Association of Serum Antibody Levels against TUBB 2 C with Diabetes and Cerebral Infarction Integrative Biomedical Sciences

Takaki Hiwasa1*, Xiao-Meng Zhang1, Risa Kimura1, Toshio Machida2, Kenichiro Kitamura3, Rika Yamazoe4, Mitoshi Kunimatsu5, Seiichiro Mine6,7, Eiichi Kobayashi6, Yasuo Iwadate6, Naokatsu Saeki6, Minoru Takemoto8, Kazuki Kobayashi8, Harukiyo Kawamura8, Ryoichi Ishibashi8, Ken-ichi Sakurai8, Masaki Fujimoto8, Koutaro Yokote8, Yo Iwata9, Takashi Nakayama10, Jun-ya Harada10, Yoshio Kobayashi10, Mikiko Ohno11, Po-Min Chen11, Eiichiro Nishi11, Mitsuhiro Yokota12, Ikuo Kamitsukasa13, Takeshi Wada14, Akiyo Aotsuka14, Masahiro Mori15, Akiyuki Uzawa15, Mayumi Muto15, Kazuo Sugimoto1,15, Satoshi Kuwabara15, Ken-ichiro Goto1,16, Ryutaro Matsumura17, Takao Sugiyama18, Hirotaka Takizawa19, Hideaki Shimada20, Masaaki Ito20, Hao Wang1, Akiko Taira1, Emiko Arita1, Katsuro Iwase1, Takashi Kudo21, Hirofumi Doi21, Rika Nakamura1,22, Go Tomiyoshi1,22, Natsuko Shinmen1,22 and Hideyuki Kuroda22


Introduction
Cerebrovascular disease occupies approximately 10% of the cause of mortality in Japan.After the onset of a stroke, patients have suffered from prognostic symptoms for a long time, and the disease is a major cause of being bedridden.Despite of the development of treatment with urokinase and anticoagulants, the effect is limited [1][2][3].Thus, it is absolutely important to treat before the onset of a cerebral infarction (CI) such as at the stage of transient ischemic attack.For the diagnosis of CI, various risk factors have been developed including body mass index (BMI), low density lipoprotein (LDL) cholesterol, uric acid, smoking habits and family history [4][5][6].
On the other hand, diabetes mellitus (DM) is a major risk factors of CI and cardiovascular diseases (CVD) [7].DM frequently induces atherosclerosis, which is accompanied by complications such as CI and CVD.Therefore, prevention of CI and CVD has been attempted by using DM markers, which are blood glucose, glycoalbumin and glycohemoglobin (HbA1c) [8].However, despite frequent application for diagnosis, prevention of the onset remains unsuccessful.Thus, the development of more sensitive markers is awaited.
We have performed a large scale screening of esophageal squamous cell carcinoma by serological identification of antigens by recombinant cDNA expression cloning (SEREX), and identified many SEREX antigens [9][10][11].We also reported that the antibody levels against SEREX antigens were excellent and sensitive for diagnosis of tumors.Although many autoantibodies have been reported for cancer [12,13] and autoimmune diseases [14], only few have been identified for atherosclerosis-related diseases.Among the ones reported were phospholipid [15], apolipoprotein A-1 [16] and oxidized low-density lipoprotein [17] and heat shock proteins (Hsps), for CVD [18], Hsp60 for stroke [19], and GAD, insulin and protein tyrosine phosphatase for DM [20].
We introduced the SEREX screening method to identify autoantigens responsible to atherosclerosis and reported anti-RPA2 antibody as a CI-related marker [21].In this report, further screening identified novel TUBB2C as an antigen, against which the serum antibody levels were higher in patients with DM or CKD than in HD.

