- Email: [email protected]
Identification of chitotriosidase isoforms in plasma of Gaucher disease patients by two dimensional gel electrophoresis
Biochimica et Biophysica Acta 1764 (2006) 1292 – 1298 http://www.elsevier.com.locate/bba
Identification of chitotriosidase isoforms in plasma of Gaucher disease patients by two dimensional gel electrophoresis Lucía Quintana a , Alberto Monasterio b , Kepa Escuredo b , Jokin del Amo b , Pilar Alfonso a , Felix Elortza c , Simon Santa Cruz b , Laureano Simón b , Antonio Martínez b , Pilar Giraldo a , Miguel Pocoví a , José Luis Castrillo d,⁎ a
c
Instituto Aragonés de Ciencias de la Salud, Edificio CEA, Avda. Gómez Laguna 25, 50009 Zaragoza b Proteomika SL, Parque Tecnológico Bizkaia, Edificio 801B, 48160 Derio Spain Unidad de Proteómica, CIC-bioGUNE, Parque Tecnológico de Bizkaia, Edificio 801A, 48160 Derio, Spain d Centro de Biología Molecular “Severo Ochoa” (CSIC-UAM) Cantoblanco, 28049 Madrid, Spain Received 30 March 2006; received in revised form 19 May 2006; accepted 26 May 2006 Available online 6 June 2006
Abstract Chitotriosidase protein (ChT) is the most important biochemical marker described for Gaucher disease (GD). ChT activity is increased several hundred-fold in plasma of GD patients and shows a strong positive correlation with the severity of the disease. However, a recessively inherited enzyme deficiency, with an incidence of about 6% in the Caucasian population, means that not all patients with GD can be monitored by measuring ChT activity. Applying two-dimensional gel electrophoresis (2-DE) technology this study describes the localization and identification of five ChT isoforms in 2-DE images obtained from plasma of GD patients. All these isoforms were unequivocally identified using MALDI-TOF mass spectrometry (MS) and validated by western blot analysis. The features of each ChT isoform separated by 2-DE in plasma from GD patients homozygous for the wild-type ChT allele, carriers of one defective allele and patients homozygous for the mutant allele are presented. We also show the correlation between each ChT isoform and the plasma ChT enzymatic activity of the GD patients sampled in this study. © 2006 Elsevier B.V. All rights reserved. Keywords: Gaucher disease; Chitotriosidase; Biomarker; 2-DE
1. Introduction Gaucher disease (GD) is a lysosomal glycolipid storage disease caused by recessive inherited mutations in the acid β-glucosidase gene. This defect causes a deficiency in the glucocerebrosidase enzyme and leads to the accumulation of glucocerebroside in the lysosomes of macrophage cells. According to the presence and evolution of neurological symptoms, the disease is classified into three clinical types. Type 1, the most common, is characterized by the absence of neuropathic symptoms [1]. Many biomarkers show altered accumulation in the plasma of GD patients and can be used for disease diagnosis and follow-up. The measurement of these molecules is therefore extremely Abbreviations: GD, Gaucher disease; ChT, chitotriosidase ⁎ Corresponding author. Fax: +34 914979097. E-mail address: [email protected] (J. Luis Castrillo). 1570-9639/$ - see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.bbapap.2006.05.009
important in studies of the pathophysiology of GD and for monitoring patient response to therapy. Among them, the chitotriosidase (ChT) enzyme is produced in large amounts by the lipidladen macrophages in GD patients, and has become the most important biomarker of GD [2]. ChT activity is known to be increased several hundred-fold in the plasma of GD patients and high levels of enzyme activity are correlated with disease severity. However, a 24 base pair (bp) duplication in the ChT gene results in the absence of detectable enzyme activity in the plasma of individuals homozygous for the mutant allele [3]. About 6% of the Caucasian population is deficient for ChT activity, while 35% are carriers of the mutant allele and show reduced levels of ChT enzyme activity. Efforts to detect the mutant ChT protein in plasma have been unsuccessful [3]. Moreover, three differently sized isoforms of ChT of approximately 50, 39 and 40 kDa, respectively, have been described although cultured macrophages predominantly produce and
L. Quintana et al. / Biochimica et Biophysica Acta 1764 (2006) 1292–1298
secrete the 50 kDa form [4] which itself appears to exist in different forms as isoelectric point (pI) values of about 6.5 and 7.2 were reported. Renkema et al. [4] concluded that the acidic pI of the 50 kDa ChT isoform secreted by human macrophages was due to the presence of O-linked glycans with variable amounts of sialic acid. Also, they described that pI values obtained by measuring ChT activity after isoelectric focusing of GD plasma samples showed greatest activity at pH 7.2 with minor amounts of activity at pH 6.0 and 8.0 [5]. To our knowledge, this is the first report of 2-DE images and 2DE western blotting of plasma ChT from GD patients. Five isoforms of ChT in 2-DE of GD plasma protein samples were detected and they were studied according to ChT genotypes in GD patients. A connection between the quantitative image analysis of each ChT isoform and the plasma ChT enzymatic activity in each patient is proposed. 2. Materials and methods 2.1. Patients This study was made with plasma samples from 21 type 1 GD patients and 23 healthy subjects. All patients are on the Spanish GD Registry [6]. Written informed
1293
consent was obtained from all patients and from legal representatives of patients aged less than 18 years. Our clinical protocol conformed to both the directives of the Ethical Committee of the Miguel Servet University Hospital, Spain, and the 1983 revision of the Helsinki declaration of 1975. Diagnosis of GD was based on low activity of acid β-glucosidase in leukocytes [7].
2.2. Chitotriosidase enzyme assay and genotyping The ChT enzyme activity assay was based on the method described by Hollak et al. [2] with minor modifications as described by Giraldo et al. [8]. Genotyping of ChT was performed as described by Giraldo et al. [8].
2.3. 2-DE Two-dimensional gel electrophoresis is based on standard procedures with minor modifications for improved protein resolution [9]. Plasma samples were purified using the Aurum serum protein mini kit (Bio-Rad, Hercules, CA, USA) to remove albumin and class G immunoglobulins. For samples from GD patients with high levels of immunoglobulins [10], an additional step in plasma preparation was included by passing samples through HiTrap protein A columns (Amersham Biosciences, GE Healthcare, Uppsala, Sweden) prior to use of the Aurum columns in order to further deplete immunoglobulins. Samples were then precipitated with 5% trichloroacetic acid/acetone followed by centrifugation (15 m, 13,000×g, 4 °C). Pellets were resuspended in 100 μL of rehydration solution containing 7 M urea, 2 M thiourea, 4% CHAPS and 1% dithiotreitol (DTT). Protein content of the resuspended samples was determined by Bradford assay using Bio-Rad Protein Assay (Bio-Rad) following the manufacturer's
Fig. 1. Silver stained 2-DE gels of plasma proteins from GD patients and healthy (non-GD) individuals. (A) Plasma proteins in GD patients were resolved by 2-DE using pH 3–10 NL gradient IPG strips in the first dimension and a 12.5% polyacrylamide SDS gel in the second dimension. The area of interest is framed. (B) The five spots identified as ChT protein are highlighted and numbered in the GD panel. (C) Identical region of 2-DE gel image in non-GD control sample.
