Plant Growth Regulation

, Volume 60, Issue 2, pp 163–168

Heterologous expression of a Ammopiptanthus mongolicus late embryogenesis abundant protein gene (AmLEA) enhances Escherichia coli viability under cold and heat stress


  • Ruiling Liu
    • National Engineering Laboratory for Tree Breeding, College of Biological Sciences and BiotechnologyBeijing Forestry University
  • Meiqin Liu
    • National Engineering Laboratory for Tree Breeding, College of Biological Sciences and BiotechnologyBeijing Forestry University
  • Jie Liu
    • National Engineering Laboratory for Tree Breeding, College of Biological Sciences and BiotechnologyBeijing Forestry University
  • Yuzhen Chen
    • National Engineering Laboratory for Tree Breeding, College of Biological Sciences and BiotechnologyBeijing Forestry University
  • Yiyin Chen
    • National Engineering Laboratory for Tree Breeding, College of Biological Sciences and BiotechnologyBeijing Forestry University
    • National Engineering Laboratory for Tree Breeding, College of Biological Sciences and BiotechnologyBeijing Forestry University
Original Paper

DOI: 10.1007/s10725-009-9432-6

Cite this article as:
Liu, R., Liu, M., Liu, J. et al. Plant Growth Regul (2010) 60: 163. doi:10.1007/s10725-009-9432-6


A late embryogenesis abundant protein gene, AmLEA from Ammopiptanthus mongolicus, was introduced into Escherichia coli using the IMPACT™-TWIN system to analyze the possible function of AmLEA under heat and cold stresses. A fusion protein about 38 kD was expressed in E.coli cells harboring pTWIN-LEA after the induction of IPTG by SDS–PAGE analysis and the accumulation of the fusion protein peaked 3 h after IPTG addition when cultured at 37°C. Compared with control cells, the E. coli cells expressing AmLEA fusion protein showed improved chilling and heat resistence, illuminating the protein may play a protective role in cells under stress conditions. These results suggested the natively unstructured protein, similar to other members of LEA proteins, has high capacity for binding water and potential protective function against dehydration or action similar to the cold shock chaperones.


Ammopiptanthus mongolicusAmLEA protein expressionHeat and cold stress


LEA (late-embryogenesis abundant) proteins were initially identified in plant seeds (Dure et al. 1981). Later, several members of LEA proteins were identified in bacteria, moss, plants and nematodes (Tunnacliffe and Wise 2007). The molecular weight of LEA proteins are mainly low (10–30 kDa) and at least six groups of lea genes have been identified, based on conserved amino acid sequence motifs, which probably define functional domains (Cuming 1999). Although there is a strong association of LEA proteins with abiotic stress tolerance particularly dehydration and cold stress, for most of the past time, their function has been entirely obscure. The challenge now facing researchers investigating these enigmatic proteins is to make sense of the various in vitro defined functions in the living cell (Tunnacliffe and Wise 2007).

LEA genes were first identified as genes that are expressed during the maturation and desiccation phases of seed development (Baker et al. 1988). Among the groups of LEA proteins, role of only the group 1, 2 and 3 LEA genes has been investigated in imparting abiotic stress tolerance (Tunnacliffe and Wise 2007). Previously, our group isolated a cold-induced AmLEA (Ammopiptanthus mongolicus late-embryogenesis abundant) cDNA by different screenings based on a modified solid-phase subtraction hybridization technique from cotyledons of two-week-cold-acclimated A. mongolicus seedlings(Liu et al. 2005, 2006), the only evergreen broadleaf shrub endemic to the Alashan desert of Inner Mongolia of China where the climate is arid and cold in winter (Liu et al. 2007; Wang et al. 2008).This cDNA encoded 98 amino acids with a predicated molecular mass of 10.6 kD. Amino acid sequence analysis revealed that this polypeptide was hydrophilic according to Kyte and Doolittle method and was soluble with the average hydrophobicity −0.344898 according to SOSUI system. The predicted secondary structure showed more than 60% of total amino acid residues formed random coil. Amino acid sequence comparison shows 71% identity with late-embryogenesis protein of mung bean (T10900) and 40–70% similarity with leas in various plants. The gene was designated as AmLEA with accession number AY843523 in GenBank. Whereas the result of searching in HMMs (local models) and Pfam HMMs (global models) databases shows this AmLEA is a member of lea 3 superfamily. Members of this family are similar to late embryogenesis abundant proteins and have been isolated in a number of different screens. However, the molecular function of these proteins remains obscure (Liu et al. 2009).

