A novel immunochromatographic strips assay for rapid and simple detection of systemic lupus erythematosus

Systemic lupus erythematosus (SLE) is a complex multi-system autoimmune disease. Detection of anti-nuclear antibodies (ANA) is fundamental for the diagnosis of SLE. In the present study, we found that the level of core fucosylation catalyzed by α1,6-fucosyltransferase (Fut8) is markedly up-regulated on immunoglobulin G (IgG) in the sera of SLE patients detected by Aspergillus oryzae lectin (AOL) blot. In sandwich Dot enzyme-linked immunosorbent assay (Dot-ELISA), the core fucosylation level was also found significantly increased in the sera from SLE patients with a higher ANA titer. To establish a rapid and sensitive laboratory test for the diagnosis of SLE, we prokaryotically expressed AOL and C3-D1-C3-D2-C3 of protein G (SpG3), and generate AOL-conjugated colloid gold immunochromatographic strips (ICS). The detection limit of core fucosylated IgG was 10 μg/mL for AOL-conjugated colloid gold ICS. As well as indirect immunofluorescence, the AOL-conjugated colloid gold ICS showed reliable results in the serum of 39 SLE patients. Our results indicated that the AOL-conjugated colloid gold ICS could serve as a rapid test for the detection of ANA and suspected cases of SLE.

Scientific RepoRtS | (2020) 10:14178 | https://doi.org/10.1038/s41598-020-71137-0 www.nature.com/scientificreports/ N 297 of IgG is important to maintain the conformation of Fc fragment and plays a critical role in the biological function of IgGs 10 , and thereby may critically contribute to development of autoimmune pathology. For instance, sialylation of IgGs attenuates the development of autoimmune disease 11 . De-galactosylated IgG autoantibodies are associated with severity of autoimmune disorders, such as rheumatoid arthritis (RA) and SLE 9,12 . In multiple sclerosis, polymorphisms in the gene coding for the glycosylation enzyme MGAT5 were correlated to disease severity 13 . In IgA nephropathy (IgAN), galactose-deficient IgA1 is targeted by anti-IgA1 auto-antibodies and form autoimmune complexes which induce renal toxicity by complement activation 14 .
Immunochromatographic strips (ICS) is a rapid, convenient detection of analytes and easily visual endpoint. It is wildly used in medical fields for disease screening 19 . In the present study, we established a rapid and sensitive ICS using prokaryotically expressed Aspergillus oryzae lectin (AOL) and C3 domain repeated protein G (SpG3), which is specific for L-fucose recognition of serum IgG 20 for the diagnosis of SLE.