Patients and healthy donor sera
The Local Ethical Review Board of the Chiba University, Graduate School of Medicine as well as those of co-operating hospitals approved this study.Sera were collected from patients after they had given written informed consent.Each serum sample was centrifuged at 3,000×g for 10 min, and then the supernatant was stored at -80˚C until use.Repeated thawing and freezing of samples were avoided.
The serum samples of CI were obtained from Chiba Prefectural Sawara Hospital, Chiba Rosai Hospital and Chiba Aoba Municipal Hospital.Samples of CVD and DM were obtained from Chiba University Hospital, and those of CKD were from Kumamoto University.Serum samples of CVD were obtained from the Kita-Nagoya Genomic Epidemiology (KING) Study, which is a prospective population-based observational survey of aged Japanese [22].The sera of healthy donors were obtained from Chiba University, Shimoshizu National Hospital and Port Square Kashiwado Clinic.

Screening by expression cloning
Recombinant DNA studies were performed with the official permission of the Chiba University Graduate School of Medicine and were carried out in accordance with the rules of the Japanese government.We used a commercially available human microvascular endothelial cell cDNA library (Uni-ZAP XR Premade Library, Stratagene, La Jolla, CA) to screen for clones that were immunoreactive against serum IgG from patients with severe carotid stenosis as described previously [9][10][11]21].Escherichia coli (E.coli) XL1-Blue MRF' was infected with Uni-ZAP XR phage and the expression of resident cDNA clones was induced after blotting the infected bacteria onto NitroBind nitrocellulose membranes (Osmonics, Minnetonka, MN) that had been treated with 10 mM isopropyl-β-Dthiogalactoside (IPTG, Wako Pure Chemicals, Osaka, Japan) for 30 min.The membranes with bacterial proteins were rinses 3 times with TBS-T [20 mM Tris -HCl (pH 7.5), 0.15 M NaCl and 0.05% Tween-20], and non-specific binding was blocked by incubation with 1% protease-free BSA (Wako Pure Chemicals) in TBS-T for 1 h.The membranes were exposed to 1:2000-diluted plasma for 1 h.After 3 washes with TBS-T, the membranes were incubated for 1 h with 1:5000-diluted alkaline phosphatase-conjugated goat anti-human IgG (Jackson ImmunResearch Laboratories, West Grove, PA).Positive reactions were developed using 100 mM Tris-HCl (pH 9.5) containing 100 mM NaCl, 5 mM MgCl 2 , 0.15 mg/mL of 5-bromo-4chloro-3-indolylphosphate and 0.3 mg/mL of nitro blue tetrazolium (Wako Pure Chemicals).Positive clones were re-cloned twice until obtaining monoclonality.
Monoclonal phage cDNA clones were converted to pBluescript phagemids by excision in vivo using the ExAssist helper phage (Stratagene).Plasmid pBluescript containing cDNA was obtained from the E. coli SOLR strain after transformation by the phagemid.Sequences of cDNA inserts were searched for homology with identified genes or proteins within the public sequence database (http://blast.ncbi.nlm.nih.gov/Blast.cgi).

Expression and purification of GST-fusion proteins
Recombinant proteins tagged with glutathione-Stransferase (GST) were constructed by recombining the insertion sequences of pBluescript into pGEX-4T (GE Healthcare Life Sciences, Pittsburgh, PA) vector plasmids.The pBluescript plasmids were digested using a combination of EcoRI and XhoI or SmaI and XhoI.The inserted DNA fragments were ligated in frame to SmaI/ XhoI-digested pGEX-4T-1 or EcoRI/XhoI-digested pGEX-4T-3 using Ligation Convenience Kits (Nippon Gene, Toyama, Japan).The ligation mixtures were used to transform ECOS™ competent E. coli BL-21 (Nippon Gene) and appropriate recombinants were confirmed by DNA sequencing as well as protein expressions.Treating the transformed E. coli with 0.1 mM IPTG for 3 h induced the expression of the GST-fusion proteins.The GST recombinant proteins were purified by glutathione-Sepharose column chromatography according to the manufacturer's instructions (GE Healthcare Life Sciences) and dialyzed against PBS as described [9,10,21].

