Introduction

As a microelement, iron (Fe) is an indispensable element for plants to sustain their metabolic processes such as enzymatic activities, respiration and photosynthesis (Marschner 1995). The deficiency or toxicity of iron impedes plant growth and development resulting in loss of crop yield and quality. Particularly, the iron toxicity gives rise to formation of reactive oxygen species (ROS) as a result of fenton reactions which may have lethal effects for cells (Thomine and Vert 2013). Therefore, to shed light on the metabolic pathway of iron and identification of genes and transcription factors who are responsible for uptake of iron in plants are crucial for plant growth, development and improvement of biofortified crops.

Plants have two distinctive mechanisms to take up iron from soils: Strategy I and II. Strategy I is based on reduction of iron from ferric iron (Fe+3) to ferrous iron (Fe+2) to solubilize iron complexes, used by the plant species, nongraminaceous (non-poaceae or non-grass) family, when soil pH is neutral. On the other hand, the chelation of iron, strategy II, is related to formation of Fe chelates through phytosiderophores to generate soluble Fe complexes by graminaceous plants (Brumbarova et al. 2014; Liang et al. 2017). For nongraminaceous plants to solubilize iron during iron deficient conditions, soil pH in rhizosphere should be reduced by upregulation of AHA2 proton pump gene. Solubilization of iron is followed by activation of ferric-iron reductase gene which reduces ferric iron to ferrous iron. Once ferrous iron was formed, it was transported into epidermis cells of roots through iron regulated transporter 1 (IRT1) (Colangelo and Guennoti 2004; Sivitz et al. 2012; Brumbarova et al. 2014, Vatansever et al. 2015; Liang et al. 2017). Contrary to nongraminaceous plants like Arabidopsis thaliana, the iron uptake by graminaceous is dependent on formation of phytosiderophores which chelate ferric iron (Colangelo and Guennoti 2004). After chelation, chelate bound iron was carried into apoplast of root epidermis through IRT1 (Brumbarova et al. 2014). IRT1 also reported to be involved in Fe uptake in above-ground tissues of plant (Kim and Guerinot 2007). Regulation of iron uptake genes in Arabidiopsis which are responsible for acquisition of iron from soil during iron limited conditions is modulated by FIT (FER-like Iron-deficiency-induced Transcription Factor), a bHLH protein (Liang et al. 2017). Additionally, another bHLH transcription factor, POPEYE (PYE), is positively regulated under iron deficiency and control mobilization of iron between plant organs (Long et al. 2010; Brumbarova et al. 2014). FIT was assumed to form heterodimers with bHLH38, bHLH39, bHLH100 and bHLH101 to modulate different subset genes and the expression of AHA2, FRO2 and IRT1 genes were thought to be induced by FIT under low-iron conditions (Celma et al. 2012). However, increasing evidences suggested that the upregulation of bHLH038, bHLH039, bHLH100, bHLH101, PYE, NRAMP4, ORG1, PYE, NAS4 were FIT independent (Wang et al. 2007; Sivitz et al. 2012).

A basic helix-loop-helix (bHLH) was first discovered in animals and later was identified in most of eukaryotic organisms. As a transcription factor, bHLH super family proteins have crucial role in many regulatory and development processes in animals and plants. This protein family is approximately 60 amino acids long and has two conserved and distinct domains, basic region and helix-loop-helix. The basic region is composed of 10–15 amino acids while HLH is made up of 40 amino acids (Pires and Dolan 2010; Filiz et al. 2017). The binding of HLH region to DNA is modulated through basic domain whereas formation of dimeric structures such as homo or hetero dimers through interaction of bHLH proteins is regulated by HLH region (Murre et al. 1994). The binding of bHLH proteins to DNA take place with recognition of hexanucleotide sequence defined as E box (Toledo-Ortiz et al. 2003). Palindromic G-box is the most common form of E box where is recognized by many Ib subgroup bHLH proteins (Jones 2004).

The studies of characterization of bHLH proteins in Arabidopsis showed that there are as many as 139 or 147 bHLH genes (Toledo-Ortiz et al. 2003; Heim et al. 2003). bHLH genes regulate different transcription factors and induce upregulation of numerous genes. For example, bHLH34 and bHLH104 modulates continuation of Fe deficiency response in Arabidopsis (Li et al. 2014). Besides, bHLH115, bHLH38, bHLH39, bHLH100 and bHLH101 also play important role in Fe homeostasis. Therefore, in this study, Ib bHLH genes/proteins of bHLH38, bHLH39, bHLH100 and bHLH101 were analyzed comparatively at genome-wide scale in Arabidopsis, tomato, rice, maize and soybean genomes using bioinformatics tools to have more information of these genes in plant metabolism, particularly in iron homeostasis.

