) were identified. Since we used sativa as a reference for genome structure and evolution analysis, only four homologs corresponding to rice, i.e., MIR172 (A–D), were included in the study. Chromosomal locations and coordinates of each homolog (Table 1, Figure 1a–d) and the number of predicted homologs per MIR172 are provided in

Table 2. Prediction of gene number and homologs of MIR172 in respective grasses with Arabidopsis as outliers. Notably, four homologs in Oryza, five in maize and Arabidopsis, and six in sorghum were identified with gene content as low as 1–2 in O. sativa (MIR172D) and as high as 25–27 in S. bicolor (MIR172D).

	MIRNA		Plant Species
			Oryza sativa	Oryza barthii	Oryza glaberrima	Oryza glumaepatula	Oryza rufipogon	Oryza brachyantha	Oryza punctata	Sorghum bicolor	Triticum aestivum	Zea mays	Arabidopsis thaliana
No. of genes predicted	MIR172	MIR172A	23	22	23	17	23	27	22	20	0	14	27
		MIR172B	23	21	18	18	17	9	25	21	0	15	28
		MIR172C	21	25	22	13	28	21	22	18	0	12	29
		MIR172D	1–23, 2–19	22	23	19	18	17	22	27	0	19	26
		MIR172E	0	0	0	0	0	0	0	25	0	14	34
		MIR172F	0	0	0	0	0	0	0	19	0	0	0
No. of homologs predicted		MIR172A	1	1	1	1	1	1	1	1	0	1	1
		MIR172B	1	1	1	1	1	1	1	1	0	1	1
		MIR172C	1	1	1	1	1	1	1	1	0	1	1
		MIR172D	2	1	1	1	1	1	1	1	0	1	1
		MIR172E	0	0	0	0	0	0	0	1	0	1	1
		MIR172F	0	0	0	0	0	0	0	1	0	0	0

. The presence of four homologs in rice, five in maize and Arabidopsis, and six in sorghum was confirmed by the BLAST result in this study. All the homologs of MIR172 viz., MIR172A, B, C, and D were successfully amplified, and the nucleotide sequences have been submitted as Supplementary Data S1.

3.2. Expression Analysis of MIR172

MIR172 is known to play a cardinal role in vegetative to reproductive-stage transition in plants [7]. To determine the variations in the expression pattern of MIR172 during development, an expression profile of mature MIR172 was compared in six RWR and other poaceae members. Accumulation of mature MIR172 increased with the growth of the plant, gradually increasing from the 2-leaf stage (0.582-fold) to the 10-leaf stage (8.54-fold), reaching its maximum in the flag-leaf stage in rice (25.97-fold, Figure 2). However, there was no increase in M1IR172 expression in roots in rice. Similarly, in sorghum, the expression of MIR172 increased from the 3-leaf stage (0.934-fold) to growth point differentiation (7.35-fold), reaching a maximum in the flag-leaf stage (24.9-fold, Figure 2). A similar pattern was observed in maize, with maximum expression in the flag leaf (23.9-fold). In the case of Arabidopsis, the expression level gradually increased from the 2-rosette leaf stage (0.745-fold) to the complete rosette growth stage (22.615-fold, Figure 2). Flag leaf or complete rosette growth marks the transition stage between the vegetative and reproductive stages. During the reproductive growth phase, increased accumulation of MIR172 was observed in the booting panicle stage in rice (13.465-fold), maize (13.546-fold), and sorghum (10.918-fold); and developing inflorescence in Arabidopsis (11.22-fold). Expression of MIR172 gradually decreased during panicle development. Mature MIR172 exhibited a similar expression profile in all Oryza spp. with minor variations in the magnitude of expression among different species (Figure 2).

3.3. Conservation and Divergence in Mature and Precursor Sequence of MIR172

Mature MIR172A and MIR172D of all the poaceae members and Arabidopsis were found to be highly conserved (Figure 3a,d), while mature MIR172B homologs of all the seven Oryza spp and sorghum were highly conserved. However, in mature MIR172B sequence of maize and Arabidopsis, multiple substitutions at 1^st (G→A), 3^rd (A→C), 5^th (T→C), 7^th (T→A), 9^th (G→C in maize and G→T in Arabidopsis), 11^th (T→A), 15th (G→T) and 21^st (T→C) positions were detected along with deletions and insertions (Figure 3b). A single nucleotide insertion was observed at the zero position in Arabidopsis and at the 22^nd position in maize. Three nucleotide (TGC) deletions were observed in Arabidopsis and maize between the 16^th and 20^th positions (Figure 3b). MIR172C homologs revealed the highly conserved nature of all the rice MIR172C homologs. In the case of sorghum, maize, and Arabidopsis, two base substitutions at the 1^st (T→A) and 21^st (C→T) positions were observed (Figure 3c).

