A quick and easy method to detect the GDF-15 protein (Growth Differentiation Factor-15) has been developed, which utilize the magnetic response of magnetic nanoparticles by switching magnetic fields. The magnetic nanoparticles and GDF-15 are bound by an antigen-antibody reaction and aggregated into a spherical shape using a needle-shaped magnetic yoke. The density of GDF-15 changed as follows: 0, 1 ng/ml, 10 ng/ml, and 100 ng/ml. The increase of GDF-15 aggregated the magnetic nanoparticles and enhanced the signal-to-noise ratio. We also tried the sandwich-type bound method using a primary and secondary antibody with additional magnetic nanoparticles and obtained the enhancement of the magnetic signal in a lower concentration (under 10 ng/ml) of GDF-15. The cross-bridges between magnetic nanoparticle and the protein may strengthen the magnetic couplings of nanoparticles.

There is substantial demand for the rapid detection of proteins which can act as biomarkers for diagnosing metastatic cancer,1–3 mitochondrial disease,4–6 hypertension, etc. Western Blotting,7 the Enzyme-Linked Immunosorbent Assay (ELISA),8 and Mass Spectrometry9 usually carry out these protein detections. However, these methods require large and expensive equipment, as well as highly skilled technicians who can operate them in a laboratory over a long period. Therefore, we aimed to develop a fast, simple, low-cost, highly accurate protein detector. Magnetic nanoparticles are useful in biomedical applications, as magnetic fields can easily be manipulated remotely. For this reason, many magnetic methods have been studied to detect magnetic nanoparticles and evaluate antigen-antibody reactions using various magnetic sensors.10–23 

In a previous study, we developed a detection system of oral bacteria based on a magnetic immunoassay using a magnetoimpedance (MI) sensor.24,25 However, the system could not be used to detect several samples simultaneously. On the other hand, in our research, the sample did not conjugate with the MI sensor. Our system differs from the conventional detection method, which currently use a GMI sensor12–20 at this point. Later, we succeeded in detecting bacteria (Fusobacterium Nucleatum, etc.) using the magnetic properties of antigen-antibody reaction aggregates26 and obtained a high correlation with a Polymerase Chain Reaction (PCR).27 

Now, we are trying to detect proteins using the magnetic response from the aggregates, including magnetic nanoparticles and an antigen (GDF-15). GDF-15 (Growth Differentiation Factor-15, a stress response cytokine) is a well-known biomarker used for early identification of mitochondrial diseases.4–6 Mitochondrial diseases reduce cellular activity by reducing mitochondria function and are designated as intractable diseases. Conventionally, this protein is evaluated using PCR, ELISA, etc., but there is a strong need for a faster and simpler evaluation. However, since the proteins are far smaller than bacteria, seeing differential magnetic signals with and without the protein is more challenging.

In the present study, we have demonstrated a clear tendency for specific detection of the concentration of GDF-15. We have found a new phenomenon wherein magnetic nanoparticles aggregate as the protein concentration increases, leading to an enhanced magnetic signal. The cross-bridges between magnetic nanoparticles and proteins may decrease the distance between magnetic nanoparticles, thereby enhancing the magnetic coupling. In the case of bacteria, the size is around micrometers, which is far larger than proteins, so when the amount of antigen increased, the magnetic interactions between magnetic nanoparticles decreased.26 The results in this protein detection exhibit a tendency opposite to that of antigen-antibody reactions in bacteria, revealing a new mechanism. We have also developed a protein detector for six samples. An advantage of the developed system is that the protein concentration is measured and evaluated without post-treatment, such as the washing process after the antigen-antibody reaction, and this allows for rapid detection. The results of this research can be applied for detecting many proteins other than GDF-15.

Figure 1 shows the schematic diagram of the developed system. The hardware configuration is based on a previous paper24,25 and is expanded for six sample evaluations simultaneously, which results in applicability for many samples and saves measuring time. By evaluating six samples simultaneously, we are now able to evaluate a large number of samples in a shorter period of time. This method is also thought to be effective in evaluating reproducibility and improving the signal-to-noise ratio. It consists of two MI sensors (developed by JNS Co. Ltd.), six NdFeB magnets (5 × 4 × 3 mm3), six drive coils (3000 turns), six power supplies, and a rotor. Figure 2 shows a flowchart illustrating the process for evaluating the magnetic signal of protein aggregates using the prototype device shown in Fig. 1. Firstly, the samples were magnetized by the NdFeB magnets with a surface magnetic flux density of 21 mT for about 300 s. Subsequently, the aggregated nanoparticles were magnetized above the needle-type NiFe yokes for 300 s. After the magnetization and aggregation processes mentioned above, the Hdc was changed from 0 mT to 1.56 mT with a step interval of 0.06 mT in the upward and downward direction (i.e., Hdc was +0.06 mT, −0.06 mT, …, +1.56 mT, and −1.56 mT). Each Hdc was applied for 30 s, and the magnetic moments of nanoparticles tended to align in the direction of the applied magnetic field. The sample was then rotated slowly, close to the MI sensor, with a rotation speed of 100 deg/s. The sampling rate of the sensor was 50 kHz. The distance between the microtube and the sensor was around 200 µm.