Materials and methods

Retrieving bHLH gene sequences

At3g56970 (bHLH38), At3g56980 (bHLH39), At2g41240 (bHLH100), At5g04150 (bHLH101) genes were retrieved from UniProtKB database (uniprot.org/) and supplied to Phytozome v12.1.4 for Blastp analysis to identify Ib bHLH genes against rice, tomato, soybean and maize (phytozome.jgi.doe.gov/pz/portal.html; Goodstein et al. 2012). Protein domains were queried in Pfam 31.0 database (http://pfam.xfam.org/; Sonnhammer et al. 1997).

Sequence and conserved motif analyses

Coding sequence (CDS), exon and intron organization, peptide sequences, the length of amino acid residues, chromosome numbers of bHLH genes were retrieved from Phytozome database (phytozome.jgi.doe.gov/pz/portal.html; Goodstein et al. 2012). ProtParam server was employed to find physio-chemical features of bHLH genes in this study (http://web.expasy.org/protparam/; Gasteiger et al. 2005). CELLO server was used prediction of subcellular localization of bHLH genes (http://cello.life.nctu.edu.tw/; Yu et al. 2006). Conserved motifs were analyzed using MEME server with five motifs having 5–60 motifs wide as maximum and minimum motif width respectively (meme-suite.org/tools/meme; Timothy et al. 2009).

Phylogenetic analysis

Alignments of Ib bHLH proteins were made in Bioedit V7.0.5 with Clustal W method (Hall 1999). Phylogenetic tree and distance matrix table of bHLH genes were generated in MEGA 7 using maximum likelihood (ML) method, based on James-Taylor-Thornton (JTT) model and 1000 bootstrap, and pair wise method respectively (Kumar et al. 2016).

Promoter sequence and co-expression analyses

Cis-regulatory elements were found by using 1000 bp upstream sites of bHLH genes in Phytozome and supplied to PlantCare (http://www.bioinformatics.psb.ugent.be/webtools/plantcare/html/; Lescot et al. 2002). The digital expression analyses of bHLH genes were performed by using microarray data from Genevestigator database (https://genevestigator.com/gv/index.jsp; Hruz et al. 2008). ATTED-II plant co-expression database (http://atted.jp) was used to construct co-expression network of bHLH genes in Arabidopsis (Aoki et al. 2016).

3D structures of bHLH proteins

3D structures of studied proteins were predicted by Phyre2 server at intensive mode (sbg.bio.ic.ac.uk/phyre2/; Kelley et al. 2015). 3D structures of bHLH38/39/100/101 proteins were aligned through CLICK server by superimposing protein pairs to have alignment details (mspc.bii.a-star.-edu.sg/minhn/; Nguyen et al. 2011). Quality assessments of 3D structures were made by Vadar server (http://vadar.wishartlab.com, Willard et al. 2003). Molecular cavities of 3D structures of bHLH proteins were computed with Beta-Cavity server using beta-complex, a construct derived from Voronoi diagram of atoms (http://voronoi.hanyang.ac.kr/betacavityweb/about.html; Kim et al. 2015).

Results and discussion

Identification of bHLH38, bHLH39, bHLH100 and bHLH101 genes

To identifty Ib subgroup bHLH genes in rice, tomato, soybean and maize, amino acid sequences of bHLH38/39/100/101 in Arabidopsis were used as reference sequences. A total of 13 genes were found at the end of investigation for rice (one), tomato (three), soybean (four) and maize (one) (Table 1). The shortest open reading frame (ORF) was detected in Glyma.19G132600.1 (354 bp) whereas Solyc10g079660.1.1 had the longest ORF (2094 bp). The polypeptide chains length of bHLH genes ranged from 117 to 697 amino acids with 13412.40–78715.51 kDa molecular weights. The maximum number of exons was identified as 13 in Solyc10g079660.1.1 whereas the number of exons of other queried genes ranged from two to four. All genes had the same protein domain structure, helix-loop-helix DNA binding domain (PF00010). Theoretical isoelectric points (pI) of genes showed a wide variation ranging from 5.63 to 9.77. Sub-cellular localization of putative genes was usually predicted in nuclear but Solyc10g079660.1.1 gene may also be in plasma membrane. Niu et al. (2017) stated that bHLH genes in Brachypodium distachyon (BdbHLH) were orthologous in rice, maize and sorghum and it showed a wide range of variation for pI, polypeptide chains length, and molecular weight. Similarly, bHLH genes in Salvia miltiorrhiza and Arabidopsis were reported to have similarity in number and pI values of Salvia miltiorrhiza were between 4.8 and 9.9 (Zhang et al. 2015b). Collectively, the four Ib subgroup bHLH genes had variation in terms of studied properties of proteins and genes indicating functional diversities of bHLH genes.