Multiple sequence alignment of precursor sequences of MIR172 revealed the non-occurrence of SNPs in the MIR172A precursor sequences of all poaceae members and Arabidopsis. However, insertions and deletions were detected in almost all poaceae species under study. An insertion of 11 nucleotides (ATCCGAACCAC) was observed between the 41^st and 42^nd positions in maize and Arabidopsis. Further, an eight-nucleotide (ATCCTCGG) insertion was observed between the 41^st and 42^nd positions in sorghum. A five nucleotides (TATGT) insertion between the 69^th and 70^th positions was also observed in sorghum. A deletion of the first 76 nucleotides (GTGTTTGCGGGCGTGGCATCATCAAGATTCACATCCATGCATATATCACAAGACGCACATATACATCCGATTTGGC) was observed in barthii and glaberrima, while deletion of first 41 nucleotides (GTGTTTGCGGGCGTGGCATCATCAAGATTCACATCCATGCA) was observed in rufipogon. Similarly, the first 62 nucleotide (GTGTTTGCGGGCGTGGCATCATCAAGATTCACATCCATGCATATATCACAAGACGCACATAT) deletion was also observed in punctata. A nine nucleotide (CGCAGACAA) deletion was also observed at the 100th position in barthii and glaberrima. A 17-nucleotide deletion (CACATATACATCCGATT) between the 55^th and 73^rd position was observed in maize. In addition to this, a ten-nucleotide (CCGCAGACAA) deletion at the 99^th position in sorghum and maize and a six-nucleotide (AGACAA) deletion at the 103^rd position in Arabidopsis was also observed (Figure 4a).

In MIR172B precursors, specific single nucleotide substitution was detected at the 203^rd position, where G was replaced with A; at the 205^th position, G was replaced with A, and at the 208^th position, G was replaced with A was observed in maize, and Arabidopsis in comparison to rice. Specific single base substitutions were also observed in sorghum at the 197^th (A→G) and the 199^th (A→G) positions. In Arabidopsis, similar single nucleotide substitutions were observed at the 196^th (T→G) and 201^st (T→G) positions. In addition to single nucleotide substitutions, insertions and deletions at multiple locations were also observed. A genome-specific two-nucleotide (CC) insertion between the 71^st and 72^nd positions was observed in sorghum. An insertion of six nucleotides (GATGAG) between the 204^th and 205^th positions was observed in maize. A deletion of the first 195 nucleotides (GTGATTTCTGACGTGGCATCATCAAGATTCACACATTACATTGCATGCATGCATATTCTATGCATCTTTGAGCTTGTTGTTCTGATCTCAACAACCTAGCTAGCTATATTTCTCTCCTGGCCCTGACCTGCATGATGCATGGTTGCACGCATGGAGAGAGAAGAGAGAGATCGAAGCTAA TT AAACGCATGTG) in brachyantha and the first 193 nucleotides (GTGATTTCTGACGTGGCATCATCAAGATTCACACATTACATTGCATGCATGCATATGTTATGCATCTTTGA GCTTGTTGTTCTGATCTCAACAACCTAGCTAGCTAATATTTCTCTCCTGGCCCT GA CCTGCATGATGCATGGTTGCACGCATGGAGAGAGAAGAGAGAGATCGAAGCTA ATTAAACGCATG) in punctata was observed. The first ten nucleotides (GTGATTTCTG) were also found to be deleted in sorghum and maize, and the first eight nucleotides (GTGATTTC) in Arabidopsis were found to be deleted. In sorghum, (ATTAC) deletions between the 34^th and 40^th, (TG) deletions between the 49^th and 52^nd, (TCTCCTG) deletions between the 114^th and 122^nd, (GTTGCACGCATGGAGAGAGAAGAGAGAGATCGAAGC) deletions between 143^rd and 180^th and (CGGA) at 228^th positions were observed. In maize, 50^th and 52^nd positions (G), 70^th and 168^th (GAGCTTGTTGTTCTGATCTCAACAACCTAGCTAGCTAATATTTCTCTCTGGCCCTGACCTGCATGATGCATGGTTGCACGCATGGAGAGAGAA GAGAG) deletions between and (CGGA) deletions between 228^th were observed. In Arabidopsis, single deletion of 130 nucleotides (CATTACATTGCATGCATGCATATGTCTATGCATCTTTGAGCTTGTTGTTCTGATCTCAACAACCTAGCTAGCTAATATTTCT CTCCTGGCCCTGACCTGCATGATGCATGGTTGCACGCATGGAGAGAGA) between the 34^th and 164^th position was observed (Figure 4b).