FIG. 1.

Schematic (bottom) and photo (top) of a laboratory-made system for detecting protein in a liquid phase using magnetic nanoparticles.

FIG. 1.

Schematic (bottom) and photo (top) of a laboratory-made system for detecting protein in a liquid phase using magnetic nanoparticles.

Close modal
FIG. 2.

Flowchart.

1. Experiment 1 antigen-antibody reaction

Four samples were prepared by combining magnetic nanoparticles and antibodies. Magnetic nanoparticles, nanomag®-D (250 nmϕ) coated with protein A (micromod Partikeltechnologie GmbH), were prepared at a concentration of 50 μg/ml in PBS-T and then dispersed using an ultrasonic cleaner for 1 min. Antibodies were prepared at 4 μg/ml in PBS-T using a rabbit monoclonal antibody (abcam ab206414). The antibody was then mixed in equal volumes with the magnetic nanoparticle solution and allowed to react in a rotary mixer for 30 min at room temperature. The magnetic nanoparticles were then resuspended in PBS-T and mixed in equal volumes with a GDF-15 solution (containing 0, 1, 10, and 100 ng/ml in PBS-T) for subsequent measurements, following the same procedure as described above.

2. Experiment 2—The sandwich method

Magnetic nanoparticles with a size of 250 nmϕ were employed, and a rabbit monoclonal antibody was used as the primary antibody. Similar to Experimet 1, these magnetic nanoparticles were conjugated with GDF-15 (0, 1, 10, 100 ng/ml in PBS-T). Pre-conjugating rabbit polyclonal antibodies conducted a sandwich assay as the secondary antibody to magnetic nanoparticle, specifically nanomag®-D (50 nmϕ, coated with protein A). The secondary reaction’s magnetic nanoparticle/antibody concentration (μg/ml) ratio was set at 250/50, 100/20, 50/10, 50/7, and 50/4, and the optimal conditions were investigated.

The output waveforms and an optical micrograph of magnetic nanoparticle/antigen aggregates at a diameter of 250 nm after the measurement are presented in Fig. 3. The magnetic signal waveforms indicated that the blue series represents odd-numbered outputs. At the same time, the yellow-to-red series represents even-numbered outputs. The waveform’s color is shown as changing from blue to yellowish-green for odd rotations and from red to yellow for even rotations as the count increases. It is observed that the signal magnitude and variations are small during odd rotations, but they progressively increase with each even rotation. Furthermore, in the case of monoclonal antibodies, the waveform magnitude tends to increase with higher GDF-15 concentrations during even rotations as shown in Fig. 3. As described in Fig. 2, a switched magnetic field is applied to the sample every odd and even rotation, and the reversed magnetization was measured with a sensor. The larger the magnetic signal of the sample after an even number of rotations was, the greater was the magnetized reversal caused by the switched magnetic field, which indicated that the magnetic signal of the aggregate increased. Therefore, the amplitude of the output voltage corresponds to the strength of the magnetic signal (magnetization reversal).

FIG. 3.

Output waveform and optical micrographs of magnetic nanoparticle/antigen aggregates.

FIG. 3.

Output waveform and optical micrographs of magnetic nanoparticle/antigen aggregates.

Close modal
As shown in the photographs of Fig. 3, the brown area of aggregates expanded as the amount of antigen (GDF-15) increased. The intensity of the brown color in Fig. 3 was determined from the optical microscope image using Eq. (1).28 
Brown=0.45Red+0.31Green+0.19Blue
(1)

According to the graph, it can be seen that the higher the amount of antigen was, the darker the brown color was. On the other hand, the magnetic signal from the aggregates also increased with the increase of the antigen, which suggested that the aggregation of the magnetic nanoparticles progress with the rise of the protein. This may be because the cross-linking between magnetic nanoparticles and proteins reduces the distance between magnetic nanoparticles and enhances magnetic coupling.

Figure 4 shows the amplitude of the output voltage for each rotation when the monoclonal antibody was used. As the magnetic field increased, the output signal tended to become stronger at higher GDF-15 concentrations of 10 and 100 ng/ml. The amplitude of 100 ng/ml and 10 ng/ml became larger in that order, suggesting quantitativeness to a certain extent. However, the plots for 0 and 1 ng/ml were overlapped and could not be distinguished even if the magnetic field increased. Considering practicality, a measurement system at least at the ng/ml level was considered necessary, and an attempt was made to improve the sample preparation method. The increase in magnetization reversal in response to an increase in the magnetic field has been reproduced in multiple evaluations, and the error bars are small, so it is not a drift.

FIG. 4.

Output voltage as a function of rotating number and magnetic field in each concentration of GDF-15.

FIG. 4.

Output voltage as a function of rotating number and magnetic field in each concentration of GDF-15.

Close modal

We attempted the sandwich assay to enable the measurement of low concentrations of GDF-15. The examination of the magnetic nanoparticle/antibody concentration ratio in the secondary reaction revealed a significant increase [Fig. 5(a)]. At ratios of 250/50, 50/7, and 50/4, it was possible to distinguish between 0 ng/ml and 1 ng/ml, which could not be discerned using the primary antibody alone. Even at the same magnetic nanoparticle concentration of 50 μg/ml, the results varied significantly depending on the antibody concentration, highlighting the importance of the ratio. Considering the cost savings and the clear distinctions achieved, the conditions tested in this study suggest that a ratio of 50/7 is optimal. Figure 5(b) shows the output voltage when the sandwich method was applied as a function of rotating number and magnetic field in each concentration of the GDF-15. Changes at low concentrations (0–1 ng/ml) increased compared to experiments using only the primary antibody. Figure 6 shows the comparison of two procedures as a function of the concentration of the GDF-15. Monoclonal antibody methods showed significant changes at high concentrations (more than 1 ng/ml). On the other hand, in the sandwich method, the change was significant at low concentrations (below 10 ng/ml), and the signal intensity increased because more magnetic nanoparticles could be integrated. Figure 7(a) shows the antigen (GDF-15) aggregates with the magnetic nanoparticles (250 nmϕ) and the primary antibody after the antigen-antibody reaction. The antigen and the magnetic nanoparticles may have formed a cross-bridge, the distance between the magnetic nanoparticles became closer, and the magnetostatic coupling increased. In addition to Figs. 7(a) and 7(b) added secondary antibodies and magnetic nanoparticles (50 nmϕ). Because the aggregates also contained small-sized magnetic nanoparticles, the magnetic signal may increase in lower antigen concentrations as shown in Fig. 6.

FIG. 5.

Results of sandwich type methods. (a) Effect of the magnetic nanoparticle/antibody concentration in a secondary reaction on the output voltage in each concentration of GDF-15 when the MNPs/antibody (μg/ml) was 50/7. (b) Output voltage when the sandwich method was applied as a function of the rotating number and magnetic field in each concentration of GDF-15.

FIG. 5.

Results of sandwich type methods. (a) Effect of the magnetic nanoparticle/antibody concentration in a secondary reaction on the output voltage in each concentration of GDF-15 when the MNPs/antibody (μg/ml) was 50/7. (b) Output voltage when the sandwich method was applied as a function of the rotating number and magnetic field in each concentration of GDF-15.

Close modal
FIG. 6.

Comparison of each method as a function of the concentration of the GDF-15.

FIG. 6.

Comparison of each method as a function of the concentration of the GDF-15.

Close modal
FIG. 7.

Diagram of the magnetic nanoparticles and aggregations with and without GDF-15 (antigen). The cross bridges may reduce the distance between magnetic nanoparticles and strengthen the magnetostatic interaction. (a) antigen-antibody reaction. (b) sandwich type method.

FIG. 7.

Diagram of the magnetic nanoparticles and aggregations with and without GDF-15 (antigen). The cross bridges may reduce the distance between magnetic nanoparticles and strengthen the magnetostatic interaction. (a) antigen-antibody reaction. (b) sandwich type method.

Close modal

An attempt to detect bacteria using the sandwich method with magnetic nanoparticles has been reported.29 However, measuring protein or bacteria using this magnetic response is novel. This study utilized protein A-coated magnetic nanoparticles for primary and secondary reactions.

We have developed a protein detection system which can detect six-samples simultaneously in the liquid phase for point-of-care diagnostics based on magnetic immunoassay. We analyzed the magnetic response of an antigen-antibody aggregation by switching the magnetic field. We found a clear tendency of the concentration of the GDF-15 from 0 ng/ml to 100 ng/ml. The cross-bridge between the magnetic nanoparticles and the antigen (GDF-15) may promote the aggregation of magnetic nanoparticles. The sandwich method using primary and secondary antibodies increased the magnetic signal at several ng/ml of low antigen concentrations.

The authors would like to thank Prof. Kazutaka Murayama and Prof. Yoshikazu Tanaka of Tohoku University for their advice on the consideration of magnetic nanoparticles and protein aggregates. This work was supported in part by the Adaptable and Seamless Technology Transfer Program (A-STEP JPMJTM22AB) from the Japan Science and Technology Agency (JST), Japan Agency for Medical Research and Development (AMED) (JP21zf0127001), and the TERUMO LIFE SCIENCE FOUNDATION (22-Ⅱ1011).

The authors have no conflicts to disclose.

Shin Yabukami: Conceptualization (equal); Data curation (equal); Formal analysis (equal); Funding acquisition (equal); Investigation (equal); Methodology (equal); Project administration (equal); Resources (equal); Software (equal); Supervision (equal); Validation (equal); Visualization (equal); Writing – original draft (equal); Writing – review & editing (equal). Toru Murayama: Conceptualization (equal); Data curation (equal); Formal analysis (equal); Investigation (equal); Methodology (equal); Writing – original draft (equal); Writing – review & editing (equal). Koki Kaneko: Data curation (equal); Formal analysis (equal). Junichi Honda: Data curation (equal); Formal analysis (equal). Loi Tonthat: Data curation (equal); Formal analysis (equal). Kazuhiko Okita: Data curation (equal); Formal analysis (equal).

The data that support the findings of this study are available from the corresponding author upon reasonable request.

1.
L.
Wu
and
X.
Qu
, “
Cancer biomarker detection: Recent achievements and challenges
,”
Chem. Soc. Rev.
44
(
10
),
2963
2997
(
2015
).
2.
M. K.
Masud
,
J.
Na
,
M.
Younus
,
M. S. A.
Hossain
,
Y.
Bando
,
M. J. A.
Shiddiky
, and
Y.
Yamauchi
, “
Superparamagnetic nanoarchitectures for disease-specific biomarker detection
,”
Chem. Soc. Rev.
48
(
24
),
5717
5751
(
2019
).
3.
V. S. P. K. S. A.
Jayanthi
,
A. B.
Das
, and
U.
Saxena
, “
Recent advances in biosensor development for the detection of cancer biomarkers
,”
Biosens. Bioelectron.
91
,
15
23
(
2017
).
4.
K.
Sawalha
,
N. B.
Norgard
,
B. M.
Drees
, and
A.
López-Candales
, “
Growth differentiation factor 15 (GDF-15), a new biomarker in heart failure management
,”
Curr. Heart Failure Rep.
20
,
287
299
(
2023
).
5.
Y.
Wan
and
J.
Fu
, “
GDF15 as a key disease target and biomarker: Linking chronic lung diseases and ageing
,”
Mol. Cell. Biochem.
24
,
1
14
(
2023
).
6.
A.
Assadi
,
A.
Zahabi
, and
R. A.
Hart
, “
GDF15, an update of the physiological and pathological roles it plays: A review
,”
Pfluegers Arch.
472
,
1535
1546
(
2020
).
7.
C.
Favrot
,
M.
Linek
,
J.
Fontaine
,
L.
Beco
,
A.
Rostaher
,
N.
Fischer
,
N.
Couturier
,
S.
Jacquenet
, and
B. E.
Bihain
, “
Western blot analysis of sera from dogs with suspectedfood allergy
,”
Vet. Dermatol.
28
,
180
(
2017
).
8.
D. M.
Rissin
,
C. W.
Kan
,
T. G.
Campbell
,
S. C.
Howes
,
D. R.
Fournier
,
L.
Song
,
T.
Piech
,
P. P.
Patel
,
L.
Chang
,
A. J.
Rivnak
,
E. P.
Ferrell
,
J. D.
Randall
,
G. K.
Provuncher
,
D. R.
Walt
, and
D. C.
Duffy
, “
Single-molecule enzyme-linked immunosorbent assay detects serum proteins at subfemtomolar concentrations
,”
Nat. Biotechnol.
28
,
595
599
(
2010
).
9.
K.
Tanaka
,
H.
Waki
,
Y.
Ido
,
S.
Akita
,
Y.
Yoshida
,
T.
Yoshida
, and
T.
Matsuo
, “
Protein and polymer analyses up to m/z 100 000 by laser ionization time-of-flight mass spectrometry
,”
Rapid Commun. Mass Spectrom.
2
,
151
153
(
1988
).
10.
K.
Enpuku
et al, “
Liquid phase immunoassay using AC susceptibility measurement of magnetic markers
,”
Appl. Phys. Express
2
,
037001
(
2009
).
11.
H.
Shao
et al, “
Magnetic nanoparticles for biomedical NMR-based diagnostics
,”
Beilstein J. Nanotechnol.
1
,
142
154
(
2010
).
12.
T.
Wang
et al, “
Ultrasensitive determination of carcinoembryonic antigens using a magnetoimpedance immunosensor
,”
RSC Adv.
5
,
51330
51336
(
2015
).
13.
Z.
Yang
et al, “
A biosensor was developed for detection of Escherichia coli (E. coli) O157
,”
Biomed. Microdevices
17
,
5
(
2015
).
14.
T.
Wang
et al, “
Magnetic impedance biosensor: A review
,”
Biosens. Bioelectron.
90
,
418
435
(
2017
).
15.
G. V.
Kurlyandskaya
and
V.
Levit
, “
Advanced materials for drug delivery and biosensors based on magnetic label detection
,”
Mater. Sci. Eng.: C
27
,
495
503
(
2007
).
16.
H.
Chiriac
et al, “
Microwire array for giant magneto-impedance detection of magnetic particles for biosensor prototype
,”
J. Magn. Magn. Mater.
311
,
425
428
(
2007
).
17.
F.
Blanc-Béguin
et al, “
Cytotoxicity and GMI bio-sensor detection of maghemite nanoparticles internalized into cells
,”
J. Magn. Magn. Mater.
321
,
192
197
(
2009
).
18.
G. V.
Kurlyandskaya
et al, “
Nanostructured materials for magnetic biosensing
,”
Biochim. Biophys. Acta, Gen. Subj.
1861
,
1494
1506
(
2017
).
19.
D.
de Cos
et al, “
Study of the influence of sensor permeability in the detection of a single magnetotactic bacterium
,”
J. Magn. Magn. Mater.
500
,
166346
(
2020
).
20.
A.
Kumar
et al, “
Magnetoimpedance biosensor for Fe3O4 nanoparticle intracellular uptake evaluation
,”
Appl. Phys. Lett.
91
,
143902
(
2007
).
21.
V. D.
Krishna
et al, “
Giant magnetoresistance-based biosensor for detection of influenza A virus
,”
Front. Microbiol.
7
,
400
(
2016
).
22.
Y.
Wu
et al, “
Rapid detection of Escherichia coli O157:H7 using tunneling magnetoresistance biosensor
,”
AIP Adv.
7
,
056658
(
2017
).
23.
T.
Mizoguchi
et al, “
Highly sensitive third-harmonic detection method of magnetic nanoparticles using an AC susceptibility measurement system for liquid-phase assay
,”
IEEE Trans. Appl. Supercond.
26
(
5
),
1602004
(
2016
).
24.
L.
Tonthat
et al, “
A simple and rapid detection system for oral bacteria in liquid phase for point-of-care diagnostics using magnetic nanoparticles
,”
AIP Adv.
9
(
12
),
125325
(
2019
).
25.
S.
Yabukami
,
T.
Murayama
,
S.
Takahashi
,
S.
Ohno
,
J.
Washio
, and
N.
Takahashi
, “
A detection and analysis of Fusobacterium utilizing changes in the magnetic properties of magnetic nanoparticles-antibody-antigen aggregates
,”
IEEE Trans. Magn.
58
,
5300306
(
2022
).
26.
Y.
Pu
,
H.
Zhao
,
T.
Murayama
,
L.
Tonthat
,
K.
Okita
,
Y.
Watanabe
, and
S.
Yabukami
, “
Method for rapid detection of bacteria using magnetic nanoparticle aggregates
,”
J. Magn. Soc. Jpn.
47
,
66
69
(
2023
).
27.
K.
Okita
,
P.
Youcheng
,
L.
Tonthat
,
T.
Murayama
,
S.
Yabukami
,
Y.
Ozawa
,
S.
Asamitsu
,
H.
Okamoto
, and
T.
Kamei
, “
Magnetic susceptibility-based detection of fusobacterium Nucleatum in human saliva
,”
IEEE Magn. Lett.
14
,
1
5
(
2023
).
29.
H.
Kuang
,
G.
Cui
,
X.
Chen
,
H.
Yin
,
Q.
Yong
,
L.
Xu
,
C.
Peng
,
L.
Wang
, and
C.
Xu
, “
A one-step homogeneous sandwich immunosensor for Salmonella detection based on magnetic nanoparticles (MNPs) and quantum dots (QDs)
,”
Int. J. Mol. Sci.
14
,
8603
8610
(
2013
).