Printer Friendly

Multilectin assay for detecting fibrosis-specific glyco-alteration by means of lectin microarray.

It is estimated that 350 million and 170 million people are chronically infected with the hepatitis B virus (HBV) [4] and hepatitis C virus (HCV), respectively (1,2). People with HBV or HCV have an increased risk of developing of chronic hepatitis, cirrhosis, and hepatocellular carcinoma. Globally, 57% of cirrhosis cases are attributable to either HBV (30%) or HCV (27%), and 78% of hepatocellular carcinoma (HCC) cases are attributable to HBV (53%) or HCV (25%) (3). In Japan, >90% of HCC patients had chronic hepatitis associated with HBV (13%) or HCV (81%) during the period 1996-2000 (4). Although HCV infection is asymptomatic for about 10 years before chronic hepatitis develops, it causes cirrhosis and irreversible fibrosis during this interval. According to the association between staging fibrosis (F0-F4) and carcinogenesis, 7.9% of F4 stage fibrosis patients in Japan will develop HCC each year, whereas the corresponding rate for hepatitis patients with the primary F0/F1 stage is much lower (0.5%) (5). An accurate method for monitoring the progression of fibrosis is needed to identify F4 stage fibrosis patients as a high-risk group for the development of HCC, especially in the case of hepatitis C.

HCV patients are currently subjected to continuous monitoring for chronic inflammation by using biochemical markers, such as aspartate aminotransferase and alanine aminotransferase, to evaluate inflammatory activity and platelet count to assess the progression of fibrosis. However, platelet count is affected by physical status and some medical treatments. Consequently, despite the invasiveness and high risk, liver biopsies are regarded as the conventional method of evaluating the progression of fibrosis in patients with chronic hepatitis (6). As an alternative to liver biopsy, Imbert-Bismut et al. developed a method that involves analysis of a combination of biochemical markers (7, 8). Transient elastography (e.g., the Fibroscan), a noninvasive method based on physiological principles (9), has been considered to have a diagnostic value superior to that of the biochemical markers (10). However, transient elastography is not free of shortcomings (11), since its diagnostic success rate depends on the operator's experience and the BMI of the patient. In fact, the American Association for the Study of Liver Diseases recommended that these noninvasive tests should not replace liver biopsies in routine clinical practice. Consequently, liver biopsy is widely regarded as the gold standard for defining liver disease status (12). Other more convenient assays using serum markers are still needed for frequent monitoring.

Recent omics-based technologies have facilitated the discovery of disease-specific biomarkers (13).We focused our attention on glycoproteomics, which involves analysis of the glycosylation patterns of glycoproteins, because almost all secreted proteins except albumin are eventually glycosylated. Fucosylation analysis of [alpha]-fetoprotein (the AFP-L3 fraction) improved the diagnostic specificity of this marker of HCC. As a result, the AFP-L3% test was approved by the U.S. Food and Drug Administration in 2005. More recently, several clinical glycomics technologies have been proposed for diagnosing liver cirrhosis and HCC (14-16).

In the present study, we applied an antibody-assisted lectin profiling technology to select an optimal lectin set for monitoring the progression offibrosis (see Fig. 1 in the Data Supplement that accompanies the online version of this article at http://www.clinchem. org/content/vol57/issue1) (17, 18). We focused on the glyco-alteration of [alpha]1-acid glycoprotein (AGP), a well-known acute-phase protein that is mainly secreted by the liver and of which glyco-alteration has been reported to occur during cirrhosis and during acute and chronic inflammation (19, 20).

Materials and Methods


Since January 1999, we obtained long-term serial samples from 213 patients at Nagoya City University Hospital with chronic hepatitis B or C (21, 22). To optimize the procedure for the comparative assay of AGP glycosylation by means of lectin microarray, and to estimate the reproducibility, we collected control specimens from 20 randomly selected healthy volunteers aged 30-60 years. Exclusion criteria for healthy volunteers included hepatitis, other hepatic diseases, renal failure, clinical cardiovascular disease, and malignancies. The institutional ethics committees at Nagoya City University Hospital and the National Institute of Advanced Industrial Science and Technology (AIST) approved this study, and informed consent for the use ofclinical specimens was obtained from all individuals before collection at the Nagoya City University Hospital and AIST. The general characteristics of the patient populations are presented in Table 1. Chronic hepatitis was defined as a persistent increase in serum alanine aminotransferase activity (>1.5 times the upper limit of the normal range [35 U/L] over a 6-month period, with at least 3 readings at 2-month intervals) accompanied by histological documentation of grading and staging scores (23). Fibrosis was graded according to the guidelines for fibrosis in the Histologic Activity Index by using biopsy or surgical specimens for pathological diagnosis. Biopsy specimens were classified as follows: F0, no fibrosis; F1, portal fibrosis without septa; F2, few septa; F3, numerous septa without cirrhosis; and F4, cirrhosis. Three diagnostic targets in this study were defined as significant fibrosis: F2 + F3 + F4; severe fibrosis: F3 + F4; and cirrhosis: F4. Inflammation was also classified as follows: A0, no histological activity; A1, mild activity; A2, moderate activity; and A3, severe activity. Cirrhosis was also determined by ultrasonography (coarse liver architecture, nodular liver surface, and blunt liver edges) and evidence of hypersplenism (splenomegaly on ultrasonography) and/or a platelet count <100 000/[mm.sup.3].


