Isolation and characterization of allergens from curvularia lunata 2

2.1.3 Identification of allergens by western blotting Although, the two methods mentioned earlier are robust and are well known for high throughput identification of proteins, they iden

CHAPTER 2:

IDENTIFICATION OF Curvularia lunata ALLERGENS

2.1 INTRODUCTION

Amongst the various fungal aero allergens found in the Singapore environment, the

genus Curvularia (as explained earlier) was found to be a fungus of great importance Several studies carried out previously on Curvularia suggest it to be an important allergenic fungus of medical importance (Gupta et al., 1999; Gupta et al., 2000; Chew

et al., 2000; Asero and Botazzi, 2001; Schroeder et al., 2002; Bisht et al., 2002; Green

et al., 2003; Calhoun, 2004) Although much literature has described Curvularia to be

an important fungus, very few reports (two studies) have actually tried to isolate and

characterize individually, the underlying allergenic components of Curvularia in

detail The first study describes the amino terminal sequence (GLTQKSAPWGLGADTIVAVELDSY) of a glycoprotein allergen (Cur l 1) showing

similarity in sequence and activity with serine proteases (Gupta et al., 2004) The latest study on Curvularia lunata describes cloning, expression and characterization of a 48 kDa recombinant enolase allergen, named as Cur l 2 (Sharma et al., 2006)

The first and foremost step for generating a total recombinant allergen repertoire from

C.lunata, high throughput identification of allergens is required Various genomics and proteomics methods for rapid and high throughput identification of proteins can be utilized for this purpose On the genomics side, methods such as whole genome short-

gun sequencing (Venter et al., 1992), genome microarrays (Liu et al., 2006) and Expressed Sequence Tagging (EST) (Adams et al., 1991) are commonly used On the

proteomics side, protein microarrays (Petrik, 2006), high performance liquid chromatography (HPLC), Surface-Enhanced Laser Desorption Ionization - Time Of Flight (SELDI-TOF) (Elek and Lapis, 2006), Two Dimensional Sodium Do-decyl

Sulfate Poly Acrylamide Gel Electrophoresis (2D SDS PAGE) followed by mass

spectrometry (Lee, 2001), Isotope-Coded Affinity Tags (ICAT) (Allison et al., 2006),

methods are being used A combinatorial method exploiting more than one of the above mentioned techniques can prove more useful for better allergen identification and isolation

2.1.1 Expressed Sequence Tagging for rapid allergen transcript identification

The sequence tagging approach is one of the most effective approaches towards large scale expressed proteome profiling In this technology, a library of directionally cloned partial DNA sequences from randomly selected cDNA clones (termed as ESTs) is generated These clones are then sequenced and the generated sequences are aligned with the available known nucleotide and protein databases for putative matches Single pass sequencing of these clones creates high throughput expressed proteome sequences

for a particular organism spanning various regions of the proteome (Adams et al.,

1991; 1993) This approach has been successfully used in discovering novel expressed

genes in many cell/tissue/organ types (Gong et al., 1994; Gross et al., 2001; Jia et al.,

2001; Escribano and Coca-Prados, 2002) It also provides the profile and abundance of

the expressed genes in the source cDNA library (Adams et al., 1995) The cDNA

library being representative of the expressed genes, ESTs provide a powerful technique for indirect genome identification The deduced amino acid sequences from the cDNAs corresponding majority of the mRNAs help in elucidating the primary structure of the expressed proteins (Yamamoto and Sasaki, 1997) Moreover, identification of differentially expressed genes between two states (e.g normal and

diseased, early and late, juvenile or adult) is possible by using this technique (Schmitt

et al., 1999)

To date, there are over 250 reported publications which used ESTs in identification of fungal genes The majority of them concentrate on identification of differentially expressed genes in pathogenic fungi in order to find out possible pathogenesis related

genes (Mammadov et al., 2005; Sexton et al., 2006); identification of novel enzymes

or other biochemicals for various biotechnological interests (Morrita et al., 2006; Shibuya et al., 2006) Recently, ESTs were used for identification of fungal allergens from Beauveria bassiana (Westwood et al., 2006) EST approach by itself does not

provide information about specific proteins and hence, other techniques in combination with EST are required

2.1.2 Identification of allergens by 1D and 2D SDS PAGE followed by tandem mass spectrometry

Proteomic analysis has been one of the most powerful methods for identification of novel proteins as well as in studying protein expression in organisms under different

environmental conditions (Elinbaum et al., 2002) Along with the transcriptome

analysis, it reveals post-translational regulation and modifications of extracellular

proteins (Oda et al., 2006) In this approach, a 1D or 2D SDS PAGE of the total

protein is used to generate a proteomic profile of the organism These bands/spots as generated by the protein gels are cut, trypsinized and sent for tandem mass spectrometric identifications by Matrix Assisted Laser Desorption/Ionization – Time

