WO2015059057A2 - Novel class of antifreeze proteins - Google Patents

Abstract

The present invention relates to a novel class of antifreeze proteins. More specifically the invention relates to the use of proteins from spider mite, preferably from Tetranychus urticae, with a specific CXX based antifreeze signature that is different from all known antifreeze proteins from insects, fish or micro-organisms, as antifreeze proteins. The invention relates further to the use of said antifreeze proteins to protect food and/or living organisms such as plants from freeze damage.

Description

Novel class of antifreeze proteins

The present invention relates to a novel class of antifreeze proteins. More specifically the invention relates to the use of proteins from spider mite, preferably from Tetranychus urticae, with a specific CXX based antifreeze signature that is different from all known antifreeze proteins from insects, fish or micro-organisms, as antifreeze proteins. The invention relates further to the use of said antifreeze proteins to protect food and/or living organisms such as plants from freeze damage.

Antifreeze proteins (AFPs) (also called thermal hysteresis proteins (THPs) or ice structuring proteins (ICPs) are proteins that have affinity for ice. They lower the freezing point of water without significantly altering the melting point. AFPs have been isolated from several different organisms such as Arctic and Antarctic fish, insects, plants and sea ice organisms. They form a structurally divergent group of proteins: apart from the antifreeze glycoproteins (AFGPs), there are four classes of AFPs in fish, and three classes in insects (Barrett, 2001 ). AFGPs are characterized by AAT repeats, the fish AFP type I is A rich (60%); type II is cysteine rich (9%) and disulfide stabilized; type III has no specific amino acid bias, and type IV is glutamine/glutamate rich (26%) (Barett, 2001 ; Venketesh and Dayananda, 2008). Insect AFPs are characterized by T and C rich 12 or 13 amino acid repeats, or by a S an T rich sequence. Plant AFPs are different from other AFPS and characterized by a much weaker thermal hysteresis effect which makes them less interesting for industrial applications. Recently, AFPs have been detected in cold tolerant diatoms (Bayer-Giraldi et al., 2010), bacteria (Raymond et al., 2007) and fungi (Hoshino et al., 2003; Raymond and Janech, 2009).

Diapause is a hormonally controlled dormancy, common across a range of organisms and widespread amongst insects. Diapause helps the organisms to survive in adverse conditions, such as cold stress. Several studies have been carried out to identify genes that are involved in diapause induction and cold survival, but the genes identified till now are mainly heat shock proteins (HSPs) (Clarck and Worland, 2008), and no specific thermal hysteresis proteins have been identified. Bale and Hayward indicate that, besides HSPs, cryoprotectants such as Unsaturated Fatty Acids (UFAs) and glycerol may play a role, but the authors do not cite a possible role for AFPs.

Surprisingly we found that, during diapause, spider mites are expressing proteins that function as antifreeze proteins. Even more surprisingly, these proteins are characterized by continuous stretches of CXX repeats, wherein at least 50% of the CXX repeats is a CXN sequence. This signature is different from all antifreeze proteins known.

A first aspect of the invention is the use of protein, comprising a continuous repeated sequence with at least 8 CXX repeats, preferably at least 9 CXX repeats, of which at least 4, preferably at least 5 repeats are CXN repeats as an antifreeze protein. Even more preferably, said protein comprises SEQ ID N°1 , more preferably said protein comprises SEQ ID N° 2, even more preferably, said protein comprises SEQ ID N° 3. Preferably, said protein is a mite protein, even more preferably it is a protein from Tetranychus urticae. Most preferably said protein comprises a sequence, preferably consist of a sequence selected from the group consisting of SEQ ID N° 4 - SEQ ID N° 24. "Use of a protein... as antifreeze protein", as used here, means any use of an antifreeze protein known to the person skilled in the art, including, but not limited to the use for cryostabilization of food or technical solutions such as cooling liquids, induction of cold tolerance and/or freeze protection by addition of and/or expressing the protein in oocytes, spermatozoids, microorganisms, cells, animals or plants

Another aspect of the invention is the use of a protein comprising a continuous repeated sequence with at least 8 CXX repeats, preferably at least 9 CXX repeats, of which at least 4, preferably at least 5 repeats are CXN repeats for protecting food against freeze damage. Even more preferably, said protein comprises SEQ ID N°1 , even more preferably, said protein comprises SEQ ID N° 2, most preferably said protein comprises SEQ ID N° 3. Preferably, said protein is a mite protein, even more preferably it is a protein from Tetranychus urticae. Most preferably said protein comprises a sequence, preferably consist of a sequence selected from the group consisting of SEQ ID N° 4 - SEQ ID N° 24, or the processed sequence without signal sequence. Signal sequences are predicted according to the SignalP program (Emanuelsson et al., 2007; SignalP 4.0, accessible at the TargetP1 .1 server, Technical university of Denmark at http://www.cbs.dtu.dk/services/TargetP/)

Another aspect of the invention is the use of a protein comprising a continuous repeated sequence with at least 8 CXX repeats, preferably at least 9 CXX repeats, of which at least 4, preferably at least 5 repeats are CXN repeats for protecting cells against freeze damage. Even more preferably, said protein comprises SEQ ID N°1 , even more preferably, said protein comprises SEQ ID N° 2, most preferably said protein comprises SEQ ID N° 3. Preferably, said protein is a mite protein, even more preferably it is a protein from Tetranychus urticae. Most preferably said protein comprises a sequence, preferably consist of a sequence selected from the group consisting of SEQ ID N° 4 - SEQ ID N° 24, or the processed sequence without signal sequence. Said cell can be any cell, including but not limited to plant cells, microbial cells, and animal cells including, but not limited to fish cells and mammalian cells. Preferably, said cells are plant cells or microbial cells. Preferably said cells are recombinant cells protected by expressing said protein on sequence suitable for the expression of the protein. Said sequence can be a recombinant expression vector, or the expression sequence can be integrated in the genome.