We purified monoclonal anti-AGP IgG1 (clone AGP47) from mouse ascites fluid (Sigma-Aldrich) by using a HiTrap Protein G HP column (GE Healthcare BioScience). We then biotinylated the obtained IgG1 with the Biotin Labeling Kit-N[H.sub.2] (Dojindo Laboratories) and used it as a biotinylated anti-AGP antibody. All sera were diluted 1:10 with PBS containing 0.2% SDS and then heated at 95 [degrees]C for 20 min. The resulting solution (5 [micro]L) was reacted with antibody solution (40 [micro]L) containing 100 ng of the biotinylated anti-AGP antibody at 4 [degrees]C for 2 h. After the incubation, we added 10 [micro]L streptavidin-immobilized magnetic bead (Dynabeads MyOne[TM] Streptavidin T1; DYNAL Biotech ASA) conjugate to the reaction tube and incubated it at 4 [degrees]C for 1 h with vigorous shaking. The beads, which captured AGP, were washed 3 times with 500 [micro]L 1% Triton X-100 in TBS (TBSTx), and then the bound AGP was eluted with 10 [micro]L of the elution buffer (TBS containing 0.2% SDS) by heat denaturation at 60 [degrees]C.


The purified proteins were diluted to 60 [micro]L with PBS containing 1% TritonX-100 (PBSTx) and then applied to the lectin array containing triplicate spots of 43 lectins [5] as described previously (24). After incubation at 20 [degrees]C for 12 h or longer, an excess amount of blocker glycoprotein (20 [micro]g nonlabeled human serum polyclonal IgG) was added to the glass slide before detection by the biotinylated anti-AGP antibody. After incubation at 20 [degrees]C for 30 min, the reaction solution was discarded and the glass slide was washed 3 times with PBSTx. A 60-[micro]L volume of biotinylated antibody solution in PBSTx was applied to the array, which was then incubated at 20 [degrees]C for 1 h. After 3 washes with PBSTx, 60 [micro]L Cy3-labeled streptavidin (GE Healthcare BioScience) solution in PBSTx was added to the array and incubated at 20 [degrees]C for 30 min. The glass slide was rinsed with PBSTx and scanned by using an evanescent-field fluorescence scanner (Glycostation[TM]; Moritex). We analyzed all data by using the Array Pro Analyzer, version 4.5 (Media Cybernetics). The net intensity value for each spot was calculated by subtracting the background value from the signal intensity values of 3 spots. We selected scanning data under appropriate gain conditions, which provided the intensities of all positive spots <40 000. The relative intensities of positive lectins were determined from their ratio to the fluorescent intensity of an internal-standard lectin DSA.


We used either Dr. SPSS II software for Windows (SPSS) or Origin version 7.5 software for Windows (OriginLab) to perform the statistical analysis of each lectin signal. To estimate the correlation of lectin signals with liver fibrosis, the relative intensities of lectins in samples from 125 patients, and their platelet counts, were plotted as box plots that described the median and interquartile range. We estimated the abilities of lectins to monitor fibrosis progression from P values calculated by a nonparametric test (the Mann-Whitney U-test). To assess the classification efficiencies of the various lectin signals as markers of significant fibrosis, severe fibrosis, and cirrhosis as described by Leroy et al. (25), receiver operating characteristic (ROC) curve analysis was also done to determine the area under curve (AUC) values. To classify the patients, we used cutoff values obtained from Youden's index (26). Diagnostic accuracy of each test was expressed as the overall accuracy, diagnostic odds ratio, and AUC (27).



For differential analysis of fibrosis-associated changes, we initially used sera from healthy volunteers having normal liver function and patients with liver cirrhosis or chronic hepatitis. Sera from patients with rheumatoid arthritis were included as inflammation controls. AGPs were enriched from 0.5-[micro]L aliquots of sera by immunoprecipitation through use of a biotinylated anti-AGP monoclonal antibody conjugated with streptavidin-coated magnetic beads. The heating temperature for elution of the captured AGP was fixed at 60 [degrees]C, since contamination of streptavidin and biotinylated antibody was minimized while maintaining high elution efficiency and reproducible glycan profiles (see Fig. 2 in the online Data Supplement). An aliquot of the eluate containing nanogram quantities of AGP was then subjected to the antibody-overlay lectin microarray (18). We obtained glycan profiles with signal spots for 15 lectins of the 43-lectin array, namely, AAL, AOL, MAL, SNA, SSA, TJA-I, PHA-L4, RCA120, PHA-E4, DSA, ABA, LEL, STL, UDA, and WGA (Fig. 1). The pattern of signals enabled us to distinguish between the cirrhosis group and the noncirrhosis group (chronic hepatitis, normal liver function, and rheumatoid arthritis). In particular, the fucose-binding lectins (Fucbinders) AOL and AAL preferentially bound to the AGPs of the cirrhosis group compared with those of the noncirrhosis group. These results are consistent with previous reports showing that a-1,3 fucosylation of a lactosamine unit, the Lewis X antigen, is increased in AGP molecules (19, 28).