Peptide array
The possible epitope sites in the candidate antigen proteins can be predicted using the program ProPred (http://www.imtech.res.in/raghava/propred/) as described [23].We designed 100 peptides derived from seven candidate antigens.These peptides together with positive or negative controls were synthesized onto the cellulose membranes using F-moc amino acids (Auto spot robot ASP222; ABIMED Analysen-Technik GmbH, Langenfeld, Germany) as previously described [24].Membranes loaded with 150 peptides were washed with PBS-T-BSA [phosphate-buffered saline containing 1% (w/v) bovine serum albumin and 0.05% Tween-20 and 0.05% NaN 3 ] for 30 min five times, and then incubated with the sera of patients at 1:200 dilutions for 18 h.The membranes were washed with PBS-T-BSA five times, and then treated with FITC-conjugated goat anti-human IgG (Jackson ImmunoResearch, West Grove, PA) at 1:10000 dilutions for 1-2 h.After washing, the fluorescence of the peptide spots were detected using Typhoon 9400 Imager (Amersham Biosciences, Stockholm, Sweden) with a 488 nm/520 nm filter as described previously [25].

Peptide synthesis
N-terminal biotinylated 14mer peptides corresponding to the amino acid sequences of the positive spots of peptide arrays were synthesized and used for AlphaLISA.The purity of each peptide was determined to be higher than 90%.

Statistical analyses
Fisher's exact (two-sided) probability test and the Mann-Whitney U test were used to determine the significance of the differences between the two groups.Correlation was examined by Spearman's correlation analysis.All statistical analyses were carried out using the GraphPad Prism 5 (GraphPad Software, La Jolla, CA).P values lower than 0.05 were considered statistically significant.

Identification of epitope peptides by peptide array
A total of 100 peptides were designed as possible epitope sites based on the amino acid sequences of seven candidate antigen proteins (Table 1).These peptides as well as positive or negative controls were synthesized directly onto the cellulose membranes, which were then treated with sera followed by treatment with FITC-labeled second antibodies and detection of fluorescence.Figure 1 shows the representative results after treatment with sera of patients #1 (a) and #2 (b) or with that of HD (c).
Based on the results using sera of 22 patients and 10 HD, we selected three spots, numbers E6 (LOC729260-337), F6 (TUBB2C-375) and F19 (KIAA0020-477), of which the fluorescence was more potent after treatment with patients' sera as compared with that with HD sera.No peptide spot was selected from peptides derived from the sequence of RPA2, LRPAP1, EEF1A1 and SPARC.
N-terminal biotinylated peptides, bLOC729260-337, bTUBB2C-375 and bKIAA0020-477, were synthesized with the purity higher than 90%, and used as antigens in the following study.

The levels of s-TUBB2C-Abs increased in patients with CI
We examined the serum antibody levels against selected antigens using specimens of CI and CVD, which have been obtained from Chiba Rosai Hospital, Chiba Aoba Municipal Hospital, Chiba Prefectural Sawara Hospital, Kita-Nagoya Genomic Epidemiology (KING) Study and Chiba University Hospital.HD sera were obtained from Shimoshizu National Hospital and Chiba University.AlphaLISA was introduced to measure the serum antibody levels because of the low background levels and small variation.When control GST was used as an antigen, serum antibody levels were not apparently different between HD and patients with CI or CVD (Figure 2a).On the other hand, the levels of s-TUBB2C-Abs were significantly higher in patients with CI than in HD (Figure 2b).When the cut-off value was determined as the average +2SD of HD, the positive rates of s-TUBB2C-Abs in HD and CI patients were 1.6% and 14.8%, respectively (Table 2).s-TUBB2C-Abs levels were somewhat higher in patients with CVD as compared with that in HD, yet the difference was not significant.The levels of anti-LOC729260 antibodies were not altered between patients with CI or CVD and HD, and anti-LRPAP1 antibody levels were slightly higher in patients with CI but not in patients with CVD as compared to those in HD (Table 2).Thus, these antigens were not used in the following study.
The levels of serum anti-biotinylated TUBB2C-375 peptide antibodies (s-bTUBB2Cpep-Abs) and antibiotinylated LOC729260-337 peptide antibodies were not apparently different between HD and patients with CI (Table 2).