Of Flight (MALDI-TOF) and Mass Spectrometry (MS-MS) Mass spectrometry generates peptide mass fingerprints and peptide fragment ion data which are then used

to search for protein candidates in the NCBI database as well as other locally available

or generated databases Proteomic identification of the fungal proteins is been used

extensively for various purposes (Brosson et al., 2006; Carberry et al., 2006; Kalari et

al, 2006; Oda et al., 2006) Proteomics method of 2D SDS PAGE followed by mass

spectrometry was used for identification of fungal allergens by a group of researchers from Taipei Veterans General Hospital, Taiwan Not so long ago, a serine protease

allergen (Rho m 2) from Rhodotorula mucilaginosa (Chou et al., 2005), enolase from Penicillium citrinum and Aspergillus fumigatus (Lai et al., 2002) and a 33kDa heat- labile alkaline serine protease-like allergen from P citrinum (Shen et al., 1997) were

identified using this method

2.1.3 Identification of allergens by western blotting

Although, the two methods mentioned earlier are robust and are well known for high throughput identification of proteins, they identify the underlying genes or proteins by homology alignments (BLASTX in case of ESTs and BLASTP in case of Proteomics) Homology alignment may give a clue about the identity of the protein but cannot be used as confirmative to prove a protein to be an allergen Hence, immunochemistry is commonly combined with 1D or 2D SDS PAGE As the allergens are IgE binding proteins, they are detected by using patients` sera (containing IgEs) which specifically bind to the respective allergenic proteins separated on a 1D of 2D protein gel The bands/spots are then cut, digested with trypsin to generate random peptides and are sent for mass spectrometric identification Western blotting has been commonly used

for identification of molecular weights of the allergenic proteins Recently, Barbieri et

al (2005) used this technique to identify the allergenic components from the fungus

Metarhizium anisopliae Although this technique can identify the putatively allergenic components in total protein extract, protein identification is not possible

Hence, in the present study, we combined all the three techniques: Western blotting, Proteomics and ESTs were combined to obtain the confirmed identity of the allergenic components Firstly, putative allergens were obtained by generating ESTs Total

protein extracts from Curvularia were run on a 1D SDS PAGE and the components

were separated by their molecular weights These were then transferred to a nitrocellulose membrane and western blotting (using allergic patients` sera) was carried out in order to identify the allergenic components Simultaneously, 2D SDS PAGE was run to separate these proteins by molecular weights as well as isoelectric point (PI) Then, these bands/spots (from the corresponding 1D and 2D gels) were cut and sent for mass spectrometric identification The generated peptides were then compared with the in-house generated ESTs as well as with the global protein databases in order to establish the allergen protein identity as well as to know the amino acid/nucleotide sequence

2.2 MATERIALS AND METHODS

2.2.1 Expressed Sequence Tagging of C lunata for allergen identification

2.2.1.1 Fungal culture and raw material

A pure culture of Curvularia lunata (obtained in-house previously) was cultured in

Erlenmeyer flasks (1L) containing 200 ml of 3% Sabouraud`s liquid medium (Oxoid)

at 28˚C for 12-15 days until sufficient sporulation occurred This method was preferred

as it was known to yield a highly potent and allergenic extract (Gupta et al., 1999) At

the end of the incubation period, the spore-mycelial mass (fungal mat) was collected in

a 50 ml (Falcon) tubes The mat was then washed thoroughly with distilled water to remove spent medium and was lyophilized overnight

2.2.1.2 Bacterial strains

The following bacterial strains (E.coli) were used for the preparation of C.lunata

cDNA library and ESTs:

XL1-Blue [N1] ∆(mcrA) 183 ∆(mcrCB-hsdSMR-mrr)173 end A1 supE44 thi-1

recA1gyr 1A96 relA1 lac[F’proAB lacI q Z ∆M15Tn10(Tetr)]

SOLRTM me14-(McrA - ) ∆(mcrCB-hsdSMR-mrr)171 sbcC recB recJ uvrC

umuC::Tn5(Kan r )lac gyrA96 relA1 thi-1 endA1 λR

[F’ proAB lacI q Z

∆m15) c

Su BL-21 (DE3) F-ompThsdSB (r - B m - B )galdcm(DE3)pLysS

-ExAsist ® interference-resistant helper phage (~1.0 x 1010 pfu/ml) Single-strand size

is 7.3kb [co-migrates with ~5kb of double-strand linear DNA on 1% (w/v) agar]

2.2.1.3 Curvularia lunata mRNA extraction

One gm of the dried fungal mat was powdered with liquid nitrogen RNA extraction was performed using RNeasy mini kit (QIAGEN) as per manufacturer’s protocol The eluted total RNA was used for further isolation of mRNA using Poly (A) Quick mRNA isolation kit (Stratagene) as per manufacturer’s protocol

2.2.1.4 Curvularia lunata λZAPII cDNA library

The cDNA library of the extracted C.lunata mRNA (as mentioned above) was

prepared (with the help of Ms Wong Fei Ling) using uni-ZAP (Stratagene) XR vector system (Figure 2.1) as per manufacturer’s protocols A primary library of 105 phage

was amplified to generate a higher titer of 109 pfu Inserts of lengths between 0.5 and 2.5 kb were found on preliminary survey

2.2.1.5 Curvularia lunata EST clones

Exassist helper phage was used for pBluescript phagemid excision (Figure 2.2) from

λZAP using the host E.coli, XL1-MRF strain The single-stranded phagemid was converted to the double-stranded one using SOLR E.coli strain Isolated individual

colonies with the phagemid with (cloned cDNA) insert were subcultured onto plates containing 2% Luria Bertani (LB)-agar (DIFCO) and allowed to grow A total of 3,000 colonies were picked from the plates and kept as glycerol stock (15% glycerol) at -80˚C till further use

These colonies were then inoculated in 5ml of 2% LB liquid medium and cultured for 16-20 hours at 37˚C Plasmid extraction was performed using QIAprep kit (QIAGEN) These plasmids were then stored at -20˚C till use The inserts from the extracted plasmids were then sequenced from the 5` end

2.2.1.6 Sequencing of the inserts

Sequencing of the inserts was carried out using ABI PrismTM dye terminator cycle sequencing ready reaction kit (Applied Biosystems) Each 20µl PCR reaction involved

a mixture containing 4µl of BigDyeTM, 2.5X sequencing buffer (Applied Biosystems), 250-500ng template DNA, 3.2pmol T3 primer and sterile double distilled water to make up the volume Thermal cycling steps (30 cycles) were as follows: denaturation – 96˚C for 30s, annealing - 50˚C for 15s and extension – 60˚C for 4 min Sequencing was carried out using PTC-100TM thermal Controller (MJ Research)

Figure 2.1: Map of Uni-ZAP XR insertion vector

Figure 2.2: Map of pBluescript SK (+/-) phagemid

Precipitation of the PCR product after sequencing was carried out using 2µl of 3M sodium acetate, pH 4.6, 50µl of 95% ethanol, 2µl of 125mM EDTA and 10µl of sterile distilled water The mixture was centrifuged at 13,000g for 20min after incubation at -20˚C for 30 min The pellet was washed with 500µl of 70% ethanol and air dried before loading it on a sequencer

Purified products were subjected to ABI Prism (ABI 3100) automated DNA sequencer (Applied Biosystems) The sequencing services were provided by DNA Sequencing Laboratory (DSL), Department of Biological Sciences, National University of Singapore

2.2.1.7 Sequence analyses of the inserts using various softwares

The electrophoreograms (.ABI files) for various sequenced ESTs were analyzed using the Phred-Phrap-Cross_Match software package program (Version 10.0) by CodonCode Corporation (USA) This software package helps in analyzing the EST electrophoreogram sequences for base calling, sequence assembly and comparisons by classifying the sequences into various contigs Firstly, the sequences were subjected to

Phred (Ewing and Green, 1998; Ewing et al., 1998; Green and Ewing © 1993-1996)

for reading the DNA sequencing trace files, base calling and assigning sequence quality value to each called base The quality value is an error probability (log-transformed) given by the formula; Q= -10 log10 (Pe), where Q is the quality value and

Pe is the error probability of a particular called base PHD2FASTA software then extracted information from the Phred (.phd) files and created input files for next program Briefly, this software transformed all the sequences from ABI files to FASTA format Further, these transformed sequences were analyzed using

Phrap/Cross_Match (Green, © 1994-1996) software Briefly, the Cross_Match software compares a set of reads to a set of vector sequence and produces vector-masked versions of the reads screens and trims it The edited sequences are then analyzed by Phrap to generate contig sequence (mosaic of the highest quality read segments) rather than generating consensus sequences providing extensive assembly information which then aids in trouble-shooting assembly problems and ability to handle large datasets The sequences were assembled and grouped into different contigs as explained above Sequences that were not in any contigs were considered as singletons

2.2.1.8 Redundancy analysis of the analyzed EST sequences

Redundancy reflects the nature of the generated ESTs The % Redundancy Vs No of ESTs plot reflects the trend of ESTs being classified into contigs When this plot reaches a plateau, it suggests that the ESTs are getting more and more redundant This means that the chance of getting novel sequences is getting lesser and lesser Hence, when the plot saturates, it is advisable to stop further sequencing as it will just yield redundant sequences

The assembled sequences were analyzed for contigs and reads with sets of 100 sequences with subsequence increments of 100 sequences in following sets in order to find the redundant sequences Percentage redundancy Vs number of analyzed ESTs was plotted in order to obtain the % redundancy for the analyzed ESTs, where

%Redundancy = (The total ESTs represented by all contigs – No of contigs) / Total

no of analyzed ESTs

2.2.1.9 Sequence homology search for the ESTs and cataloguing into various biochemical groups

Sequence similarity may aid in identification of the putative function of the generated ESTs; as sequence identity may also infer functional identity Hence, the edited sequences (after vector sequence trimming) were analyzed against a non-redundant protein database in the GenBank using BLASTX (translated query vs protein database) sequence alignments for putative functions Identification was generally based on high sequence identity over a long length of sequence So, results with E-values <0.001, Bit score <100 and 6-8 contiguous amino acid similarity were considered as significant identities The sequence alignments were done with the help

of National Center for Biotechnology Information (NCBI) site (www.ncbi.nlm.nih.gov/BLAST) The sequences with significant identities (after BLASTX) were catalogued according to their putative biological functions Various

biochemical categories used in this catalogue were based on Adams et al., (1993)

classification system with some modifications (a category with sequences showing

similarity to allergens was also included)

2.2.2 Identification of allergens by Proteomics and Western Blots

2.2.2.1 Total protein extraction of the cultured fungus

Total protein extraction was carried out using trichloroacetic acid (TCA) / acetone method 1g of dried fungal mat was powdered with liquid nitrogen 10 ml of TCA extraction solution (10% TCA, 0.007% DTT) was added to the powder and incubated

at -20˚C for 1-2 hours The pellet, after centrifuging (35, 000g for 20 min at 4˚C) was

washed three times with sample washing solution (0.007% DTT in acetone) with an hourly interval of incubation in -20˚C between washes The pellet was then lyophilized and incubated at -20˚C until use

2.2.2.2 One-dimensional sodium dodecyl sulfate – polyacrylamide gel electrophoresis (1D SDS-PAGE)

SDS-PAGE (17cm) was performed as per Laemilli`s method (Laemilli, 1970) under reducing conditions The resolving gel contained 14% (w/v) acrylamide, 0.04% (w/v) bis-acrylamide, 375mM Tris-HCl pH 8.8 and 0.1% (w/v) SDS Ammonium persulfate (1mg/ml) and 0.04% (v/v) TEMED was used for polymerization The stacking gel consisted of 4.5% (w/v) acrylamide, 0.12% (w/v) bis-acrylamide, 125mM Tris-HCl

pH 6.7 and 0.1% (w/v) SDS and was polymerized as mentioned above for the resolving gel Extracted protein samples (from the fungal extracts) were dissolved in protein sample buffer (10mM Tris-HCl pH 6.8, 1% SDS, 1% β-mercaptoethanol, 1% glycerol and 0.01% bromophenol blue) for 8-10 min The denatured extracts were then loaded onto the gel and electrophoresis was carried out in SDS-PAGE electrophoresis buffer (25mM tris-base, 0.19M glycine pH 8.3, 0.1% (w/v) SDS) at 80V for 15min, 100V till the gel was completely run After electrophoresis, protein gels were stained with Coomassie Brilliant Blue R250 [0.25% (w/v) Coomassie brilliant blue in methanol: glacial acetic acid: water ::: 10:10:80] For half an hour followed by de-staining overnight in 10% acetic acid (v/v) and 10% methanol (v/v) The separated proteins were compared with broad range protein marker mix (Bio-Rad Laboratories)

2.2.2.3 Two-dimensional sodium dodecyl sulfate – polyacrylamide gel electrophoresis (2D SDS-PAGE) and staining

For the first dimension, isoelectric focusing (IEF) was carried out Total protein extraction was carried out using TCA, as mentioned earlier Around 600µg protein sample was then dissolved in sample buffer [9M urea, 4% (w/v) CHAPS, 100mM Dithiothreitol, 0.2% (v/v) Bio-Lytes ampholytes pH 3-10 (Bio-Rad Laboratories), 35

mM tris base] After centrifugation at 10, 000 g for 10 min, the supernatant was dissolved in rehydration buffer [8M Urea, 10mM dithiothreitol, 0.5% (w/v) CHAPS and 0.2% (v/v) Bio-lytes ampholytes pH 3-10 (Bio-Rad Laboratories)] and applied to IEF with an immobilized pH gradient gel (IPG) strip [17 cm long ReadyStrip IPG strips, pH 3-10 NL (Bio-Rad Laboratories)] IEF was carried out using PROTEAN® IEF Cell (Bio-Rad Laboratories) according to the manufacturer’s protocol The IEF steps were as follows: active rehydration for 12-16 hours at 50V, 250V for 15 min, 8,000V for 4 hours followed by linear voltage ramping to reach 80, 000Vh Reduction and alkylation of the proteins on the strips was achieved by incubating with 130mM Dithiothreitol and 135mM Iodoacetamide respectively in equilibration buffer [6M urea, 0.375M tris-HCl, pH 8.8, 2% (w/v) SDS, 20% (v/v) glycerol] at room temperature for 15 min each After IEF, the strip was loaded onto second dimension separation of proteins which was performed by running SDS-PAGE as described earlier The separated proteins were compared with broad range protein marker mix (Bio-Rad Laboratories) After running the gel, it was incubated overnight in Fixative solution [50% (v/v) methanol and 10% (v/v) acetic acid] Instead of staining the gels with Coomassie Brilliant Blue, silver staining was carried out The gels were removed

from the fixative solution and washed three times with distilled water for 15 min each with gentle shaking The gels were then washed with 0.02% (w/v) Sodium thiosulfate reagent followed by two washes with distilled water for 1 min each After washing, the gels were stained with Silver nitrate reagent [0.2 % (w/v) silver nitrate, 0.02 % (v/v) formaldehyde], followed by two washes with distilled water for 1 min each Development of the color was achieved by using development solution [3% (w/v) sodium carbonate, 0.05% (v/v) formaldehyde] After staining the gels, 1.6% (w/v) EDTA was used to stop the color development Gels were incubated in this solution for 10 min with gentle shaking followed by three washes with distilled water for 10 min each The gels were then incubated in distilled water till further use

Western Blotting to identify the IgE binding proteins in the fungal extract

The proteins separated in SDS-PAGE were electro-blotted (Towbin et al., 1976) using

transfer buffer (25mM tris-base, 192mM glycine, 20% (v/v) methanol, pH 8.3) on PolyVinylidine DiFluoride (PVDF) membrane (Hybond-PVDF, Amersham Biosciences) overnight on 30V at 4˚C) The membrane was then blocked with 5% (w/v) skimmed milk (Anlene) in PBS [0.8% (w/v) NaCl, 0.02% (w/v) KCl, 0.144 % (w/v) Na2HPO4 and 0.024% (w/v) KH2PO4, pH 7.4] for 1 h Following the blocking step, the membranes were washed three times (15 min, 10 min and 7 min respectively) with PBST [PBS with 0.05% (v/v) Tween 20] at room temperature The membranes were then incubated with atopic patients` sera as well as controls overnight at 4˚C After washing three times with PBST as mentioned earlier, the membrane was incubated with 1:1000 diluted horse radish peroxidase (HRP) conjugated anti-human IgE secondary antibody (Sigma A9667) for 1 h at room temperature IgE binding

protein bands were visualized using ECLTM Western blotting detection reagents (Amersham) as per manufacturer’s protocol

2.2.2.4 Tandem Mass Spectrometric analyses

The bands corresponding to the IgE binding bands as obtained from the 1D western blots (as described above) were cut from the simultaneously run and Coomassie stained 1D SDS PAGE Also, various spots were cut from the 2D SDS PAGE The excised protein bands/spots were then digested with 0.1µg/µl of modified, sequencing grade Trypsin (Promega) The details of trypsin digestion protocol can be obtained from http://www.dbs.nus.edu.sg/research/ppc/index.htm MALDI-TOF-TOF mass spectrometric analysis of the generated tryptic peptides was carried out at The Proteins and Proteomics Centre (PPC), Department of Biological Sciences, National University

of Singapore, Singapore (http://www.dbs.nus.edu.sg/research/ppc/index.htm) Analysis was performed using an intranet version of MASCOT 1.7 (MATRIX SCIENCE), with the peptide masses assumed to be monoisotopic and protonated ions, allowing some peptide modifications viz cysteine carbamidomethylation, protein N-acetylation and methionine oxidation Scores greater than 78 were considered as significant (p<0.05) The maximum number of missed cleavages and the peptide mass tolerance was set to 1 and ±110 ppm respectively Fragment mass tolerance was set to ±0.2 Da Some of the randomly selected peptides were further sent for MS/MS analysis in order to obtain a peptide summary report which would give a better picture of the results The processed data was then searched against NCBI database as well as the in-house generated

C.lunata EST sequences via a Mascot search engine in order to find the identity as

well as the exact cDNA sequence of the obtained peptides and hence the excised bands/spots on 1D as well as 2D SDS PAGE

2.3 RESULTS AND DISCUSSION

2.3.1 Expressed Sequence Tagging of C lunata for allergen identification

2.3.1.1 Curvularia lunata cDNA library

A cDNA library represents information of the encoded mRNA giving a brief picture of the pattern of expression for the organism/state/condition under study The cDNA

library of C.lunata was made with a mixture of mycelial fragments as well as spores

This was done to ensure that none of the allergenic proteins expressed specifically in a

particular stage would be missed and hence a full repertoire of C.lunata expressed

allergens would be obtained A non-normalized library was used in order to know the expression levels of various transcripts as well as possible variants present in the fungal genome

2.3.1.2 Sequencing of C.lunata ESTs

Single pass sequencing of each EST (5` to 3`) was carried out Sequencing from 5`-3` end would help in identifying transcription/translation start sequence for individual Open Reading Frames (ORFs) and avoiding 3` untranslated regions (3` UTRs) as well

as the polyadenylation signals providing with a possibility of getting a full-length

sequence for a particular ORF Moreover, 5` ESTs are considered as gene family

specific ESTs as the genes belonging to the same family tend to have a conserved

functional motif and hence might be conserved at their respective 5` ends (Hillier et

al., 1996) Out of the sequenced ESTs, a total of 1683 ESTs passed the criteria of Phred and Phrap/Cross_Match analysis

2.3.1.3 Assembly of the ESTs into contigs/singletons and redundancy analysis

For the 1600 ESTs analyzed, 891 (55.7%) ESTs were represented in 201 contigs

Remaining 709 (44.3%) ESTs were classified as Unigenes Unigenes contained

contigs with one EST (125) or singletons (584) Phrap may consider a contig with one EST when the EST is homologous with other contig, but the homologous score is lower than the allowed score to assemble it into that contig Hence, the contigs which

remained with single EST were then considered as unigenes The largest contig

contained 98 ESTs whilst the smallest contig had 2 ESTs

EST redundancy percentage was calculated with the following formula, %Redundancy

= [(Total no of ESTs represented by all contigs – No of contigs)/Total no of ESTs analyzed] x 100 The details of the % redundancy calculations for consecutive 100 ESTs analyzed are as shown in Table 2.1 and Figure 2.3 As seen from the Table, %

redundancy for C.lunata was found to be around 43% This means that there was a 43% probability that a new EST from C.lunata obtained after this would already be

represented in the current data set Hence, no more sequencing of ESTs was carried out and the available 1683 ESTs were used for further cataloguing and analysis

C.lunata redundancy was similar to that of the dust mite Dermatophagoides farinae

ESTs (~43%) generated in-house The redundancy rate of other in-house ESTs from

the dust mites Blomia tropacalis and Tyrophagus putrescentiae was comparatively

lower (~30%) This could be due to the difference in the quality of cDNA libraries constructed or due to the normalization of the libraries carried out by pre-hybridization

Table 2.1: % Redundancy rate for C.lunata ESTs

*No of Unigenes = No of contigs with single EST + No of singletons

No of contigs

ESTs represented

by all contigs

% Redundancy

of the cDNA library with highly redundant clones Although pre-hybridization would allow the occurrence of the poorly expressed genes, a non-normalized library (in case

of C.lunata) can be explored for the presence of gene polymorphisms, alternative transcripts and for differential levels of gene expression (Lee et al., 1995; Burke et al.,

1998; Buetow et al., 1999)

2.3.1.4 BLASTX homology alignments of ESTs to search for putative function

All the trimmed ESTs after the Phrap/Phred analysis were subjected to BLASTX search alignments at NCBI website (www.ncbi.nlm.nih.gov/BLAST) BLASTX compares 6 frames of a translated query nucleotide sequence (ESTs) against GenBank non-redundant protein sequence database Identification of the homologs was based on high sequence similarity over a contiguous stretch of amino acids The significance of the given alignment with score (S) is represented by the expect value (E-value) E-being the expected number of chance alignments with a score S or better is inversely proportional to S An E-value of 10-3 was used as an optimal cut-off This cut-off was previously standardized in-house by comparing the percentage of non-significant identity with various E-values as possible cut-offs for a set of 1000 ESTs

2.3.1.5 Putative biological function assignments to the respective ESTs

The putative allergens were classified into 11 biochemical groups based on BLASTX match with a known protein The 11 groups are as follows: 1) Allergens, 2) Defense and homeostasis related proteins, 3) Gene expression and protein synthesis, 4) Hypothetical proteins, 5) Metabolism related proteins, 6) Nucleotide biosynthesis related proteins, 7) Proteases and inhibitors, 8) Structure, cell surface and motility related proteins, 9) Cell signaling and communication related proteins, 10)

Unclassified proteins and 11) Proteins with unknown homology (Unknown proteins) Figure 2.4 as well as Table 2.2 demonstrates the frequencies of ESTs falling into 11

different biochemical groups As per Adams et al., (1995), a useful library had at least

50% new genes, a broad variety of the transcripts and not more than 20% of

uninformative sequences C.lunata EST library satisfies these criteria where there were

around 800 (47.5 %) of the ESTs which had unknown homology The rest of 52.7% sequences had significant matches (E-value < 10-3) Moreover, the matched ESTs could be classified into various biochemical groups which suggested a broad variety of the generated transcripts The majority of genes which matched a known protein fall into the category of general house-keeping proteins like metabolism (21.5%), gene expression/protein synthesis (7.5%) and nucleotide biosynthesis (3.2%) related proteins, suggesting their high redundancy to be a reflection of high level of expression of these house-keeping genes rather than an artifact of the library A total of

77 (4.6%) ESTs belonged to structural proteins and 22 (1.3%) ESTs belonged to the group of proteins involved in cell signaling and communication Being a known plant

pathogen, C.lunata was thought to have abundance of proteases and proteins related

defense related proteins inhibitors This is so because the proteases aid in the entry of the pathogen by dissolving the host membranes and other matrix proteins On the contrary, only 9 (0.6%) as well as 13 (0.8%) ESTs belonged to proteases/inhibitors and defense related proteins respectively; which was surprising About 87 (5.2%) ESTs showed similarity to hypothetical ORFs as well as proteins from other organisms Due

to the increasing number of fungal as well as non-fungal genomes being sequenced, many putative ORFs generated are attributed as ‘hypothetical proteins’

Figure 2.4: Classification of C.lunata ESTs (1683) into biochemical groups (Adapted from Adams et al., 1993)

32 22 77 13

126 87

364 54

9 99

800

Allergens Cell signaling & Communication

Structure/Cell surface/Motility

Defense and Homeostasis

Gene expression/Protein Synthesis

Hypothetical proteins

Metabolism Nucleotide biosynthesis

Proteases and Inhibitors

This is the reason why some of the C.lunata ESTs showed matches with the

hypothetical proteins Similarly around 6% of the ESTs were labeled as ‘Unclassified proteins’ These ESTs had significant match with a protein in the NCBI database but the protein did not have a function (cysteine rich proteins) and hence were kept under one group as unclassified proteins Although such proteins are currently classified as

‘hypothetical’ or ‘unclassified’, with time and detailed studies of such proteins, it would be possible to annotate functional attributes to such proteins A high number of unknown proteins (47.5%) suggest that there are still many genes of interest present in

C.lunata which could be further studied in detail Hence, the high number of unknown proteins suggests that the EST strategy serves as a very good tool for identifying novel expressed genes from an organism A total of 32 (around 2%) ESTs were classified as

‘Allergens’ as they showed significant sequence similarity with known allergens

2.3.1.6 Putative allergenic proteins obtained from C.lunata EST database

Due to the availability of EST catalogues for C.lunata, many putative allergen

homologs could be identified Out of the 32 different allergen hits obtained, 14 different types of putative allergens were identified The identified allergen types could be classified into fungal as well as non-fungal allergen hit types (Table 2.3) As expected, 12 different types of fungal allergen homologs were obtained This is due to the conserved phylogeny amongst different fungi which might be responsible in generating similar fungal allergenic proteins Two non-fungal (pollen) allergen hits were obtained

Table 2.3: ESTs of C.lunata showing similarities to the known allergens

P

: Partial EST sequence * : ESTs showing sequence similarity with more than one

known allergens U : EST hits with allergens with unknown biochemical functions

No Identity to known allergens

No of ESTs

Variants

Fungal Homologs

2 Asp f 6 (Manganese Superoxide Dismutase, MnSOD)

[Aspergillus fumigatus]

4 Asp f 15U precursor (Asp f 13) [Aspergillus fumigatus] 1

5 Pen n 18/Asp f 1 P* (Vacuolar Serine Protease)

[Penicillium notatum, Aspergillus fumigatus]

6 Pen c 19/Cla h 4P* (Heat Shock Protein 70)

[Penicillium citrinum, Cladosporium herbarum]

1

7 Alt a 10/Cla h 3P* (Aldehyde Dehydrogenase)

[Alternaria alternata, Cladosporium herbarum]

8 Can a 1P (Alcohol Dehydrogenase) [Candida albicans] 2 -

10 Mal f 4P (Malate Dehydrogenase) [Malassezia furfur] 2 2

11 Asp f 11 / Bet v 7* (Cyclophilin) [Malassezia

sympodialis, Betula verrucosa]

12 Tri r 4P (Serine Protease) [Trichophyton rubrum] 1

Pollen Homologs

13 Jun o 2 (Ca+2 binding protein) [Juniperus oxycedrus] 1

14 Par j 3/Hev b 8 * (Profilin) [ Parietaria judaica, Hevea

brasiliensis]

Total 32

Most of the allergen hits showing similarity with known fungal allergens showed

similarity to the allergens from Aspergillus fumigatus viz Asp f 2, Asp f 6, Asp f 7,

Asp f 11 and Asp f 15 Nine out of the obtained 32 ESTs with similarity to putative

allergens belonged to A fumigatus allergens Moreover, the majority of the obtained putative allergens showed similarity to allergens from Aspergillus, Cladosporium and Alternaria species as they all (including C lunata) are ascomycetous fungi Some allergen homologs of a basidiomycete fungus (Malassezia furfur) were also obtained

Among the fungal allergen homologs obtained, the highest number of ESTs (10) matched Cop c 2 (Thioredoxin) allergen suggesting higher expression levels of these proteins This might be due to the fact that thioredoxins play multiple roles in cellular

processes such as proliferation, apoptosis and gene expression (Cho et al., 2001)

The biochemical functions for the allergens Asp f 2, 7 and 15 have not been characterized while the rest of the allergens have been characterized for their biochemical functions as shown in Table 2.3 ESTs with sequence similarities to Asp f

7, Pen n 18, Pen c 19, Can a 1, Alt a 10, Mal f 4 as well as Tri r 4 were found to have partial sequences (due to the truncations at the 3` ends) The rest of all the sequences were found to be full-length protein sequences bearing start codon (ATG) as well as the stop codons at the 5` and 3` ends respectively The 2 non-fungal allergen hits matched to those of pollen allergens (Jun o 2 as well as Par j 3/Hev b 8)

Enolase has been known be an important allergenic protein in various fungi e.g Asp f

22w (Aspergillus fumigatus) 46, Pen c 22w (Penicillium citrinum), Alt a 6 45(Alternaria alternata), Cla h 6 (Cladosporium herbarum) [Achatz et al., (1995)] Recently, enolase from C lunata has also been isolated, cloned, expressed and purified

by Sharma et al., 2006 Surprisingly, no Enolase allergen homologs were obtained from the generated C lunata ESTs Similarly, various other commonly found fungal

allergens such as ribosomal proteins P1 and P2, various metalloproteases, glutathione

S-transferases were not found to be present in the C lunata ESTs A possible

explanation could be due to the variable levels of expression of these proteins in

various fungi present in different niches

Possible multiple variants (isoallergens) for a particular protein were looked for in the

obtained C.lunata ESTs As reported in http://www.allergen.org, members of an

allergen group which have >67% amino acid sequence identity are designated as

Isoallergens Further, each isoallergen may have multiple forms of closely similar

sequences, designated as Variants Among the 14 different types of allergens obtained,

8 allergen types had more than one EST representing them However, only 3 allergen types [Asp f 6 (MnSOD), Alt a 10 (Aldehyde dehydrogenase) and Mal f 4 (Malate dehydrogenase) showed the presence of possible isoallergens For example, in the case

of C lunata MnSODs allergen homologs, out of the three ESTS (CL0021, CL0837

and CL1614), two of the ESTs (CL0021 and CL0837) had exactly identical amino acid sequence while CL1614 only had 40% amino acid sequence similarity (in the compared segment as obtained from EST) with the other two sequences, suggesting it

to be a possible isoallergen Isoallergens of polymorphisms in general might occur due

to post-translational modifications or due to convergent/divergent evolution of a particular gene Existence of such sequence variations is important in generating

allergen diversity which might influence allergenicity (Chua et al., 1993, 1996)

ESTs can also provide the means for facilitating proteomic studies and characterization

of allergenic proteins Previously, allergenic proteins screened via 2D-proteomics approach required Edman degradation (N-terminal sequencing) to determine the amino

acid sequence of the obtained protein (Le Mao et al., 1998) Availability of EST

database has greatly facilitated proteomics based protein identification as the obtained peptides can directly be compared with the available EST sequences Therefore, this approach was further used in the identification of the proteins observed on 1D as well

as 2D proteome of C lunata which would be discussed further

Thus, majority of the allergenic components that exist in C lunata can be identified

using the EST approach Furthermore, the EST approach aids in the identification of the novel allergenic components as obtained from western blot of 1D as well as 2D proteome

2.3.2 Identification of allergens by Proteomics

2.3.2.1 Identification of the allergens using 1D western blots

The resolved protein components of various Curvularia species were immobilized

onto a nitrocellulose membrane followed by incubation with a total of 12 fungal atopic

patients` sera reactive to Curvularia Protein components from 2 species of Cladosporium (Cladosporium herbarum and Cladosporium cladosporioides) and from Penicillium notatum were also blotted and developed

The western blots showed that each of the fungal components tested contained several antigenic components binding to atopic patients` IgE The majority of the tested serum samples demonstrated IgE binding to more than one component in the fungal crude