To optimize the detection system for the progression of fibrosis, we next analyzed AGPs from patients with liver cirrhosis or chronic hepatitis (n = 10 for each) with HCV infection. First, we examined the effect of various concentrations of AGP on the glycan profile. The binding signal derived from the lectin microarray analysis of 4 AGP concentrations (there was a 5-fold difference between the highest and lowest concentrations) was normalized by using the "max-normalization" procedure (29). This procedure showed that the glycan profiles of a particular patient were identical, regardless of AGP concentration. A profile based on the signal pattern of 12 selected lectins enabled us to classify the 20 patients into a cirrhosis group or noncirrhosis group. The classification was done by using a fingerprinting method to classify the 12 lectins. Half of the lectins (DSA, SNA, SSA, TJA-I, WGA, and RCA120) served as internal standards (i.e., as a "pattern formatter"), and the remaining half (AOL, AAL, PHA-E4, LEL, STL, and MAL) served as "fibrosis indicators" (see Fig. 3 in the online Data Supplement). From a technological perspective, the assay offers the following advantages: (a) eluate is directly injected onto the array glass slide without the need for a protein quantification process such as Western blotting or an ELISA, and (b) reliable classification is achieved by using multiplex lectins (i.e., it is error tolerable).

To estimate the reproducibility of the lectin microarray analysis, 45 immunoprecipitated preparations from serum of a particular healthy volunteer were subjected to the lectin microarray analysis. We obtained signal intensities of 10 lectins (i.e., MAL, SNA, SSA, TJA-I, RCA120, DSA, PHA-E4, LEL, STL, and WGA) within the dynamic range (from 1000 to 40 000). When each lectin signal was normalized by the signal of DSA, the CVs of 9 lectins ranged from 6.3% to 14.3% (mean CV 10.5%). Next, we examined the day-to-day variation. For this purpose, we enriched AGPs from 50-[micro]L aliquots of sera. Seven wells of the arrayglass slide were used for the analysis per day, and the analysis was continued for 9 days. As shown in Fig. 4 in the online Data Supplement, the relative intensity of each lectin was acquired with small day-to-day variation.


To confirm that the signal pattern of the lectin microarray correlated with the progression of fibrosis, we analyzed the AGPs of 125 HCV patients. Fibrosis was graded according to the Histologic Activity Index by using biopsies or surgical specimens. The distribution of fibrosis stage in these patients was as follows: F0 -F1, 26.4% (33 patients); F2, 25.6% (32 patients); F3, 24% (30 patients); and F4, 23.2% (29 patients). All patients with fibrosis stage 4 (F4) in this study had Child-Pugh class A cirrhosis. After normalizing the lectin signal data to the signal of an internal standard lectin (DSA), we used the Mann-Whitney U-test as a nonparametric test (P value) to identify the most reliable fibrosis indicator. As shown in Table 2, the P values of all lectins except STL were <0.0001, indicating that they were strongly correlated with liver fibrosis. Among them, we selected AOL as the best fibrosis indicator. As shown in Table 2 and Fig. 2, AOL values were exponentially increased with the progression of fibrosis and were significantly (P = 1.9 x [10.sup.-13]) higher in patients with severe fibrosis (F3 and F4). In particular, a significant difference was observed between F2 and F3 (P < 0.001). We found that MAL was also one of the candidate indicators. MAL values gradually decreased with the fibrosis progression and were significantly lower in patients with cirrhosis (F4) (P = 1.5 x [10.sup.-13]). Incidentally, we did not detect a correlation between the intensity of necroinflammatory lesions and the lectin signals (see Fig. 5 in the online Data Supplement).

We next examined AUC to characterize the diagnostic accuracy of each lectin value at each stage of fibrosis, i.e., significant fibrosis (F2/F3/F4), severe fibrosis (F3/F4), and cirrhosis (F4). For the prediction of significant fibrosis, AUC, diagnostic sensitivity, and diagnostic specificity of AOL were 0.76%, 62%, and 88%, respectively (Fig. 3). These values were close to the values of the tests using the glyco-alteration in serum total IgG and specific IgG against [alpha]-Gal epitope, which are known to be high-performance biomarkers for liver fibrosis (30, 31). AOL preferentially distinguished between F2 and F3. In fact, diagnosis parameters for severe fibrosis were an AUC of 0.85, a diagnostic sensitivity of 78%, and a diagnostic specificity of 78%. On the other hand, MAL could detect cirrhosis with an AUC of 0.91, a diagnostic sensitivity of 93%, a diagnostic specificity of 76%, and a diagnostic odds ratio of 42, which was better than AOL (AUC 0.87, diagnostic sensitivity 83%, and diagnostic specificity 85% for cirrhosis).

Additionally, we analyzed a series of sera from 2 patients with fibrosis stage F4 (blood specimens of these HCV patients were collected at 0.5-year or 1.0year intervals). This retrospective analysis showed that signal changes gradually occurred in AOL and MAL, accompanied by the progression of fibrosis, whereas the platelet count had no correlation with the progression of fibrosis (see Fig. 6 in the online Data Supplement). It is noteworthy that the signal changes for both AOL and MAL in the HCC patient with advanced fibrosis invariably preceded the corresponding changes in the patient with liver cirrhosis. These results suggest that a test based on the ratio of the AOL or MAL signal to the DSA signal on the lectin microarray would enable the progression of liver fibrosis, as well as the development of cirrhosis, to be monitored.