Therefore, still another aspect of the invention is a freeze tolerant recombinant host cell, comprising a recombinant protein with at least 8 CXX repeats, preferably at least 9 CXX repeats, of which at least 4, preferably at least 5 repeats are CXN repeats. Even more preferably, said recombinant protein comprises SEQ ID N°1 , even more preferably, said recombinant protein comprises SEQ ID N° 2, most preferably, said recombinant protein comprises SEQ ID N°3. Preferably, said recombinant protein is a mite protein, even more preferably it is a recombinant protein derived from Tetranychus urticae. Derived, as used here, means that the protein occurs in some growth phases of Tetranychus urticae, but that the sequence, encoding said protein is adapted for the expression in said host cell. As a non- limiting example, said adaptation may be placing the coding sequence under control of a suitable promoter, and/or optimization of the coding usage to improve the expression in said host cell. Most preferably said recombinant protein comprises a sequence, preferably consist of a sequence selected from the group consisting of SEQ ID N° 4 - SEQ ID N° 24, or the processed sequence without signal sequence. A typical signal sequence is indicated in SEQ ID N°4, but sequence variations are of the signal are known to the person skilled in the art (Petersen et al., 201 1 ). A preferred embodiment is a freeze tolerant cell, wherein said cell is a plant cell.

Still another aspect of the invention is a freeze tolerant recombinant plant, comprising a recombinant protein with at least 8 CXX repeats, preferably at least 9 CXX repeats, of which at least 4, preferably at least 5 repeats are CXN repeats. Even more preferably, said recombinant protein comprises SEQ ID N°1 , even more preferably, said recombinant protein comprises SEQ ID N° 2, most preferably it comprises SEQ ID N°3. Preferably, said recombinant protein is a mite protein, even more preferably it is a recombinant protein derived from Tetranychus urticae. Most preferably said recombinant protein comprises a sequence, preferably consist of a sequence selected from the group consisting of SEQ ID N° 4 - SEQ ID N° 24, or the processed sequence without signal sequence.

BRIEF DESCRIPTION OF THE FIGURES

Figure 1 : Ribbon illustration of a putative T. urticae antifreeze protein (AFP).

Side (A) and end-on (B) view of the predicted 3D-structure of a putative spider mite AFP (tetur22g03033) with β-sheets indicated by blue arrows. Asn, Asp and Thr side-chains are indicated in orange while Cys side-chains are indicated in yellow. The N-and C-terminus of the T. urticae AFP structure are indicated with the letters N and C, respectively. The 3D-model for tetur22g03033 was created using the Phyre² server (with pdb-model: dl ezga) (Kelly and Sternberg, 2009) and further edited with Swiss PDB viewer (Guex and Peitsch, 1997) Figure 2: Expression levels of genes coding for putative T. urticae antifreeze proteins.

qPCR quantification of expression levels of putative AFP genes in T. urticae. Green, orange and blue bars represent the relative mean expression in non-diapausing spider mites at 17°C, diapausing spider mites at 17°C and non-diapausing spider mites at 5°C, respectively, relative to expression in non-diapausing mites of the LS-VL strain at 24°C. Error bars represent the standard error of the calculated mean based on three biological replicates. Asterisks indicate significantly differential expressed genes (random reallocation test) compared to the reference condition (green, non-diapausing LS-VL strain at 24°C).

Figure 3: expression constructs for the antifreeze proteins.

GST=glutathione S-transferase tag, His= histidine tag, Trx= thioredoxine tag, TEV= Tobacco Etch Virus site, TuAFP= Tetranychus urticae Antifreeze protein

Figure 4: Production and purification of tetur22g02640 (A) SDS-PAGE analysis: lane 1 : marker 1 ; lane 2: BSA, lane 3: tetur22g02640 (B) Western Blot analysis using Anti- His antibody: lane 1 : marker 2; lane 2: tetur22g02640. Figure 5: Production and purification of tetur22g02780 (A) SDS-PAGE analysis: lane 1 : marker 1 ; lane 2: tetur22g02780; lane 3: BSA (B) Western Blot analysis using Anti- His antibody: lane 1 : marker 2; lane 2: tetur22g02780

Figure 6: Production and purification of tetur22g02690 (A) SDS-PAGE analysis: lane 1 : tetur22g02690; lane 2: marker 1 , lane 3: BSA (B) Western Blot analysis using Anti- His antibody: lane 1 : marker 2; lane 2: tetur22g02690

Figure 7: production and purification of antifreeze protein in Pichia pastoris. Lane 2 and 4 are the marker sequences, lane 1 represents 10mg of antifreeze protein in presence of β-mercaptoethanol, lane 3 the same amount of antifreeze protein without β-mercaptoethanol.