Results
Core fucosylation is significantly upregulated on the IgG of SLE patients. The fucosylation modification was significantly increased in sera from AD patients 6,21 . We also detected the level of core fucosylation in the sera of AD patients and found that core fucosylation was significantly increased in AD patients, evidenced by lectin blot with AOL, which is specific for L-fucose recognition (*p < 0.05) ( Fig. 2A,B). The value of AOL/IgG was markedly increased in AD patients (**p < 0.01) (Fig. 2B). The core fucosylation on the IgGs from SLE patients was higher than those from RA patients (**p < 0.01), albeit the core fucosylation was significantly increased in the serum of RA and SLE patients compared with the healthy controls (Fig. 2C).
To further investigate the level of core fucosylated IgG in the SLE sera, we detected the sera of the SLE patients (n = 39) by AOL blot. The level of core fucosylation was significantly increased in SLE serum (****p < 0.0001), and the expression of IgGs was also upregulated (*p < 0.05) (Fig. 3A,B). The value of AOL/IgG was dramatically increased in SLE patients (**p < 0.01) (Fig. 3C). Moreover, IgGs from the serum of SLE patients and healthy controls were purified with Protein A Magnetic Beads. Indeed, the core fucosylated N-glycans were increased on the IgGs of the SLE patients compared with those of healthy controls by electrospray ionization mass spectrometry (ESI-MS) ( Supplementary Fig. 1). These results implied that the hyper core fucosylation on the IgG is characterized in the serum of SLE patients.
Next, the Dot-ELISA was developed for the detection of core fucosylated IgG in the SLE sera. Dot-ELISA showed that 2 μg/dot Protein G concentration, 1:100 dilution of serum, 1:1,000 dilution of biotin-conjugated AOL and 1:3,000 dilution of HRP-conjugated streptavidin were optimized (Fig. 3D). Among the SLE clinical samples, we grouped the samples according to the ANA titer, and performed Dot-ELISA. Rates of detection of core fucosylated IgG by the Dot-ELISA were 100%, while those of healthy controls were indicated as negative result without any false-positive results (Fig. 3E).
Construction of pET28a-FleA and expression of AOL. Given that there are six fucose-binding sites in the structure of AOL, which has the most specific binding capability to core fucose 20,22 , we prokaryotically expressed the AOL-encoding gene, FleA, from Aspergillus oryzae. Moreover, because the surface of colloidal www.nature.com/scientificreports/ gold harbors electrostatic characteristics and hydrophobic properties 23 , we added 10 lysine (lys) into pET28a-FleA expression system to improve the ability to conjugate with colloidal gold (Fig. 4A). Pet28a-FleA was confirmed by restriction enzyme (EcoR I and Xho I) digestion and a positive clone was sequenced to show the identity of FleA gene (Fig. 4B).
The AOL was expressed after induction of IPTG for 5 h (h) in E. coli BL21 (DE3) transformed with the plasmid pET-28a (+)-FleA (Fig. 4C). The AOL was purified by Ni 2+ -NTA affinity chromatography. SDS-PAGE results verified the successful purification of the AOL with nearly 95% purity (Fig. 4D). AOL corresponding to 36 KDa was recognized by anti His-tag Ab (Fig. 4E).
Expression of C3-D1-C3-D2-C3 domains of protein G (SpG3). Protein G is spontaneously released from streptococci, especially in group G streptococci, which has high affinity for binding to IgG. The C3 domain of protein G possesses a higher binding affinity to sites on the Fc portion of IgG 24 . Thus, we designed C3-D1-C3-D2-C3 domains of protein G (SpG3). Moreover, to improve the affinity for immobilizing with the NC membrane, we insert five alanine (Ala) gene after the initiation codon (ATG) and six histidine (His) gene before the terminal codon (TAA) into PU57 cloning vector (Fig. 5A). Recombined SpG3 fragments were digested with EcoR I and Xho I, and then inserted into EcoR I and Xho I-digested pET21a (+), yielding pET21-SpG3. PET21a-SpG3 was confirmed by restriction enzyme digestion (Fig. 5B). The pET21a-SpG3 vector was induced by IPTG (Fig. 5C) and a 21 KDa protein was detected by anti His-tag Ab (Fig. 5D).
Establishment of AOL-conjugated colloidal gold Dot-ELISA. Preparation of high-quality AOLconjugated colloid gold solution is an essential step to ensure the peculiarity and sensibility of the immunological diagnosis. NaCl causes the aggregation of colloidal gold solution and changes the color from red to blue. If the color does not vary with the addition of NaCl, the AOL-immobilized on the colloidal gold solution contains the optimal concentration of AOL for fully covering the surface of all the colloidal gold solution. To obtain the AOLconjugated colloidal gold solution, 100 μL of colloidal gold nanoparticles were mixed with 1, 2, 4, 6, 8, 10, 20, 40, 60 and 80 μg/mL AOL expressed in the addition of 10% NaCl. The optimum concentration of AOL adsorption was determined to be 40 μg/mL. At this concentration, AOL was confirmed to be the minimum amount for stabilizing colloidal gold solution (Fig. 6A,B).
FUT8 is the sole enzyme responsible for catalyzing the core fucosylation in human. As no core fucosylation can be detected in the sera of Fut8 −/− mice, we detected the sensitivity of this test by the different dilution of Fut8 +/+ serum and found that the detection limit was 1:1,000 dilution in Fut8 +/+ sera (10 μg/mL), while negative result was in Fut8 −/− mice sera (Fig. 6C). AOL-conjugated colloidal gold Dot-ELISA showed that 2 μg/dot SpG3 concentration, 1:100 dilution of sera, 0.2 μg of AOL-conjugated colloidal gold were optimized (Fig. 6D). Among www.nature.com/scientificreports/ the clinical samples from 39 SLE patients, we grouped the samples according to the ANA titer of 1:100, 1:320 and 1:1,000, and performed AOL-conjugated colloidal gold Dot-ELISA. The core fucosylation is higher in the serum of 1:1,000 ANA titer than those of 1:100 or 1:320 ANA titers, indicated the sensitivity of the assay for detection of core fucosylated IgG (Fig. 6D). We also found that the core fucosylation level was significantly increased in sera from SLE patients by AOL-conjugated colloidal gold Dot-ELISA (Fig. 6E).

Detection of SLE by AOL-conjugated colloidal gold ICS.
We have used novel ICS concept; this is dependent on the transport of a reactant to its binding partner immobilized on the surfaces of the membrane. Capillary action of aqueous-medium through membrane pones is applied for the transfer. This helps to separate the unbound reactant from the binding complex formed at the liquid-solid interface 25 . ICSs were assembled in sequence, which was laminated by the sample application pad, AOL-conjugated colloidal gold pad, NC membrane and the absorption pad (Fig. 7A). Samples were recorded as positive when two clear red lines appeared. And samples were considered negative when detection band was not colored but control band turned red. We firstly proved the specificity of AOL-conjugated colloidal gold ICS by the Fut8 +/+ serum and Fut8 −/− serum. The serum of Fut8 −/− mice indicated only C line, while the serum of Fut8 +/+ mice had both test (T) and control (C) lines (Fig. 7B). Similarly, the AOL-conjugated colloidal gold ICS for SLE sera showed positive results, but those were not in the healthy control sera (Fig. 7C). Moreover, the AOL-conjugated colloidal gold ICS showed no  www.nature.com/scientificreports/ cross-reactivity with the other ADs, such as RA, IgAN, connective tissue disease (CTD), indicated that the AOLconjugated colloidal gold ICS established are highly specific to SLE (Fig. 7C).