We validated the relative intensity of AOL and MAL signals normalized by DSA signal (AOL/DSA and MAL/DSA). To establish a cutoff value for each signal, we used a data set of 125 patients mentioned previously. We compiled ROC curves to estimate diagnostic sensitivity, diagnostic specificity, and cutoff values (8.0% for AOL/DSA and 11.8% for MAL/DSA; Table 3). Through use of these cutoff values, we tested the accuracy of the immunoassay for the diagnosis of liver cirrhosis by using a validation data set comprised of another 88 patients (liver cirrhosis, n = 45; chronic hepatitis, n = 43) in a single-blind trial. AOL/DSA and MAL/DSA detected liver cirrhosis with 86% (37/43) and 91% (41/45) diagnostic sensitivity, 91% (39/43) and 89% (40/5) diagnostic specificity, and 89% (78/ 88) and 90% (79/88) overall accuracy, respectively (Table 3 and online Table 1). However, the likelihood ratios (LRs) of a positive test result (LR+), defined as the ratio of a positive test result among liver cirrhosis patients to the same result in chronic hepatitis patients, i.e., sensitivity/(1-specificity), and a negative test result (LR-), defined as ratio of a negative test result among liver cirrhosis patients to the same result in chronic hepatitis patients, i.e., (1-sensitivity)/ specificity, were not achieved to >10 and <0.1 in the both cases (27). To enhance the diagnostic accuracy by combining both the lectin signals, we developed the following:

[L.sub.Comb] = ([NI.sub.AOL] x 1.5 - [NI.sub.MAL])/[NI.sub.DSA] (1)

where NI is net intensity ([L.sub.Comb] is the value for combinational use of 2 lectins).

Each value ([L.sub.Comb]) given in equation 1 was subsequently used for ROC analysis. As shown in Table 3, the AUC for [L.sub.Comb] was 0.90, which was midway between that for MAL/DSA (0.91) and AOL/DSA (0.87). By using the most efficient cutoff value (-2), we were able to distinguish between liver cirrhosis and non-liver cirrhosis patients with 95% (41/43) diagnostic sensitivity, 91% (41/45) diagnostic specificity, and 93% (82/88) accuracy (Table 3 and online Table 1). Notably, LR+, LR-, and diagnostic odds ratio were 11, 0.051, and 210, respectively. This demonstrates that cirrhosis can be accurately diagnosed by the combined use of the 3 lectin signals on the lectin microarray.


The best way to monitor the progression of hepatitis is to establish an accurate serological method for quantitative evaluation of fibrosis. However, none of the noninvasive methods are superior to liver biopsy in accuracy. Although the latter is an invasive, and thus, high-risk procedure, it has been widely regarded as the gold standard for defining liver disease status. In the present study, we aimed to develop alternative convenient serological biomarkers for daily monitoring. We focused our analysis on a serum glycoprotein, AGP, which is mainly synthesized in the liver. AGP is a well-characterized glycoprotein with the highest carbohydrate content, representing 45% of the molecular weight attached in the form of 5 highly branched complex-type N-glycans. Glyco-alteration of AGP (e.g., desialylation, increased branching, and increased fucosylation) has been reported to occur during cirrhosis and during acute and chronic inflammation (19). In particular, increased fucosylation detected by lectin antibody sandwich ELISA was evidently confirmed to be a cirrhosis marker with a high diagnostic accuracy (20). Furthermore, because of the relatively high abundance of AGP in serum (from 0.5 to 1.0 g/L), high detection of glyco-alteration is expected, enabling the construction of an assay system by using a small volume of serum. In fact, our lectin microarray detected many aspects of glyco-alteration of AGP enriched from 0.5 [micro]L serum in the verification phase (step 1 in Fig. 1 in the online Data Supplement). We statistically evaluated the relationships between each lectin signal and the progression of fibrosis diagnosed histopathologically. The analysis identified the most useful indicator lectins for the evaluation of the fibrosis progression. In the subsequent blind test, we confirmed that the combinational use of 3 lectin signals allowed us to identify F4 patients as having liver cirrhosis. The diagnostic accuracy was greatly improved compared with a previous report (20). We must construct a rapid and simple sandwich ELISA kit with the lectin set we optimized in this study, before the next validation phase toward a much larger cohort. The optimized assay could be compared with other recently introduced glyco-based test approaches by using the same sample set (30, 31).

An additional key point of our system is the ability to quantify the lectin signals, which would greatly improve the reliability of the assay and enable dynamic monitoring of patients. The developed assay system allowed us to carry out a retrospective analysis of a series of patient sera to evaluate the correlation between changes in the indicator lectin signals (biochemical parameter) and the progression of liver fibrosis (clinical parameter; see Fig. 6 in the online Data Supplement). The proposed monitoring method might facilitate classification of chronic hepatitis patients with respect to disease progression and thereby enable intensive surveillance and follow-up of the chronic hepatitis patients to estimate their risk of developing HCC. The current standard treatment, pegylated interferon-a plus ribavirin, does not provide sustained virologic response in all patients with HCV. Two SNPs near the gene IL28B on chromosome 19 are associated with null virological response in the treatment of patients with HCV genotype 1 (32). Studies on new antiviral agents and treatment approaches are required. Our study is the first to apply a lectin array-based sandwich immunoassay system to the clinical evaluation of glycobiomarkers. Because the glycosylation features strongly depend on tissue origin and the presence of malignancy, blood glycoproteins consist of a considerable mixture of heterogeneous molecules derived from many sources, although their protein moieties are the same. Thus, detection of serum glycoproteins exhibiting cancer-associated glyco-alteration can be extremely useful in biomarker investigation (33). Until now, however, there was no reliable assay targeting the glycoprotein biomarkers. The absence of a "rational" system for verification of glyco-biomarker candidates and selection of the most efficient lectin was a major impediment to the development of a reliable sandwich immunoassay. In the present study, we applied an antibody-assisted lectin profiling system to identify the most useful lectin for the validation of fibrosis marker AGP. The glyco-alteration of AGP was completely depicted by using the most reliable lectin set, which was rationally identified by using the lectin array-based sandwich immunoassay before the validation (step 5 in Fig. 1 in the online Data Supplement). These processes should contribute to reducing the current labor of the time-consuming glyco-marker development and thus expedite future innovation in the field of biomarkers.