The levels of s-TUBB2C-Abs were responsible to DM
Atherosclerosis is closely related to DM, and thus, we then examined specimens of DM as well as another set of CI and HD which were obtained from Chiba University Hospital and Port Square Kashiwado Clinic.Both of s-TUBB2C-Abs and s-bTUBB2Cpep-Abs increased significantly and markedly in patients with DM as compared with those in HD (Figure 3 and Table 3).The positive rates of s-TUBB2C-Abs in HD and DM patients were 2.3% and 25.6%, respectively (Table 3).The levels of s-TUBB2C-Abs and s-bTUBB2Cpep-Abs were significantly higher in patients with CVD than in HD with the positive rates of 7.5% for both antigens.Anti-KIAA0020-477 peptide antibodies showed a marginal increase in DM but not in CVD.

The levels of s-TUBB2C-Abs were elevated in patients with chronic kidney disease
Chronic kidney disease (CKD) was divided into three groups; type 1: diabetic kidney disease, type 2: nephrosclerosis, and type 3: glomerulonephritis.s-TUBB2C-Ab levels in CKD patients were higher than those in HD (Figure 4).The positive rates of CKD type 1, type 2 and type 3 were 22.1%, 28.1% and 11.4%, respectively (Table 4).The average positive rate of CKD was 18.3% (55 positive specimens out of 300).Similar results were obtained when bTUBB2C-375 peptide was used as an antigen.

The levels of s-TUBB2C-Abs were not associated with cancer
Autologous antibodies developed frequently in cancer patients [13].We then examined the specimens  Columns A and B are positive or negative controls, and columns C -F are synthetic peptides corresponding to the amino acid sequences predicted as epitopes of seven candidate proteins.The selected three peptides are marked in color.  of patients with benign glioma, malignant glioma or esophageal squamous cell carcinoma obtained from Toho University Hospital and Chiba University Hospital.The levels of s-TUBB2C-Abs were not significantly altered between HD and cancer patients (Figure 5).

Correlation between s-TUBB2C-Abs and atherosclerosis indices
We then performed Spearman correlation analysis between the antibody marker levels and the information of the subject persons including age, gender, height, weight, BMI, smoking habit, artery stenosis degrees such as plaque score, maximum intima-media thickness (maxIMT), cardio ankle vascular index (CAVI) and ankle brachial pressure index (ABI), blood test data such red blood cell, white blood cell, platelet, hemoglobin, hematocrit, total protein, albumin, uric nitrogen, uric acid, creatinin, Na, K, Cl, Ca, P, Fe, Mg, AST, ALT, LD, gamma-GTP, alkaline phosphatase, total billirubin, amylase, creatin kinase, total cholesterol, HDL-cholesterol, LDL-cholesterol, triglyceride, and C-reactive protein.Total of 384 specimens were analyzed.The levels of s-TUBB2C-Abs showed a positive correlation with plaque score, max IMT and CAVI (Table 5), suggesting the close association between s-TUBB2C-Ab levels and atherosclerosis.s-TUBB2C-Ab levels were also correlated with AST but not with ALT.This implicates that the high s-TUBB2C-Abs levels are responsible to some disorder in cardiac muscle, skeletal muscle or erythrocytes but not of liver.s-TUBB2C-Abs also showed positive correlation with C-reactive protein, age and triglyceride but inverse correlation with the levels of albumin and HDLcholesterol.

Discussion
Through the phage expression cloning, we have identified six candidate antigens which can be recognized by serum IgG antibodies in patients with atherosclerosis.Replication protein A2 (RPA2) which plays a role in the repair of DNA double-strand breaks [26] has been identified in our previous study Shown are average, SD, cut-off values (average + 2SD), total sample numbers, the number of positive sera of which the antibody levels were higher than the cut-off value, and the positive rate (%) of HD; and average, SD, total sample number, number of positive sera of which the antibody levels were higher than the cut-off value, and the positive rate (%) of patients; and P value of comparison between HD and patients.Antigens used were purified GST-fusion proteins such as EEF1A1, LOC729260, TUBB2C and LRPAP1 and synthetic peptides such as bLOC729260-337 and bKIAA0020-477.P values lower than 0.05 and positive rates higher than 10% were marked in color.