1294
L. Quintana et al. / Biochimica et Biophysica Acta 1764 (2006) 1292–1298
Table 1 Proteins identified by MALDI-TOF analysis Spot UniProt No. of Mascot Sequence RMS b MS/MS (#) (Accession peptides Score a coverage peptide number) matched/ % (Ion Score) c Total peptides observed 1
Q13231
8/17
116
18
39
2
Q13231
8/21
121
18
30
3 4 5
Q13231 Q13231 Q13231
13/26 6/9 8/14
164 85 107
25 12 12
11 48 28
QTFVNSAIR (28) ADGLYPNPR (35)
a Mascot Score: Mascot Score, Protein scores greater than 76 are significant ( P < 0.05). b RMS: RMS (root mean square) error of the set of matched mass values in ppm. c MS/MS peptide (Ion score): peptide amino acid sequence and ion mascot score of the chosen precursor ion in brackets.
instructions. 250 μg of purified samples were loaded to 24 cm 3–10 NL IPG strips (Amersham Biosciences). The first dimension (isoelectrofocusing) was performed using an IPG-Phor system (Amersham Biosciences). Rehydration of the IPG strips was carried out at 50 V for 11–12 h, and proteins were focused by gradually increasing the voltage across the IPG strips to 8000 V and maintaining this voltage for at least 70–80 kVh. After IEF, IPG strips were soaked once in equilibration buffer (50 mM Tris–HCl pH 8.8, 6 M urea, 30% glycerol, 2% sodium dodecyl sulphate and 0.002% bromophenol blue) containing 2% DTT and subsequently in equilibration buffer containing 2.5% iodoacetamide. Separation of proteins in the second dimension was performed by denaturing SDS-PAGE on 12.5% polyacrylamide gels. Following electrophoresis, proteins were stained with a Silver Staining Kit (Amersham Biosciences). For protein identification, a silver staining method compatible with MS was used.
Recovered peptides were purified prior to MALDI analysis by home made nano-columns as described by Gobom et al. [12] with some modifications. A column consisting of 100–300 nL of POROS R2 material (PerSeptive Viosystems, Framingham, MA) or activated charcoal from Sigma (C-5510) was packed in a constricted GELoader tip (Eppendorf, Hamburg, Germany) [13]. A 1 mL syringe was used to force liquid through the column by applying air pressure. The column was equilibrated with 0.1% TFA and the bound peptides subsequently eluted directly onto the MALDI target with 0.5 μL aCyano-4-hydroxycinnamic acid (CHCA) solution (20 μg/μL in CAN, 0.1% TFA, 70:30, vol/vol). Peptide mass fingerprinting was performed on a Bruker Ultraflex TOF/TOF mass spectrometer (Bruker-Daltonics, Bremen, Germany). Positively charged ions were analyzed in reflector mode, using delayed extraction. Spectra were obtained by randomly scanning the sample surface. Typically 200–300 spectra were averaged to improve the signal to noise ratio. Spectra were externally calibrated resulting in a mass accuracy of <70 ppm. Protein identification was performed by searching in a non-redundant protein database (NCBI) using Mascot searching engine (http://www.matrixscience.com). The following parameters were used for database searches: missed cleavages 1, allowed modifications carbamido-methylation of cysteine (complete) and oxidation of methionine (partial).
2.6. Production of rabbit polyclonal antibodies against human full-length recombinant chitotriosidase Human RNA was prepared from whole blood using a PAXgene RNA extraction kit following the manufacturer's instructions (PreAnalytiX, Hombrechtikon, CH). The full length ChT cDNA was obtained by reverse transcription and PCR with primers specific for the ChT gene (5′-CACCATGGTGCGGTCTGTGG3′ and 5′-TCAATTCCAGGTGCAGCATTT-3′) using human mRNA as template. PCR was performed using Pfx proofreading polymerase (Invitrogen, Carlsbad,
2.4. Gel imaging and analysis Images of silver stained gels were acquired using an ImageScanner (Amersham Biosciences). Image analysis was performed using Progenesis PG220 software (Nonlinear Dynamics, Newcastle-upon-Tyne, UK). Protein spots were detected using automatic spot detection and the background was subtracted using the mode of non-spot method. The images were also subjected to spot filtering in order to remove artifacts. The total intensity of pixels within each spot (the integrated intensity) was determined by the software. The integrated intensity of each spot (normalized volume) was expressed as percentual fractions of the total integrated intensity of all spots within the region of analysis of the gel (see Fig. 1). This normalizes the amount of any given spot and gives relative protein abundance values for each sample. Significant differences in protein expression levels were determined by Student's t test with a set value of P < 0.05. Those proteins whose expression was up- or down-regulated by 2-fold or greater were excised from gels and identified by Peptide Mass Fingerprinting (PMF).
2.5. Protein identification by mass spectrometry The spots representing proteins of interest were cut from 2D silver stained electrophoresis gels and subjected to in-gel tryptic digestion according to Shevchenko et al. [11], with minor modifications. The gel pieces were swollen in a digestion buffer containing 50 mM NH4HCO3 and 12.5 ng/μL of trypsin (Roche Diagnostics, recombinant, proteomics grade trypsin, Penzberg, Germany) in an ice bath. After 30 min the supernatant was discarded, 20 μL of 50 mM NH4HCO3 were added to the gel piece and the digestion allowed to proceed at 37 °C overnight. The supernatant was transferred to an empty Eppendorf tube and basic and acidic peptide extraction was performed on the gel pieces. Supernatants of each sample were pooled, and dried by vacuum centrifugation. Prior to MS analysis, pellets were resuspended in 10 μL of 0.1% trifluoroacetic acid (TFA).
Fig. 2. 2-DE western blot analysis of plasma protein samples from GD patients and healthy (non-GD) individuals. 2-DE western blot analysis of plasma protein sample from a GD patient before and after neuraminidase incubation. (A) NC membrane stained with colloidal gold after blotting of protein from a 2-DE gel run with a plasma protein sample from a GD patient and probed with the antiserum raised against ChT. The five white spots correspond to the points where levels of ChT were saturating. The area of interest is framed. (B) 2-DE western blot of a plasma protein from a healthy (non-GD) control. (C) 2-DE western blot of a plasma protein sample from a GD patient showing the five immunoreactive spots detected by the antibody raised against recombinant human ChT, this image is identical to the 2-DE blot of the GD patient control in the neuraminidase experiment. (D) 2-DE western blot of the neuraminidase treated plasma protein sample from a GD patient. No signal from ChT isoforms 1 and 2 is detected but a new spot appears above isoform ChT 3. The positions of ChT isoforms 4 and 5 remain unchanged.
L. Quintana et al. / Biochimica et Biophysica Acta 1764 (2006) 1292–1298
1295
the viscosity of the extract. Recombinant histidine-tagged ChT was purified by affinity chromatography using the ProBond Purification System (Invitrogen). After incubation of the cell lysate with the Ni-resin, non-specifically bound proteins were removed by sequential washes with urea buffer of pH ranging from 7.8 to 5.3. His-fusion ChT was eluted in urea buffer pH 4.0 and dialysed against phosphate-buffered saline (PBS) prior to immunisation of rabbits. Each of two New Zealand White rabbits were immunised with 400 μg of recombinant ChT as a 1:1 mixture with Freunds complete adjuvant, followed by three booster injections of 150 μg of protein given at three weekly intervals. Finally, rabbits were exsanguinated and approximately 40 mL of serum per rabbit were collected and frozen in aliquots.
2.7. 2-DE western blot
Fig. 3. Silver stained 2-DE gels showing the region of the gel where ChT isoforms 1 and 2 are located (marked by arrows) in different plasma protein samples. (A) Plasma protein sample from a GD patient homozygous for the wild-type allele of the gene encoding ChT showing ChT isoforms 1 and 2. (B) Plasma protein sample from a GD patient heterozygous for the mutant allele of the ChT gene showing only ChT isoform 2 and two non-related spots in the region of ChT isoform 1 (marked by a circle). (C) Plasma protein sample from a GD patient homozygous for the mutant allele of the ChT gene lacking ChT isoforms 1 or 2. (D) Plasma protein sample from a healthy (non-GD) individual. Note the presence of the two overlapping spots (marked by a circle) and the absence ChT isoform 1 at panels B, C–D.
For western blotting small format 2-DE gels were run. Briefly, 20–25 μg of protein were loaded to 7 cm 3–10 NL IPG strips (Amersham Biosciences) and isoelectric focusing was performed as described above but using only 20 kVh. Equilibration and second dimension were carried out as described previously. Unstained proteins were electroblotted to a nitrocellulose (NC) membrane (Millipore Corporation, Billerica, MA, USA). Non-specific binding to NC was blocked by soaking membranes in a solution of 6% fat-free milk in PBS/Tween 0.2% (PBST) for 1 h. After rinsing with PBST three times, blots were incubated with the anti-ChT primary antibody. After rinsing again with PBST, blots were incubated with a horseradish peroxidase-conjugated donkey anti-rabbit antibody. Horseradish peroxidase was detected using Supersignal West Dura Extended Duration Substrate (Pierce Biotechnology, Inc., Rockford, IL). The chemiluminescent signal was subsequently detected by exposure to Hyperfilm MP (Amersham Biosciences). Blotted membranes were stained with Protogold (BBInternational, Cardiff, UK) to reveal the full protein map.