Author Contributions: All authors confirmed they have contributed to the intellectual content of this paper and have met the following 3 requirements: (a) significant contributions to the conception and design, acquisition of data, or analysis and interpretation of data; (b) drafting or revising the article for intellectual content; and (c) final approval of the published article.

Authors' Disclosures or Potential Conflicts of Interest: Upon manuscript submission, all authors completed the Disclosures of Potential Conflict of Interest form. Potential conflicts of interest:

Employment or Leadership: None declared.

Consultant or Advisory Role: None declared.

Stock Ownership: None declared.

Honoraria: None declared.

Research Funding: This work was supported in part by a grant from New Energy and Industrial Technology Development Organization of Japan.

Expert Testimony: None declared.

Role of Sponsor: The funding organizations played no role in the design of study, choice of enrolled patients, review and interpretation of data, or preparation or approval of manuscript.

Acknowledgments: We thank H. Ando, T. Shikanai, K. Saito, Y. Kubo, J. Murakami, S. Shirakawa, N. Matsuura (AIST), and H. Naganuma (Nagoya City University) for technical assistance. We also thank H. Kaji for critical discussion. We are grateful to Prof. N. Hamajima and F. Kurbanov for critical suggestions on statistical study and writing the manuscript.


(1.) Lavanchy D. Hepatitis B virus epidemiology, disease burden, treatment, and current and emerging prevention and control measures. J Viral Hepatol 2004;11:97-107.

(2.) Szabo E, Lotz G, Paaska C, Kiss A, Schaff Z. Viral hepatitis: new data on hepatitis C infection. Pathol Oncol Res 2003;9:215-21.

(3.) Perz JF, Armstrong GL, Farrington LA, Hutin YJ, Bell BP. The contributions of hepatitis B virus and hepatitis C virus infections to cirrhosis and primary liver cancer worldwide. J Hepatology 2006; 45:529-38.

(4.) Yoshizawa H, Tanaka J, Miyakawa Y. National prevention of hepatoce u ar carcinoma in Japan based on epidemiology of hepatitis C virus infection in the general population. Intervirology 2006;49:7-17.

(5.) Yoshida H, Shiratori Y, Moriyama M, Arakawa Y, Ide T, Sata M, et al. Interferon therapy reduces the risk for hepatocellular carcinoma: national surveillance program of cirrhotic and noncirrhotic patients with chronic hepatitis C in Japan. IHIT Study Group. Inhibition of Hepatocarcinogenesis by Interferon Therapy. Ann Intern Med 1999;131: 174-81.

(6.) Afdhal N, McHutchison J, Brown R, Jacobson I, Manns M, Poordad F, et al. Thrombocytopenia associated with chronic liver disease. J Hepatol 2008;48:1000-7.

(7.) Imbert-Bismut F, Ratziu V, Pieroni L, Charlotte F, Benhamou Y, Poynard T. Biochemical markers of liver fibrosis in patients with hepatitis C virus infection: a prospective study. Lancet 2001;357: 1069-75.

(8.) Poynard T, Imbert-Bismut F, Munteanu M, Messous D, Myers RP, Thabut D, et al. Overview of the diagnostic value of biochemical markers of liver fibrosis (FibroTest, HCV FibroSure) and necrosis (ActiTest) in patients with chronic hepatitis C. Comp Hepatol 2004;3:1-12.

(9.) Sandrin L, Fourquet B, Hasquenoph JM, Yon S, Fournier C, Mal F, et al. Transient elastography: a new noninvasive method for assessment of hepatic fibrosis. Ultrasound Med Biol 2003;29: 1705-13.

(10.) Colletta C, Smirne C, Fabris C, Toniutto P, Rapetti R, Minisini R, et al. Value of two noninvasive methods to detect progression of fibrosis among HCV carriers with normal aminotransferases. Hepatology 2005;42:838-45.

(11.) Kettaneh A, Marcellin P, Douvin C, Poupon R, Ziol M, Beaugrand M, et al. Features associated with success rate and performance of fibroscan measurements for the diagnosis of cirrhosis in HCV patients: a prospective study of 935 patients. J Hepatol 2007;46:628-34.

(12.) Ghany MG, Strader DB, Thomas DL, Seeff LB. Diagnosis, management, and treatment of hepatitis C: an update. Hepatology 2009;49:1335-74.

(13.) Aebersold R, Anderson L, Caprioli R, Druker B, Hartwell L, Smith R. Perspective: a program to improve protein biomarker discovery for cancer. J Proteome Res 2005;4:1104-9.

(14.) Callewaert N, Van Vlierberghe H, Van Hecke A, Laroy W, Delanghe J, Contreras R. Noninvasive diagnosis of liver cirrhosis using DNA sequencer-based total serum protein glycomics. Nat Med 2004;10:429-34.

(15.) Liu X-E, Desmyter L, Gao CF, Laroy W, Dewaele S, Vanhooren V, et al. N-glycomic changes in hepatocellular carcinoma patients with liver cirrhosis induced by hepatitis B virus. Hepatology 2007; 46:1426-35.