EXAMPLES

Materials and methods to the examples Mite rearing

The strain LS-VL of Tetranychus urticae that was originally collected in October 2000 near Ghent (Van Leeuwen et al., 2004) was maintained on potted kidney bean plants Phaseolus vulgaris L. var. Prelude in controlled conditions at 26 ±0.5°C and 60% RH with a 16:8h light:dark photoperiod. Subsequently 500 adult females were transferred to the same bean plants and kept in the same conditions during 4 days until larvae came out. Next the plants were moved in diapause conditions at 15±0.5°C, 80% RH with a 8:16h light:dark photoperiod. RNA preparation

In order to extract total RNA, 250 adult female mites with distinguishing diapause characteristics were collected and homogenized according to the protocol of RNeasy mini kit (Qiagen) in four replicates. The control group consisted of 100 adult females that did not enter diapause but lived in the same diapause conditions. An additional washing step with one volume of chloroform:isoamylalcohol (24:1 ) was performed to decrease the amount of carotenoid pigments in the supernatant. The quality and quantity of the total RNA was analyzed by NanoDrop ND-1000 spectrophotometer (NanoDrop Technologies) and by running the sample on a 1 % agarose gel.

Microarray construction

A custom Sureprint genome wide G3 Gene Expression 8x60K microarray was designed using the Agilent eArray platform (Agilent Technologies) based on the T. urticae gene annotation file frozen in August 201 1 including coding sequences of 18317 predicted unigenes (Grbic et al. 201 1 ). Manually annotated genes belonging to known detoxification gene families (CCEs, GSTs, CYPs, ABC-transporters) were represented in the input file in two copies. To potentially increase the specificity of the probes, we extended the coding sequences of all genes with 100 bp of their predicted 3' UTR. Where UTRs were not predicted or predicted shorter, 100 bp downstream the stop-codon was added to the coding sequence. The probe design aimed at 3 probes of 60 nucleotides per gene with a Tm of 80°C and parameters set to "best probe design" and "3' bias". From the 18317 target genes, 99.72% were covered by at least 3 probes. The final array design included 61 1 1 1 probes that were validated using custom scripts (mapping, cross-hybridization potential, overlap). Standard Agilent features such as spike-ins were added (IS-62976-8-V2_60kby8_GX_EQC_20100210). As a positive control, we selected 182 unique probes that mapped to T. urticae genes expressed across 4 developmental stages, as identified by RNAseq experiments, with different ranges of expression (normalised read- counts) ([30-50]=>65,[60-100]=>50, [100-200]=>52,[150-400]=>62). These probes were randomly distributed in 10 to 15 copies per array, and can be used for array normalisation.. The current slide layout consists of eight arrays per slide, permitting comparison of two treatments with fourfold replication on each slide (design ID 028213).

Microarray preparation, hybridization and analysis

One hundred nanograms of RNA was used to generate Cy3- and Cy5-labeled cRNA from respectively non-diapausing and diapausing mites, using the Agilent Low input Quick amplification labeling kit (version 6.5, Agilent Technologies),. RNA spike-in controls (Agilent Technologies) were added to RNA samples before cRNA synthesis. The labeled cRNA was purified with the RNeasy mini kit (Qiagen). The dye content and concentration of cRNA was measured by NanoDrop ND-1000 spectrophotometer (NanoDrop Technologies). Cy3- and Cy5-labeled cRNAs were pooled and hybridized using the Gene Expression Hybridization Kit (Agilent Technologies) for 17 h in a rotating hybridization oven at 10 r.p.m and 65°C. After hybridization, slides were washed using the Gene Expression Wash Buffer kit (Agilent Technologies), treated with Stabilization and Drying solution (Agilent Technologies), protected by an Ozone-Barrier cover (Agilent Technologies) until scanned by an Agilent Microarray High Resolution Scanner with default settings for 8 x 60K G3 microarrays. Data were then normalized by the Agilent Feature Extraction software version 10.5 (Agilent Technologies) with default parameter settings for gene expression two-color microarrays (protocol GE2_107_SEP09) and data was transferred to GeneSpring GX 1 1 .0 software (Agilent Technologies) for further statistical evaluation. Experiments were built from these microarray data by GeneSpring GX 1 1 .0. Next, probes were flag filtered (only probes that had flag-value 'detected' in 50% of all replicates of each experiment were retained) and linked to the most recent T. urticae genome annotation file (September 201 1 ) using the 'Create New Gene-Level Experimenf-option. Genes that showed a > 2.0-fold change were selected for t-test p-value (p < 0.05 after implying the Benjamin-Hochberg false discovery rate correction for multiple testing).

T. urticae antifreeze proteins (AFPs) analysis

Gene expression analysis revealed the upregulation of 14 genes belonging to the same hypothetical protein family. A tBLASTn analysis (E-value cutoff = 1 e^"5) using these hypothetical proteins as queries was conducted against the T. urticae genome (http://bioinformatics.psb.ugent.be/orcae/overview/Tetur). Gene models were refined or created on the basis of homology, RNA-seq and/or EST support and RT-PCR. A BLASTp search, using these hypothetical proteins as queries, was performed against the NCBI nr protein sequence database to detect homologues in other organisms. Identity and similarity matrices were calculated using MatGAT 2.0 (Campanella et al., 2003) while an alignment was created using MUSCLE (Edgar, 2004). T. urticae AFP structures were predicted using the Phyre² server (Kelley and Sternberg, 2009), in order to create a 3D model for the query sequence, and further edited in Swiss-Pdb viewer (Guex and peitsch, 1997). The top match produced by Phyre2 is the protein sequence that showed the highest raw alignment score with the query sequence and is based on the number of aligned residues and the quality of alignment. This first match was selected to fit our protein to a 3D model. SignalP 4.0 was used for prediction of signal peptides of AFP protein sequences using default cutoff values for eukaryotes (Petersen et al.,201 1 ).