Table 1 The details of bHLH genes/proteins in Arabidopsis, rice, tomato, soybean and maize
Full size table

Conserved motifs of Ib bHLH proteins

Based on the results of conserved motif analysis of MEME server, the most conserved five motifs of bHLH proteins were shown in Table 2. Motif widths ranged from 21 to 50 amino acids. The motif 1 was detected in all proteins and was related to the HLH domain family. All five motifs were detected in Glyma.03G130600.1, Glyma.03G130400.1 and in Glyma.19G132500.1 whereas Glyma.19G132600.1 had only two motifs, motif 1 and motif 3, respectively. Apart from motif 1, other motifs showed different distribution in bHLH proteins (Supplemental Fig. 1). Furthermore, conserved motifs of bHLH38/39/100/101 proteins in queried species were analyzed by multiple sequence alignment program of Clustal W in Bioedit. As a result, 22 amino acids in bHLH regions were detected as shown in Fig. 1. The conserved residues in this study were lysine (K-Lys), leucine (L-Leu), histidine (H-His), asparagine (N-Asn), alanine (A-Ala), glutamic acid (E-Glu), arginine (R-Arg), serine (S-Ser), proline (P-Pro), threonine (T-Thr), tyrosine (Y-Tyr) and isoleucine (I-Ile). Interestingly motif 1, found all examined genes, was only lack of Thr, instead of having Aspartic acid (Asp-D). Plants were reported to have more conserved Ile-20, Asn-21, Leu-24, Gln (Glutamine-Q)-28, Lys-36, Asp-38, Ile-43, Val (Valine-V)-51 and Leu-54 amino acid sequences in bHLH regions (Niu et al. 2017). However, this suggestion differed from our results in terms of presence of conserved Gln, Asp and Val amino acids. Glu-13, Arg-14, Arg-16 and Leu-27 amino acid residues were reported to be conserved in rice, B. distachyon and Arabidopsis (Niu et al. 2017). Furthermore, Arg-16, Leu-27, and Leu-61 amino acids were found to have highly conserved in tomato (Sun et al. 2015). Hudson and Hudson (2015) reported that soybean had all but one of bHLH subfamily proteins and its bHLH subfamily proteins had high coherence with Arabidopsis. Consequently, they had more similar functionality in metabolic processes. Glu-13 and Arg-16/Arg-17 amino acid residues in the basic region of bHLH were suggested to be involved in DNA binding whereas Leu-27 and Leu-61 residues in helix region was reported to act in dimerization (Sun et al. 2015). Similarly, Pires and Dolan (2010) reported that Ile, Leu and Val amino acids were conserved in most of plant and animal bHLH proteins and stabilization of dimerization were provided by these hydrophobic residues. Also, the first helix was reported to be broken by conserved Pro amino acid. In summary, bHLH proteins in this study conserved Ile, Glu, Arg, and Leu amino acids in their HLH regions and these residues might be involved in acquisition of metal ions, particularly iron. Additionally, Met (methionine) and Val were rather found in basic regions of bHLH genes but not in all plant species in this study.