(16.) Comunale MA, Wang M, Hafner J, Krakover J, Rodemich L, Kopenhaver B, et al. Identification and development of fucosylated glycoproteins as biomarkers of primary hepatocellular carcinoma. J Proteome Res 2009;8:595-602.

(17.) Kuno A, Uchiyama N, Koseki-Kuno S, Ebe Y, Takashima S, Yamada M, et al. Evanescent-field fluorescence-assisted lectin microarray: a new strategy for glycan profiling. Nat Methods 2005; 2:851-6.

(18.) Kuno A, Kato Y, Matsuda A, Kaneko MK, Ito H, Amano K, et al. Focused differential glycan analysis with the platform antibody-assisted lectin profiling for glycan-related biomarker verification. Mol Cell Proteomics 2008;8:99-108.

(19.) Turner GA. N-glycosylation of serum proteins in disease and its investigation using lectins. Clin Chim Acta 1992;208:149-71.

(20.) Ryden I, Pahlsson P, Lindgren S. Diagnostic accuracy if [alpha]1-acid glycoprotein fucosylation for liver cirrhosis in patients undergoing hepatic biopsy. Clin Chem 2002;48:2195-201.

(21.) Tanaka Y, Hanada K, Mizokami M, Yeo AE, Shih JW, Gojobori T, et al. A comparison of the molecular clock of hepatitis C virus in the United States and Japan predicts that hepatocellular carcinoma incidence in the United States will increase over the next two decades. Proc Natl Acad SciUSA2002;99:15584-9.

(22.) Orito E, IchidaT, Sakugawa H, Sata M, Horiike N, Hino K, et al. Geographic distribution of hepatitis B virus (HBV) genotype in patients with chronic HBV infection in Japan. Hepatology 2001;34: 590-4.

(23.) Desmet VJ, Gerber M, Hoofnagle JH, Manns M, Scheuer PJ. Classification of chronic hepatitis: diagnosis, grading and staging. Hepatology 1994; 19:1513-20.

(24.) Uchiyama N, Kuno A, Tateno H, Kubo Y, Mizuno M, Noguchi M, et al. Optimization of evanescent-field fluorescence-assisted lectin microarray for high-sensitivity detection of monovalent oligosaccharides and glycoproteins. Proteomics 2008; 8:3042-50.

(25.) Leroy V, Halfon P, Bacq Y, Boursier J, Rousselet MC, Bourliere M, et al. Diagnostic accuracy, reproducibility and robustness of fibrosis blood tests in chronic hepatitis C: a meta-analysis with individual data. Clin Biochem 2008;41:1368-76.

(26.) Akobeng AK. Understanding diagnostic tests 3: receiver operating characteristic curves. Acta Paediatr 2007; 96:644-7.

(27.) Glas AS, Lijmer JG, Prins MH, Bonsel GJ, Bossuyt PM. The diagnostic odds ratio: a single indicator of test performance. J Clin Epidemiol 2003;56: 1129-35.

(28.) Hashimoto S, AsanoT, Takahashi J, Yagihashi Y, Nishimura T, Saniabadi AR, et al. [alpha]1-Acid glycoprotein fucosylation as a marker of carcinoma progression and prognosis. Cancer 2004;101: 2825-36.

(29.) Kuno A, Itakura Y, Toyoda M, Takahashi Y, Yamada M, Umezawa A, et al. Development of a data-mining system for differential profiling of cell glycoproteins based on lectin microarray. J Proteom Bioinform 2008;1:68-72.

(30.) Vanderschaeghe D, Laroy W, Sablon E, Halfon P, Van Hecke A, Delanghe J, et al. GlycoFibroTest is a highly performant liver fibrosis biomarker derived from DNA sequencer-based serum protein glycomics. Mol Cell Proteomics 2009;8:986-94.

(31.) Mehta AS, Long RE, Comunale MA, Wang M, Rodemich L, Krakover J, et al. Increased levels of galactose-deficient anti-Gal immunoglobulin G in the sera of hepatitis C virus-infected individuals with fibrosis and cirrhosis. J Virol 2008;82:1259-70.

(32.) Tanaka Y, Nishida N, Sugiyama M, Kurosaki M, Matsuura K, Sakamoto N, et al. Genome-wide association of IL28B with response to pegylated interferon-alpha and ribavirin therapy for chronic hepatitis C. Nat Genet 2009;41:1105-9.

(33.) Narimatsu H, Sawaki H, Kuno A, Kaji H, Ito H, Ikehara Y. A strategy for discovery of cancer glyco-biomarkers in serum using newly developed technologies for glycoproteomics. FEBS J 2010;277:95-105.

Atsushi Kuno, [1] Yuzuru Ikehara, [1] Yasuhito Tanaka, [2] Takashi Angata, [1] Sachiko Unno, [1] Maki Sogabe, [1] Hidenori Ozaki, [1] Kiyoaki Ito, [3] Jun Hirabayashi, [1] Masashi Mizokami, [2,3] and Hisashi Narimatsu [1] *

[1] Research Center for Medical Glycoscience (RCMG), National Institute of Advanced Industrial Science and Technology (AIST), Tsukuba, Ibaraki, Japan; [2] Department of Virology and Liver Unit, Nagoya City University Graduate School of Medical Sciences, Kawasumi, Mizuho, Japan; the [3] Research Center for Hepatitis and Immunology, National Center for Global Health and Medicine, Ichikawa, Japan.