The LS-VL strain was used for the additional qPCR experiments assessing the effect of temperature on expression levels T. urticae AFP genes. For the samples of 24 °C, adult female mites were collected from the culture maintained at standard conditions. The group which experienced a cold shock consisted of female mites that were maintained on a leaf disk at 5°C during 7 days. Subsequently, RNA extraction and cDNA preparation was carried out as abovementioned.

Example 1 : Identification of the antifreeze proteins

In order to investigate the process of diapause in Tetranychus urticae, an experiment was set up to compare the total gene expression from diapausing and non diapausing mites. The samples for the diapausing mites were collected on the basis of clearly distinguishing diapause characteristics (morphology and color). The non diapausing mites were cultured in the same conditions without entering diapause. Subsequently, a dual color gene expression microarray experiment was performed and a set of differentially expressed genes defined. In this set, one striking example was a gene that showed to be 164 times upregulated in diapausing mites and belongs to a gene cluster (orthoMCL clustering) where all genes showed a > 2.0-fold change up regulation.

One of the most striking differentially expressed genes in our analysis consisted of a small family of 'hypothetical proteins' (SEQ ID N° 4-24). This family was manually annotated in the genome database of T. urticae, and consisted of a set of 20 genes and 2 pseudogenes (tetur22g03073 and 63g00100). Of those, only 16 genes were represented by probes on the microarray and 14 of these were upregulated in diapausing spider mites, with fold changes varying between 2 and 164. Gene loci are distributed over three genomic scaffolds: 22 (15), 63 (4) and 283 (1 ) and all genes except tetur22g03033 are intronless. Genes on scaffold 22 are tightly clustered within a 90 kb region, but the proteins showed only moderate sequence similarity, suggesting their proliferation in the T. urticae genome is the result of "ancient" tandem duplication events. Tetur283g00030, tetur63g00030, tetur63g00050, tetur63g00070 and tetur63g00090 showed very high identity values (96.7% - 100%) with tetur22g02550, tetur22g02640, tetur22g02670, tetur22g02730 and tetur22g02690, respectively. This might suggest a recent duplication event, although the possibility that the genes on scaffold 22 and 63/283 actually represent the same gene due to assembly issues caused by allelic variants cannot be excluded at this stage.

All members of this hypothetical protein family, except tetur22g02690, tetur22g03063 and tetur63g00090, have a predicted signal peptide (Petersen et al., 201 1 ) and hence are probably secreted by cells. Strikingly, an InterPro-scan revealed that the majority (16/20) of these hypothetical proteins contained the "Insect Antifreeze Protein" motif (IPR016133 with an E- value between 1 e^"04 and 1 e^"06). A BLASTp search against the non-redundant protein database of NCBI also hit to antifreeze proteins of Coleoptera with a low to moderate E-value (between 1 e^"02 and 1 e^"6). Furthermore, all members from this family were predicted as an insect AFP by AFP-Pred, a recently developed software tool using a "random forest" approach for the prediction of antifreeze proteins (Kandaswamy et al., 201 1 ).

Accordingly, we aligned and compared the T. urticae mite sequences with well-studied insect AFP sequences of two beetles: Dendroides canadiensis (AAF86362 and AAB94303) and Tenebrio molitor (1 L1 I_A). These insect AFPs consist of 78 to 148 amino acids and their cysteine (Cys) content ranges from 15 to 19%, whereas the spider mite sequences measure between 92 and 210 amino acids and consist for 19 to 25 % of Cys. In addition, typical insect AFPs are characterized by 7 repeats of 12-or 13-mer repeats (Thr-Cys-Thr-X-Ser-X-X-Cys-X- X-Ala-X) with at least every sixth residue a cysteine (Graham et al., 1997; Liou et al., 2000). The Cys residues can form stable disulfide bridges throughout the protein and are flanked by Thr residues, which are thought to be responsible for the ice binding sites (Graether et al., 2000). In T. urticae, the predicted AFPs comprise a continuous repeated sequence (Asn-Cys- Thr-X-Cys-X-X-Cys-X-Asn-Cys-X). It is clear that two of the Cys (at the sixth residue, involved in disulfide bridging) are conserved, but two other Cys are found at positions where in beetles a conserved Ser and Ala residues is found. Furthermore, the conserved Cys are not only flanked by Thr, but a variety of Asp, Asn and Thr. These hydrophilic residues are capable of ice-binding and inhibition of ice crystal growth (Garnham et al., 201 1 ; Zhang et al., 2004). The specific arrangement and position of these residues is however fundamental for the activity of AFPs in order to bind to the ice surface with a particular orientation. Hence, all 20 protein sequences were submitted to Phyre² web service for structure prediction. Six out of 20 query sequences (tetur22g02640, tetur22g02690, 22g03033, tetur22g03063, tetur63g00030 and tetur63g00090) had a top scoring match with an insect AFP template structure (pdb: dl ezga) in the Phyre² library. Subsequently, 2-D and 3-D models for these six T. urticae proteins were returned based on this match. For example, the 3D model of tetur22g03033 (53% of the tetur22g02640 protein sequence has been modeled with 97.1 % confidence, Additional File 15) showed the typical configuration of cysteine rich insect AFP consisting of six β-strands containing the typical Cys-residues (Figure 5) (at six AA distance) producing disulfide bridges. The important ice-binding amino acids (Asn, Asp and Thr) are located on the outside of the protein. Interestingly, the extra Cys-residues (at three AA distance) specific for mite AFP sequences, are pointing inwards and may form alternative cysteine bridges in vivo. The typical ice-binding surface (Marshall, 2002; Scotter et al., 2006; Graham et al., 2007) was not apparent in the 3D homology model of tetur22g03033, but these outfacing hydrophilic residues might still be involved in ice binding. Recently, divergent tertiary structures were also reported for Rhagium inquisitor (Hakim et al., 2013) and snow fleas (Lin et al., 2007). In conclusion, sequence alignments, and in silico predictions including a homology model strongly suggest that this small family of proteins might represent a new class of arthropod AFPs. Nevertheless, functional expression combined with activity tests using nanoliter osmometry should provide formal evidence of the activity and ice binding properties of these proteins (Nicodemus et al., 2006; Lin et al., 201 1 ; Venketesh and Dayananda, 2008).