Multiple sequence alignment of MIR172C precursor sequences revealed the occurrence of only one SNP. Single nucleotide polymorphism was observed at the 101^st position, where ‘T’ in rice was replaced with ‘G’ in Arabidopsis. Although many insertions and deletions were also observed, prominent amongst them is the insertion of five nucleotides (AGCTA) at the zero position, 19 nucleotides (TTAGATTTTTGATGTATGT) between the 70^th and 71^st position, and five nucleotides (TGGCT) at 111^th position in Arabidopsis. Sorghum, a seven-nucleotide insertion (ATGCATG) between the 62^nd and 63^rd, and 28 nucleotides insertion (TGGCTCGCAGTTGCTATATATGCATATG) between the 70^th and 71^st positions were observed. A deletion of the first 73 nucleotides (CTTGTTGCGGGTGCAGCGTCATCAAGATTCACGTGTGCCGCACGGCACACGTATCGGTTTTCAAGTGTAGTCA) and the first 81 nucleotides (CTTGTTGCGGGTGCAGCGTCATCAAGATTCACGTGTGCCGCACGGCAC ACGTATCGGTTTTCAAGTGTAGTCATCGTGCGT) was observed in brachyantha and punctata, respectively. Additionally, the deletion of four nucleotides (AGAG) at the 107^th position was observed in punctata. A deletion of ten nucleotides (CTGCAATCAG) at the 101st position was observed in sorghum and maize (Figure 5a).

In MIR172D precursors, genome-specific single nucleotide substitutions were detected in sorghum at 23^rd (C→T), 24^th (A→G), 28^th (C→T), 43^rd (T→A), 57^th (A→G) and 71^st (T→A) positions. Four consecutive single nucleotide substitutions were detected at the 59^th (A→T), 60^th (C→T), 61^st (G→T), and 62^nd (A→G) positions in S. bicolor. In maize, two consecutive nucleotide substitutions were observed at the 56^th (T→C) and 57^th (A→C) positions and at the 61^st (G→T) and 62^nd (A→C) positions. In Z. mays, two SNPs at 59^th (A→G) and 71^st (T→C) positions, and in Arabidopsis, SNPs were observed at 25^th (G→A), 28^th (C→T), 30^th (A→T), 41^st (C→T), 43^rd (T→A) and 55^th (C→T) positions. Prominently, four successive SNPs were detected at the 67^th (G→T), 68^th (A→T), 69^th (G→T), and 70^th (G→A) positions in Arabidopsis (Figure 5b).

In addition to the substitutions, insertions and deletions were also observed in MIR172D precursor sequences. A single nucleotide insertion at the zero position was observed in brachyantha. An insertion of 10 nucleotides (CAAATAAACC) in sorghum and a nine-nucleotide insertion (CTTATGCCT) in Arabidopsis between the 53^rd and 54^th positions was observed. Similarly, deletion of the first 22 nucleotides (AAACAGTCGGTGCTTGCAGGTG) in sorghum, the first 19 nucleotides (AAACAGTCGGTGCTTGCAG) in maize and the first four nucleotides (AAAC) in Arabidopsis were observed. A deletion of nine nucleotides (GAGTTCATC) between the 44^th and 53^rd positions was observed in maize. Another deletion of 59 nucleotides (TGGCTGACTATATGTGATGAGAATCTTGATGATGCTGCATCAGCAAACGCTCGACTACT) from the 71^st to the 130^th position was observed in punctata. An eleven-nucleotide deletion (CTGACTATATG) in sorghum between the 74^th and 85^th position and an eight-nucleotide deletion (CTGACTAT) between the 74^th and 83^rd position in maize and Arabidopsis were also observed. A 19-nucleotide deletion (CAGCAAACGCTCGACTACT) from the 111^th to 130^th position in sorghum and maize was observed. A three-nucleotide deletion (ACT) at the 128^th to 130^th position was also observed in Arabidopsis (Figure 5b).