Group 3 LEA function was extensively studied in transgenic plants (Bahieldin et al. 2005; Park et al. 2005; Xiao et al. 2007). However, less is known about how LEA proteins might affect prokaryotic stress tolerance. Liu and Zheng (2005) have shown that Escherichia coli expressing recombinant PM2, a group 3 LEA protein from soybean (Glycine max), were more tolerant of salt (0.5 M NaCl or KCl). E. coli cells expressing soybean late embryogenesis abundant (LEA) genes either PM11 or PM30 showed a shorter lag period and improved growth when transferred to LB (Luria–Bertani) liquid media containing 800 mmol l−1 NaCl or 700 mmol l−1 KCl or after 4°C treatment (Lan et al. 2005). In the present study, an effective prokaryotic recombinant AmLEA expression system was established in order to provide a solid basis for future functional studies and biotechnological applications of AmLEA. The main finding is that AmLEA expressed in E. coli enhances cell viability not only under cold stress but also at heat temperatures.

Materials and methods

Construction of E. coli strains expressing Ssp intein—AmLEA fusion proteins

The expression vector (pTWIN1) used for gene fusion construction was the IMPACT™-TWIN System (New England BioLabs, USA). Coding region of about 0.3 kb was prepared by PCR with the 5′ end primer (5′-AGAATTCATGGCTCGCTCCTTCACTAAC-3′) to add a EcoR I site upstream from the start codon ATG and with the 3′ end primer (5′-AGGATCCCATGAAGCAAA GAGACCCACAG-3′) to add an BamH I site after the stop codon TAA using PCR amplification.

The PCR was 40 s at 94°C, 40 s at 54°C, 1 min at 72°C for 30 cycles followed by 10 min at 72°C. The PCR products were ligated into T-vector (Pormega). The vector was then digested using an EcoRI/BamHI double digestion, and the resulting DNA was gel-purified and subcloned into the pTWIN1 vector linearized by a double digestion with the same restriction enzymes to construct pTWIN-LEA plasmid. The target gene AmLEA was fused with Ssp intein in pTWIN-LEA instead of Mxe intein and CBD of pTWIN1 vector (Fig. 1). All DNA manipulation were according to standard procedure (Sambrook and Russell 2001) and product instruction of New England BioLabs, and the AmLEA coding region and the junction sequences were confirmed by DNA sequencing. The pTWIN-LEA plasmid was transformed into applicable E. coli host strain BL21 to produce BA cells.
Fig. 1

Construction and identification of pTWIN-LEA prokaryotic expression vector. a Schematic diagram of the features of the Ssp-AmLEA region of the pTWIN-LEA vector constructed from pTWIN1. CBD, chitin biding domain; Mxe intein,Mycobacterium xenopi GyrA intein gene; Ssp intein, Synechocystis sp. DnaB intein gene, AmLea, coding region of AmLea; T7,T7 promoter. b Identification of recombinant plasmid pTWIN-LEA by polymerase chain reaction (PCR) analysis. Lane M, Molecular Weight Markers; Lanes 1–5, PCR amplification products of pTWIN-LEA using insert-flanking primers

Growth of E. coli cells and expression of Ssp intein—AmLEA fusion protein in E. coli

Ttransformed E. coli cells were grown in Luria–Bertani (LB) broth containing 100 mg ml−1 of ampicillin at 37°C overnight. The overnight cultures were diluted 1,000-fold using fresh LB broth and incubation continued at 37°C until mid-log phase (3–4 h, OD600 = 0.6). Isopropyl b-D-thiogalactopyranoside (IPTG) was then added to a final concentration of 1 mM, and incubation was continued at 37°C for 3 h. Cells were then harvested by centrifugation and resuspended in Buffer 1 [20 mM Tris–HCl (pH 8.5), 500 mM NaCl, 1 mM EDTA] and lysed by sonication. The lysate was centrifuged at 12,000 rpm for 10 min to collect clarified cell extract for SDS–PAGE analysis.

Thermophylactic and cryophylactic experiments with transformed E. coli cells

Cell cultures were grown as described above. IPTG was added to mid-log phase cultures (OD600 = 0.6) to a final concentration of 1 mM, and incubation was continued at 37°C for 3 h. After IPTG induction, the cultures were diluted and transferred to 50°C. Samples (100 μl) were taken at 0, 30, 60, 90, 120 and 240 min, and serial dilutions were plated in triplicate onto LB plus ampicillin plates. Cell viability was estimated by counting the number of colony-forming units after incubation of the plates overnight at 37°C (Soto et al. 1999). For cold treatments, appropriate dilutions from induced cultures were plated onto Luria–Bertani agar supplemented with ampicillin and 1 mM IPTG. Plates were then incubated at 0°C for 0, 2, 4, 6, 8, 10, and 12 day. Cell viability was estimated as described above. For both treatments (heat and cold), the means of three experiments were determined from at least two independent transformants (with SD being less than 5% in all cases).

Statistical analysis

Data are indicated as means ± SE. Statistical analyses were performed with ANOVA and significant differences between means were determined by Duncan’s multiple-range test. Differences were considered statistically significant when P < 0.05.