The production of antifreeze proteins is a protective mechanism against the effects of lowered environmental temperatures and freezing (Clark and Worland, 2008). Typically, cysteine-rich insect antifreeze proteins adsorb to seeded ice crystals and inhibit these crystals in growing bigger (Scotter et al., 2006) which causes a difference between the freezing temperature and the melting point, a process that is known as thermal hysteresis. Thermal hysteresis antifreeze proteins have also been found in vertebrates, invertebrates, fungi, bacteria and plants (Barrett, 2001 ). At least fifty species of insects and many terrestrial arthropods (Venketesh and Dayananda, 2008; Duman, 2001 ) are known to produce AFPs and although many have low levels of thermal hysteresis, some insects produce hyperactive antifreeze proteins. The common yellow mealworm beetle, T. molitor, contains antifreeze proteins that can account a thermal hysteresis of 5.5 °C at a concentration of 1 mg/ml(Graham et al., 1997). Comparative results are found in Hypogastrura harveyi (Graham and Davies, 2005; Qin et al., 2007) and C. fumiferana who display similar thermal hysteresis properties (Barrett, 2001 ). The magnitude of thermal hysteresis effect occurring in a species depends not only on the structural type of proteins, but can differ greatly depending on the moment of sampling or the level of acclimatization of insects (Qin et al., 2007).

We performed additional qPCR experiments to investigate the association of AFPs with diapause and/or cold stress of 5 typical AFP genes {tetur22g02690, tetur22g02640, tetur22g02730, tetur22g02790 and tetur22g02670). Expression levels were compared between non-diapausing mites at 24 °C (standard rearing conditions), and mites that were submitted to a cold shock of 7 days, together with diapausing and non-diapausing mites at 17 °C (the conditions of the microarray experiments). Results showed that the expression of these genes was overall more affected by the diapause condition, than the reduction in temperature in T. urticae, indicating that for some genes the expression is probably regulated by physiological, rather than environmental changes (Figure 6). All genes were significantly high upregulated in diapausing mites at 17 °C when compared to the expression of these genes in active mites at 17 °C, confirming microarray results. Only one gene, tetur22g02690, was equally high expressed both in diapause mites at 17°C and in mites treated by cold shock. Of the 4 other genes tested, two did respond to cold stress (tetur22g02670, tetur22g02730) but with small changes in expression, while tetur22g02790 and tetur22g02640 were only upregulated to high levels in diapausing mites.

It was previously shown that the production of antifreeze proteins is not exclusively associated with diapause. In T. molitor, D. Canadensis (Duman et al., 1998), R. inquisitor (Kristiansen et al., 1999) and H. harveyi (Graham and Davies, 2005), antifreeze proteins could be collected from insects that were exposed to cold. In the spruce budworm, transcripts of antifreeze proteins were found in both the first larval instar and in the second diapausing instar (Qin et al., 2007). This indicates that both cold hardening and diapause can trigger the production of AFPs in insects, similar to what was shown for T. urticae in this study. It was demonstrated previously in T. urticae that diapausing forms had a lower super cooling point than non- diapausing forms, even if those active forms were cold acclimated for 10 days at 5 or 0 °C [27]. Furthermore, the 5 °C cold acclimated active mites showed a lower super cooling point than the not acclimated active mites, indicating cryoprotection caused by cold and diapause, but the involvement of mite AFPs identified in this study was not assessed and should be further studied. Of particular note, homologues of spider mite AFPs were not found in other mite and tick species for which a draft genome or transcriptome is available, such as Varroa destructor, Metaseiulus occidentalis , Ixodes scapularis and Panonychus citri.

Example 2: Full length amplification of antifreeze genes

Specific oligonucleotide primers were designed to amplify the full length (or nearly full length) of genes encoding antifreeze proteins. Two microlitres of cDNA template was used in the amplification using 51Ι/μΙ of Taq DNA Polymerase (Invitrogen) and 20 pM of specific primers in a 50 μΙ reaction mix. The cDNA template was synthesized from RNA of diapausing mites (prepared as above) using the Maxima first strand cDNA synthesis kit (Fermentas). Next, the PCR products were visualized on a 1 % agarose gel and subsequently purified with the E.Z.N. A. Cycle Pure Kit (Omega Bio-Tek) and either directly sequenced or cloned into the pJET vector according to the CloneJet PCR Cloning Kit protocol (Fermentas). Plasmids were purified with the E.Z.N.A Plasmid mini Kit II (Omega Bio-Tek). Sequencing was performed at LGC Genomics. Example 3: Expression of antifreeze proteins

Representative antifreeze proteins were functionally expressed using an Escherichia coli expression system. Because previous attempts for expression of antifreeze proteins in this system resulted in inclusion bodies or soluble but inactive material (Bar et al., 2006), we fused the antifreeze proteins with different tags. The following constructs were tested for expression of antifreeze proteins in E.coli.

These three systems based on fusions with GST and Trx are known to be successful in promoting correctly folded and soluble antifreeze proteins in the cytoplasm (Bar et al., 2006; Liou et al., 2000; Mao et al., 201 1 ; Tyshenko et al., 2006; Yue & Zhang, 2009).

First the genes were synthesized and inserted in the vector pUC57 and further subcloned in into three distinct expression vectors with a Histag, a Histag-GST and a Trx-Histag. Subsequently the Trx-Histag was selected as the best expression vector and the protein was purified with a Ni-HiTrap column and visualized by SDS-PAGE after Coomassie Blue staining. Western blot analysis using anti-His antibodies confirmed the presence of pure and intact protein. A TEV cleavage site allows unleashing the antifreeze protein from the fusion partner. The constructions and the results of the expression are shown in Figure 3-6. Example 4: Expression of recombinant antifreeze protein in Pichia

The pAOXZalfaH-HSA-C vector (which is a derivate of the pPICZa vector from Life Technologies) was transformed in the wild type GS1 15 Pichia pastoris strain. The expression vector is provided with the AOX1 promoter fused to the omating factor prepro signal sequence followed by respectively a His-tag, HSA fusion protein and the AFP protein. Between the HSA fusion protein and the AFP protein, a mCaspase3 recognition site is introduced. The final protein sequence is given as SEQ ID No.25. After selection of an appropriate expression clone, a production was performed in baffled shake flasks on a level of 6 liters (24 x 250 ml/2 liter flask) (Schoonooghe et al., 2012). The medium fraction was isolated by centrifugation at 18,000 x g for 30 min at 4°C and diafiltered to 20 mM NaH2P04 pH 7.5, 500 mM NaCI, 20 mM imidazole and 1 mM PMSF. The clear supernatant was applied to a Ni-Sepharose 6 FF column (GE Healthcare), equilibrated with 20 mM NaH2P04 pH 7.5, 500 mM NaCI, 20 mM imidazole, 1 mM PMSF. After loading, the column was washed with the same buffer in presence of 0.1 % empigen as detergent over 20 column volumes. Before the elution, the column was equilibrated with the equilibration buffer without detergent over 5 column volumes. The column was eluted with 20 mM NaH2P04 pH 7.5, 20 mM NaCI, 400 mM imidazole, 1 mM PMSF after an intermediate elution step with 50 mM of imidazole in the same buffer. The elution fraction was diluted 1/20 with 20 mM Tris pH 8.0 and loaded on a Source 15Q column (GE Healthcare) to remove contaminants. After equilibration, the protein of interest was eluted by a linear gradient over 20 column volumes of NaCI from 0 to 1 M in 20 mM Tris pH 8.0. Finally, the recombinant protein was injected on a Superdex 200 gelfiltration column with PBS as running solution for formulation and to remove minor contaminants. The obtained fractions were analyzed by SDS-PAGE as shown in Figure 7 and the concentration was determined using the Micro-BCA assay (Pierce). Example 5: Antifreeze capacity of the isolated proteins

Determination of ice recrystallization inhibition by the antifreeze protein

Ice recrystallization is defined as the growth of large ice crystals at the expense of small ones.

Ice recrystallization can cause cold stress in organisms, and is causing spoilage of frozen foods.

Inhibition of ice recrystallization is essentially measured as described by Tomczak et al. (2003). Alternatively, ice recrystallization is measured by the splat assay (Knight et al., 1988). All T. urticae antifreeze proteins tested show a significant inhibition of ice recrystallization. Measurement of Thermal hysteresis

Thermal hysteresis induced by the antifreeze proteins is essentially tested as described by Hansen and Baust (1989). All T. urticae antifreeze proteins tested with a nanolitre osmometer show a significant increase in thermal hysteresis in a concentration dependent manner (up to 10 mg/ml).

Cryoprotection assay

The cryoprotection assay of the antifreeze for improving freeze resistance of E. coli is based on the method described by Yue & Zhang (2009). Colonies of E. coli are grown overnight at 37°C in LB medium. The cells are treated with different concentrations of antifreeze proteins ranging from 50 to 500 μg ml. Next, samples are frozen at -20°C for 24h, 48h and 72h and spread on LB plates. After 16h of incubation at 37 °C, the colonies are counted. All antifreeze concentrations show a cryoprotective property compared to the untreated E.coli sample.

REFERENCES

Bale, J.S. and Hayward, S.A.L. (2010). Insect overwintering in a changing climate. J. Exp. Biol. 213,980-994.

- Bar, M., Bar-Ziv, R., Scherf, T., & Fass, D. (2006). Efficient production of a folded and functional, highly disulfide-bonded beta-helix antifreeze protein in bacteria. Protein expression and purification, 48, 243-52.

- Barrett, J. (2001 ). Thermal hysteresis proteins. Int. J. Biochem. Cell Biol. 33, 105-1 17.

- Bayer-Giraldi, M., Uhlig, C, John, U., Mock, T., and Valentin, K. (2010). Antifreeze proteins in polar sea ice diatoms: diversity and gene expression in the genus Fragilariopsis. Environ. Microbiol. 12, 1041 -1052.

Campanella JJ, Bitincka L, Smalley J: MatGAT: an application that generates similarity/identity matrices using protein or DNA sequences. BMC Bioinformatics 2003, 4:29.

Clark, M.S. and Worland, M.R. (2008). How insects survive the cold: molecular mechanisms - a review. J. Comp. Physiol. B 178, 917-933.

Duman JG: Antifreeze and ice nucleator proteins in terrestrial arthropods. Annu Rev Physiol 2001 , 63:327-357.

- Duman JG, Li N, Verleye D, Goetz FW, Wu DW, Andorfer C a, Benjamin T, Parmelee DC: Molecular characterization and sequencing of antifreeze proteins from larvae of the beetle Dendroides canadensis. J Comp Physiol B 1998, 168:225-32.

Edgar RC: MUSCLE: multiple sequence alignment with high accuracy and high throughput. Nucleic Acids Res 2004, 32:1792-7.

Emanuelsson, O., Brunak, S., von Heijne, G. and Nielsen, H. (2007). Locating proteins in the cell using TargetP, SignalP and related tools. Nature protocols 2, 953-971 .

- Garnham CP, Campbell RL, Walker VK, Davies PL: Novel dimeric β-helical model of an ice nucleation protein with bridged active sites. BMC Struct Biol 201 1 , 1 1 :36.

- Graether SP, Kuiper MJ, Gagne SM, Walker VK, Jia Z, Sykes BD, Davies PL: β-Helix structure and ice-binding properties of a hyperactive antifreeze protein from an insect. Nature 2000, 406:325-328.

- Graham L, Davies P: Glycine-rich antifreeze proteins from snow fleas. Science 2005,

310:461 .

Graham LA, Liou YC, Walker VK, Davies PL: Hyperactive antifreeze protein from beetles. Nature 1997, 388:727-8.

Graham LA, Qin W, Lougheed SC, Davies PL, Walker VK: Evolution of hyperactive, repetitive antifreeze proteins in beetles. J Mol Evol 2007, 64:387-98.

Grbic M, Van Leeuwen T, Clark RM, Rombauts S, Rouze P, Grbic V, Osborne EJ,

Dermauw W, Ngoc PC, Ortego F, Hernandez-Crespo P, Diaz I, Martinez M, Navajas M, Sucena E, Magalhaes S, Nagy L, Pace RM, Djuranovic S, Smagghe G, Iga M, Christiaens O, Veenstra JA, Ewer J, Villalobos RM, Hutter JL, Hudson SD, Velez M, Yi SV, Zeng J, Pires-daSilva A, Roch F, Cazaux M, Navarro M, Zhurov V, Acevedo G, Bjelica A, Fawcett JA, Bonnet E, Martens C, Baele G, Wissler L, Sanchez-Rodriguez A, Tirry L, Blais C, Demeestere K, Henz SR, Gregory TR, Mathieu J, Verdon L, Farinelli L,

Schmutz J, Lindquist E, Feyereisen R, Van de Peer Y. (201 1 ). The genome of Tetranychus urticae reveals herbivorous pest adaptations. Nature 479, 487-492.

- Guex N, Peitsch MC: SWISS-MODEL and the Swiss-Pdb Viewer: An environment for comparative protein modeling. Electrophoresis 1997, 18:2714-2723.

- Hakim A, Nguyen JB, Basu K, Zhu DF, Thakral D, Davies PL, Isaacs FJ, Modis Y,

Meng W: Crystal structure of an insect antifreeze protein and its implications for ice binding. J Biol Chem 2013, 288:12295-304.

Hinosho, T., Kiriaki, M., Ohgyiya, S., Fujiwara, M., Hondo, H., Nishimiya, Y. (2003).

Antifreeze proteins frim snow mold fungi. Can. J. Bot. 81 , 1 175-1 181 .

- Kandaswamy KK, Chou K-C, Martinetz T, Moller S, Suganthan PN, Sridharan S,

Pugalenthi G: AFP-Pred: A random forest approach for predicting antifreeze proteins from sequence-derived properties. J Theor Biol 201 1 , 270:56-62.

Kelley LA, Sternberg MJE: Protein structure prediction on the Web: a case study using the Phyre server. Nat Protoc 2009, 4:363-71 .

- Knight, C.A., Hallet, J. and De Vries, A.L. (1988). Solute effects on ice recrystallization: an assessment technique. Cryobiology, 25, 55-60.

Kristiansen E, Pedersen S, Raml0v H, Zachariassen KE: Antifreeze activity in the cerambycid beetle Rhagium inquisitor. J Comp Physiol B 1999, 169:55-60.

Lin F-H, Graham L a, Campbell RL, Davies PL: Structural modeling of snow flea antifreeze protein. Biophys J 2007, 92:1717-23.

Lin X, Wisniewski ME, Duman JG: Expression of two self-enhancing Antifreeze Proteins from the beetle Dendroides canadensis in Arabidopsis thaliana. Plant Mol Biol Report 201 1 , 29:802-813.

- Liou, Y. C, Daley, M. E., Graham, L. a, Kay, C. M., Walker, V. K., Sykes, B. D., & Davies, P. L. (2000). Folding and structural characterization of highly disulfide-bonded beetle antifreeze protein produced in bacteria. Protein expression and purification, 19, 148-57.

Liou YC, Tocilj A, Davies PL, Jia Z: Mimicry of ice structure by surface hydroxyls and water of a beta-helix antifreeze protein. Nature 2000, 406:322-4.

- Mao, X., Liu, Z., Ma, J., Pang, H., & Zhang, F. (201 1 ). Characterization of a novel β- helix antifreeze protein from the desert beetle Anatolica polita. Cryobiology, 62, 91 -9. Marshall C: Identification of the ice-binding face of antifreeze protein from Tenebrio molitor. FEBS Lett 2002, 529:261-267.

Nicodemus J, O'tousa JE, Duman JG: Expression of a beetle, Dendroides canadensis, antifreeze protein in Drosophila melanogaster. J Insect Physiol 2006, 52:888-96.

- Petersen TN, Brunak S, von Heijne G, Nielsen H: SignalP 4.0: discriminating signal peptides from transmembrane regions. Nat Methods 201 1 , 8:785-6.

Qin W, Doucet D, Tyshenko MG, Walker VK: Transcription of antifreeze protein genes in Choristoneura fumiferana. Insect Mol Biol 2007, 16:423-34.

Raymond, J.A., Fritsen, C, and Shen, K. (2003). An ice binding protein from an Antarctic sea bacterium. FEMS Microbiol. Ecol. 61 , 214-212.

Raymond, J.A. and Janech, M.G. (2009). Ice-binding proteins from enoki and shiitake mushrooms. Cryobiology 58, 151 -156.

Schoonooghe, S., Leoen, J. and Haustraete, J. (2012). Production of antibody derivatives in the methylotrophic yeast Pichia pastoris. Methods Mol. Biol. 907, 325-340 - Scotter AJ, Marshall CB, Graham L a, Gilbert J a, Garnham CP, Davies PL: The basis for hyperactivity of antifreeze proteins. Cryobiology 2006, 53:229-39.

- Tomczak, M.M., Marshall, C.B., Gilbert, J.A. and Davies, P.L. (2003). A facile method for determining ice recrystallization inhibition by antifreeze proteins. Biochem. Biophys. Res. Com. 31 1 , 1041 -1046.

- Tyshenko, M. G., d' Anjou, M., Davies, P. L., Daugulis, A. J., & Walker, V. K. (2006).

Challenges in the expression of disulfide bonded, threonine-rich antifreeze proteins in bacteria and yeast. Protein expression and purification, 47, 152-61.

Van Leeuwen, T., Stillatus, V and Tirry,L. (2004). Genetic analysis and cross-resistance spectrum of a laboratory-selected chlorfenapyr resistant strain of two-spotted spider mite (Acari: Tetranychidae). Exp. Appl. Acarol. 32, 249-261 .

- Venketesh, S. and Dayananda, C. (2008). Properties, Potentials and prospects of Antifreeze proteins/ Crit. Rev. Biotechnol. 28, 57-82.

Yue, C.-W., & Zhang, Y.-Z. (2009). Cloning and expression of Tenebrio molitor antifreeze protein in Escherichia coli. Molecular biology reports, 36, 529-36.

- Zhang D-Q, Liu B, Feng D-R, He Y-M, Wang S-Q, Wang H-B, Wang J-F: Significance of conservative asparagine residues in the thermal hysteresis activity of carrot antifreeze protein. Biochem J 2004, 377(Pt 3):589-95.

Claims

1 . Use of a protein, comprising a continuous repeated sequence with at least eight CXX repeats, of which at least 4 are CXN sequences, as an antifreeze protein.

2. The use of a protein according to claim 1 , wherein said protein comprises SEQ ID N° 1 .

3. The use of a protein according to claim 1 , wherein said protein comprises SEQ ID N° 2.

4. The use of a protein according to claim 1 , wherein said protein comprises SEQ ID N°3.

5. The use of a protein according to any of the previous claims, wherein said protein is a mite protein.

6. The use of a protein according to claim 5 wherein said mite is Tetranychus urticae

7. The use of a protein according to claim 1 , wherein said protein comprises a sequence, selected from the group consisting of SEQ ID N°4 - SEQ ID N° 24.

8. The use of a protein according to any of the claims 1 -7 for protecting food against freeze damage.

9. The use of a protein according to any of the claims 1 -7 for protecting cells against freeze damage.

10. A freeze tolerant recombinant host cell, comprising a recombinant protein with at least eight CXX repeats, of which at least 4 are CXN sequences.

1 1 . A freeze tolerant recombinant plant, comprising a protein with at least eight CXX repeats, of which at least 4 are CXN sequences.