Table 2:
Comparison of serum antibody levels between HD and patients with CI or CVD examined by AlphaLISA and reported [21].Low density lipoprotein-related protein-associated protein 1 (LRPAP1) functions as a chaperone protein in the intracellular transport of lowdensity lipoprotein-related receptor and in involved in anti-aging [27].Because low-density lipoprotein is closely related to atherosclerosis [28], LRPAP1 may also be involved in the related disease.In fact, the serum levels of anti-LRPAP1 antibodies were higher in patients with CI but not CVD than in HD (Table 2).Eukaryotic translation elongation factor 1 alpha 1 (EEF1A1) is one of elongation factors and inhibits p53-dependent apoptosis [29].Autoantibody against EEF1A1 was elevated in patients with Felty's syndrome [30], which shows similar symptoms as rheumatoid arthritis (RA) and systemic lupus erythematosus (SLE).
It should be noted that RA and SLE are high risk groups of CVD [31].We observed higher levels of anti-EEF1A1 antibodies in patients with CVD as compared with those in HD (Table 2).Sparc/osteonectin, cwcv and kazal-like domains proteoglycan 1 (SPOCK1) is a Ca 2+binding proteoglycan and contains Kazal-like proteaseinhibitory domains, with no further information on the function [32].KIAA0020 protein contains a sequence of a minor histocompatibility antigen, HA-8 [33].The function of LOC729260 is unknown.
Tubulin beta 2C (TUBB2C) is highly expressed in heart and testis [34].TUBB2C is of much interest because s-TUBB2C-Abs levels were higher in patients with CI, DM, CVD or CKD (Figures 2-4 and Tables 2-4).The most prominent difference of the levels was observed between HD and patients with DM or CKD.Similar results were obtained by using bTUBB2C-375 peptide as an antigen.The positivity of s-TUBB2C-Abs for CI was 14.8% ( Antigens used were purified GST-fused TUBB2C protein and bTUBB2C-375 and bKIAA0020-477 peptides.Shown numbers are as described in Table 2.Both levels of s-TUBB2C-Abs and s-bTUBB2Cpep-Abs were higher in CVD specimens than in those of HD (Figure 3a and b and Table 3), whereas their levels showed no significant difference between CVD and HD specimens (Figure 2 and Table 2).Most specimens (92 out of 128 specimens) were old myocardial infarction in Figure 2  CKD types 1, 2 and 3 were diabetic kidney disease, nephrosclerosis and glomerulonephritis, respectively.Antigens used were purified GST-TUBB2C protein and bTUBB2C-375 peptide.Shown numbers are as described in Table 2.