Table 2 The bHLH proteins with the most conserved five motifs
Full size table
Fig. 1
figure 1

Multiple sequence alignment of amino acid sequences of bHLH 38/39/100/101 of four species such as Arabidopsis, rice, tomato, soybean, and maize. Conserved motifs of sequences were framed in red and blue colors, respectively. The residues of motif 1 were shown in red frame whereas residues of unidentified motif were pointed out with blue rectangular. Identical residues were shown in black columns while similar residues were in gray shading. (At: A. thaliana; LOC: O. sativa, Solyc: S. lycopersicum, Glyma: G. max, GRMZM: Z. mays)

Full size image

Phylogenetic analysis

Phylogenetic tree of bHLH proteins were shown in Fig. 2a. As can be seen, phylogenetic analysis was not revealed a clear distinction between monocot and dicot plants. bHLH proteins were dived into two main groups, group A and B. Group A was composed of two subgroups, A1 and A2. Interestingly, Arabidiopsis bHLH101 was separated from Arabidopsis’ other proteins and clustered together with tomato with 85% bootstrap value in subgroup A1. This may be explained by similarity rates between bHLH proteins whereby blastp of bHLH38 were showed 89.5, 73.0 and 67.8% similarity with bHLH39/100/101 genes in question, respectively. Similarly, Sivitz et al. (2012) reported that bHLH38 and bHLH39 are tandemly situated in chromosome 3 with 79% identity and 89% similarity, respectively. However, bHLH100 and bHLH101 are located in different chromosomes and they are 39% identical and 69% similar to each other. To better shed light on this separation we constructed a new phylogenetic tree including BRUTUS (BTS) and PYE genes (Supplemental file, Fig. 2). As a result, bHLH101 was classified into the same cluster with BTS and PYE proteins along with Soly10g079680.1.1. This result showed that bHLH101 may be outparalogous to PYE, BTS and Soly10g079680.1.1 genes. Additionally, the internal cluster of tomato was found between Soly10g079650.1.1 and Soly10g079660.1.1 with 91% bootstrap value. The duplication event in tomato bHLH genes may be resulted in this paralogous homology. As for Soly10g079680.1.1 and bHLH101, they were orthologous to each other. On the other hand, bHLH100 separated from bHLH38/39 at 87% bootstrap in A2. Whereas all bHLH proteins of dicot soybean were observed in B2, the monocot plants such as LOC_Os01g72370.1 and GRMZM2G057413_T01 were grouped into B1 with 99% bootstrap value.

Fig. 2
figure 2

Phylogenetic tree (a) and exon–intron distribution (b) of bHLH38/39/100/101 genes/proteins of Arabidopsis, tomato, rice, maize and soybean. The tree was generated in MEGA 7.0 software using maximum likelihood (ML) method with 1000 bootstrap replicates. The yellow filled bars and black lines show the exon and intron structures, respectively in Fig. 2B

Full size image

Gene structure analysis

In terms of exon–intron patterns of bHLH genes (Table 1 and Fig. 2b), At5g04150 (bHLH101) Solyc10g079650.1.1 and Solyc10g079680.1.1 had two introns while intriguingly Solyc10g079660.1.1 had 12 intron regions. Hudson and Hudson (2015) stated that three intron regions were conserved in majority of bHLH genes. Similarly, the bHLH intron numbers in tomato were reported to be ranged from null to three and differed from those in Arabidopsis in spite of conserved regions of bHLH protein domains (Sun et al. 2015). In this study all bHLH genes, except At5g04150, in Arabidopsis had one intron. GRMZM2G057413_T0 along with all soybean genes had two introns whereas three introns were observed in LOC_Os01g72370.1. Soybean intron structure in bHLH subgroups were reported to be coherent with those in rice and Arabidopsis. In this respect, bHLH family showed null to three conserved introns showing nine patterns depending on intron positions in soybean (Hudson and Hudson 2015). Also, the relationship of bHLH genes in Arabidopsis, rice, tomato, maize and soybean was complicated. The intron regions of bHLH genes in this study showed coherence with the results of other studies of bHLH genes, except Solyc10g079660.1.1. In addition, the distinction of bHLH genes in monocot and dicot were not found in phylogenetic tree. According to gene structures and phylogenetic relationships of studied bHLH genes, it can be suggested that relationships of bHLH genes in Arabidopsis, rice, tomato, maize and soybean was complicated.

Promoter region analysis

Cis-acting elements are located in the upstream of gene coding regions within in gene promoters and the regulation of transcription is related to finding motifs in promoter region (Garcia and Finer 2014). To identify cis-regulatory elements related to bHLH genes, PlantCare database was supplied with 1000 bp upstream DNA sequences of each bHLH genes in this study. As a result, a total of 61 cis-elements were identified and a heatmap were constructed to give a better insight into cis-regulatory elements in associated with bHLH genes. Cis-regulatory elements whose functions are unknown were excluded from the Heatmap (Fig. 3) and absent or present elements were shown in red and green colors, respectively. The highest number of cis- elements was identified in GRMZM2G057413_T01 and Glyma.03G130600.1 genes. In this sense, CAAT and TATA boxes were found in all bHLH genes of each species investigated. Skn-1_motif was present, except LOC_Os01g72370.1, in all genes under investigation. TATA box is part of core promoter region of protein coding genes and is a binding site for TATA binding protein (TBP) which make up transcription initiation factor (TFIID) (Garcia and Finer 2014). CAAT box is a binding site for transcription complex with which modulation of transcription of genes occur (Laloum et al. 2013). Skn-1 motif is required for endosperm expression and has an important role with other cis-acting elements such as 5UTR-Py-rich stretch, ABRE, Box-4, G-Box, MBS in the upregulation of genes acting in in oxidative defense pathway in rice (Yousefi et al. 2012). G-Box and G-box were found in more than half of bHLH genes in the study, showing agreement with the report that G-box (CACGTG) is a common motif for bHLH and bZIP protein families (Wong et al. 2017). G box is also identified as a MYC recognition site along with MYB, having role in upregulation rd22 drought-induced gene. Moreover, bHLH associated protein AtMYC2 and MYB connected protein AtMYB2 together binds to MYB and MYC binding sites to upregulate ABA inducible genes during drought stress. ABRE sequence was reported to be a promoter region for ABA inducible genes (Abe et al. 2003). In parallel to these reports, MBS and ABRE elements also were found as cis-elements in this study, involving in MYB binding site involved in drought-inducibility and ABA responsiveness respectively. Apart from these findings, the other promoter sequences, having role in hormone metabolism for bHLH genes, were found as TCA-element (salicylic acid responsiveness), CGTCA-motif (the MeJA-responsiveness), TGA-element (auxin-responsive element), TATC-box (gibberellin-responsiveness), CE3 (cis-acting element involved in ABA and VP1 responsiveness), TGACG-motif (involved in the MeJA-responsiveness), GARE-motif (gibberellin-responsive element) and ERE (ethylene-responsive element). Also, a considerable number of cis-regulatory elements for bHLH genes in the study involved in light responsiveness (Box I, Box 4, AE-box, GA-motif, TCT-motif, GT1-motif,I-box,L-box, Sp1, TCCC-motif, TCCC-motif, as-2-box, ACE, chs-CMA1a, MNF1, rbcS-CMA7a, ATCT-motif, CATT-motif, MRE, AAAC-motif, GATA-motif, chs-Unit 1 m1, Box II, 3-AF1 binding site, AT1-motif, C-box, GTGGC-motif). The other cis-regulatory elements of bHLH genes can be grouped into three classes: tissue-specific, stress responsive elements and other elements. Cis-regulatory elements, revealed in the abiotic stress responsiveness, were ARE (anaerobic induction), LTR (low-temperature responsiveness), TC-rich repeats (defense and stress responsiveness), GC-motif (enhancer-like element involved in anoxic specific inducibility) and HSE (heat stress responsiveness). As for tissue-specific cis-regulatory elements, they were identified as CCGTCC-box (related to meristem specific activation), RY-element (seed-specific regulation), CAT-box (meristem expression), as1 (root-specific expression), AACA_motif (endosperm-specific negative expression) and GCN4_motif (endosperm expression). Lastly, several cis-regulatory elements of bHLH genes in this study were identified functioning in fungal elicitor responsiveness (Box W1), regulation of circadian control (circadian) and regulation of zein metabolism (O2-site). Overall, the diverse functions of cis-regulatory elements of bHLH genes in various metabolic pathways have shown a complex system which regulates numerous metabolic processes in investigated plants.

Fig. 3
figure 3

The heatmap of cis-elements in bHLH38/39/100/101 genes from Arabidopsis, rice, soybean, tomato and maize, respectively (green: present and red: absent)

Full size image

Co-expression network analyses of bHLH genes in Arabidopsis

The four A. thaliana genes such as bHLH38 (At3g56970), bHLH39 (At3g56980), bHLH100 (At2g41240) and bHLH101 (At5g04150) were investigated in ATTED-II co-expression database to shed a light on interactions among them and on their roles as transcription factors in synthesis of other proteins which modulates numerous biological processes. For this purpose, an interactome map was constructed (Fig. 4). At3g56980 and At5g04150 genes were found in co-expressed gene network whereas no co-expression data was found for At3g56970 and At2g41240. At3g56980 (ORG3 or OBP3-responsive gene 3), At5g04150, At3g56970 (ORG2 or OBP3 responsive gene 3) and At2g41240 was reported to have functions in the regulation of transcription iron ion homeostasis and cellular response to iron ion starvation (Xiang 2015). At2g41240 was suggested to be expressed in response to drought stress (Rasheed et al. 2016). In co-expression network, At3g56980 (ORG3) putatively interacts with At5g04150 (bHLH101, another transcription factor), At1g23020 (FRO3) and At1g47400 (841144) gene, which has currently not been annotated. At1g23020 (FRO3) is defined as iron ion transport, protein involving in oxidation–reduction process and cellular response to iron ion starvation (Xiang 2015). Furthermore, this gene was reported to encode superoxide-producing enzyme NADPH oxidase (Rizhsky et al. 2003). NADPH oxidase, also known as respiratory burst oxidase homologs (Rbohs), named AtRohA to J, has 10 members in Arabidopsis, each of them have different roles and expression patterns. For example, AtRbohD and AtRbohF were stated to regulate numerous abiotic stresses and pathogen defense response (He et al. 2017). As for At5g04150 (bHLH101), it supposedly interacts with At3g56980 transcription factor and At1g12030 (DUF 506), At5g05250 (830407), At2g30760 (817627) and At1g47400 (841144) genes whose annotations are not available.

Fig. 4
figure 4

The co-expression network of bHLH101 and bHLH38 (ORG3) genes in Arabidopsis by ATTED-II server

Full size image

There are two networks regulating iron uptake mechanisms: POPEYE (PYE) network and FIT network. PYE is a bHLH protein involves in redistribution of iron ions in plants (Brumbarova et al. 2014). It downregulates NICOTIANAMINE SYNTHASE 4 (NAS4), FRO3, and ZINC-INDUCED FACILITATOR1 (ZIF1) (Liang et al. 2017) and interacts bHLH105 (ILR3) and bHLH115. Meanwhile, the dimers of bHLH104 and bHLH105 or bHLH34 and bHLH105 involve in modulation of PYE (Conorton et al. 2017). bHLH105, bHLH104 and bHLH34 act in the regulation of bHLH38, bHLH39, bHLH100 and bHLH101 (Liang et al. 2017). FIT is orthologue of FER, a bHLH transcription factor in tomato with At2g28160 gene number, known also as FIT1 or FRU FIT, expressed in roots under low iron conditions and was suggested as the key regulator of iron mechanism (Wang et al. 2007). Upregulation of FRO2 or IRT1 genes were reported not to be dependent on constitutively upregulation FIT but homo or hetero-dimerization of FIT with bHLH038, bHLH039, bHLH100, bHLH101 genes, whose expression hindered by excessive iron medium (Xing et al. 2015; Zhang et al. 2015a). Transcription of these genes were reported to occur FIT-independent way (Wang et al. 2007; Sivitz et al. 2012; Xing et al. 2015). Furthermore, although bHLH38/39/100/101 were described as functionally redundant genes in Arabidopsis, based on results obtained from knockout experiments, the functional redundancy of bHLH101 with bHLH38 was not found in a melon mutant, named fefe, under adequate iron conditions (Ramamurthy and Waters, 2017). Similarly, another knockout experiment in Arabidopsis showed that presence of bHLH38 and bHLH101 or bHLH38 and bHLH100 genes did not affect iron uptake mechanism under iron deficiency (Wang et al. 2013). Overall, though bHLH38, bHLH39, bHLH100 and bHLH101 genes play important roles in iron uptake mechanism, bHLH39 and bHLH101 may be suggested as more crucial genes regulating metal homeostasis, particularly iron uptake and regulation metabolism in plants.

3D structures of bHLH proteins

Understanding or prediction of 3D structure of a protein is crucial to know how it functions in any processes in metabolic pathways. Comprehension the structures of bHLH38/39/100/101 proteins will give insights into their features such as conformational flexibilities, allosteric regulations, catalytic activities and posttranslational modifications (Eisenhaber 2006). In this respect, 3D structures of bHLH38/39/100/101 proteins were constructed by Phyre2 server (Fig. 5). Firstly, the seconder structure analyses and quality assessment of protein models were made by Vadar server. The seconder structure analyses revealed that bHLH proteins varied 22–42% for α-helices, 0–16% for β-strand, 48–61% for coils, and 4–22% for turns. Furthermore, Ramachandran analysis showed that 81 and 94% of residues were in allowed region, indicating that the quality of models was good. The 3D models were superimposed as pairs to reveal their similarities and divergences based on overlap values. The calculations were made superimposition of each bHLH (bHLH 38/39/100/101) proteins of Arabidopsis against bHLHs in other plant species in this study. Also, to shed light on phylogenetic relationship of bHLH proteins, the bHLH proteins under the same internal clad in phylogenetic tree was superimposed to each other (refer to phylogenetic section). Thus relying on overlap values, showing similarities and divergences of 3D structures, how functions and structures of these proteins during evolutionary process were changed was estimated (Supplemental file Table 2). In this respect, 49.57, 42.98 and 39.38% overlap values were obtained by superimposition of At2g41240 (bHLH100) with Glyma.19G132600.1, Solyc10g079650.1.1 and Solyc10g079680.1.1 respectively. The superimposion of At3g56970 (bHLH38) with Glyma.19G132600.1, At5g04150 (bHLH101) and Glyma.19G132500.1 showed 52.99, 38.75 and 37.92% similarity respectively. The 47.86, 45.60 and 36.10% overlap values were obtained when At3g56980 (bHLH39) superimposed Glyma.19G132600.1, Solyc10g079680.1.1 and Glyma.03G130600.1 respectively. At5g04150 (bHLH101) showed highest similarities with Glyma.19G132600.1 (51.28%), Glyma.19G132500.1 (40.83%) and At3g56970 (bHLH38) (%39.17). These results showed that Arabidopsis bHLH38/39/100/101 were structurally most similar to Glyma.19G132600.1. However, analysis of phylogenetic tree and proteins’ structures were not coherent. Though these overlap values found low, the overlap values of Arabidopsis genes to other bHLH genes in this study were still above the twilight zone (> 30%). In the meantime, to present 3D molecular structures of studied proteins better the molecular cavities’ functions were computed (Fig. 5). Properties of cavities (or voids) along with channel numbers also were reported to be important parameters to determine the interactions of polypeptide with other molecules in their environment (Kim et al. 2015). According to channel numbers, bHLH proteins showed a considerable variation ranged from 3 to 11 in all species investigated (Supplemental file Table 3). All data considered, 3D structural variation may be explained by functional diversities of bHLH genes in plant metabolic processes.

Fig. 5
figure 5

The predicted 3D structures of bHLH proteins by Phyre2 server. The putative channels were predicted by Beta-Cavity server and shown as red on 3D structures

Full size image

Analyses of bHLH38/39/100/101 gene expressions in Arabidopsis

The co-regulation of bHLH genes was examined in terms of developmental stages, perturbations and anatomical parts. In this respect, microarray data of each bHLH genes were retrieved from Genevestigator platform. Based on developmental stage gene expression data, bHLH38 highly expressed in flowering stage whereas bHLH39 was expressed at moderate level from seedling through flowers and siliques stage. The main increase of bHLH39 expression was observed in transition from germination to seedling stage. bHLH100 was highly expressed in flowering stage while bHLH101 expression was fluctuated between low and high levels from seedling through flowering stages but showed decrease, after flowering, through senescence stage. Interestingly, although expression levels differed, bHLH38 and bHLH100 had same similar expression patterns with regard to expression tendencies as was the case for bHLH39 and bHLH101 (Fig. 6). This data made us infer that iron should be supplied to Arabidopsis particularly in seedling and flowering stages since upregulation of four Ib subgroup bHLH genes were dependent on iron limiting stress conditions (Xing et al. 2015; Zhang et al. 2015a).

Fig. 6
figure 6

Gene expression levels of bHLH genes in different developmental stages of Arabidopsis, including germinated seed, seedling, young rosette, developed rosette, bolting, young flower, developed flower, flower and siliques, mature siliques, and senescence

Full size image

Gene expression data of bHLH genes in different anatomical parts were presented in Fig. 7. Both bHLH38 and bHLH100 genes were expressed high in adult leaves. On the other hand, bHLH39 and bHLH101 genes showed more complicated expression profiles in more anatomical parts. For example, bHLH39 gene was highly expressed in root cell and shoot stele cell along with their lower components such as radicle and hypocotyl, suggesting that iron’s involvement in photosynthesis together with maintenance of chloroplast structure and function (Rout and Sahoo 2015). All things considered, the expression patterns of bHLH genes in anatomical parts in this study showed that seed parts, having role in emergence and germination, and adult leaves were the main anatomical parts in which bHLH genes may play important roles.