Construction of E. coli strains expressing Ssp intein—AmLEA fusion proteins

The IMPACT-TWIN (Intein Mediated Purification with an Affinity Chitin-binding Tag-Two Intein) system, which utilizes the inducible self-cleavage activity of protein splicing elements (termed inteins) to separate the target protein from the affinity tag, was adopted. Fusion of the Ssp intein-tag to the N-terminus of the target protein allows one-column protein purification with a pH and temperature shift. The coding region of target protein AmLEA was cloned into the pTWIN1 expression vector to fuse with Ssp intein-tag in stead of Mxe intein and CBD (Fig. 1a) and to construct pTWIN-AmLEA vector for fusion protein expression. Then the pTWIN-LEA plasmid was transformed into BL21 to produce BA cells. The pTWIN1 vector alone was also introduced into BL21 to produce Bp cells as a control.

In order to subclone the AmLEA coding region corresponding to 98 amino acids into the pTWIN1 vector, we used two primers designed to amplify only the ORF. The recombinant plasmid pTWIN-LEA (Fig. 1a) was transformed into E. coli and 9 clones were selected by growth in ampicillin-containing medium. The clones containing the construct were identified by PCR, and then confirmed by nucleotide sequencing using insert-flanking primers. The length of all PCR products was approximately 300 bp (Fig. 1b). The cDNA and the plasmid insert-flanking regions were completely identified. The inserted fragment showed 100% similarity with the corresponding region of AmLea cDNA previously described (GenBank accession number AY843523). Based on nucleotide sequencing of the expression vectors produced, two clones were selected for expression of the recombinant protein.

Induction of fusion protein in transformed E. coli cells

The supernatants of sonicated cell lysates from the 1 mM IPTG induced transformed cells were investigated to detect the presence of fusion protein by SDS–PAGE analysis. According to the construction (Fig. 1a), the Bp cells should produce Ssp-Mxe inteins fusion protein (about 55 kD), whereas BA cells produce an about 38 kD Ssp-AmLEA fusion protein. A band about 55 kD fusion protein was clearly observed in gels containing Bp cell supernatant, indicating this vector works normally. A fusion protein about 38 kD was expressed in BA cells after the induction of IPTG, and the accumulation of the fusion protein reached maximal levels 3 h after IPTG addition when cultured at 37°C (Fig. 2). This result showed that Ssp-AmLEA fusion protein is soluble with no inclusion formation even accumulated at 37°C, which is in coincidence with the predicted result that AmLEA is hydrophilicity.
Fig. 2

SDS–PAGE analysis of AmLEA overexpression in E.coli BL21 M, Molecular weight marker is denoted as M; B, E. coli cell BL21; BP, BL21 cells with TWIN1 vector transformed; BA1, BA2, BL21 cells with TWIN-LEA plasmid transformed from different clones; +, IPTG added; −, No IPTG. Black arrows show the 38 kD AmLEA fusion protein. The white arrow shows the 55 kD Ssp-Mxe inteins fusion protein

Cold resistance of BA cells producing AmLEA fusion protein

We tested whether recombinant AmLEA might be relevant for cell viability at chilling temperatures, as hypothesized for Brassica napusBnLEA4-1 (Dalal et al. 2009). For these experiments aliquots from IPTG-induced cultures were plated and kept at 0°C. At different times cell viability was measured by counting colony-forming units in triplicate plates. As shown in Fig. 3, both control cells (including Bp cells, BA cells without IPTG addition and empty host strain BL21) and cells overexpressing AmLEA lost viability upon storage in the cold, although at significantly different rates.The survival rate of BA cells (with IPTG addition) was more than 70% after storage at 0°C for 8 days. Whereas the control cells died with a half-life of 8 days, and after 12 days at 0°C only less than 28–20% remained alive; conversely, approximately 51% of the cells producing the AmLEA fusion protein (BA cells with IPTG induction) survived after the same period.
Fig. 3

Effect of recombinant AmLEA on cell viability (a) and growth performance (b) of E. coli transformed with pTWIN-LEA and pTWIN1 constructs subjected to 0°C treatment. Cultures grown normally and induced with IPTG for 3 h at 37°C were diluted and plated onto Luria–Bertani agar supplemented with 100 mg ml−1 ampicillin and 1 mM IPTG. Plates were then kept at 0°C. At the times indicated after temperature downshift, plates were transferred to 37°C and cell viability is plotted as the percentage of colony-forming units relative to the starting number of colonies at time 0. Means of three independent experiments are shown (SD was less than 5%). B, E. coli cell BL21; BP, BL21 cells with pTWIN vector transformed; BA, BL21 cells with TWIN-LEA plasmid transformed. +, IPTG added; −, No IPTG

Thermotolerance of BA cells producing Ssp-AmLEA fusion protein

The result of thermophylactic experiment was similar to that of cryophylactic experiments. The measured survival rates were significantly higher in cells overexpressing the AmLEA fusion protein (BA cells with IPTG addition). The cells without expressing AmLEA fusion protein died with a half-life after 60 min at 50°C, whereas BA cells (with IPTG induction) died less than 30%. After 240 min at 50°C, approximately 12% BA cells (with IPTG addition) survived, while almost no cells remained alive in control cultures(including Bp cells, BA cells without IPTG addition and empty host strain BL21; Fig. 4).
Fig. 4

Viability (a) and rowth performance(b) of E. coli transformants for pTWIN-LEA and pTWIN1 constructs subjected to 50°C treatments. At the times indicated after the temperature shift, culture samples were taken, serially diluted, and plated onto Luria–Bertani plus ampicillin plates. Plates were transferred to 37°C overnight and cell viability was estimated as in Fig. 3. Means of three independent experiments are shown (SD was less than 5%). B, E. coli cell BL21; BP, BL21 cells with pTWIN1 vector transformed; BA, BL21 cells with TWIN-LEA plasmid transformed; +, IPTG added; −, No IPTG


Higher plant adaptation to changing environment has many mechanisms involved at different levels, one of which is LEA protein gene expression and regulation formed during long evolutionary history of natural selection and artificial selection. LEA proteins mainly play functions in dehydration tolerance and storage of seeds and in whole-plant stress resistance to drought, salt, and cold (Shao et al. 2005). The LEA proteins are a subset of stress-induced proteins that can be divided into several groups based on their amino acid sequence. Because different LEA groups display conservation of unique primary structures, it is reasonable to predict that at least part of the putative protective function is the result of individual contributions by each LEA protein group and that those contributions can be demonstrated and analyzed independently (Swire-Clark and Marcotte 1999).

Hydrophilicity and heat stability are common characteristics of LEA proteins. It has been suggested that LEA-type proteins act as multi-functional water-binding molecules, in ion sequestration, in macromolecule and membrane stabilization, or even as enzyme protectant and antioxidant (Wang et al. 2003; Tunnacliffe and Wise 2007). However, their exact function remains unclear. Functional analyses of LEA proteins have been performed both in vivo and in vitro. A group 3 LEA protein, namely HVA1, reportedly displayed increased tolerance to salt stress and water deficit in both a S. cerevisiae expression system and transgenic rice plants (Xu et al. 1996; Zhang et al. 2000). Overexpression of a group 4 LEA BnLEA4-1 cDNA in E. coli conferred salt and chilling tolerance to the transformed cells. And transgenic Arabidopsis plants overexpressing BnLEA4-1 showed enhanced tolerance to salt and drought stresses (Dalal et al. 2009).

In earlier studies, we cloned and characterized the AmLEA cDNA encoding a LEA protein which is rich in Ala (11.22%), Lys (10.20%), Ser (11.22%), Val (9.18%) and lacks Cys and His residues. Pfam research results showed AmLEA belongs to superfamily lea3 (pfam03242, LEA_3, Late embryogenesis abundant protein, 3e-25) with similar domain architecture, although non-specific hits were found. Commonly, group 3 LEA proteins contain a repeat of an 11 mer amino acid motif with the consensus sequence TAQAAKEKAGE. A possible amphiphilic helix structure of the 11 mer repeating unit may allow dimerization of the polypeptides via binding of their hydrophobic faces, and a right-handed coiled coil would be formed (Dure 1993). However, AmLEA lacks a repeat of an 11-mer amino acid motif, indicating AmLEA is a LEA protein but not a typical lea3 protein (Liu et al. 2009).

To investigate the possible function of AmLEA in vivo, we introduced its coding sequence into E. coli using the pTWIN1 expression vector. As shown in Fig. 3, we found that overexpression of AmLEA in E. coli was correlated with maintenance of viability under cold-stress conditions. Experimental evidence here also conformed that recombinant AmLEA is important in maintaining cell viability at heat temperatures. pTWIN-LEA cells overexpressing the AmLEA protein died more slowly after treatment at 50°C than control cells (Fig. 4). Although the precise reasons why bacterial cells die at 0°C or 50°C have not yet been discovered, the protective effect of AmLEA might be due to the maintenance of proteins in a functional conformation. Thus, it can be speculated that AmLEA might retain water molecules and prevent crystallization of cellular components under water-deficit, which results from cold and heat stresses (Park et al. 2005). Although the thermostabilizing and cryophylactic function of AmLEA was conformed by the present experiment, the exact biochemical mechanisms by which AmLEA, and probably other LEAs, attenuate both heat- and cold-induced cell damage remain to be determined.


This work was supported by the National Science Foundation of China (Grant No. 30671476).

Copyright information

© Springer Science+Business Media B.V. 2009