[4] Nonstandard abbreviations: HBV, hepatitis B virus; HCV, hepatitis C virus; HCC, hepatocellular carcinoma; AGP, [alpha]1-acid glycoprotein; AIST, National Institute of Advanced Industrial Science and Technology; PBSTx, PBS containing 1% Triton X-100; ROC, receiver operating characteristic; AUC, area under the curve; LR, likelihood ratio.

[5] Lectins: AAL, Aleuria aurantia lectin; ABA, Agaricus bisporus agglutinin; ACA, Amaranthus caudatus agglutinin; AOL, Aspergillus oryzae lectin; BPL, Bauhinia purpurea alba lectin; ConA, Canavalia ensiformis agglutinin; DBA, Dolichos biflorus agglutinin; DSA, Datura stramonium agglutinin; ECA, Erythrina cristagalli agglutinin; EEL, Euonymus europaeus lectin; GNA, Galanthus nivalis agglutinin; GSL-I, Griffonia simplicifolia lecti n I; GSL-I[A.sub.4], GSL-I isolectin [A.sub.4]; GSL-IB4, GSL-I isolectin [B.sub.4]; GSL-II, Griffonia simplidfolia lectin II; HHL, Hippeastrum hybrid lectin; HPA, Helix pomatia agglutinin; Jacalin, Artocarpus integrifolia (jackfruit) lectin; LCA, Lens culinaris agglutinin; LEL, Lycopersicon esculentum lectin; LTL, Lotus tetragonolobus lectin; MAH, Maackia amurensis haemagglutinin; MAL, Maackia amurensis leukoagglutinin; MPA, Maclura pomifera agglutinin; NPA, Narcissus pseudonarcissus agglutinin; PHA-E4, phytohemagglutinin E-subunits from Phaseolus vulgaris; PHA-L4, phytohemagglutinin L-subunits from Phaseolus vulgaris; PNA, peanut agglutinin; PSA, Pisum sativum agglutinin; PTL-I, Psophocarpus tetragonolobus lectin I; PWM, pokeweed mitogen; RCA120, Ricinus communis agglutinin 120 kDa fraction; SBA, soybean agglutinin; SNA, Sambucus nigra agglutinin; SSA, Sambucus sieboldiana agglutinin; STL, Solanum tuberosum lectin; TJA-I, Trichosanthes japonica agglutinin I; TJA-II, Trichosanthes japonica agglutinin II; UDA, Urtica dioica agglutinin; UEA-I, Ulex europaeus agglutinin I; VVA, Vicia villosa agglutinin; WFA, Wisteria floribunda agglutinin; WGA, wheat germ agglutinin.

* Hisashi Narimatsu, Research Center for Medical Glycoscience (RCMG), National Institute of Advanced Industrial Science and Technology (AIST), 1-1-1 Umezono, Tsukuba, Ibaraki 305-8568, Japan. Fax +81-29-861-3191; e-mail

Received June 1, 2010; accepted October 11, 2010.

Previously published online at DOI: 10.1373/clinchem.2010.151340
Table 1. General characteristics of the patient population. (a)

         n    % Male   Age, years      ALT, U/L        Platelet,

Total   125    46.4    56.5 (12.1)   79.9 (149.9)     155.2 (63.1)
F0-F1   33     42.4    50.9 (13.7)   109.9 (278.5)    203.2 (60.8)
F2      32     43.8    53.9 (12.2)    50.9 (39.2)     171.2 (49.9)
F3      31     41.9    58.7 (9.7)     84.6 (58.6)     145.4(44.6)
F4      29     58.6    63.6 (7.9)     72.8 (57.5)     93.2 (37.6)

(a) Data are means (SD) unless otherwise indicated.

Table 2. Selection of lectins as fibrosis indicators. (a)

n = 125            Significant fibrosis (F2-F4)

Lectin    Relative intensity (%) (a)

             F0-F1         F2-F4
           (n = 33)      (n = 92)        U-test (P)

AOL        1.8 (2.0)    13.4 (17.7)   7.2 x [10.sup.-6]
MAL       16.9 (4.1)    11.4 (6.8)    6.9 x [10.sup.-5]
PHA-E4    28.5 (3.8)    21.9 (8.0)    2.6 x [10.sup.-5]
AAL       16.8 (8.2)    33.3 (22.2)   3.5 x [10.sup.-5]
LEL       85.1 (10.8)   68.9 (25.6)   3.7 x [10.sup.-4]
STL       39.9 (6.3)    36.1 (9.1)    3.4 x [10.sup.-2]

n = 125            Severe fibrosis (F3-F4)

Lectin    Relative intensity (%) (a)

             F0-F2         F3-F4
           (n = 65)      (n = 60)         U-test (P)

AOL        2.5 (3.7)    18.8 (18.9)   1.9 x [10.sup.-13]
MAL        16.4(4.6)     9.0 (6.4)    2.0 x [10.sup.-10]
PHA-E4    27.8 (4.9)    19.1 (7.7)    9.0 x [10.sup.-11]
AAL       17.7 (10.0)   41.0 (22.6)   3.1 x [10.sup.-12]
LEL       82.8 (15.6)   62.7 (26.6)   2.2 x [10.sup.-6]
STL       39.3 (7.2)    34.7 (9.5)    2.0 x [10.sup.-3]

n = 125                     Cirrhosis (F4)

Lectin    Relative intensity (%) (a)

             F0-F3          F4
           (n = 96)      (n = 29)         U-test (P)

AOL        4.9 (8.5)    28.5 (19.8)   6.0 x [10.sup.-11]
MAL       15.1 (5.5)     5.5 (4.5)    1.5 x [10.sup.-13]
PHA-E4    26.1 (6.2)    15.5 (6.7)    3.4 x [10.sup.-11]
AAL       22.1 (13.4)   51.4 (24.9)   1.2 x [10.sup.-9]
LEL       81.6 (17.5)   45.1 (19.8)   1.6 x [10.sup.-13]
STL       39.2 (7.7)    30.2 (7.9)    1.8 x [10.sup.-6]

(a) Data are mean (SD) unless otherwise indicated.