2.8. Glycosylation analysis CA) with primer annealing at 62 °C. Gel purified PCR products were cloned into pET100/D-polyhistidine TOPO bacterial expression vector (Invitrogen) in frame with an N-Terminal (6XHis) tag. The integrity of the full-length clones obtained was verified by nucleotide sequencing of selected plasmids. Bacterial culture medium was inoculated with E. coli BL21 (DE3): pET100/D-TOPO-ChT and grown at 37 °C. Expression of target protein was induced by supplementing culture medium with a final concentration of 2 mM IPTG and after incubation for a further 90 min cells were harvested. Bacterial cell pellets were resuspended in a guanidinium-based lysis buffer (6 M Guanidine HCl, 20 mM NaPO4 pH 7.8 and 500 mM NaCl) followed by sonication in order to lyse bacterial cells and reduce
GD plasma samples were purified by passing them through Aurum columns (Bio-Rad) and acidified with 3 M sodium acetate pH 5.2, to adjust the pH to the optimal pH for neuraminidase activity. In a separate aliquot of the same GD purified plasma sample 1 μL of neuraminidase was added. Clostridium perfringens neuraminidase (Roche Diagnostics Corporation, Indianapolis, IN, USA) was prepared by dissolving the enzyme in ultrapure water to a final concentration of 5 U/mL. Plasma samples with and without neuraminidase were then incubated for 2 h at 37 °C [14]. After incubation, samples were precipitated and a 2-DE western blot was performed as described above.
Fig. 4. Silver stained 2-DE gels showing a detailed region of ChT isoforms 3, 4 and 5 (marked by arrows) in different plasma protein samples. (A) Plasma protein sample from a GD patient homozygous for the wild-type allele of the ChT gene. (B) Plasma protein sample from a GD patient heterozygous for the mutant allele of the ChT gene. (C) Plasma protein sample from a GD patient homozygous for the mutant allele of the ChT gene without that lacks ChT isoform 3. (D) Plasma protein sample from a healthy (non-GD) individual.
1296
L. Quintana et al. / Biochimica et Biophysica Acta 1764 (2006) 1292–1298
3. Results 3.1. Identification of ChT isoforms in 2-DE maps from GD plasma samples In order to identify the ChT protein using 2-DE, we compared the plasma proteome profile of GD patients with
that of healthy subjects. As shown in Fig. 1, we found five GDassociated spots after gel image analysis. The spots were all excised and prepared for MALDI-TOF analysis as described in Materials and methods. Table 1 summarises the peptide mass mapping data obtained for each protein. In the case of samples 1 and 2, MS/MS analysis was performed for the ADGLYPNPR (1002.51 m/z)
Fig. 5. Relative levels of ChT isoforms and plasma ChT enzymatic activity in the samples analyzed in this study. Bars in panels A–C show the relative abundance of each ChT isoform, as the integrated intensity of every spot expressed as a percentage of the total integrated intensity of all spots in the region of analysis. The total plasma ChT enzymatic activity is also plotted for each sample.
L. Quintana et al. / Biochimica et Biophysica Acta 1764 (2006) 1292–1298
and QTFVNSAIR (1035.55 m/z) precursors, respectively. Ion scores for the MS/MS analysis are shown in Table 1. These two ions were detected in all five proteins analyzed and all the spots corresponded to the chitotriosidase 1 enzyme (CHIT1_human UniProt entry, primary accession number Q13231) and named: isoforms ChT 1, ChT 2, ChT 3, ChT 4 and ChT 5, respectively. 2-DE western blot analysis was also performed using an antibody raised against recombinant ChT protein. The results are shown in Fig. 2A, B and C. The antibody reacted positively with the identified ChT isoforms. 3.2. Glycosylation analysis Neuraminidase treatment of plasma proteins followed by 2DE and western blotting indicate that isoforms ChT 1 and 2 lost sialic acid linked to their O-glycans after incubation with neuraminidase. This loss produced a decrease in the protein molecular weight (Mw) and a shift in the pI to a more basic form (Fig. 2C and D). Also a new spot appeared in Fig. 2D above ChT 3, which could corresponds to ChT1 without sialic acid residues. Probably the ChT2 without sialic acid is masked by the ChT3 spot. Isoforms ChT 3, 4 and 5 appeared to remain unchanged regarding their Mw and pI within the limits of resolution imposed by 2-DE (Fig. 2C and D). 3.3. Characterization of ChT isoforms in 2-DE maps The ChT isoforms 2, 3, 4 and 5 were clearly located in all 2-DE images from GD patients (Figs. 3A, B, 4A and B). Also, the ChT isoform 1 was clearly identified in GD patients with high values of plasma ChT activity (Fig. 3A). However, the identification of ChT isoform 1 was more difficult in patients with low to medium values of plasma ChT activity because two new spots appear in the same region of the 2-DE gel images (Fig. 3B). These two spots also appear in the 2-DE gels of the plasma samples from three GD patients homozygous for the ChT mutation and without plasma ChT activities (Fig. 3C) as well as appearing in all 2-DE gels run with samples from 23 healthy controls (Fig. 3D). To elucidate if these two spots were ChT protein, they were excised from a 2-DE gel for MALDI-TOF analysis. Both spots were identified as hemopexin (UniProt Acc No. P02790). Moreover, 2-DE western blot analysis with anti-ChT primary antibody of plasma samples from healthy individuals and GD patients homozygous for the 24 bp duplication did not detect any ChT signal (data not shown). We conclude that the ChT isoform 1 present in GD patients with high plasma ChT activity masks the two hemopexin spots present in samples from GD patients with low to medium values of ChT enzyme activity and samples from healthy controls. There is an absence of detectable ChT isoforms in samples from patients homozygous for the ChT mutation (Figs. 3C and 4C) and from healthy controls (Figs. 3D and 4D). Fig. 5 shows enzyme activity measurements plotted against relative spot intensities for each of the five ChT isoforms.
1297
4. Discussion 4.1. Identification of ChT isoforms in 2-DE maps from GD plasma samples In the Swiss-Prot database (http://www.expasy.org), the theoretical values for Mw and pI of ChT protein are 49,376 kDa and 6.29, respectively. According to our 2-DE gel data obtained from GD plasma samples, we estimated a Mw for the identified ChT isoforms of 52 kDa for ChT isoforms 1 and 4, 51 kDa for ChT isoform 2 and 50 kDa for ChT isoforms 3 and 5, respectively. We also estimated that pI for the five ChT isoforms ranged from 6 to 6.5. These results do not coincide exactly with the previously reported pI values for ChT isoforms of 6.0, 7.2 and 8.0 obtained by preparative flat-bed isoelectric focusing [5]. We consider this could be due to intrinsic limitations of the different technologies used to resolve proteins based on pI. The possibility that additional ChT isoforms remained undetected due to incomplete resolution of basic proteins cannot be ruled out. Applying specific protocols to resolve basic proteins may allow us to resolve this issue [15,16]. Immunoblot results confirm the identity of the five ChT isoforms characterized in this study. The failure to detect the mutant form of ChT protein by immunoblotting in GD patients homozygous for the 24 bp duplication in ChT gene (data not show) suggest that levels of accumulated protein must be extremely low, possibly due to rapid turnover of the mutant protein [3]. 4.2. Glycosylation analysis In chitinases from other species, O-linked glycosylation had been noted [17]. In 1997, Renkema et al. [4] described different ChT activities in pI fractionated medium of culture of macrophages which were dependent on the neuraminidase treatment, suggesting the presence of post-translational modifications in ChT protein. Our characterization data of ChT isoforms in plasma using 2DE gels before and after neuraminidase treatment support the finding published by the cited authors. 4.3. Characterization of ChT isoforms in 2-DE maps A strong correlation between the total plasma ChT activity and the integrated intensities of ChT isoforms was observed. The ChT 3 isoform showed the highest intensity/activity correlation. In the case of ChT 1 isoform, relative spot intensity is only shown in GD patients 1, 2, 3, 5 and 7, because 2-DE images of samples from these patients were the only ones in which the identity of ChT 1 was unambiguous. The 2-DE localization of the ChT 4 isoform produced a problem in its identification and quantification due to the fact that in plasma from some of the GD patients ChT 4 isoform comigrated with immunoglobulin heavy chains.
1298
L. Quintana et al. / Biochimica et Biophysica Acta 1764 (2006) 1292–1298
ChT enzymatic activity in plasma samples from healthy individuals ranged from 0.5 to 118 nmol/mL h, in contrast to the increased activity levels detected in GD patients. Moreover, none of the ChT isoforms were detected in 2-DE maps obtained from healthy subjects as happened with GD patients homozygous for mutant ChT (patients 19, 20 and 21, enzyme activities 0 nmol/mL h). These results are consistent with the failure of previous efforts to detect mutant ChT protein in enzyme-deficient individuals. Cultured macrophages of ChT enzyme-deficient individuals synthesized only minor amounts of a short-lived 47 kDa ChT [3]. Also, transfection of COS-1 cells with mutant ChT expression vector led to modest synthesis of 47 kDa protein that was degraded intracellularly [3]. 4.4. Concluding remarks To our knowledge, this is the first study that presents 2-DE gel images of plasma from GD patients. Here we have described and characterized five ChT isoforms in 2-DE maps of plasma samples from GD patients that show a Mw of between XXX and pI values ranging from 6.0 to 6.5. The observed shift in protein migration in 2-DE western blots of samples after incubation with neuraminidase verifies the hypothesis that posttranslational glycosylation explains some of the differences between ChT isoforms. Future experiments with enzymatic deglycosylation, glycoprotein purification and detection will resolve the differences between the different ChT isoforms. Furthermore, we compared the 2-DE images of each ChT spot in plasma protein samples from 21 healthy controls, 13 GD patients with wild-type ChT protein, and 8 GD patients either homozygous or heterozygous for the 24 bp duplication. Interestingly, no ChT isoforms appeared in healthy controls and GD patients homozygous for the mutation even in immunoblotting experiments. We have proposed a connection between the quantitative image analysis of each ChT isoform and the enzymatic activity of ChT in each patient. Further efforts will be needed to measure enzyme activity levels for each of the different ChT isoforms identified in GD patients and establish whether all isoforms show similar enzymatic activities. The implications of the apparent heterogeneity in chitotriosidase remain still unclear and intriguing: do the various species stem from/reflect heterogeneous glycosylation, or are they the result of spontaneous deglycosylation in plasma? In addition, another issue of future interest will be the impact of glycan composition on liver clearance and steady state plasma levels. Finally, the possibilities of proteomic technology in the investigation of biological biomarkers of Gaucher disease and the nature of the ChT protein are demonstrated. Acknowledgments This work was supported by grants from Fondo de Investigación Sanitaria (FIS P03/1290 and FIS P04/2476), REDEMETH G03/054 and Fundación Española para el Estudio y la Terapeútica de la Enfermedad de Gaucher (FEETEG). L.Q.
is a recipient of a fellowship from Fondo de Investigación Sanitaria. References [1] E. Beutler, G.A. Grabowski, in: C.R. Scriver, A.L. Beaudet, W.S. Sly, D. Valle (Eds.), The Metabolic and Molecular Basis of Inherited Disease, 8a ed., McGraw-Hill, New York, 2001, pp. 3635–3668. [2] C.E.M. Hollak, S. van Weely, M.H.J. van Oers, J.M.F.G. Aerts, Marked elevation of plasma chitotriosidase activity, a novel hallmark of Gaucher disease, J. Clin. Invest. 93 (1994) 491–502. [3] R.G. Boot, G.H. Renkema, M. Verhoek, A. Strijland, J. Bliek, T.M.A.M.O. de Meulemeester, M.M.A.M. Mannens, J.M.F.G. Aerts, The human chitotriosidase gene, nature of inherited enzyme deficiency, J. Biol. Chem. 273 (1998) 25680–25685. [4] G.H. Renkema, R.G. Boot, A. Strijland, W.E. Donker-Koopman, M. Van den Berg, A.O. Muijsers, J.M.F.G. Aerts, Synthesis, sorting, and processing into distinct isoforms of human macrophage chitotriosidase, Eur. J. Biochem. 244 (1997) 279–285. [5] G.H. Renkema, R.G. Boot, A.O. Muijsers, W.E. Donker-Koopman, J.M.F.G. Aerts, Purification and characterization of human chitotriosidase, a novel member of the chitinase family of proteins, J. Biol. Chem. 270 (1995) 2198–2202. [6] P. Giraldo, M. Pocoví, J.I. Pérez-Calvo, D. Rubio-Félix, M. Giralt, Report of the Spanish Gaucher's disease registry: clinical and genetic characteristics, Haematologica 85 (2000) 792–799. [7] S.S. Raghavan, J. Topol, E.H. Kolodny, Leukocyte beta-glucosidase in homozygotes and heterozygotes for Gaucher disease, Am. J. Hum. Genet. 32 (1980) 158–173. [8] P. Giraldo, A. Cenarro, P. Alfonso, J.I. Pérez-Calvo, D. Rubio-Félix, M. Giralt, M. Pocoví, Chitotriosidase genotype and plasma activity in patients with type 1 Gaucher's disease and their relatives (carriers and noncarriers), Haematologica 86 (2001) 977–984. [9] A. Görg, C. Obermaier, G. Boguth, A. Harder, B. Scheibe, R. Wildgruber, W. Weiss, The current state of two-dimensional electrophoresis with immobilized pH gradients, Electrophoresis 21 (2000) 1037–1053. [10] A. Brautbar, D. Elstein, G. Pines, A. Abrahamov, A. Zimran, Effect of enzyme replacement therapy on gammopathies in Gaucher disease, Blood Cells, Mol. Dis. 32 (2004) 214–217. [11] A. Sevchenko, M. Wilm, O. Vorm, M. Mann, Mass spectrometric sequencing of proteins silver-stained polyacrylamide gels, Anal. Chem. 68 (1996) 850–858. [12] J. Gobom, E. Nordhoff, E. Mirgorodskaya, R. Ekman, P. Roepstorff, Sample purification and preparation technique based on nano-scale reversed-phase columns for the sensitive analysis of complex peptide mixtures by matrix-assisted laser desorption/ionization mass spectrometry, J. Mass Spectrom. 34 (1999) 105–116. [13] M.R. Larsen, S. Cordwell, P. Roepstorff, Graphite powder as an alternative or supplement to reversed-phase material for desalting and concentration of peptide mixtures prior to matrix-assisted laser desorption/ionizationmass spectrometry, Proteomics 2 (2002) 1277–1287. [14] S.B. Rosalki, A. Ying Foo, Incubation with neuraminidase and affinity electrophoresis with wheat-germ lectin compared for separating and quantifying alkaline phosphatase isoenzymes in plasma, Clin. Chem. 31 (1985) 1198–1200. [15] K. Pennington, E. McGregor, C.L. Beasley, I. Everall, D. Cotter, M.J. Dunn, Optimization of the first dimension for separation by twodimensional gel electrophoresis of basic proteins from human brain tissue, Proteomics 4 (2004) 27–30. [16] S.H. Bae, A.G. Harris, P.G. Hains, H. Chen, D.E. Garfin, S.L. Hazell, Y.K. Paik, B.J. Walsh, S.J. Cordwell, Strategies for the enrichment and identification of basic proteins in proteome projects, Proteomics 3 (2003) 569–579. [17] J.A. Fuhrman, J. Lee, D. Dalamagas, Structure and function of a family of chitinase isozymes from Brugian microfilariae, Exp. Parasitol. 80 (1995) 672–680.
Identification of chitotriosidase isoforms in plasma of Gaucher disease patients by two dimensional gel electrophoresis Lucía Quintana a , Alberto Monasterio b , Kepa Escuredo b , Jokin del Amo b , Pilar Alfonso a , Felix Elortza c , Simon Santa Cruz b , Laureano Simón b , Antonio Martínez b , Pilar Giraldo a , Miguel Pocoví a , José Luis Castrillo d,⁎ a
c
Instituto Aragonés de Ciencias de la Salud, Edificio CEA, Avda. Gómez Laguna 25, 50009 Zaragoza b Proteomika SL, Parque Tecnológico Bizkaia, Edificio 801B, 48160 Derio Spain Unidad de Proteómica, CIC-bioGUNE, Parque Tecnológico de Bizkaia, Edificio 801A, 48160 Derio, Spain d Centro de Biología Molecular “Severo Ochoa” (CSIC-UAM) Cantoblanco, 28049 Madrid, Spain Received 30 March 2006; received in revised form 19 May 2006; accepted 26 May 2006 Available online 6 June 2006
Abstract Chitotriosidase protein (ChT) is the most important biochemical marker described for Gaucher disease (GD). ChT activity is increased several hundred-fold in plasma of GD patients and shows a strong positive correlation with the severity of the disease. However, a recessively inherited enzyme deficiency, with an incidence of about 6% in the Caucasian population, means that not all patients with GD can be monitored by measuring ChT activity. Applying two-dimensional gel electrophoresis (2-DE) technology this study describes the localization and identification of five ChT isoforms in 2-DE images obtained from plasma of GD patients. All these isoforms were unequivocally identified using MALDI-TOF mass spectrometry (MS) and validated by western blot analysis. The features of each ChT isoform separated by 2-DE in plasma from GD patients homozygous for the wild-type ChT allele, carriers of one defective allele and patients homozygous for the mutant allele are presented. We also show the correlation between each ChT isoform and the plasma ChT enzymatic activity of the GD patients sampled in this study. © 2006 Elsevier B.V. All rights reserved. Keywords: Gaucher disease; Chitotriosidase; Biomarker; 2-DE
1. Introduction Gaucher disease (GD) is a lysosomal glycolipid storage disease caused by recessive inherited mutations in the acid β-glucosidase gene. This defect causes a deficiency in the glucocerebrosidase enzyme and leads to the accumulation of glucocerebroside in the lysosomes of macrophage cells. According to the presence and evolution of neurological symptoms, the disease is classified into three clinical types. Type 1, the most common, is characterized by the absence of neuropathic symptoms [1]. Many biomarkers show altered accumulation in the plasma of GD patients and can be used for disease diagnosis and follow-up. The measurement of these molecules is therefore extremely Abbreviations: GD, Gaucher disease; ChT, chitotriosidase ⁎ Corresponding author. Fax: +34 914979097. E-mail address: [email protected] (J. Luis Castrillo). 1570-9639/$ - see front matter © 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.bbapap.2006.05.009
important in studies of the pathophysiology of GD and for monitoring patient response to therapy. Among them, the chitotriosidase (ChT) enzyme is produced in large amounts by the lipidladen macrophages in GD patients, and has become the most important biomarker of GD [2]. ChT activity is known to be increased several hundred-fold in the plasma of GD patients and high levels of enzyme activity are correlated with disease severity. However, a 24 base pair (bp) duplication in the ChT gene results in the absence of detectable enzyme activity in the plasma of individuals homozygous for the mutant allele [3]. About 6% of the Caucasian population is deficient for ChT activity, while 35% are carriers of the mutant allele and show reduced levels of ChT enzyme activity. Efforts to detect the mutant ChT protein in plasma have been unsuccessful [3]. Moreover, three differently sized isoforms of ChT of approximately 50, 39 and 40 kDa, respectively, have been described although cultured macrophages predominantly produce and
L. Quintana et al. / Biochimica et Biophysica Acta 1764 (2006) 1292–1298
secrete the 50 kDa form [4] which itself appears to exist in different forms as isoelectric point (pI) values of about 6.5 and 7.2 were reported. Renkema et al. [4] concluded that the acidic pI of the 50 kDa ChT isoform secreted by human macrophages was due to the presence of O-linked glycans with variable amounts of sialic acid. Also, they described that pI values obtained by measuring ChT activity after isoelectric focusing of GD plasma samples showed greatest activity at pH 7.2 with minor amounts of activity at pH 6.0 and 8.0 [5]. To our knowledge, this is the first report of 2-DE images and 2DE western blotting of plasma ChT from GD patients. Five isoforms of ChT in 2-DE of GD plasma protein samples were detected and they were studied according to ChT genotypes in GD patients. A connection between the quantitative image analysis of each ChT isoform and the plasma ChT enzymatic activity in each patient is proposed. 2. Materials and methods 2.1. Patients This study was made with plasma samples from 21 type 1 GD patients and 23 healthy subjects. All patients are on the Spanish GD Registry [6]. Written informed
1293
consent was obtained from all patients and from legal representatives of patients aged less than 18 years. Our clinical protocol conformed to both the directives of the Ethical Committee of the Miguel Servet University Hospital, Spain, and the 1983 revision of the Helsinki declaration of 1975. Diagnosis of GD was based on low activity of acid β-glucosidase in leukocytes [7].
2.2. Chitotriosidase enzyme assay and genotyping The ChT enzyme activity assay was based on the method described by Hollak et al. [2] with minor modifications as described by Giraldo et al. [8]. Genotyping of ChT was performed as described by Giraldo et al. [8].
2.3. 2-DE Two-dimensional gel electrophoresis is based on standard procedures with minor modifications for improved protein resolution [9]. Plasma samples were purified using the Aurum serum protein mini kit (Bio-Rad, Hercules, CA, USA) to remove albumin and class G immunoglobulins. For samples from GD patients with high levels of immunoglobulins [10], an additional step in plasma preparation was included by passing samples through HiTrap protein A columns (Amersham Biosciences, GE Healthcare, Uppsala, Sweden) prior to use of the Aurum columns in order to further deplete immunoglobulins. Samples were then precipitated with 5% trichloroacetic acid/acetone followed by centrifugation (15 m, 13,000×g, 4 °C). Pellets were resuspended in 100 μL of rehydration solution containing 7 M urea, 2 M thiourea, 4% CHAPS and 1% dithiotreitol (DTT). Protein content of the resuspended samples was determined by Bradford assay using Bio-Rad Protein Assay (Bio-Rad) following the manufacturer's
Fig. 1. Silver stained 2-DE gels of plasma proteins from GD patients and healthy (non-GD) individuals. (A) Plasma proteins in GD patients were resolved by 2-DE using pH 3–10 NL gradient IPG strips in the first dimension and a 12.5% polyacrylamide SDS gel in the second dimension. The area of interest is framed. (B) The five spots identified as ChT protein are highlighted and numbered in the GD panel. (C) Identical region of 2-DE gel image in non-GD control sample.
1294
L. Quintana et al. / Biochimica et Biophysica Acta 1764 (2006) 1292–1298
Table 1 Proteins identified by MALDI-TOF analysis Spot UniProt No. of Mascot Sequence RMS b MS/MS (#) (Accession peptides Score a coverage peptide number) matched/ % (Ion Score) c Total peptides observed 1
Q13231
8/17
116
18
39
2
Q13231
8/21
121
18
30
3 4 5
Q13231 Q13231 Q13231
13/26 6/9 8/14
164 85 107
25 12 12
11 48 28
QTFVNSAIR (28) ADGLYPNPR (35)
a Mascot Score: Mascot Score, Protein scores greater than 76 are significant ( P < 0.05). b RMS: RMS (root mean square) error of the set of matched mass values in ppm. c MS/MS peptide (Ion score): peptide amino acid sequence and ion mascot score of the chosen precursor ion in brackets.
instructions. 250 μg of purified samples were loaded to 24 cm 3–10 NL IPG strips (Amersham Biosciences). The first dimension (isoelectrofocusing) was performed using an IPG-Phor system (Amersham Biosciences). Rehydration of the IPG strips was carried out at 50 V for 11–12 h, and proteins were focused by gradually increasing the voltage across the IPG strips to 8000 V and maintaining this voltage for at least 70–80 kVh. After IEF, IPG strips were soaked once in equilibration buffer (50 mM Tris–HCl pH 8.8, 6 M urea, 30% glycerol, 2% sodium dodecyl sulphate and 0.002% bromophenol blue) containing 2% DTT and subsequently in equilibration buffer containing 2.5% iodoacetamide. Separation of proteins in the second dimension was performed by denaturing SDS-PAGE on 12.5% polyacrylamide gels. Following electrophoresis, proteins were stained with a Silver Staining Kit (Amersham Biosciences). For protein identification, a silver staining method compatible with MS was used.
Recovered peptides were purified prior to MALDI analysis by home made nano-columns as described by Gobom et al. [12] with some modifications. A column consisting of 100–300 nL of POROS R2 material (PerSeptive Viosystems, Framingham, MA) or activated charcoal from Sigma (C-5510) was packed in a constricted GELoader tip (Eppendorf, Hamburg, Germany) [13]. A 1 mL syringe was used to force liquid through the column by applying air pressure. The column was equilibrated with 0.1% TFA and the bound peptides subsequently eluted directly onto the MALDI target with 0.5 μL aCyano-4-hydroxycinnamic acid (CHCA) solution (20 μg/μL in CAN, 0.1% TFA, 70:30, vol/vol). Peptide mass fingerprinting was performed on a Bruker Ultraflex TOF/TOF mass spectrometer (Bruker-Daltonics, Bremen, Germany). Positively charged ions were analyzed in reflector mode, using delayed extraction. Spectra were obtained by randomly scanning the sample surface. Typically 200–300 spectra were averaged to improve the signal to noise ratio. Spectra were externally calibrated resulting in a mass accuracy of <70 ppm. Protein identification was performed by searching in a non-redundant protein database (NCBI) using Mascot searching engine (http://www.matrixscience.com). The following parameters were used for database searches: missed cleavages 1, allowed modifications carbamido-methylation of cysteine (complete) and oxidation of methionine (partial).
2.6. Production of rabbit polyclonal antibodies against human full-length recombinant chitotriosidase Human RNA was prepared from whole blood using a PAXgene RNA extraction kit following the manufacturer's instructions (PreAnalytiX, Hombrechtikon, CH). The full length ChT cDNA was obtained by reverse transcription and PCR with primers specific for the ChT gene (5′-CACCATGGTGCGGTCTGTGG3′ and 5′-TCAATTCCAGGTGCAGCATTT-3′) using human mRNA as template. PCR was performed using Pfx proofreading polymerase (Invitrogen, Carlsbad,
2.4. Gel imaging and analysis Images of silver stained gels were acquired using an ImageScanner (Amersham Biosciences). Image analysis was performed using Progenesis PG220 software (Nonlinear Dynamics, Newcastle-upon-Tyne, UK). Protein spots were detected using automatic spot detection and the background was subtracted using the mode of non-spot method. The images were also subjected to spot filtering in order to remove artifacts. The total intensity of pixels within each spot (the integrated intensity) was determined by the software. The integrated intensity of each spot (normalized volume) was expressed as percentual fractions of the total integrated intensity of all spots within the region of analysis of the gel (see Fig. 1). This normalizes the amount of any given spot and gives relative protein abundance values for each sample. Significant differences in protein expression levels were determined by Student's t test with a set value of P < 0.05. Those proteins whose expression was up- or down-regulated by 2-fold or greater were excised from gels and identified by Peptide Mass Fingerprinting (PMF).
2.5. Protein identification by mass spectrometry The spots representing proteins of interest were cut from 2D silver stained electrophoresis gels and subjected to in-gel tryptic digestion according to Shevchenko et al. [11], with minor modifications. The gel pieces were swollen in a digestion buffer containing 50 mM NH4HCO3 and 12.5 ng/μL of trypsin (Roche Diagnostics, recombinant, proteomics grade trypsin, Penzberg, Germany) in an ice bath. After 30 min the supernatant was discarded, 20 μL of 50 mM NH4HCO3 were added to the gel piece and the digestion allowed to proceed at 37 °C overnight. The supernatant was transferred to an empty Eppendorf tube and basic and acidic peptide extraction was performed on the gel pieces. Supernatants of each sample were pooled, and dried by vacuum centrifugation. Prior to MS analysis, pellets were resuspended in 10 μL of 0.1% trifluoroacetic acid (TFA).
Fig. 2. 2-DE western blot analysis of plasma protein samples from GD patients and healthy (non-GD) individuals. 2-DE western blot analysis of plasma protein sample from a GD patient before and after neuraminidase incubation. (A) NC membrane stained with colloidal gold after blotting of protein from a 2-DE gel run with a plasma protein sample from a GD patient and probed with the antiserum raised against ChT. The five white spots correspond to the points where levels of ChT were saturating. The area of interest is framed. (B) 2-DE western blot of a plasma protein from a healthy (non-GD) control. (C) 2-DE western blot of a plasma protein sample from a GD patient showing the five immunoreactive spots detected by the antibody raised against recombinant human ChT, this image is identical to the 2-DE blot of the GD patient control in the neuraminidase experiment. (D) 2-DE western blot of the neuraminidase treated plasma protein sample from a GD patient. No signal from ChT isoforms 1 and 2 is detected but a new spot appears above isoform ChT 3. The positions of ChT isoforms 4 and 5 remain unchanged.
L. Quintana et al. / Biochimica et Biophysica Acta 1764 (2006) 1292–1298
1295
the viscosity of the extract. Recombinant histidine-tagged ChT was purified by affinity chromatography using the ProBond Purification System (Invitrogen). After incubation of the cell lysate with the Ni-resin, non-specifically bound proteins were removed by sequential washes with urea buffer of pH ranging from 7.8 to 5.3. His-fusion ChT was eluted in urea buffer pH 4.0 and dialysed against phosphate-buffered saline (PBS) prior to immunisation of rabbits. Each of two New Zealand White rabbits were immunised with 400 μg of recombinant ChT as a 1:1 mixture with Freunds complete adjuvant, followed by three booster injections of 150 μg of protein given at three weekly intervals. Finally, rabbits were exsanguinated and approximately 40 mL of serum per rabbit were collected and frozen in aliquots.
2.7. 2-DE western blot
Fig. 3. Silver stained 2-DE gels showing the region of the gel where ChT isoforms 1 and 2 are located (marked by arrows) in different plasma protein samples. (A) Plasma protein sample from a GD patient homozygous for the wild-type allele of the gene encoding ChT showing ChT isoforms 1 and 2. (B) Plasma protein sample from a GD patient heterozygous for the mutant allele of the ChT gene showing only ChT isoform 2 and two non-related spots in the region of ChT isoform 1 (marked by a circle). (C) Plasma protein sample from a GD patient homozygous for the mutant allele of the ChT gene lacking ChT isoforms 1 or 2. (D) Plasma protein sample from a healthy (non-GD) individual. Note the presence of the two overlapping spots (marked by a circle) and the absence ChT isoform 1 at panels B, C–D.
For western blotting small format 2-DE gels were run. Briefly, 20–25 μg of protein were loaded to 7 cm 3–10 NL IPG strips (Amersham Biosciences) and isoelectric focusing was performed as described above but using only 20 kVh. Equilibration and second dimension were carried out as described previously. Unstained proteins were electroblotted to a nitrocellulose (NC) membrane (Millipore Corporation, Billerica, MA, USA). Non-specific binding to NC was blocked by soaking membranes in a solution of 6% fat-free milk in PBS/Tween 0.2% (PBST) for 1 h. After rinsing with PBST three times, blots were incubated with the anti-ChT primary antibody. After rinsing again with PBST, blots were incubated with a horseradish peroxidase-conjugated donkey anti-rabbit antibody. Horseradish peroxidase was detected using Supersignal West Dura Extended Duration Substrate (Pierce Biotechnology, Inc., Rockford, IL). The chemiluminescent signal was subsequently detected by exposure to Hyperfilm MP (Amersham Biosciences). Blotted membranes were stained with Protogold (BBInternational, Cardiff, UK) to reveal the full protein map.
2.8. Glycosylation analysis CA) with primer annealing at 62 °C. Gel purified PCR products were cloned into pET100/D-polyhistidine TOPO bacterial expression vector (Invitrogen) in frame with an N-Terminal (6XHis) tag. The integrity of the full-length clones obtained was verified by nucleotide sequencing of selected plasmids. Bacterial culture medium was inoculated with E. coli BL21 (DE3): pET100/D-TOPO-ChT and grown at 37 °C. Expression of target protein was induced by supplementing culture medium with a final concentration of 2 mM IPTG and after incubation for a further 90 min cells were harvested. Bacterial cell pellets were resuspended in a guanidinium-based lysis buffer (6 M Guanidine HCl, 20 mM NaPO4 pH 7.8 and 500 mM NaCl) followed by sonication in order to lyse bacterial cells and reduce
GD plasma samples were purified by passing them through Aurum columns (Bio-Rad) and acidified with 3 M sodium acetate pH 5.2, to adjust the pH to the optimal pH for neuraminidase activity. In a separate aliquot of the same GD purified plasma sample 1 μL of neuraminidase was added. Clostridium perfringens neuraminidase (Roche Diagnostics Corporation, Indianapolis, IN, USA) was prepared by dissolving the enzyme in ultrapure water to a final concentration of 5 U/mL. Plasma samples with and without neuraminidase were then incubated for 2 h at 37 °C [14]. After incubation, samples were precipitated and a 2-DE western blot was performed as described above.
Fig. 4. Silver stained 2-DE gels showing a detailed region of ChT isoforms 3, 4 and 5 (marked by arrows) in different plasma protein samples. (A) Plasma protein sample from a GD patient homozygous for the wild-type allele of the ChT gene. (B) Plasma protein sample from a GD patient heterozygous for the mutant allele of the ChT gene. (C) Plasma protein sample from a GD patient homozygous for the mutant allele of the ChT gene without that lacks ChT isoform 3. (D) Plasma protein sample from a healthy (non-GD) individual.
1296
L. Quintana et al. / Biochimica et Biophysica Acta 1764 (2006) 1292–1298
3. Results 3.1. Identification of ChT isoforms in 2-DE maps from GD plasma samples In order to identify the ChT protein using 2-DE, we compared the plasma proteome profile of GD patients with
that of healthy subjects. As shown in Fig. 1, we found five GDassociated spots after gel image analysis. The spots were all excised and prepared for MALDI-TOF analysis as described in Materials and methods. Table 1 summarises the peptide mass mapping data obtained for each protein. In the case of samples 1 and 2, MS/MS analysis was performed for the ADGLYPNPR (1002.51 m/z)
Fig. 5. Relative levels of ChT isoforms and plasma ChT enzymatic activity in the samples analyzed in this study. Bars in panels A–C show the relative abundance of each ChT isoform, as the integrated intensity of every spot expressed as a percentage of the total integrated intensity of all spots in the region of analysis. The total plasma ChT enzymatic activity is also plotted for each sample.
L. Quintana et al. / Biochimica et Biophysica Acta 1764 (2006) 1292–1298
and QTFVNSAIR (1035.55 m/z) precursors, respectively. Ion scores for the MS/MS analysis are shown in Table 1. These two ions were detected in all five proteins analyzed and all the spots corresponded to the chitotriosidase 1 enzyme (CHIT1_human UniProt entry, primary accession number Q13231) and named: isoforms ChT 1, ChT 2, ChT 3, ChT 4 and ChT 5, respectively. 2-DE western blot analysis was also performed using an antibody raised against recombinant ChT protein. The results are shown in Fig. 2A, B and C. The antibody reacted positively with the identified ChT isoforms. 3.2. Glycosylation analysis Neuraminidase treatment of plasma proteins followed by 2DE and western blotting indicate that isoforms ChT 1 and 2 lost sialic acid linked to their O-glycans after incubation with neuraminidase. This loss produced a decrease in the protein molecular weight (Mw) and a shift in the pI to a more basic form (Fig. 2C and D). Also a new spot appeared in Fig. 2D above ChT 3, which could corresponds to ChT1 without sialic acid residues. Probably the ChT2 without sialic acid is masked by the ChT3 spot. Isoforms ChT 3, 4 and 5 appeared to remain unchanged regarding their Mw and pI within the limits of resolution imposed by 2-DE (Fig. 2C and D). 3.3. Characterization of ChT isoforms in 2-DE maps The ChT isoforms 2, 3, 4 and 5 were clearly located in all 2-DE images from GD patients (Figs. 3A, B, 4A and B). Also, the ChT isoform 1 was clearly identified in GD patients with high values of plasma ChT activity (Fig. 3A). However, the identification of ChT isoform 1 was more difficult in patients with low to medium values of plasma ChT activity because two new spots appear in the same region of the 2-DE gel images (Fig. 3B). These two spots also appear in the 2-DE gels of the plasma samples from three GD patients homozygous for the ChT mutation and without plasma ChT activities (Fig. 3C) as well as appearing in all 2-DE gels run with samples from 23 healthy controls (Fig. 3D). To elucidate if these two spots were ChT protein, they were excised from a 2-DE gel for MALDI-TOF analysis. Both spots were identified as hemopexin (UniProt Acc No. P02790). Moreover, 2-DE western blot analysis with anti-ChT primary antibody of plasma samples from healthy individuals and GD patients homozygous for the 24 bp duplication did not detect any ChT signal (data not shown). We conclude that the ChT isoform 1 present in GD patients with high plasma ChT activity masks the two hemopexin spots present in samples from GD patients with low to medium values of ChT enzyme activity and samples from healthy controls. There is an absence of detectable ChT isoforms in samples from patients homozygous for the ChT mutation (Figs. 3C and 4C) and from healthy controls (Figs. 3D and 4D). Fig. 5 shows enzyme activity measurements plotted against relative spot intensities for each of the five ChT isoforms.
1297
4. Discussion 4.1. Identification of ChT isoforms in 2-DE maps from GD plasma samples In the Swiss-Prot database (http://www.expasy.org), the theoretical values for Mw and pI of ChT protein are 49,376 kDa and 6.29, respectively. According to our 2-DE gel data obtained from GD plasma samples, we estimated a Mw for the identified ChT isoforms of 52 kDa for ChT isoforms 1 and 4, 51 kDa for ChT isoform 2 and 50 kDa for ChT isoforms 3 and 5, respectively. We also estimated that pI for the five ChT isoforms ranged from 6 to 6.5. These results do not coincide exactly with the previously reported pI values for ChT isoforms of 6.0, 7.2 and 8.0 obtained by preparative flat-bed isoelectric focusing [5]. We consider this could be due to intrinsic limitations of the different technologies used to resolve proteins based on pI. The possibility that additional ChT isoforms remained undetected due to incomplete resolution of basic proteins cannot be ruled out. Applying specific protocols to resolve basic proteins may allow us to resolve this issue [15,16]. Immunoblot results confirm the identity of the five ChT isoforms characterized in this study. The failure to detect the mutant form of ChT protein by immunoblotting in GD patients homozygous for the 24 bp duplication in ChT gene (data not show) suggest that levels of accumulated protein must be extremely low, possibly due to rapid turnover of the mutant protein [3]. 4.2. Glycosylation analysis In chitinases from other species, O-linked glycosylation had been noted [17]. In 1997, Renkema et al. [4] described different ChT activities in pI fractionated medium of culture of macrophages which were dependent on the neuraminidase treatment, suggesting the presence of post-translational modifications in ChT protein. Our characterization data of ChT isoforms in plasma using 2DE gels before and after neuraminidase treatment support the finding published by the cited authors. 4.3. Characterization of ChT isoforms in 2-DE maps A strong correlation between the total plasma ChT activity and the integrated intensities of ChT isoforms was observed. The ChT 3 isoform showed the highest intensity/activity correlation. In the case of ChT 1 isoform, relative spot intensity is only shown in GD patients 1, 2, 3, 5 and 7, because 2-DE images of samples from these patients were the only ones in which the identity of ChT 1 was unambiguous. The 2-DE localization of the ChT 4 isoform produced a problem in its identification and quantification due to the fact that in plasma from some of the GD patients ChT 4 isoform comigrated with immunoglobulin heavy chains.
1298
L. Quintana et al. / Biochimica et Biophysica Acta 1764 (2006) 1292–1298
ChT enzymatic activity in plasma samples from healthy individuals ranged from 0.5 to 118 nmol/mL h, in contrast to the increased activity levels detected in GD patients. Moreover, none of the ChT isoforms were detected in 2-DE maps obtained from healthy subjects as happened with GD patients homozygous for mutant ChT (patients 19, 20 and 21, enzyme activities 0 nmol/mL h). These results are consistent with the failure of previous efforts to detect mutant ChT protein in enzyme-deficient individuals. Cultured macrophages of ChT enzyme-deficient individuals synthesized only minor amounts of a short-lived 47 kDa ChT [3]. Also, transfection of COS-1 cells with mutant ChT expression vector led to modest synthesis of 47 kDa protein that was degraded intracellularly [3]. 4.4. Concluding remarks To our knowledge, this is the first study that presents 2-DE gel images of plasma from GD patients. Here we have described and characterized five ChT isoforms in 2-DE maps of plasma samples from GD patients that show a Mw of between XXX and pI values ranging from 6.0 to 6.5. The observed shift in protein migration in 2-DE western blots of samples after incubation with neuraminidase verifies the hypothesis that posttranslational glycosylation explains some of the differences between ChT isoforms. Future experiments with enzymatic deglycosylation, glycoprotein purification and detection will resolve the differences between the different ChT isoforms. Furthermore, we compared the 2-DE images of each ChT spot in plasma protein samples from 21 healthy controls, 13 GD patients with wild-type ChT protein, and 8 GD patients either homozygous or heterozygous for the 24 bp duplication. Interestingly, no ChT isoforms appeared in healthy controls and GD patients homozygous for the mutation even in immunoblotting experiments. We have proposed a connection between the quantitative image analysis of each ChT isoform and the enzymatic activity of ChT in each patient. Further efforts will be needed to measure enzyme activity levels for each of the different ChT isoforms identified in GD patients and establish whether all isoforms show similar enzymatic activities. The implications of the apparent heterogeneity in chitotriosidase remain still unclear and intriguing: do the various species stem from/reflect heterogeneous glycosylation, or are they the result of spontaneous deglycosylation in plasma? In addition, another issue of future interest will be the impact of glycan composition on liver clearance and steady state plasma levels. Finally, the possibilities of proteomic technology in the investigation of biological biomarkers of Gaucher disease and the nature of the ChT protein are demonstrated. Acknowledgments This work was supported by grants from Fondo de Investigación Sanitaria (FIS P03/1290 and FIS P04/2476), REDEMETH G03/054 and Fundación Española para el Estudio y la Terapeútica de la Enfermedad de Gaucher (FEETEG). L.Q.
is a recipient of a fellowship from Fondo de Investigación Sanitaria. References [1] E. Beutler, G.A. Grabowski, in: C.R. Scriver, A.L. Beaudet, W.S. Sly, D. Valle (Eds.), The Metabolic and Molecular Basis of Inherited Disease, 8a ed., McGraw-Hill, New York, 2001, pp. 3635–3668. [2] C.E.M. Hollak, S. van Weely, M.H.J. van Oers, J.M.F.G. Aerts, Marked elevation of plasma chitotriosidase activity, a novel hallmark of Gaucher disease, J. Clin. Invest. 93 (1994) 491–502. [3] R.G. Boot, G.H. Renkema, M. Verhoek, A. Strijland, J. Bliek, T.M.A.M.O. de Meulemeester, M.M.A.M. Mannens, J.M.F.G. Aerts, The human chitotriosidase gene, nature of inherited enzyme deficiency, J. Biol. Chem. 273 (1998) 25680–25685. [4] G.H. Renkema, R.G. Boot, A. Strijland, W.E. Donker-Koopman, M. Van den Berg, A.O. Muijsers, J.M.F.G. Aerts, Synthesis, sorting, and processing into distinct isoforms of human macrophage chitotriosidase, Eur. J. Biochem. 244 (1997) 279–285. [5] G.H. Renkema, R.G. Boot, A.O. Muijsers, W.E. Donker-Koopman, J.M.F.G. Aerts, Purification and characterization of human chitotriosidase, a novel member of the chitinase family of proteins, J. Biol. Chem. 270 (1995) 2198–2202. [6] P. Giraldo, M. Pocoví, J.I. Pérez-Calvo, D. Rubio-Félix, M. Giralt, Report of the Spanish Gaucher's disease registry: clinical and genetic characteristics, Haematologica 85 (2000) 792–799. [7] S.S. Raghavan, J. Topol, E.H. Kolodny, Leukocyte beta-glucosidase in homozygotes and heterozygotes for Gaucher disease, Am. J. Hum. Genet. 32 (1980) 158–173. [8] P. Giraldo, A. Cenarro, P. Alfonso, J.I. Pérez-Calvo, D. Rubio-Félix, M. Giralt, M. Pocoví, Chitotriosidase genotype and plasma activity in patients with type 1 Gaucher's disease and their relatives (carriers and noncarriers), Haematologica 86 (2001) 977–984. [9] A. Görg, C. Obermaier, G. Boguth, A. Harder, B. Scheibe, R. Wildgruber, W. Weiss, The current state of two-dimensional electrophoresis with immobilized pH gradients, Electrophoresis 21 (2000) 1037–1053. [10] A. Brautbar, D. Elstein, G. Pines, A. Abrahamov, A. Zimran, Effect of enzyme replacement therapy on gammopathies in Gaucher disease, Blood Cells, Mol. Dis. 32 (2004) 214–217. [11] A. Sevchenko, M. Wilm, O. Vorm, M. Mann, Mass spectrometric sequencing of proteins silver-stained polyacrylamide gels, Anal. Chem. 68 (1996) 850–858. [12] J. Gobom, E. Nordhoff, E. Mirgorodskaya, R. Ekman, P. Roepstorff, Sample purification and preparation technique based on nano-scale reversed-phase columns for the sensitive analysis of complex peptide mixtures by matrix-assisted laser desorption/ionization mass spectrometry, J. Mass Spectrom. 34 (1999) 105–116. [13] M.R. Larsen, S. Cordwell, P. Roepstorff, Graphite powder as an alternative or supplement to reversed-phase material for desalting and concentration of peptide mixtures prior to matrix-assisted laser desorption/ionizationmass spectrometry, Proteomics 2 (2002) 1277–1287. [14] S.B. Rosalki, A. Ying Foo, Incubation with neuraminidase and affinity electrophoresis with wheat-germ lectin compared for separating and quantifying alkaline phosphatase isoenzymes in plasma, Clin. Chem. 31 (1985) 1198–1200. [15] K. Pennington, E. McGregor, C.L. Beasley, I. Everall, D. Cotter, M.J. Dunn, Optimization of the first dimension for separation by twodimensional gel electrophoresis of basic proteins from human brain tissue, Proteomics 4 (2004) 27–30. [16] S.H. Bae, A.G. Harris, P.G. Hains, H. Chen, D.E. Garfin, S.L. Hazell, Y.K. Paik, B.J. Walsh, S.J. Cordwell, Strategies for the enrichment and identification of basic proteins in proteome projects, Proteomics 3 (2003) 569–579. [17] J.A. Fuhrman, J. Lee, D. Dalamagas, Structure and function of a family of chitinase isozymes from Brugian microfilariae, Exp. Parasitol. 80 (1995) 672–680.