Fig. 7
figure 7

The expression levels of bHLH38/39/100/101 genes in different anatomical parts of Arabidopsis

Full size image

The expression profiles of bHLH genes investigated in this study showed similar pattern to anatomical parts and developmental stages in terms of perturbation studies (Fig. 8). Depending on perturbation studies in Genevestigator, bHLH101 and bHLH39 were clearly upregulated by numerous stimuli and particularly by iron deficiency. Sivitz et al. (2012) reported that IRT1 and FRO2, MYB72, and At3g07720 were found to be targets of FIT and bHLH100/101 mutants showed chlorosis under iron deficiency. Also, they investigated iron regulated genes such as bHLH038, bHLH039, ZIF1 and MTP3 in wild-type and double mutant plants. These genes were not reported to be deregulated in the double mutant. Therefore, they concluded that these genes act in iron uptake mechanism FIT independent way. bHLH39 and bHLH101 were reported to be upregulated in exchange for iron deficiency. Dinneny et al. (2008) found that after 24 h low iron conditions triggered different levels of a high number of gene expressions involving in iron metabolism similar to the salt stress in Arabidopsis; however, these genes were mostly upregulated in seeds where iron stored and distributed in matured plants. Buckhout et al. (2009) found that 65 and 79 genes were expressed in roots following 6 and 24 h of removal of iron from hydroponic system and 65% of these genes were identified as the same genes such as bHLH101 and bHLH39. Long et al. (2010) identified PYE transcription factor along with BTS gene in iron uptake mechanism in Arabidopsis. They suggested that bHLH39 and bHLH101 were transcribed after Arabidopsis exposed to 24 h of iron deficient medium. Stein and Waters (2012) reported that Arabidopsis genotypes showed different gene expression profiles under iron deficient conditions, suggesting that genes having role in iron deficiency may have secondary or new roles in regulating iron homeostasis and acquisition of iron. The exposure of these ecotypes to iron deficiency for 24 h increased expression levels of Fe-deficiency-regulated genes such as bHLH39 and bHLH101. Iyer-Pascuzzi et al. (2011) reported sulphur deficient media specifically increased bHLH39 expression levels. To better understand the mechanism of iron uptake under iron deficient conditions, it is necessary to reveal the relationship between these perturbation experiments and iron uptake as well. In this regard, hormones and some small molecules whose responses were led by environmental conditions were studied by many authors in last years to shed light on iron acquisition mechanism. As a result of these efforts, ethylene, auxin, brassinosteroids, jasmonic acid, ABA, and cytokinins were reported having important roles in iron homeostasis. Lastly, gibberellic acid was suggested to modulate synthesis of BHLH038, BHLH039, FRO2, and IRT1 proteins (Brumbarova et al. 2014). Also, particularly involvement of bHLH38 in iron uptake mechanism was found subtler. This may be emanated from post-transcriptional changes of bHLH38 protein (Ramamurthy and Waters 2017). In summary, these results suggest that four Ib subgroup bHLH genes were co-regulated many genes involved in more than one pathway.

Fig. 8
figure 8

The expression levels of Arabidopsis bHLH38/39/100/101 genes in response to different perturbations

Full size image

Conclusion

In this study, four Ib subgroup bHLH genes into iron acquisition were investigated in Arabidopsis, soybean, tomato, rice and maize. Our analyses showed that these genes were not clearly separated, particularly bHLH101 were evolutionary more distant to other bHLH genes. Although bHLH38/39/100/101 genes were identified as functionally redundant in iron uptake, their secondary and tertiary structures were not similar to each other. Also, bHLH39 and bHLH101 may be more acted in iron uptake under iron deficient conditions. In Arabidopsis, these genes were upregulated in seedling stage. Also, seed parts and adult leaves were anatomical parts in which these two genes were highly transcribed. Finally, it can be proposed that bHLH38/39/100/101 genes play important roles in regulating iron homeostasis in studied plants. In addition, these genes can be used as efficient tools for biofortification studies in crops, particularly enriched iron contents.