Table 3. Diagnostic values of lectin signal tests
for liver cirrhosis. (a)
                               AOL         MAL      2 Lectins

Area under the ROC curve      0.87        0.91        0.90
95% CI                      0.79-0.96   0.86-0.96   0.84-0.96
Cutoff line (b)                 8         11.7         -2
False positive                  4           5           4
True negative                  41          40          41
Specificity, %                 91          89          91
True positive                  37          39          41
False negative                  6           4           2
Sensitivity (%)                86          91          95
Overall accuracy (%)           89          90          93
Likelihood ratio of a          9.7         8.2         11
  positive test
  result (LR+) (c)
Likelihood ratio of a         0.15        0.10        0.051
  negative test
  result (LR-) (c)
Diagnostic odds ratio (c)      63          78          210

(a) n = 88 (chronic hepatitis: n = 45, liver cirrhosis: n = 43).

(b) Optimal cutoff point was determined by using the maximum
Youden's index.

(c) All indexes were calculated as described by Glas et al. (27).

Fig. 2. Correlation of lectin signals with liver fibrosis.

Box and whisker plots show the mean (X), the 75th and 25th
percentiles (the top and bottom of the box, respectively),
the median (line through the middle of the box), and the
95th and 5th percentiles (the top and bottom of the whiskers,
respectively). P value was calculated with a nonparametric
test (the Mann-Whitney U-test) (bottom).

         F0,1     F2      F3      F4

n         33      32      31      29
Median   1.19    1.20    4.61    30.08
SD       2.04    4.81    12.70   19.78

         F0,1     F2      F3      F4

n         33      32      31      29
Median   17.25   16.97   12.90   5.10
SD       4.13    5.09    6.10    4.53

         F0,1     F2      F3      F4

n         33      32      31      29
Median   20.8    16.75   13.8     8.4
SD       6.08    4.99    4.29    3.71

U-test (P)      F0,1 vs F2            F2 vs F3

AOL                0.66           1.6 x [10.sup.-4]
MAL                0.66          2.4 x 10[10.sup.-2]
Platelet     1.6 x [10.sup.-2]    5.7 x [10.sup.-2]
ALT                0.57           6.9 x [10.sup.-3]

                 F3 vs F4            F0,1 vs F3

AOL          2.0 x [10.sup.-4]    1.8 x [10.sup.-6]
MAL          1.4 x [10.sup.-5]    4.0 x [10.sup.-3]
Platelet     3.5 x [10.sup.-2]    7.4 x [10.sup.-5]
ALT                0.44           3.2 x [10.sup.-2]

                 F2 vs F4            F0,1 vs F4

AOL          1.5 x [10.sup.-9]   2.1 x [10.sup.-10]
MAL          9.3 x [10.sup.-11]  1.1 x [10.sup.-13]
Platelet     3.7 x [10.sup.-9]   8.6 x [10.sup.-11]
ALT          6.7 x [10.sup.-2]   4.1 x [10.sup.-2]

Fig. 3. Diagnostic accuracy of prediction of liver fibrosis.

ROC curves for AOL and MAL values obtained from lectin
microarray analysis were used to predict progression of liver
fibrosis. The area under the ROC curve (AUC) is depicted in
parentheses. Diagnostic values of lectin signal tests for
detecting significant fibrosis (F2/F3/F4), severe fibrosis
(F3/F4), and cirrhosis (F4) are shown.

Significant fibrosis

          AUC       sensitivity   Specificity

       (95% CI)         (%)           (%)       LR+   LR-    OR
         0.76           62            88        5.2   0.43   12
AOL   (0.67-0.84)
         0.73           46            97        15    0.56   28
MAL   (0.64-0.82)

Severe fibrosis

          AUC       Sensitivity   Specificity   LR+   LR-    OR

       (95% CI)         (%)           (%)
AOL   (0.79-0.92)       78            78        3.5   0 28   13
MAL   (0.74-0.89)       68            86        4.9   0.37   13


          AUC       Sensitivity   Specificity

       (95% CI)         (%)           (%)       LR+   LR-    OR
AOL   (0.79-0.96)       83            85        5.5   0.2    28
MAL   (0.86-0.96)       93            76        3.9   0.09   42
COPYRIGHT 2011 American Association for Clinical Chemistry, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2011 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:Proteomics and Protein Markers
Author:Kuno, Atsushi; Ikehara, Yuzuru; Tanaka, Yasuhito; Angata, Takashi; Unno, Sachiko; Sogabe, Maki; Ozak
Publication:Clinical Chemistry
Article Type:Report
Geographic Code:9JAPA
Date:Jan 1, 2011
Previous Article:Corin is present in the normal human heart, kidney, and blood, with pro-B-type natriuretic peptide processing in the circulation.
Next Article:Development of a homogeneous assay for measurement of small dense LDL cholesterol.

Terms of use | Privacy policy | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters