FASEB J. Cell Migration Consortium
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published as doi: 10.1096/fj.06-7303com.
 (The FASEB Journal. 2007;21:1324-1334.)
© 2007 FASEB
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
fj.06-7303comv1
21/7/1324    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Lee, S.-H.
Right arrow Articles by Lee, J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Lee, S.-H.
Right arrow Articles by Lee, J.

A novel approach to ultrasensitive diagnosis using supramolecular protein nanoparticles

Sung-Hyun Lee*,1, Hyewon Lee*,1, Jin-Seung Park*, Hyoung Choi*,2, Kyung-Yeon Han*, Hyuk-Seong Seo*, Keum-Young Ahn*, Sung-Sik Han{dagger}, Yunjung Cho{ddagger}, Kee-Hyoung Lee{ddagger} and Jeewon Lee*,3

* Department of Chemical and Biological Engineering,

{dagger} School of Life Sciences and Biotechnology,

{ddagger} College of Medicine, Korea University, Seoul 136-713, South Korea

3Correspondence: Department of Chemical and Biological Engineering, Anam-Dong 5-1, Seoul 136-713, South Korea. E-mail: leejw{at}korea.ac.kr


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
We report on the ultrasensitive protein nanoprobe system that specifically captures disease marker (autoantibodies of Type I diabetes in this case) with attomolar sensitivity. The system relies on supramolecular protein nanoparticles that bind a specific antibody [65 kDa glutamate decarboxylase (GAD65)-specific autoantibody, i.e., the early marker of Type I diabetes]. The ultrasensitive detection of early marker of Type I diabetes during the early phase of pancreatic ß-cell destruction is important because individuals at high risk of developing Type I diabetes can be identified several years before the clinical onset of the ailment. The bacterial expression of chimera genes encoding N-[human ferritin heavy chain (hFTN-H)]::[specific antigenic epitope]-C produces supramolecular nanoparticles with uniform diameters (10–15 nm), owing to self-assembly activity of hFTN-H. Each nanoparticle, formed by intermolecular self-assembly between the chimera protein molecules, is subjected to carrying a large number (presumably, 24) of epitopes with a homogeneous and stable conformation per autoantibody binding, thereby allowing substantial enhancement of sensitivity. The sensitivity was finally boosted to 3 attomolar concentration of the autoantibodies, 4–9 orders of magnitude more sensitive than conventional immunoassays. Also, this ultrasensitive protein nanoprobe successfully detected natural autoantibodies in the sera from Type I diabetic patients. The attomolar sensitivity was successfully reproduced on the detection of other antibodies, i.e., monoclonal antibodies against hepatitis B surface antigen. With the two antibody markers above, the feasibility of simultaneous and multiplexing-mode detection was also demonstrated.—Lee, S-H., Lee, H., Park, J-S., Choi, H., Han, K-Y., Seo, H-S., Ahn, K-Y., Han, S-S., Cho, Y., Lee, K-H., Lee, J. A novel approach to ultrasensitive diagnosis using supramolecular protein nanoparticles.


Key Words: human ferritin heavy chain • nanoprobe system • attomoloar sensitivity


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
ABNORMAL CONCENTRATIONS OF certain marker proteins often indicate the presence of various cancers/diseases. However, current diagnosis methods only allow detection when protein levels become higher than critical threshold concentrations. Since at these concentrations the cancer or disease is often significantly advanced, more sensitive methods that allow for early detection of protein markers could potentially revolutionize physician treatment of various cancers/diseases and increase patient survival rates. In the area of protein diagnostics, the detection limit of current standard method [enzyme-linked immunosorbant assay (ELISA)] is only nano- to picomolar level (1) .

In this study, we propose a novel ultrasensitive system to detect antibody markers [i.e., autoantibodies of Type I diabetes (insulin-dependent diabetes mellitus, IDDM) and antibodies against small hepatitis B surface antigen (S-HBsAg)] using supramolecular protein nanoparticles. Since the IDDM autoantibodies are present even before the clinical onset, sensitive and early detection of the autoantibodies is of crucial importance to prevention and/or treatment of the ailment (2 3 4 5 6 7 8 9 10) . Intermolecular self-assembly of hFTN-H molecules forms nanoscale (~12 nm) supramolecules (i.e., self-assembled nanoparticles) with a spherical shell structure (11) and provides the distinct advantages such as small size (10–15 nm) and correspondingly large surface-to-vol ratio, high size uniformity, biological display of multiple antigenic probes on each nanoparticle, and homogeneous and stable probe conformation. Using the hFTN-H-based protein nanoparticles, sensitivity was finally boosted to a 3-attomolar concentration of the antibodies, which corresponds to 4–9 orders of magnitude higher sensitivity compared to that of other previous antibody immunoassays (1 , 12) .


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Synthesis, purification, and transmission electron microscopy (TEM) imaging of supramolecular protein nanoparticles
The plasmid expression vectors pET28a-HFTN-GADa, pET28a-HFTN-GADb, pET28a-HGAD, and pET28a-HFTN-HBsAg were constructed by inserting the genes encoding NH2-(hFTN-H)::(GAD65-C138)-COOH, NH2-(hFTN-H)::(GAD65-C341)-COOH, NH2-(GAD65-C138)-COOH, and NH2-(hFTN-H)::(S-HBsAg-MHR)-COOH, respectively, into the NdeI-XhoI site of plasmid pET28a (Novagen, San Diego, CA, USA). The other expression vectors, pT7-KFTN-GAD, pT7-FTN-GAD, and pT7-KFTN-SHBsAg, were also constructed by inserting the gene encoding NH2-(lys)10::(hFTN-H)::(GAD65-C138)-COOH, NH2-(hFTN-H)::(GAD65-C138)-COOH, and NH2-(lys)10::(hFTN-H)::(S-HBsAg-MHR)-COOH, respectively, into the NdeI-XhoI site of plasmid pT7–7. After transformation of E. coli strain BL21(DE3) (F– ompT hsdSB(rB– mB–)) with the plasmid expression vectors above, kanamycin- (for pET28a-HFTN1-GADa, pET28a-HFTN2-GADb, pET28a-HGAD, and pET28a-HFTN-HBsAg) or ampicillin-resistant (for pT7-KFTN-GAD, pT7-FTN-GAD, and pT7-KFTN-SHBsAg) transformants were selected. The experimental procedures for chimera gene expression, purification, and TEM imaging of supramolecular protein nanoparticles are well described in our previous report (11) . The particle size distribution of supramolecular protein nanoparticles was estimated through the particle size analysis using the software Gatan Digital Micrograph (Gatan Inc., Warrendale, PA, USA) with the TEM image acquired by Multiscan 600W CCD camera (Gatan Inc.).

Polymerase chain reaction (PCR)-based detection
A PVDF membrane (Immobilion-FL, IPFL 10100, Millipore, MA, USA) was prewetted with methanol for 1 min and washed with a PBS buffer (137 mM NaCl, 2.7 mM KCl, 10 mM Na2HPO4, 2 mM KH2PO4, pH 7.4) for 5–10 min. Before the PVDF membrane was completely dried, 10 µl of the elution buffer that contained the purified supramolecular nanoparticles (0.5 mg ml–1) was dropped onto a designated spot on the membrane. Then the membrane was slowly stirred for 1 h in the blocking solution (1% skim milk) and washed twice with the PBS buffer for 30 min. Autoantibodies GAD6 (mouse anti-GAD65 monoclonal antibodies, Cat. No. 559931, BD Biosciences, San Jose, CA, USA) that recognize a C-terminal linear epitope localized in the last 41 amino acids of human GAD65 (13) or mouse anti-S-HBsAg monoclonal antibodies (kindly provided by Dobeel Corp., Republic of Korea) in 10 ml of the PBS buffer or the human plasma solution (P9523 Plasma, Sigma-Aldrich, St. Louis, MO, USA) were applied to the supramolecular nanoparticles that are already immobilized onto the PVDF membrane, by slowly stirring the membrane in the autoantibody [or anti-S-HBsAg monoclonal antibody (mAb)]-containing PBS buffer or human plasma solution for 1 h. In the case of applying Type I diabetic patient serum, 250 µl of serum from each patient was used. After the membrane was washed twice with the PBS buffer for 30 min, it was slowly stirred for 1 h in the Tris buffer (10 mM Tris-HCl) that contained another type of nanoparticles tagged with polylysine, (lys)10 on the particle surface and again washed twice with the PBS buffer for 30 min. A single-stranded 30-bp DNA primer (5'-GGATCCTGTGAATCCAAACGTATTGTTGCA-3') was applied by slowly stirring the PVDF membrane in the PBS buffer that contained the DNA primer (1 µM) for 1 h. [The DNA primer used was one of two primers required for the PCR amplification of the MADI-C gene (483 bp) encoding the C-terminus (250–410 amino acids) of arginine deiminase from Mycoplasma arginini.] After the PVDF membrane was washed with the PBS buffer for 3–5 min, it was dried at room temperature. The designated spot on the dried membrane where the supramolecular nanopartices as capture probes were initially immobilized was cut out and added to the PCR premix (Bioneer, Taejon, Republic of Korea) with the DNA template [T-vector (Promega, WI, USA) containing a single copy of the MADI-C gene] and another primer (5'-AAGCTTTTATCACTTAACATCTTTACGTGA –3', 0.5 µM). After the PCR (94°C/5 min; 30 cycles of 94°C/30 s, 52°C/1 min, and 72°C/30 s; 72°C/7 min) to amplify the MADI-C gene, its PCR product was analyzed through 1% agarose gel electrophoresis.

Quantum dot (Qdot)-based detection
The procedures for the initial prewetting and washing of a PVDF membrane, immobilization of the purified supramolecular nanoparticles onto the membrane, skim-milk blocking and washing, and application of autoantibodies (or anti-S-HBsAg mAb) in the PBS buffer, the human plasma solution, or patient serum were exactly the same as those mentioned in the previous section. After the PVDF membrane was washed twice with the PBS buffer for 30 min, 1 nM of the (Qdot-CdSe)-2°Ab conjugate [Qdot® 655 Goat F(ab')2 anti-mouse IgG conjugate, Cat. No. 1102–1, Quantum Dot Corporation, Carlsbad, CA, USA] in a reagent buffer (Product No. 83–0014, Quantum Dot Corp.) were applied onto the membrane, and subsequently the membrane was washed with the PBS buffer for 30 min. Clear photoluminescence was visualized in the presence of UV light when the autoantibodies of Type I diabetes (or anti-HBsAg mAb) were sandwiched between the (Qdot-CdSe)-2°Ab conjugate and the immobilized nanoparticles. The intensity of photoluminescence was analyzed using microplate reader (GENios, Tecan, Austria) with excitation and emission at 420 and 650 nm, respectively after the spotted area of PVDF membrane was placed in Costar 96 well (Costar 3596, Corning, NY, USA).


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Supramolecular protein nanoparticles for ultrasensitive detection
Small isoforms of glutamate decarboxylase (GAD65) are the predominant target antigens of pancreatic islet cell autoantibodies in patients with Types I diabetes mellitus. Through analyses of autoimmune epitopes in GAD65, the middle and C-terminal domains in GAD65 were defined as dominant epitope in the early phases of ß-cell autoimmunity (7 , 10 , 14 15 16 17 18) . Three of five monoclonal anti-islet-cell antibodies (MICAs) derived from diabetic patients recognized the C-terminal epitope, and two MICAs were destroyed by deleting C-terminal 41 amino acids of GAD65 (15) . Furthermore, recent preliminary data suggest that an increase in anti-GAD65 autoantibodies to C-terminal epitopes distinguishes diabetic from healthy children (19) . On the basis of these clinical reports, we focused on C-terminus of GAD65, i.e., C-terminal 138 amino acids (Arg448 to Leu585) (a linear epitope named GAD65-C138) and C-terminal 341 amino acids (Ile245 to Leu585) (a conformational epitope named GAD65-C341) in the preparation of nanoparticle capture probes to detect the autoantibodies. The linear and conformational epitopes above were defined based on the analysis results of immunoreactivity of MICAs with GAD65 mutants, which were previously reported by Richter et al. (15) and Söhnlein et al (17) .

The chimera gene was constructed by replacing the epitope ORF (encoding GAD65-C138 or GAD65-C341) of the first chimera gene with another ORF encoding major hydrophilic region (MHR) of adr subtype of S-HBsAg (residues 99–169) and expressed using the same E. coli expression system. The chimera gene products NH2-(his)6::(hFTN-H)::(GAD65-C138)-COOH (named H-FTN-GAD65S) and NH2-(his)6::(hFTN-H)::(S-HBsAg-MHR)-COOH (named H-FTN-SHBsAg) in cytoplasm of E. coli BL21(DE3) were purified through Ni+2-affinity chromatography from supernatant after cell disruption, and the TEM images of the purified H-FTN-GAD65S and H-FTN-SHBsAg showed spherical nanoparticles with diameters of 10–15 nm (Fig. 1 A, B). Other chimera gene products NH2-(his)6::(hFTN-H)::(GAD65-C341)-COOH (named H-FTN-GAD65L), NH2-(lys)10::(hFTN-H)::(GAD65-C138)-COOH (named K-FTN-GAD65S), NH2-(hFTN-H)::(GAD65-C138)-COOH (named FTN-GAD65S), and NH2-(lys)10::(hFTN-H)::(S-HBsAg-MHR)-COOH (named K-FTN-SHBsAg) also formed nanoscale supramolecules with uniform sizes (data not shown).


Figure 1
View larger version (36K):
[in this window]
[in a new window]

  		Figure 1. hFTN-H-based protein nanoparticles and schematics of the ultrasensitive detection.

A) TEM images of H-FTN-GAD65S and H-FTN-SHBsAg nanoparticles. B) Particle size distribution of H-FTN-GAD65S (gray bars) and H-FTN-SHBsAg nanoparticles (black bars). C) Result of native PAGE of purified H-FTN-GAD65S (lane 1), H-FTN-SHBsAg nanoparticles (lane 2), and hFTN-H (lane 3) [M: marker, urease from Canavalia ensiformis (Jack bean) hexamer (545 kDa)]. D) Schematics of Qdot- and PCR-based detection method (Methods 1 and 2, respectively). (Polyhistidine tags present in the H-FTN-GAD65S nanoparticles were not shown to avoid too much congestion of symbols.)

Human ferritin is a class of iron storage proteins comprising 24 subunits of heavy and light chains that self-assemble to form a hollow shell structure 12 nm in diameter. It was previously reported (20 , 21) that 24 subunits of hFTN-H alone can form stable ferritin particles with an iron storage function like native one. hFTN-H monomer consists of a bundle of four antiparallel {alpha}-helices (A–D). Helices B and C are connected by a long loop traversing the whole bundle and making contacts with the A helix. The carboxy-terminal amino acids are arranged in a fifth shorter {alpha}-helix (helix E) that forms an acute angle with the main axis of the molecule (20 21 22) . The 24 subunits of hFTN-H assemble into a stable protein shell with 4–3-2 symmetry, leaving iron channels along the 3- and 4-fold axes (20 21 22) . It was proven that the carboxy-terminus, including the E helix, is neither essential for proper folding of the monomer subunit nor for assembly of the 24 mer (20) . Furthermore, the heavy chain of human ferritin can exist in two alternative conformations that differ in the position of the E helix in the 4-fold axis: "flip" conformation with the E helix pointing inside the ferritin shell and "flop" conformation with the E helix being extruded outside, while normally the native ferritin shell is constructed with the flip conformation. Luzzago and Cesareni (20) and Levi et al. (21) observed that the fusion of {alpha}-peptide of ß-galactosidase to C-terminus of hFTN-H also resulted in the formation of functional ferritin particles with the flop conformation, which was probably because the {alpha}-peptide was too large to be packed into the cage. They suggested that on the assembly of ferritin-peptide fusion molecules, the decision of whether to flip or to flop conformation is taken depending on whether the empty volume inside the shell is sufficient to contain the 24 peptides. The fusion of GAD65-C138 and S-HBsAg-MHR peptide to the C-terminus of hFTN-H successfully formed the ferritin-like nanoparticles (Fig. 1A ), and the mobility of both the nanoparticles was exactly the same as that of hFTN-H particles when electrophoresed on a native PAGE gel (Fig. 1C ), indicating that the H-FTN-GAD65S and H-FTN-SHBsAg nanoparticles must have the same conformation as hFTN-H particles that are formed by the 24-subunit assembly. Like the case of the {alpha}-peptide fusion to the ferritin C-terminus (20) , it seems that the number of antigenic epitopes (GAD65-C138 and S-HBsAg-MHR) per nanoparticle must be 24. Also, since the use of H-FTN-GAD65S and H-FTN-SHBsAg nanoparticles were very successful on the ultrasensitive diagnosis of disease marker as discussed later in this article, it is strongly suggested that the H-FTN-GAD65S and H-FTN-SHBsAg nanoparticles should have the flop conformation with the antigenic epitopes pointing outside.

Ultrasensitive detection of Type I diabetes marker
Two methods of detecting the autoantibodies were developed (Fig. 1D ). The first method uses quantum dot (CdSe-Qdot)-secondary antibody (2°Ab) conjugates that can specifically bind to the target autoantibodies that are captured by the H-FTN-GAD65S (or H-FTN-GAD65L) supramolecules immobilized on a polyvinylidene difluoride (PVDF) membrane. With the presence of UV light, sensitive detection is possible through photoluminescence emitted from (CdSe-Qdot)-2°Ab conjugates. The second method is PCR-based detection using both H-FTN-GAD65S (or H-FTN-GAD65L) and K-FTN-GAD65S nanoparticles. The autoantibodies in the sample are captured first by the H-FTN-GAD65S (or H-FTN-GAD65L) nanoparticles immobilized on a designated spot of the PVDF membrane and then sandwiched by the K-FTN-GAD65S. Since each K-FTN-GAD65S nanoparticle contains multiple polylysine tags, it can capture the single-stranded DNA primer, which corresponds to one of two primers required for PCR-amplification of the specific gene. A single and apparent DNA band observed through gel electrophoresis of the PCR product (see Materials and Methods) represents the existence of autoantibodies in the sample.

Qdot-based detection
The following autoantibody solutions were prepared: 100 pg, 100 fg, 1 fg, and 500 ag of mouse anti-GAD65 monoclonal antibodies (GAD6 recognizing C-terminal linear epitope) per ml of PBS. Clear photoluminescence was detected even at the autoantibody concentration of 500 ag ml–1 (~3 attomolar concentration) (Fig. 2 A, B). The 3 attomolar sensitivity was successfully reproduced even when the autoantibodies were present in human plasma solution (Fig. 2A , B). Photoluminescence significantly decreased when the autoantibody concentration was reduced to 100 ag ml–1 (data not shown), indicating a sensitivity limit between 100 and 500 ag ml–1. Photoluminescence was not observed, however, when only (Qdot-CdSe)-2°Ab conjugates were applied to the immobilized H-FTN-GAD65S nanoparticles without the autoantibodies (Fig. 2C ), implying that non-specific binding between the H-FTN-GAD65S nanoparticles and (Qdot-CdSe)-2°Ab was negligible.


Figure 2
View larger version (31K):
[in this window]
[in a new window]

  		Figure 2. Qdot-based detection of IDDM marker. A) Results of Qdot-based detection of anti-GAD65 autoantibodies in PBS and human plasma solution. B) Results of intensity measurement of photoluminescence shown in Fig. 2A . [Photoluminescence unit is defined as relative fluorescence unit (RFU) as per the definition of GENios (Tecan, Austria).] C) Result of a controlled experiment that applied (Qdot-CdSe)-2°Ab conjugates to H-FTN-GAD65S nanoparticles without the addition of the anti-GAD65 autoantibodies (a negative control of Qdot-based detection).

PCR-based detection
The autoantibodies at the 3 attomolar concentration were also obviously detected (lane 1 in Fig. 3 A). As shown in lane 2 (negative control 1) of Fig. 3A , no PCR product was formed when the autoantibodies were sandwiched between the H-FTN-GAD65S and polylysine-free FTN-GAD65S nanoparticles, which indicates the absence of a single-stranded DNA primer on the PCR amplification, thereby implying that the binding of the DNA primer to the K-FTN-GAD65S is very specific. In another controlled experiment [Fig. 3B (negative control 2)], an autoantibody-free human plasma solution was applied to the H-FTN-GAD65S nanoparticles immobilized on PVDF, and the K-FTN-GAD65S were subsequently added to it, followed by the addition of single-stranded DNA primer. The failure in the PCR amplification (lane 1 of Fig. 3B ) illustrates that various human proteins, including antibodies in human plasma, nonspecifically bind neither to H-FTN-GAD65S nor K-FTN-GAD65S nanoparticles. Therefore, the PCR product formation shown in lane 1 of Fig. 3A conclusively explains that both K-FTN-GAD65S and H-FTN-GAD65S nanoparticles very specifically bind to autoantibodies even in the human plasma solution, resulting in the 3-attomolar sensitivity.


Figure 3
View larger version (19K):
[in this window]
[in a new window]

  		Figure 3. PCR-based detection of IDDM marker. A) Lane 1: results of PCR-based detection of anti-GAD65 autoantibodies in PBS and human plasma solution; Lane 2: results of controlled experiments that used FTN-GAD65S nanoparticles instead of K-FTN-GAD65S nanoparticles (negative control 1 of PCR-based detection). [M: DNA markers (1.0 and 0.5 kb)]. B) Result of a controlled experiment that did not add the anti-GAD65 autoantibodies (negative control 2 of PCR-based detection). [M: DNA markers (1.0 and 0.5 kb)]. (Particles A, B, and C: H-FTN-GAD65S, K-FTN-GAD65S, and FTN-GAD65S nanoparticles, respectively).

Reproducibility of attomolar sensitivity
To prove that the attomolar sensitivity is reproducible on the detection of other antibody analytes, another nanoparticle probe (H-FTN-SHBsAg) to detect anti-S-HBsAg monoclonal antibodies was also designed and produced using the same recombinant E. coli (Fig. 1) . From Fig. 4 , the anti-S-HBsAg mAb were evidently detected with the same attomolar sensitivity by using both Qdot- and PCR-based methods. Accordingly, it seems that this probe system based on supramolecular protein nanoparticles could be applied to the detection of almost any target antibody with a known protein antigen. Also, the simple replacement of epitope-encoding ORF within a chimera gene could make it possible to easily prepare different kinds of supramolecular nanoparticle probes using the same bacterial expression system, showing another possible advantage of the supramolecular protein nanoprobe system.


Figure 4
View larger version (33K):
[in this window]
[in a new window]

  		Figure 4. Qdot- and PCR-based detection of anti-S-HBsAg antibodies. A) Results of Qdot-based detection of anti-S-HBsAg mAb in PBS and human plasma solution. (The detection method is the same as that of Fig. 2A .). B) Results of intensity measurement of photoluminescence shown in Fig. 4A . (Definition of photoluminescence unit is the same as that of Fig. 2B .) C) Lane 1: results of PCR-based detection of anti-S-HBsAg mAb in PBS and human plasma solution; Lane 2: results of controlled experiments that used FTN-SHBsAg nanoparticles instead of K-FTN-SHBsAg nanoparticles in the PCR-based detection (negative control of PCR-based detection). [M: DNA markers (1.0 and 0.5 kb)] (The detection method is the same as that of Fig. 3A .)

Simultaneous and multiplexing-mode detection of two different antibody analytes
Using the Qdot-based detection method, we investigated if simultaneous detection of two different antibody markers (i.e., anti-GAD65 and anti-S-HBsAg monoclonal antibodies at 500 ag ml–1 in phosphate buffer) is possible. When the analyte solution containing only one antibody marker (either anti-GAD65 or anti-S-HBsAg mAb) was applied to both the H-FTN-GAD65S and H-FTN-SHBsAg nanoparticles immobilized on PVDF membrane, only a single spot on the membrane emits the photoluminescence (Fig. 5 ), which implies that the antibodies bind the nanoparticles with high specificity. When the same PVDF membrane (where both the H-FTN-GAD65S and H-FTN-SHBsAg nanoparticles were immobilized at different spots) was used to capture both anti-GAD65 and anti-S-HBsAg monoclonal antibodies contained in an analyte solution (Fig. 5) , the photoluminescence emission was clearly shown at two different spots, thereby indicating that the simultaneous detection of the two different antibody analytes were successful and that multiplexing mode detection of multiple antibody analytes/markers could be feasible.


Figure 5
View larger version (22K):
[in this window]
[in a new window]

  		Figure 5. Results of simultaneous and multiplexing-mode detection of autoantibody marker of Type I diabetes and monoclonal antibodies against S-HBsAg. (The detection was implemented using Qdot-based method.)

Detection sensitivity of nonsupramolecular protein probe
Another plasmid vector was constructed for the expression of the gene encoding NH2-(His)6::(GAD65-C138)-COOH (named H-GAD65S), by deleting the hFTN-H gene from the chimera gene encoding H-FTN-GAD65S. The gene expression formed inclusion bodies of H-GAD65S in the bacterial cytoplasm, which were then recovered, solubilized, refolded, and purified through Ni+2-affinity chromatography. On the immobilization on a spot area on the PVDF membrane, the moles of H-GAD65S were maintained the same as those of immobilized H-FTN-GAD65S. That is, the number of autoantibody-binding epitopes in the immobilized protein probes was the same in both cases. On the Qdot-based detection (Fig. 6 A), the autoantibodies at 100 ng ml–1 were obviously detected, whereas the autoantibodies below the 50 ng ml–1 were not detected at all. Although the detection limit was slightly extended to 50 ng ml–1 when PCR-based detection was applied (Fig. 6B ), the autoantibodies below 1 ng ml–1 were not detected at all. Consequently, the autoantibody detection using the nonsupramolecular H-GAD65S probe was more than 6 orders of magnitude less sensitive than the supramolecular protein.


Figure 6
View larger version (22K):
[in this window]
[in a new window]

  		Figure 6. IDDM marker detection using nonsupramolecular protein probe. A) Results of Qdot-based detection of anti-GAD65 autoantibodies in PBS buffer using nonsupramolecular H-GAD65S as probes. B) Results of PCR-based detection of anti-GAD65 autoantibodies in PBS buffer using nonsupramolecular H-GAD65S as probes. [M: DNA markers (1.0 and 0.5 kb).

Detection of natural autoantibodies in Type I diabetic patients’ sera
Although the middle and C-terminal domains in GAD65 were defined as dominant epitope, serum autoantibodies of Type I diabetics consist of various MICAs, each of which recognizes a specific linear or conformational eptiope in GAD65. Richter et al. (15) examined five MICAs 1, 2, 3, 4, and 6 derived from a newly diagnosed diabetic patient to probe the autoimmune epitopes in GAD65. They showed that only MICA2 recognized a linear epitope close to the C terminus, while the other MICAs recognized conformational epitopes in the GAD65 molecule. Söhnlein et al. (17) also reported that linear epitopes were detected in only 7% sera of patients with Type I diabetes. According to the report of Richter et al. (15) , the deletion mutant GAD65 ({Delta}1–244) was detected by all the five MICAs, and hence the region from Ile245 to Leu585 of GAD65 was defined as a conformational epitope. In addition to the H-FTN-GAD65S nanoparticles containing the linear epitopes GAD65-C138, we prepared another type of nanoparticles H-FTN-GAD65L containing conformational epitopes GAD65-C341, i.e., the C-terminal 341 amino acids (Ile245 to Leu585). In this section, both types of nanoparticles were applied to the detection of serum autoantibodies of Type I diabetic patients.

With the help of pediatric department of Korea University Hospital, around 1 ml of serum was kindly donated by each of six patients (#1: female/age 19; #2: female/age 15; #3: male/age 11; #4: female/age 14; #5: male/age 12; #6: female/age 11) suffering from Type I diabetes. First, 250 µl serum from each patient was applied to the immobilized H-FTN-GAD65S nanoparticles for the PCR-based detection of autoantibodies. Only the serum of patient #2 was shown to have MICAs recognizing the C-terminal linear epitope (Fig. 7 A), while the sera of the other five patients and healthy person (male/age 45) never showed any autoimmune reactivity. This result is not surprising because reportedly the linear epitopes were detected in only 7% sera of patients with Type I diabetes. Using the same H-FTN-GAD65S nanoparticles, the autoimmune reactivity of another 250 µl serum of each patient was analyzed using the Qdot-based detection method, and it is worthy of note that the linear epitope (GAD65-C138) was also detected by the same serum of patient #2 (Fig. 7B ), thereby indicating high selectivity/specificity of this protein nanoparticle-based diagnosis. The supramolecular nanoparticles H-FTN-GAD65L containing the conformational epitopes (GAD65-C341) were also used to analyze autoimmune reactivity of the above six patients’ sera. As presented in Fig. 7C , the conformational epitope GAD65-C341 was detected by all the patient sera, which corresponds to the previous reports of Söhnlein et al. (17) and Richter et al. (15) regarding the epitope recognition of Type I diabetic sera.


Figure 7
View larger version (35K):
[in this window]
[in a new window]

  		Figure 7. Detection of natural autoantibodies in sera from Type I diabetes patients. A) PCR-based detection using nanoparticles (H-FTN-GAD65S and K-FTN-GAD65S) with linear epitopes (GAD65C-138). B) Qdot-based detection using nanoparticles (H-FTN-GAD65S) with linear epitopes (GAD65C-138). C) Qdot-based detection using nanoparticles (H-FTN-GAD65L) with conformational epitopes (GAD65C-341).


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
We emphasize here that this is the first time that supramolecular protein nanoparticles were applied to the ultrasensitive detection of antibody analytes. This detection system using supramolecular protein nanoparticles as capture probes offers several advantages over the current probe system. First, each spherical nanoparticle with a nanoscale diameter carries multiple DNA- and/or antibody-capture probes, thereby producing a high ratio of probes to binding analytes (oligonucleotide DNA or antibody markers in this case), which can substantially increase assay sensitivity. Second, the conformational orientation of target-binding probes (i.e., antigenic epitopes) is homogeneous. On the intracellular supramolecule synthesis by the self-assembly function of hFTN-H, the antigenic epitopes (GAD65-C138, GAD65-C341, or MHR of adr subtype of S-HBsAg) are subjected to localization on the nanoparticles with a homogeneous conformation. It is almost impossible to maintain a constant conformational orientation when protein probes are chemically immobilized on a solid surface, and thus the quite heterogeneous orientation of immobilized probes is known to critically reduce detection sensitivity. Third, the bacterial expression system provides protein nanoparticles with marked size uniformity. Owing to the self-assembly property of hFTN-H, the supramolecular size is kept in the range of 10–15 nm in a bacterial cytoplasm. This uniform size distribution allows all the nanoparticles, immobilized within microscale pores of PVDF membrane, to equally participate in the analyte capture and accordingly contributes to the significant enhancement of sensitivity. In addition, the assembled nanoparticles may enhance the probe stability. Using the same E. coli expression system, we synthesized the supramolecular nanoparticles of the green fluorescence protein (GFP) that showed a significantly enhanced functional stability even under extreme environments causing the denaturation of free GFP molecules (data not shown).

In this study, we showed the attomolar sensitivity for the two different target antibodies: 1) autoantibodies of Type I diabetes and 2) anti-S-HBsAg monoclonal antibodies. With these antibody analytes, we confirmed that the attomolar ultrasensitivity was evidently reproducible and also demonstrated the feasibility of multiplexing mode detection. Finally, natural autoantibodies in the sera from Type I diabetic patients were successfully detected by the nanoprobe system based on supramolecular protein nanoparticles. This approach seems to be general for almost any target marker with a known protein antigen, although the ultrasensitive detection of autoantibodies of Type I diabetes and anti-S-HBsAg monoclonal antibodies are demonstrated as proof of the concept in this case. This supramolecular protein nanoprobe system would be also useful in the early diagnosis of infectious diseases (e.g., AIDS, Hepatitis C, etc.) during the early phase of the infection when the antibody titer is too low to be detected.


   ACKNOWLEDGMENTS
 
This study was supported by the Korea Health 21 R&D Project of the Ministry of Health and Welfare of the Republic of Korea (grant no. A050750). This work was also supported by the Second Brain Korea 21 Project. Further supports from the Korea Science and Engineering Foundation (grant no. R01-2005-000-10355-0), the Korea Research Foundation (grant no. KRF-2004-005-D00057), and the Microbial Genomics and Applications Center (Taejon, Republic of Korea) are also appreciated.