Table 5:
Correlation analysis between TUBB2C antibody marker levels and the subjects' information obtained more than one month after the onset of myocardial infarction.On the other hand, all of sera used in Figure 3 and Table 3 were obtained within two weeks after the onset of acute myocardial infarction or unstable angina.Thus, some of serum antibodies may disappear in one month after the onset.Alternatively, because substantial populations of patients die within one month after the onset of myocardial infarction, patients in Figure 2 and Table 2 were selected as those with mild symptoms, resulting in lower marker levels than patients in Figure 3 and Table 3 with occasional severe symptoms.
Both s-TUBB2C-Abs and s-TUBB2Cpep-Abs levels were highly responsible to CKD almost irrespective of the types (type 1: diabetic kidney disease, type 2: nephrosclerosis, and type 3: glomerulonephritis).Thus, the marker levels may not necessarily reflect DM but renal failure itself.The highest responsibility of TUBB2C marker to this renal failure (Figure 4 and Table 4) suggests that renal failure is one of the initial events to cause the progression of atherosclerosis followed by the onset of CI or CVD.Consistently, the association of reduced glomerular filtration rate and stroke outcome was reported [35].The correlation between c-TUBB2C-Abs and blood concentrations of triglyceride and HDL-cholesterol (Table 5) suggests that these lipid metabolism may lead to the development of not only DM but also CKD.
In this study, we introduced peptide arrays in marker screening.Automated peptide synthesis onto cellulose membranes has been developed by a co-author, Mitoshi Kunimatsu [24].Successful identification of a useful marker, bTUBB2C-375, indicates that this peptide array method is effective in marker screening.
There are many factors that affect the progress of CI, and each antibody marker may be associated with the causes such as body mass index, smoking habit, atherosclerosis, uric acid, HDL-or LDL-cholesterol and triglyceride.It is probable that each marker is associated with one or few of those factors.If so, the positivity of each maker cannot be expected to be particularly high.Therefore, examination by using as more markers may provide as more precise diagnosis.
Our antibody markers were selected by using sera collected within two weeks after the onset of CI.Various antigens may appear immediately after the onset of CI whereas the antibodies are not produced until two weeks later.Thus, the antibodies specifically detected in sera immediately after the onset were present prior to the onset.By measuring the levels of these antibodies, it is possible to predict the onset, i.e., serum antibody markers can be prediction markers for the onset of CI.In most cases, CI is not induced suddenly but mediated frequently by health issues such as TIA and asymptomatic CI.It is possible for TUBB2C to leak out from infarct lesions repeatedly, which may evoke the immune system leading to development of the autoantibodies to detectable levels.Thus, diagnosis of autoantibody levels such as s-TUBB2C-Abs may enable to predict the onset of CI.

Figure 1 :
Figure 1: Results of peptide arrays.Cellulose membranes loaded with 150 peptides described in Table 1 were treated with sera from patients with atherosclerosis (#P1 and #P2) or from HD (#HD1).Antibody binding to peptide spots was detected by treatment with FITC-conjugated anti-human IgG antibody followed by imaging by a fluorescence laser scanner.Representative results are shown.Arrows indicate the position of selected peptides, numbers E6, F6 and F19.

Figure 2 :
Figure 2: Comparison of serum antibody levels of HD and patients with CI or CVD.Antigens used were control GST (a) and GST-TUBB2C proteins (b).Serum antibody levels examined by AlphalLISA are shown by box-whisker plot.The box plots display the 10th, 20th, 50th, 80th and 90th percentiles.P values versus HD specimens are shown.

Figure 3 :
Figure 3: Comparison of serum antibody levels of HD and patients with DM or CVD.Antigens used were GST-TUBB2C protein (a) and bTUBB2C-375 peptide (b).Serum antibody levels examined by AlphalLISA are shown as described in the legends of Figure 2. In Table 3, averages, SDs, cut-off values, P values, total numbers, positive numbers and positive rates (%) are shown.

Figure 4 :
Figure 4: Comparison of serum antibody levels of HD and patients with CKD types 1, 2 and 3. Antigens used were GST-TUBB2C protein (a) and bTUBB2C-375 peptide (b).Serum antibody levels examined by AlphalLISA are shown as described in the legends of Figure 2. In Table 4, averages, SDs, cut-off values, P values, total numbers, positive numbers and positive rates (%) are shown.

Figure 5 :
Figure 5: Comparison of serum antibody levels of HD and patients with cancer.Antigens used were control GST and GST-TUBB2C proteins.Serum antibody levels examined by AlphalLISA are shown as described in the legends of Figure 2. Numbers of specimens used were 111 of HD, 83 of benign glioma (bGlioma), 90 of malignant glioma (mGlioma) and 100 of esophageal squamous cell carcinoma (Eso Ca).

Table 1 :
List of peptides loaded onto a peptide array

Table 3 :
Comparison of serum antibody levels between HD and patients with DM or CVD examined by AlphaLISA

Table 4 ,
averages, SDs, cut-off values, P values, total numbers, positive numbers and positive rates (%) are shown.3and 4).Because CI is caused by atherosclerosis or cardiogenic thrombus, it is possible that s-TUBB2C-Ab marker can discriminate CI caused by atherosclerosis but not by thrombus.
and Table2, and therefore, the sera were

Table 4 :
Comparison of serum antibody levels between HD and patients with CKD examined by AlphaLISA