3.4. Spectrum of Sequence Variation in MIR172 Homologs

Tajima’s D 67 and Fu and Li’s F 68 test of neutrality of sequence polymorphisms for different MIR172 homologs, viz., MIR172A, MIR172B, MIR172C, and MIR172D revealed non-significant negative values for all four loci. Among the four pre-MIRNA homologs examined across nine poaceae species, no single nucleotide polymorphism was detected in MIR172A and, therefore, no neutrality test was conducted for MIR172A (Supplementary Figures S1 and S2). The highest negative (non-significant) Tajima’s D value (−0.48816) was found in MIR172C, followed by MIR172B (−1.0403). Sequence variation for MIR172D was the lowest (−1.45822) and non-significant (Supplementary Figures S3, S5, and S7). Nevertheless, Fu and Li’s F values were also non-significantly negative and found to be consistent with Tajima’s D value. Similarly, MIR172C exhibited the highest negative Fu and Li’s value (−0.36836) followed by MIR172B (−0.81963), with MIR172D having the lowest negative value (−1.43015) (Supplementary Figures S4, S6, and S8). The nucleotide diversity of the poaceae species studied ranged from 0.05657 at the pre-MIR172B gene to 0.05657 at the pre-MIR172C locus and 0.18291 at the pre-MIR172D locus. These findings suggest that differences in selection pressure experienced by cultivated varieties during improvement, as well as WGD events in non-Oryza species, account for sequence polymorphism at distinct MIRNA loci.

3.5. Gene Conservation and Gene Density Analysis

The number of genes predicted in the 100 kb region of each homolog of MIR172 (Table 2) revealed the presence of 23 genes around the 100 kb region MIR172A of Oryza sativa, and the highest percentage of sativa homologs were found to be conserved in rufipogon (73.91%, i.e., 17 conserved genes out of 23 detected genes). The lowest conservation (0%) was found in barthii, glaberrima, sorghum, maize, and Arabidopsis, i.e., disruptive synteny with complete loss of conservation. Similarly, the maximum gene density was found in Oryza punctata (1gene/3.703 kb), and the minimum density of genes was observed in maize (1gene/7.14 kb) (Figure 6a,b). We observed the presence of 23 genes in the 100 kb region of MIR172B of sativa, with 58.8% conservation with rufipogon (10 out of 17 detected genes), and the lowest conservation of 0% was detected in brachyantha, sorghum, and maize (disruptive synteny), including Arabidopsis. The maximum gene density was found in sorghum (1gene/3.073 kb), and the minimum density of genes was observed in punctata (1gene/5.88 kb) (Figure 6a,b).

In MIR172C, 21 genes were identified in the 100 kb region of Oryza sativa with 76.9% conservation in glumaepatula (10 out of 13 genes) and the lowest in punctata, brachyantha, sorghum, and maize (0%, i.e., disruptive synteny), including Arabidopsis. At the same time, maximum gene density was found in rufipogon and brachyantha (1gene/3.571 kb); the minimum density of genes was observed in maize (1gene/8.33 kb) (Figure 6a,b). Two homeologs of MIR172D, i.e., MIR172Di and MIR172Dii, were identified in the sativa genome, and 23 genes were identified in the 100 kb region harboring MIR172Di of sativa. The highest percentage of sativa homologs were found to be conserved in rufipogon (61.1%, i.e., 11 out of 18 genes) and the lowest in brachyantha, sorghum, Arabidopsis, and maize (0%, i.e., disruptive synteny (Figure 6a). It is observed that the maximum gene density is 1 gene/3.703 kb in sorghum, while the minimum gene density of 1 gene/5.882kb was observed in punctata (Figure 6b).

Furthermore, 19 genes were identified in the 100 kb region harboring MIR172Dii of Oryza sativa sub-species indica, and the highest percentage of sativa homologs were found to be conserved in rufipogon (72.2%, i.e., 13 out of 18 genes). However, the disruptive synteny showing a total loss of conservation was detected in brachyantha, sorghum, and maize. Further, the synteny results revealed the highest gene density of 1 gene/3.073 kb in sorghum, with punctata exhibiting the lowest gene density of 1 gene/5.88 kb.

3.6. Microsynteny Analysis

Inter-species synteny block analysis of MIRNA172 homologs revealed that out of 218 genes in MIR172A, 129 genes (59.17%) were collinear. Similarly, 88 genes (45.13%) out of 195 predicted genes in MIR172B; 77 genes (36.49%) out of predicted 211 genes in MIR172C; and 91 genes (38.72%) out of 235 predicted genes for MIR172D were collinear. The demonstration of the alignment of non-anchor genes is marked by ‘||’ in the multi-alignment of gene orders. Synteny or collinearity blocks were (Supplementary Figures S9–S45) constructed using sativa and other relevant reference genomes to develop saturated synteny/collinearity blocks for each homolog of MIRNA.

(i)

MIR172A

Although conservation of microsynteny in the 100 kb region surrounding MIR172A was detected amongst brachyantha, glumaepetula, rufipogon, and punctata; a complete loss of microsynteny was observed within barthii and glaberima, along with sorghum, maize, and Arabidopsis. An in-depth analysis of the 100 kb region in MIR172A of sativa revealed the presence of 23 genes, of which seven genes viz., Chitin elicitor receptor kinase 1-like, P-loop NTPase domain-containing protein LPA1 homolog, Glucomannan 4-beta-mannosyltransferase 1-like, probable serine/threonine-protein kinase PBL23, pollen-specific protein C13-like, dynamin-related protein 1E-like, and pentatricopeptide repeat-containing protein At5g66520-like were found to be conserved in brachyantha, glumaepetula, rufipogon, and punctata. Genes such as Retrotransposon protein, putative, unclassified, and Ethylene responsive transcription factor 1B of sativa were also found to be conserved in glumaepatula and rufipogon. Further analysis revealed that diacylglycerol lipase-beta and Ethylene-responsive transcription factor ERF096 of sativa were conserved in brachyantha and punctata. Thus, we conclude that the overall microsynteny in MIR172A was gradually lost from sativa to its wild relatives and further to dicot species (Figure 7a and Figure 8a).

(ii)

MIR172B

Similarly, conservation of microsynteny in a 100 kb region surrounding MIR172B was detected only in Oryza punctata, glumaepatula, rufipogon, barthii, and glaberima (Figure 5b and Figure 6b) with the exception of Oryza brachyantha that exhibited complete loss of microsynteny. In the 100 kb region, altogether, 23 genes were detected in Oryza sativa, of which carboxyl-terminal peptidase-like, pumilio homolog 1-like, protein FIZZY-RELATED 3, putative WRKY transcription factor 49, uncharacterized protein LOC4326395, zinc transporter 1, and probable E3 ubiquitin-protein ligase RHC2A genes were found to be conserved in all five Oryza spp. LUX gene was found to be conserved in punctata, rufipogon, barthii, and glaberima. Oryza homolog Retrotransposon protein, and Ty3-gypsy subclass were conserved in glumaepatula, and Phosphatidylinositol/phosphatidylcholine transfer protein SFH13 was conserved in barthii. However, none of the sativa homologs were detected in sorghum/maize/Arabidopsis (Figure 7b and Figure 8b), showing a complete loss of microsynteny.

(iii)

MIR172C

Analysis in a 100 kb region surrounding MIR172C revealed the conservation of microsynteny amongst glumaepatula, rufipogon, barthii, and glaberima (Figure 7c and Figure 8c). However, a complete loss of microsynteny was observed in punctata and brachyantha. Amongst 21 genes identified in the 100 kb region harboring MIR172C of sativa, six genes, namely, F-actin-capping protein subunit alpha, Ribosomal protein L23/L15e family protein, hypothetical protein OsI_25745, adenyl cyclase-like protein, G-type lectin S-receptor-like serine/threonine-protein kinase At1g34300 and Retrotransposon protein, Ty1-copia subclass were found to be conserved in glumaepatula, rufipogon, barthii, and glaberima. Another hypothetical protein gene OsI_25741 was found to be conserved in barthii, rufipogon, and glumaepatula. Similarly, hypothetical protein DAI22_07g114300 and fatty acyl-CoA reductase1 were found to be conserved in barthii, glumaepatula, and glaberrima. Genes such as retrotransposon protein, Ty3-gypsy subclass was found to be conserved in barthii and glaberrima. Amongst others, activator-like transposable element and phosphatidylinositol 4-phosphate 5-kinase 6-like are the two genes found conserved only in rufipogon while no sativa homologs were conserved in sorghum/maize/Arabidopsis indicating complete disruption of microsynteny (Figure 7c and Figure 8c).

(iv)

MIR172D

Microsynteny conservation in the 100 kb region surrounding MIR172D was detected amongst punctata, glumaepatula, rufipogon, barthii, and glaberima, but the same was completely lost in brachyantha. Two homeologs of MIR172D, i.e., OsMIR172Di and OsMIR172Dii, were identified in the sativa genome. Out of the 23 genes detected in OsMIR172Di in the 100 kb region harboring MIR172D of sativa, two genes, namely protein LUTEIN DEFICIENT 5 and pentatricopeptide repeat-containing protein At2g22410 were found to be conserved in punctata, glumaepatula, rufipogon, barthii, and glaberima. The genes heavy metal-associated isoprenylated plant protein 3-like and Pib variant protein were found to be conserved in punctata, glumaepatula, barthii, and glaberima. Another rice homolog, putative brown planthopper-induced resistance protein 1 was conserved in punctata, rufipogon, barthii, glaberima, and hypothetical protein DAI22_02g385300 was conserved in punctata, glumaepatula, rufipogon, and glaberima. Further, genes such as DNA ligase-like and probable protein S-acyltransferase 15 were conserved in all Oryza spp except punctata, glumaepatula, and brachyantha. Other genes, such as putative telomere binding protein-1; TBP1, cyst nematode resistance protein-like protein, and protein N-lysine methyltransferase METTL21A were observed to be conserved in barthii, punctata, and glumaepatula, respectively. No sativa homologs were found to be conserved in sorghum, maize, and Arabidopsis, indicating complete disruption of microsynteny (Figure 7d and Figure 8d).

Out of the 19 genes identified in the 100 kb region of OsMIR172Dii detected in sativa, four genes, pentatricopeptide repeat-containing protein At2g22410, heavy metal-associated isoprenylated plant protein 3-like, thioredoxin-like 3–2 and OTU domain-containing protein 5-A were found to be conserved in punctata, glumaepatula, rufipogon, barthii, and glaberima. Rice homolog, hypothetical protein DAI22_02g385300 was conserved in punctata, glumaepatula, and rufipogon; pumilio homolog 1-like was conserved in punctata, glumaepatula, rufipogon, barthii, and glaberima; and P-loop NTPase domain-containing protein LPA1 was conserved in glumaepatula, rufipogon, barthii, and glaberima. Further, genes named DNA ligase-like and probable protein S-acyltransferase 15 were found to be conserved in rufipogon, barthii, and glaberima, while the Pib variant protein was conserved in punctata, glaberrima, and barthii. Nonetheless, no sativa homologs were detected in sorghum, maize, or Arabidopsis, indicating complete disruption of microsynteny.

3.7. Phylogenomic Analysis

Phylogenomic analysis of MIR172 homologs revealed that they had undergone whole genome duplications and polyploidization events to segregate into different clades. Maximum-likelihood tree revealed that MIR172 sequences clustered into three clades viz. Clade-I, Clade-II, and Clade-III (Figure 9). Clade-I included sativa, brachyantha, glaberima, rufipogon, and punctata MIR172A that formed Sub-clade-IA, while, sativa, barthii, glaberima, glumaepetula, and rufipogon MIR172C formed Sub-clade-IB. Similarly, sativa, barthii, brachyantha, glaberima, glumaepetula, rufipogon, punctata, and sorghum MIR172B along with brachyantha MIR172C formed Sub-clade-IC; sorghum along with Arabidopsis MIR172D formed Sub-clade ID. Within Clade-II, sativa (1 and 2), barthii, brachyantha, glaberima, glumaepetula, rufipogon, and punctata MIR172D along with punctata MIR172C formed Sub-clade IIA. Sorghum MIR172A and C and maize MIR172B formed Sub-clade-IIB; maize MIR172C and D formed Sub-clade-IIC; barthii and glaberima MIR172A formed Sub-clade-II-D. Maize MIR172A, along with Arabidopsis MIR172C formed Clade-III (Figure 9). Maize formed a separate sub-clade IIC or was grouped with Arabidopsis in clade III. This indicates that MIR172 (A–D) share a common evolutionary descendent.