   FOOTNOTES
 
1 These authors contributed equally to this work.

2 Current address: Mogam Biotechnology Research Institute, Yongin 449-799, South Korea.

Received for publication September 24, 2006. Accepted for publication December 25, 2006.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Rosi, N. L., Mirkin, C. A. (2005) Nanostructures in biodiagnostics. Chem. Rev. 105,1547-1562[CrossRef][Medline]
  2. Baekkeskov, S., Aanstoot, H. J., Chistgau, S., Reetz, A., Solimena, M., Cascalho, M., De Camilli, P. (1990) Identification of the 64K autoantigen in insulin-dependent diabetes as the GABA-synthesizing enzyme glutamic acid decarboxylase. Nature 347,151-156[CrossRef][Medline]
  3. Kaufman, D. L., Clare-Salzler, M., Tian, J., Forsthuber, T., Ting, G. S., Robinson, P., Atkinson, M. A., Sercarz, E. E., Tobin, A. J., Lenhmann, P. V. (1993) Spontaneous loss of T-cell tolerance to glutamic acid decarboxylase in murine insulin-dependent diabetes. Nature 366,69-72[CrossRef][Medline]
  4. Solimena, M., Folli, F., Denis-Donini, S., Comi, G. C., Pozza, G., Camilli, D. P., Vicari, A. M. (1998) Autoantibodies to glutamic acid decarboxylase in a patient with Stiff-Man Syndrome, epilepsy, and type 1 diabetes mellitus. New Engl. J. Med. 318,1012-1020
  5. Notkins, A. L., Lernmark, A. (2001) Autoimmune type 1 diabetes: resolved and unresolved issues. J. Clin. Invest 108,1247-1252[CrossRef][Medline]
  6. Hagopian, W. A., Michelsen, B., Karlsen, A. E., Larsen, F., Moody, A., Grubin, C. E., Rowe, R., Peterson, J., Mcevoy, R., Lermark, A. (1993) Autoantibodies in IDDM primarily recognize the 65000-Mr reather than 67000-Mr isoform of glutamic acid decarboxylase. Diabetes 42,631-636[Abstract]
  7. Ujihara, N., Daw, K., Gianani, R., Boel, E., Yu, L., Powers, A. C. (1994) Identification of glutamic acid decarboxylase autoantibody heterogeneity and epitope region in type 1 diabetes. Diabetes 43,968-975[Abstract]
  8. Peterson, J. S., Hejns, K. R., Moody, A., Karlsen, A. E., Marshall, M. O., Hier-madsen, M., Boel, E., Michelsen, B. K., Dryberg, T. (1995) Detection of GAD65 antibodies in diabetes and other autoimmune diseases using a simple radioligand assay. Diabetes 43,459-467[CrossRef]
  9. Yoon, J. W., Yoon, C. S., Lim, H. W., Huang, Q. Q., Kang, Y., Pyun, K. H., Hirasawa, K., Sherwin, R. S., Jun, H. S. (1999) Control of autoimmune diabetes in NOD mice by GAD expression or suppression in beta-cells. Science 284,1183-1187[Abstract/Free Full Text]
  10. Tuomi, T., Rowley, M. J., Knowles, W. J., Chen, Q. Y., McAnally, T., Zimmet, P. Z., Mackay, I. R. (1994) Autoantigenic properties of native and denatured glutamic acid decarboxylas: evidence for a conformational epitope. Clin. Immunol. Immunopathol. 71,53-59[CrossRef][Medline]
  11. Ahn, J. Y., Choi, H., Kim, Y. H., Han, K. Y., Park, J. S., Han, S. S., Lee, J. (2005) Heterologous gene expression using self-assembled supramolecules with high affinity for HSP70 chaperone. Nucleic Acids Res. 33,3751-3762[Abstract/Free Full Text]
  12. Wang, J., Polsky, R., Merkoci, A., Turner, K. L. (2004) DNA-based amplified bioelectronic detection and coding of proteins. Angew. Chem. Int. Ed. 43,2158-2161[CrossRef]
  13. Kim, J., Richter, W., Aanstoot, H. J., Shi, Y., Fu, Q., Rajotte, F., Wancok, G., Baekkeskov, S. (1993) Differential expression of GAD65 and GAD67 in human, rat, and mouse pancreatic islet. Diabetes 42,1799-1808[Abstract]
  14. Richter, W., Endl, J., Eiermann, T. H., Brandt, M., Kientsch-Engel, R., Thivolet, C., Jungfer, H., Scherbaum, W. A. (1992) Human monoclonal islet cell antibodies from a patient with insulin-dependent diabetes mellitus reveal glutamate decarboxylase as the target antigen. Proc. Natl. Acad. Sci. U. S. A. 89,8467-8471[Abstract/Free Full Text]
  15. Richter, W., Shi, Y., Baekkeskov, S. (1993) Autoreactive epitopes defined by diabetes-associated human monoclonal antibodies are localized in the middle and C-terminal domains of the smaller form of glutamate decarboxylase. Proc. Natl. Acad. Sci. U. S. A. 90,2832-2836[Abstract/Free Full Text]
  16. Bu, D. F., Erlander, M. G., Hitz, B. C., Tillakaratne, N. J., Kaufman, D. L., Wagner-McPherson, C. B., Evans, G. A., Tobin, A. J. (1992) Two human glutamate decarboxylase, 65-kDA GAD and 67-kDA GAD, are each encoded by a single gene. Proc. Natl. Acad. Sci. U. S. A. 89,2115-2119[Abstract/Free Full Text]
  17. Söhnlein, P., Muller, M., Syren, K., Hartmann, U., Bohm, B. O., Meinck, H. M., Knip, M., Akerblom, H. K. (2000) Epitope spreading and a varying but not disease-specific GAD65 antibody response in type 1 diabetes. Diabetologia 43,210-217[CrossRef][Medline]
  18. Bonifacio, E., Lampasona, V., Bernasconi, L., Ziegler, A. G. (2000) Maturation of the humoural autoimmune response to epitopes of GAD in preclinical childhood type 1 diabetes. Diabetes 49,202-208[Abstract]
  19. Daw, K., Ujihara, N., Power, M. A. (1996) Glutamic acid decarboxylase autoantibodies in Stiff-Man Syndrome and insulin-dependent diabetes mellitus exhibit similarities and differences in epitope recognition. J. Immunol. 156,818-825[Abstract]
  20. Luzzago, A., Cesareni, G. (1989) Isolation of point mutations that affect the folding of the H chain of human ferritin in E. coli. EMBO J. 8,569-576[Medline]
  21. Levi, S., Luzzago, A., Franceschinelli, F., Santambrogio, P., Cesareni, G., Arosio, P. (1989) Mutational analysis of the channel and loop sequences of human ferritin H-chain. Biochem. J. 264,381-388[Medline]
  22. Santambrogio, P., Pinto, P., Levi, S., Cozzi, A., Rovida, E., Albertini, A., Artymiuk, P., Harrison, P. M., Arosio, P. (1997) Effects of modifications near the 2-, 3- and 4-fold symmetry axes on human ferritin renaturation. Biochem. J. 322,461-468[Medline]




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
fj.06-7303comv1
21/7/1324    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Lee, S.-H.
Right arrow Articles by Lee, J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Lee, S.-H.
Right arrow Articles by Lee, J.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS