Genome-wide identification, characterization and expression profiles of heavy metal ATPase 3 (HMA3) in plants

1 Introduction

Heavy metals are abundant in nature due to natural and anthropogenic causes. Heavy metals are taken up by humans through water and food-based meals and may cause serious health problems. Many of the metals, such as Fe, Cu, Zn, are essential for plants, but they need to be at an optimized level. In contrast, some of the heavy metals (Pb, Cd) are highly toxic to plants hampering photosynthesis, nutrient uptake, and yield in plants (Shanker et al., 2005). However, plants acclimatize different strategies consisting of uptake, sequestration, and chelation to regulate metal homeostasis in withstanding heavy metal toxicity (Tripathi et al., 2012). The association of different metal transporters and their binding capacity play an integral part in the cellular detoxification and maintenance of metals in plants.

The ATPases (P-type adenosine triphosphatases) are the largest superfamily of integral membrane proteins involved in transporting transition metal cations in plants. Eight P-type IB ATPases are encoded in the genome of Arabidopsis thaliana (Baxter et al., 2003). The ATPases are clustered into two groups in plants: monovalent Cu/Ag ion transporting ATPases and divalent Zn/Cd/Co/Pb ion transporting ATPases (Williams et al., 2005). Due to the distinctive N-terminal sequence, ATPases in plants are named HMA (heavy metal associated) protein. In Arabidopsis, HMA1-HMA4 and HMA5-HMA8 proteins are belonging to cluster 1 (Zn/Cd/Co/Pb) and 2 (Cu/Ag) groups (Baxter et al., 2003). In particular, HMA3 proteins participate in heavy metal ion transport and detoxification in plants. In Arabidopsis thaliana, AtHMA3 localized in tonoplast is involved in the vacuolar storage of Cd (Chao et al., 2012). Furthermore, Arabidopsis overexpressed with AtHMA3 showed increased tolerance to Cd, Zn, Pb, and Co, while AtHMA3‐knockout mutant exhibited sensitivity to Cd and Zn (Morel et al., 2009). Similarly, the overexpression of SaHMA3h in tobacco improved the Cd accumulation and tolerance of transgenic plants (Zhang et al., 2016). Further, BjHMA3 is shown to be associated with the varied Cd accumulation in leaves of Brassica rapa (Zhang et al., 2019). Again, OsHMA3 ectopic over-expression resulted in increased Cd tolerance and lower Cd concentration in leaves and grains but increased Cd concentration in rice roots (Ueno et al., 2010).

Protein topology, such as transmembrane helices, domain recognition, binding sites is crucial features for metal binding capacity in plant system. As a result, the identification of metal sites along with the components at transcriptional and Posttranslational regulation eventually determines the functions of a protein in response to metals. One of the identified candidate genes in Arabidopsis halleri was AhHMA3, which is highly similar to HMA3 in Arabidopsis thaliana (AT4G30120) (Becher et al., 2004). HMA3, located in the vacuolar membrane, participates in vacuolar sequestration of Zn, Cd, Co, and Pb in Arabidopsis (Morel et al., 2009). However, the function of HMA3 in hyperaccumulators remains unclear in planta. Although the molecular functions of Arabidopsis HMA3 are relatively well established, the analysis of HMA3 homologs and interactions with other transporters/genes are barely studied.

The characterization of HMA3 possesses the immense potential to combat metal homeostasis in plants. The in silico characterization of HMA was performed in a previous study on Brassica oleracea (Sutkovic et al., 2016) but its validation and interactions among the other species are yet to be done. The in silico characterization of HMA3 homologs may provide in-depth insight into these genes/proteins. In this study, we have searched for HMA3 homologs based on Arabidopsis heavy metal ATPase 3 (AtHMA3) referred to as AT4G30120 across different plant species. The CDS, mRNA, and protein sequences of these HMA3 homologs were taken into computational analysis with advanced bioinformatics software and online-based platforms.

2 Materials and methods

3 Results

3.4

3.4 Gene organization

ARAMEMNON analysis showed the presence of 8–9 exons among the HMA3 gene homologs located at different positions of gene ranged from 1 to 3535 base pairs (Fig. 1, Table 3). Promoter analysis showed marginal and highly like the prediction of promoter position in the gene sequence in different HMA3 homologs across the 11 plant species. The Arabidopsis thaliana HMA3 showed three different positions of promoter marginal predicted at 200, 700, and 1800 bp (Table 3). Highly predicted position of promoters are located in 2200 bp, 2100 bp and 2500 bp in XM_006303850.2 (Capsella rubella), XM_006412685.2 (Eutrema salsugineum), XM_010550291.1 (Tarenaya hassleriana), respectively (Table 3). The position of TSS varied from 22 to 36 bp if found. Also, the PolA was positioned after the coding region in all HMA3 genes showing the position at 2369–4074 bp, if found (Table 3). The identified cis-acting elements were stress, hormone, and other responsive factors. Stress responsive, anaerobic induction, and light responsive regulators were found to be the height number and most common of cis-acting elements in HMA3 genes. That projected their involvement of these activities (Table 4).

Close

Fig. 1

Gene organization of HMA3 homologs.

Export to PPT

Table 3 Organization of HMA3 genes and position features.

No.	Gene Accession	Chromosome number	Coding region	Promoter position	Position of transcriptional start site (TSS)	PolA
1	NM_119158.4	4	127-1645, 1756-2408	200 Marginal prediction700 Marginal prediction1800 Marginal prediction	24	2516
2	NM_001302990.1	4	1-2286	1700 Marginal prediction	24	2516
3	XM_006303850.2	4	38-2302	2200 Highly likely prediction	–	2453
4	XM_006412685.2	4	6-2288	1100 Marginal prediction2100 Highly likely prediction	–	2354
5	XM_013735846.1	4	1-2277	600 Marginal prediction1100 Marginal prediction1700 Marginal prediction	–	–
6	XM_018625402.1	4	71–2335	200 Marginal prediction1000 Marginal prediction1700 Marginal prediction	22	2376
7	XM_022706908.1	4	92-2341	200 Marginal prediction1100 Marginal prediction	36	2369
8	XM_009139644.3	4	123-2417	1800 Marginal prediction	–	2681
9	XM_021030625.1	2	190-4023	1000 Marginal prediction2700 Marginal prediction3200 Marginal prediction3600 Marginal prediction	–	4074
10	XM_006409021.2	2	135-4142	1100 Marginal prediction2200 Marginal prediction2600 Marginal prediction3500 Marginal prediction3900 Marginal prediction	–	4235
11	XM_010550291.1	4	155-516, 589-3535	2500 Highly likely prediction3300 Marginal prediction	–	3788

Table 4 Cis-acting element analysis of HMA3 gene promoters.

Gene Accession	Cis-acting elements
	MYC/Myc	MYB/myb	MRE	TC rich	GT1	G box/ABRE	W - box	TATC-Box	ERE	AREs	As-1/TGA	AuxRR-core	TGACG	CGTCA	TCA element	ACE	AE-Box	Box −4	GATA-Motif	GA-Motif	WUN3	WRE3	O2 Site	circadian	STRE	AT-rich element
NM_119158.4	1	4	1		2			1		2		1	1					4				1				1
NM_001302990.1	3	2					1			1	2		1	1	3			2			2	2		1
XM_006303850.2		2								1	2			2		1	1	1		1	1
XM_006412685.2	5	2		2	1	2			2	4		1				1		4		2			1
XM_013735846.1	1					4			1	2	2				1			4			1
XM_018625402.1	2	1				2	1			3		1	1		2			1			1				1
XM_022706908.1	1	1	1			1	1		2	1		1	1					2								1
XM_009139644.3	2	4	1	1		1			1	1					1			6	2	1					2	1
XM_021030625.1	1	4							1	3	2			1				1	1						1
XM_006409021.2	1	4	1		1					1	2			1			2		1					2
XM_010550291.1	4			1	2			1		1	4		1	2		1		3					1		1

Different cis- regulatory elements: MYC/Myc (Dehydration-responsive), MYB/Myb (drought-responsive), MBS (involved in drought induction), TC rich (defense and stress-responsive), GT1 (SALT-responsive), ABRE/G-box (Abscisic acid-responsive), W-box (defense responsive), TATC-Box (gibberellins responsive) ERE (ethylene-responsive), AREs (involved in anaerobic induction), As-1, and AuxRR-core, (auxin-responsive), TGACG and CGTCA (MeJA-responsive), TCA (salicylic acid-responsive), ACE, AE-box, Box-4, GATA-Motif, GA-motif, and G-Box (light-responsive), WUN3, and WRE3 (wound-responsive), O₂ site (Zein metabolism regulation), circadian (circadian control), STRE (Expression activator), and At-rich (DNA binding).

3.5

3.5 Conserved motif, sequence similarities, and phylogenetic analysis

We have used the MEME tool to search for the five most conserved motifs in identified 11 HMA3 homologs (Fig. 2). All these 5 motifs are 50 residues long located at site 11. These motifs are as follows: motif 1 (INLNGYIKVKTTALARDCVVAKMTKLVEEAQKSQTKTQRFIDKCSRYYTP), motif 2 (HPMAAALIDYARSVSVEPKPDMVENFQNFPGEGVYGRIDGQDIYIGNKRI), motif 3 (NLSHWFHLALVVLVSGCPCGLILSTPVATFCALTKAATSGFLIKTGDCLE), motif 4 (KALNQARLEASVRPYGETSLKSQWPSPFAVVSGVLLALSFLKYFYSPLEW) and motif 5 (CMZDYTEAATIVFLFSVADWLESSAAHKASTVMSSLMSLAPRKAVIAETG). Motif 1, 3 and 5 encodes pfam_fs: E1-E2_ATPase. Motif 4 is linked to freq_pat:PKC_PHOSPHO_Site, while motif 2 shows no information (Fig. 2). The HMA3 protein homologs were aligned to check the similarities of sequence across the plant species. The MTP1 proteins showed 71.7% to 100% similarities among the different plant species, in which the consensus sequence ranged from 70% to 100% (Supplementary Fig. S2). The phylogenetic tree was clustered into four groups (A, B, C, D) based on tree topologies (Fig. 3). In cluster A, HMA3 of Arabidopsis thaliana formed a cluster with the Camelina sativa and Capsella rubella, while group B consist of HMA3 protein homologs of Brassica oleracea var. oleracea, Raphanus sativus, Brassica napus and Brassica rapa. The HMA3 of Eurema salsugineum clustered alone is located in the distance from Arabidopsis thaliana homolog. The cluster D consists of Arabidopsis lyrata subsp. Lyrata HMA4, Eurema salsugineum HMA4 and Tarenaya hassleriana HMA3 (Fig. 3). In this phylogenetic tree, HMA3 of Arabidopsis thaliana, Brassica oleracea var. oleracea, Raphanus sativus, Brassica napus and Brassica rapa showed the highest 100% bootstrap value (Fig. 3).

Close

Fig. 2

Schematic representation of the 5 conserved motifs in 11 HMA3 protein homologs across 11 plant species. Scale bar corresponds to 0.1 amino acid substitution per residue. Different motifs, numbered 1–5, are displayed in different colored boxes.

Export to PPT

Close

Fig. 3

Phylogenetic trees of HMA3 protein homologs. Trees were constructed by MEGA 6 software with the maximum likelihood (ML) method for 1000 bootstrap values. Trees were used as benchmarks to analyze the clustering of 11 HMA3 sequences. A, B, C and D represents different clusters.

Export to PPT

3.6

3.6 Predicted interaction partner analysis

Interactome analysis was performed for AtHMA3 (AT4G30120) on STRING server. STRING showed five closely associated putative interaction partners of AtHMA3. These include MTPA2 (metal tolerance protein A2), ZAT (a member of the zinc transporter and cation diffusion facilitator), NRAMP3 (natural-resistance-associated macrophage protein 3), IRT1 (iron-regulated transporter 2) and NRAMP4 (natural-resistance-associated macrophage protein 4) genes (Fig. 4). Further, the analysis showed four local network clusters, including CL:28166 (nickel transport and cation efflux protein), CL:28164 (manganese ion transport and nickel transport), CL:28176 (ion influx/efflux at the host-pathogen interface), CL:28126 (transition metal ion transmembrane transporter). Lastly, reactome pathways of AtHMA3 include ion influx/efflux at host-pathogen interface, zinc efflux, and compartmentalization by the SLC30 family, peptide hormone metabolism, insulin processing, and metal ion SLC transporters (Fig. 4).

Close

Fig. 4

Gene interaction partners and gene network analysis of AtHMA3 and its homologs. Interactome was generated using Cytoscape for STRING data.

Export to PPT

3.7

3.7 Expression profiles of HMA3

The genevestigator analysis against the Affymetrix Array Platforms showed expression potential and co-expression data of HMA3 in different anatomical parts, developmental stages, and perturbations. In the anatomical part, lateral roots, cauline leaf, silique inflorescence, and radicle in shoot apex seemed to be highly potential for HMA3 expression (Fig. 5a). Further, giant root cell and sperm cells in cell culture have the potential for HMA3 expression (Fig. 5a). Besides, HMA3 has expression potential during senescence, germinated seed, seedling, young rosette, bolting, and young flower stages of development (Fig. 5b). Among the selected stress, HMA3 only showed significant upregulation under Fe deficiency; while the expression did not notably vary in other stresses, such as anoxia, cold, drought, gamma irradiation, genotoxicity, heat, osmotic stress, salt stress, shift cold stress, submerge stress, wounding stress (Fig. 5c).

Close

Fig. 5

Expression profiles of AtHMA3 in different anatomical part, developmental stage and perturbations in Genevestigator Affymetrix platform.

Export to PPT

Co-expression analysis was filtered to five closely associated genes in different anatomical parts, developmental stages and perturbations (Fig. 6). In the anatomical part, the AtHMA3 gene is closely co-expressed with AT1G30560 (putative glycerol-3-phosphate transporter), AT1G63550 (cystein-rich repeat secretory protein 9), AT3G60270 (cupredoxin superfamily protein), AT5G43370 (probable inorganic phosphate transporter 1–2) and AT3G12900 (2-oxoglutarate (2OG) and Fe(II)-dependent oxygenase superfamily protein). During the development stage, AT3G56891 (heavy metal transport/detoxification superfamily protein), AT5G19040 (adenylate isopentenyltransferase 5, chloroplastic), AT3G45410 (L-type lectin-domain containing receptor kinase I.3), AT3G29250 (short-chain dehydrogenase reductase 4) and AT2G25260 (unknown protein) co-expressed with AtHMA3 (Fig. 6). Under perturbations, the top five genes co-expressed with AtHMA3 are AT1G64480 (calcineurin B-like protein 8), AT5G17100 (cystatin/monellin superfamily protein), AT5G65980 (auxin efflux carrier family protein), AT2G28690 (protein of unknown function, DUF1635), and AT2G12190 (cytochrome P450 superfamily protein) (Fig. 6).

Close

Fig. 6

List of top 5 genes co-expressed with AtHMA3 in different anatomical parts and developmental stages of plants.

Export to PPT

4 Discussion

In recent years, the characterization of membrane transporters involved in heavy metal scavenging in plants is emerging. Prior to the wet-lab experiment, the in silico analysis is of utmost interest to narrow down the target of studies. The role of AtHMA3 in vacuolar storage of a few metals is documented (Morel et al., 2009), although the involvement of other metals and dynamic network of other associated genes and gene/protein properties are yet to extensively studies. This in silico characterization and expression profile AtHMA3 and its homologs and closely associated genes unveil significant regulatory findings that can be essential contributors to the downstream genome-editing or biotechnological approach to heavy metal studies.

In this study, we selectively blasted the AtHMA3 sequences resulted in 11 different HMA3 protein homologs having 71.73–100% percentage identity. The similarities in protein size, pI, instability index and hydrophilicity suggest that these HMA3 proteins are biochemically relevant. Domain analysis further revealed the association of these HMA3 protein homologs with E1-E2 ATPase (PF00122) localized in the plasma membrane. The proton-pumping ATPase (H⁺-ATPase) in the plasma membrane produces the proton motive force through the plasma membrane that is required to enable much of the transport of ions and metabolites (Morsomme et al., 2000). However, HMA3 proteins showed diverse cellular components in HMA3 homologs, the unique feature of these is limited to the membrane, vacuolar membrane and plasma membrane. These cellular components are crucial in mineral absorption and metal homeostasis along with salt tolerance, intracellular pH regulation and cellular expansion in plants (Morsomme et al., 2000; Logan et al., 1997). However, HMA3 proteins are predominantly associated with cation transport, metal ion transport, zinc ion transport and cadmium ion transport, as evident from our ontology analysis. HMA3 gene is involved in cadmium and lead transport along with vacuolar sequestration potentiality in a heterologous system, but not in zinc transport. Vacuolar sequestration may have a detoxification function (Gravot et al., 2004).

In predicting evolutionary relationships and functional genomics possibilities, the knowledge on the position and organization of the coding sequence of a gene is considered a critical factor. In this study, all of the identified sequences of HMA3 proteins demonstrated 4 transmembrane helices confirming similar hydropathy of these HMA3 protein homologs. The metal specificity of each subclade is determined by specific amino acids in the three transmembrane helices closest to C-terminus (Raimunda et al., 2012). In this study, all HMA3 gene homologs belonging to the 11 plant species showed 8–9 exons, suggesting that these HMA3 genes are evolutionarily closer to each other. Although promoter analysis predicted several promoter regions of each HMA3 gene, the highly likely prediction of the promoter was made at 2100–2500 bp in several plant species. Localization of exon and promoter plays an essential part in CRISPR-Cas9 and other genome editing studies in plant science (Yan et al., 2017). In addition, the identification of TSS and PolA in HMA3 homologs will be crucial in understanding the transcriptional and translational genomics. Besides, promoter analysis reveals the involvement of cis-acting elements associated with stress response, hormone, anaerobic induction, and light-responsive regulators in HMA3 genes. The architectures of 1000 bp promoter regions upstream of transcription start sites were determined using in silico analysis (New et al., 2015). In the light of our findings, the frequency with which these cis-elements appeared in each promoter varied, and the number of clusters within each promoter's core-regulatory area could indicate how susceptible each gene was to HMA regulation in plants.

Conserved motifs are identical sequences across species that are maintained by natural selection. A highly conserved sequence is of having functional roles in plants and can be a useful start point to start research on a particular topic of interest (Wong et al., 2015). Out of the five motifs, three motifs are mainly matched with the E1-E2_ATPase associated with H⁺ pumping. P-type proton ATPase is found in the plasma membranes of plants that, in turn, drives secondary active transport processes across the membrane (Palmgren et al., 2001). One of the motifs is also linked to the protein kinase C phosphorylation site that may play roles in controlling the catalytic activity, stability and intracellular localization of the enzyme (Freeley et al., 2011). Further, the phosphorylation site may be attributed to the release of Zn from intracellular stores leading to phosphorylation kinases and activation of signaling pathways (Thingholm et al., 2020). The presence of common and long-preserved residues suggests that HMA3 homologs between species may have highly conserved structures. Additionally, for sequence-specific binding sites and transcription factor analysis, this information can be targeted. In phylogenetic analysis, HMA3 protein of A. thaliana positioned in the same cluster with C. sativa and C. rubella, suggesting its close relationship during the evolutionary trend. Consistently, HMA3 protein homologs of Brassica sp. and Raphanus sp. clustered within B, suggesting the close evolutionary emergence from a common ancestor within the Brassicaceae family. It appears that the HMA3 of E. salsugineum is relatively distantly related to A. thaliana over the evolutional trends. Thus, our results might infer a functional relationship of HMA3 sequences in metal uptake across different plant species.

The interaction network of a specific gene provides information of all physical associations that can occur among family members. Global gene co-expression analysis is an emerging tool to identify the tissues and the conditions in which significant interactions occur. The interactome map analyzed in String platform showed the most close association with MTPA2, ZAT, NRAMP3, IRT2 and NRAMP2, mainly linked to metal transport in plants. Consistently, the local network of AtHMA3 implies the involvement with metal transporter. As a result, these findings might be useful to characterize HMA3 and to interpret the interactions of multiple genes linked to particular stress of interest in plants. Studies reported that Zn homeostasis is closed associated with P-type ATPase heavy metal transporters (HMA). Again, both HMA2 and HMA4 were reported to be involved with Zn homeostasis in Arabidopsis (Hussain et al., 2004). Besides, AtHMA3 showed some reactome pathways, among which ion influx, zinc influx, and metal ion SLC transporters may attribute to the metal transporter properties of this gene. Overall, this interactome finding might provide essential background for functional genomics studies of metal uptake and transport in plants.

The expression potential of a gene in different conditions is a crucial factor in determining the involvement in a particular trait. The in silico expression analysis in the Genevestigator platform showed interesting outputs concerning the expression of AtHMA3 (AT4G30120) in different anatomical, perturbations, and developmental stages. Being consistent with the AtHMA3 gene ontology, Genvestigator showed that the root is the important location where this gene showed expression potential. In a wet-lab experiment, root-specific expression of HMA3 was reported in rice (Cai et al., 2009). Further, AtHMA3 is most potentially expressed during senescence, but germinated seeds, seedlings, young rosette, bolting and young flower also possess significant potential for AtHMA3 expression. Interestingly, AtHMA3 showed a significant upregulation (>6.0 fold) in response to Fe-deficiency. Till now, HMA3 is known to induce its expression subjected to heavy metals in several plant species (Yao et al., 2018). Nevertheless, our results suggest that AtHMA3 is a potential gene that could contribute to Fe-deficiency tolerance in plants.

5 Conclusion

This in silico work identifies and characterizes 11 HMA3 homologs from each plant species. The analysis showed similar physicochemical properties, gene organization, and conserved motifs related to metal transport. The identified cis-acting elements were linked to stress response, hormone, and other responsive factors. Sequence homology and phylogenetic tree showed the closest evolutionary relationship of Arabidopsis HMA3 with Camelina sativa and Capsella rubella. In addition, the interactome map displayed some partner genes of AtHMA3 involved in metal transport in plants. It was also predicted that AtHMA3 is expressed in root tissue during senescence and was significantly upregulated in response to Fe-deficiency. These findings will provide basic theoretical knowledge for the downstream studies on HMA3 function and characterization related to metal homeostasis in various plants.