4. Discussion

Studies have shown that whole-genome duplication (WGD), in addition to the tandem duplication of MIRNA genes, are involved in their evolution, but its contribution varies from species to species [69], which means apart from their ancestral origin, plant MIRNAs may have been generated by duplication of pre-existing MIRNA genes Our study revealed the presence of single nucleotide polymorphism in precursor sequences of MIR172 homologs at multiple positions along with deletions and insertions (Figure 3, Figure 4 and Figure 5). This can be ascribed to two rounds of WGD that Arabidopsis underwent after splitting from papaya during the course of evolution [70]. However, monocots, unlike eudicots such as Arabidopsis, underwent only one shared ancestral WGD during their evolution [70]. Our analysis of the precursor sequences of MIR172A (Figure 4) in poaceae revealed that it is conserved in Oryza except in barthii, glaberrima, punctata, and rufipogon, which showed deletion in a few positions. Nonetheless, brachyantha and punctata showed deletion in MIR172B and MIR172C, while an insertion in brachyantha and a deletion in punctata was observed in MIR172D (Figure 5). These can be ascribed to the genome downsizing and resistance to genome expansion in brachyantha and glaberrima that could have led to structural variation. Similarly, in punctata and barthii both genome expansion and contraction might have led to structural variations [71]. The polymorphism in the precursor sequences of sorghum and maize could be ascribed to the WGD event that occurred 30 million years ago (MYA), separating the lineage of sorghum and maize from that of rice [72] from its common ancestors. Further, the variations in the precursor sequence of maize could be due to an extra WGD event leading to the expansion of MIRNA gene family [73]. In fact, duplicated MIRNA genes in maize underwent extensive gene loss, with approximately 35% of ancestral sites retained as duplicate homoeologous MIRNA genes [74]. Common grass triplication and genome hybridization in wheat [75] might be accountable for the complete deletion of wheat MIR172 homologs. WGDs shape up the number of MIRNA genes, but their number and pattern vary in a species-specific manner [69]. This might be responsible for the disruption of microsynteny among rice wild relatives and the complete loss of microsynteny in sorghum, maize, and Arabidopsis.

MIR172 belongs to highly conserved MIRNA families [22], and the multiple sequence alignment of mature MIR172 revealed that it is conserved amongst all the seven Oryza species. Our result is commensurate with a previous report that MIRNA genes contain fewer SNPs than their contiguous border region, and the mature sequence of the MIRNA genes contains fewer SNPs than their precursors [76,77]. The results of multiple sequence alignment of mature and precursor MIR172 revealed the same as we observed fewer SNPs in mature MIRNA sequence in comparison to precursor sequences. SNPs have a different impact on the functionality of MIRNA; for example, variants in miRNA promoter regions and other regulatory regions may result in an altered transcription rate, variants in splice sites of the host gene (for intronic miRNAs) or of the poly-cistron (clustered miRNAs) can result in aberrant expression patterns. In this regard, Tajima’s D [67] and Fu and Li’s F [68] tests were used to estimate the neutrality of sequence polymorphism for MIRNA genes. These tests detect both positive and balancing selections [78]. Our results suggest non-neutral sequence variations in all the MIRNAs. Tajima’s D (−1.45822) and Fu and Li’s F (−1.43015) values for the MIR172D locus were negative, while both test values were least negative at the MIR172C locus [Tajima’s D (−0.48816); Fu and Li’s F (−0.36836)]. Previously, negative findings in one or both tests were reported in rice and Arabidopsis [5,68]. Similarly, the nucleotide diversity was found to differ amongst MIRNA locations. As a result, MIRNAs are effective targets for the differential accumulation of variation in populations subjected to selection pressures.

Expression analysis of the mature MIR172 sequences in root and shoot tissues at different developmental stages helped in determining the spatial and temporal expression of each MIR172 member. However, discerning expression of individual MIR172 family members was unlikely as two out of four MIRNA homologs (MIR172A and D) are completely conserved, while the other two MIRs (MIR172B and C) have minimal sequence differences. Comparative analysis of MIR172 showed similar expression profiles in different developmental tissues (Figure 2). Insight into MIR172 expression revealed that it is highly expressed in late vegetative stages (10 leaf-stage-12.23-fold), flag leaf (25.97-fold), and developing panicles (13.465-fold). However, there was no increase in its expression in roots in rice. A similar trend of expression was observed in other poaceae members and Arabidopsis as well. Lower levels of MIR172 expression observed in the early vegetative stage can be ascribed to its role in the transition [79] of plant development from juvenile to the flowering stage by regulating AP2-like genes, including the target of early activation tagged 1/2/3. Higher expression of MIR172 in late vegetative and panicle (Figure 2) is consistent with its role in the acquisition of floral competence [80]. The increasing level of MIR172 is commensurate with the appearance of adult traits, as its over-expression leads to early flowering (Figure 2).

Microsynteny/collinearity analysis provides an insight into the shared ancestry of groups of genes and unravels the evolutionary history of genomes and gene families, establishing gene orthology [81]. Microsynteny analysis of MIR172 revealed that gene collinearity within the genus Oryza is conserved. However, disruption of synteny in the 100 kb region of MIR172A in barthii (AA) and glaberima (AA), of MIR172B in brachyantha (FF), and of MIR172C in punctata (BB) was observed. Oryza spp. That lost microsynteny belonged to three distinct sub-genomes. Amongst these, glaberrima is a domesticated species, while the other three are wild relatives. Evolutionary studies revealed that the AA genome diverged from FF progenitors by almost ~15 MYA [82], whereas the divergence between AA- and BB- genomes occurred at 9.11 MYA [83]. Gene loss events observed in glaberrima (AA), barthii (AA), and brachyantha (FF) plausibly led to the differences in gene family content.

The progenitors of brachyantha and punctata diverged from sativa progenitors during the course of evolution from FF- and BB- genomes to AA-Oryza genomes, in spite of a well-conserved genome organization and well-preserved gene order. Loss of microsynteny in MIR172A in barthii (AA) and glaberima (AA), MIR172B in brachyantha (FF), and MIR172C in punctata (BB) can be due to these factors. It is reported that sativa was domesticated from rufipogon (perennial wild rice) around 9000 years ago in Asia, and glaberrima was domesticated from barthii independently, around 3000 years ago in West Africa [84,85]. This corroborates our result as the highest number of sativa gene homologs were found to be conserved in rufipogon for MIR172A, B and D followed by MIR172C (Figure 7). Additionally, most of the barthii gene homologs were conserved in glaberrima (Supplementary Figures S1, S2, S10, S11, S19, S20, S28, and S29).

5. Conclusions

Our in-depth analysis of microsynteny/collinearity of MIR172 and its homologs in seven different Oryza species, sorghum, maize, and Arabidopsis is a comprehensive evolutionary study based on MIRNA sequence variation and conservation of orthologous genes. We identified the orthologous MIRNA genes of rice and, microsynteny analysis, genes harbored around 100 kb region, revealed that the gene pattern and content are conserved (conservative evolution) among Oryza species with exceptions depending upon the genome type and selection pressure during evolution/domestication. However, the microsynteny in sorghum, maize, and Arabidopsis was completely lost (disruptive evolution) during the course of evolution due to WGD events. Gain/loss of genes or chromosomal repatterning might have caused structural variation, but overall gene content and order in MIRs are maintained. Low genetic diversity at different MIRNA loci in cultivated rice revealed that the rice wild relatives are the untapped genetic reservoirs to be harnessed for crop improvement. Abiotic [86] and biotic stresses [43] due to inconsistent climate indices and population outbursts pose a threat to national food security. The green revolution has no doubt provided superior cultivars for improved food grain production, but stringent selection has created a bottleneck in the genetic variability in domesticated crops. Owing to their habitat in a robust natural environment, crop wild relatives managed to maintain a higher level of genetic variability. Utilizing the untapped genetic resources available in CWRs for crop improvement is an attractive option for further improving food production in the future.