Discussion
SLE is a sophisticated AD, and the risk of death for SLE patients is still 2 times greater than those of general population. ANAs are hallmarks of SLE. High-titer ANAs directly contribute to SLE pathogenicity by multiple mechanisms 1 . IIFA is the preferred method for SLE diagnosis in clinics. However, the method is time-consuming, very subjective and expensive 26 . Therefore, rapid and sensitive laboratory tests for the diagnosis of SLE are urgently needed to be established 27 . In the present study, we found that core fucosylation on IgG was significantly upregulated in SLE serum. The reliability and rapid method for diagnosis of SLE was successfully established by targeting the core fucosylated IgG by AOL-conjugated colloidal gold ICS. Protein glycosylation is essential for a multitude of biological processes, as exemplified by the glycosylation of immunoglobulins. The changes of glycosylation of immunoglobulins are linked to the pathophysiology of AD pathogenesis. In addition to the differences in the number of disulfide bonds and the length and flexibility of the hinge region, changes of glycosylation profiles of IgG also are related to AD pathogenesis, such as RA and SLE [28][29][30][31][32][33][34] . Hypogalactosylation of IgG in the MRL/Mp-lpr/lpr lupus mice model was reported by Mizuochi et al. 35 . Interestingly, anti-inflammatory activity was observed in lupus BXSB mice after the removal of glycan moieties on N 297 following the treatment of IgG with streptococcus endoglycosidase 36 . Similarly, another study showed that treatment by streptococcus endoglycosidase can abolish the proinflammatory properties of immune complexes from SLE patients 37 . The glycan of IgG is hyper core-fucosylated complex with bisecting N-acetylglucosamine (GlcNAc). Sjöwall 7 found that core-fucosylated glycans was increased on the IgG of SLE patients. The hyper core fucosylation is closely associated with the ANA production 18 . In the present study, we detected core fucosylated IgG in SLE patients by Dot-ELISA with high sensitivity and specificity.
ICS is efficient and one step immunochromatographic assay. It can be carried out at the sites since it does not require use of many reagents and is performed in one step only so no need of specialized facility 38 . In the present study, we established ICS with AOL-conjugated colloidal gold to detect core fucosylated IgG of SLE  25,39 . To enhance binding affinity to colloid gold particle, we added 10 lysine (lys) to prokaryotic expressed AOL, and the appropriate amount of AOL was 40 μg/mL. In addition, the porous structure of the NC membrane provides a higher binding capacity. To improve the affinity for immobilizing with the NC membrane, we expressed C3 domain repeated SpG3 with five alanine (Ala) and six His. The use of SpG3 spotted NC membrane could greatly facilitate the reproducibility and field applicability of the ICS. The optimized dilution of clinical serum of SLE patients was 1:100 for detection of ICS generated. ICS has a lot of advantages compared to traditional assay, such as needing no equipment or trained operators and having easily readable results within 5 min. Comparison of the results of ICS and IIFA yielded a relative reliability of 100% without false-positive results in healthy controls and other ADs. The hyper core fucosylation were detected in RA patients as well as SLE in the AOL blot, but not in AOL-conjugated colloidal gold ICS. We proposed that there are two reasons why the core fucosylated IgGs in RA serum showed the negative reaction in AOL-conjugated colloidal gold ICS. First, the level of core fucosylated IgGs in RA sera is significantly lower than in SLE serum (**p < 0.01). Second, about 80-90% of RA patients have circulating rheumatoid factors, it is reasonable to suppose that the rheumatoid factors bind to the IgGs to suppress the binding of IgG to the SpG3. In summary, a key finding of our study is that the core fucosylated IgGs could serve as a useful biomarker for SLE detection in clinical practice. In the present study, both AOL and SpG3 were prokaryotically expressed. Since the N-glycosylation pathways were lacked in bacteria, the sole glycan on the IgGs was from the sera of SLE patients. Therefore, the AOL-conjugated colloidal gold ICS showed high specificity and sensitivity. Our data pave the way for the future development of an adequate screening test in epidemiologic surveillance for suspected SLE.

Methods
Mice and animal immunization. Fut8 +/+ mice and Fut8 −/− mice were generated by crossing heterozygous Fut8 +/− mice in our laboratory as previously described 18 . Mice were maintained in a room illuminated for 12 h (08:00 to 20:00) and kept at 24 °C with free access to food and water in the specific pathogen-free (SPF) laboratory animal facility of Dalian Medical University, China. Mice were subcutaneously immunized with 100 µg prokaryotically expressed AOL mixed with an equal volume of complete Freund's adjuvant (Sigma). Two-week later, mice were immunized with 100 µg of AOL. The anti-AOL sera were collected, and IgG was purified by Protein A Magnetic Beads (Novus).  Antibodies. Anti His-tag antibody (Ab) (2491213) was purchased from Proteintech; biotin-conjugated AOL (A2659) was from Tokyo Chemical Industry CO, LTD; horseradish peroxidase (HRP)-conjugated donkey antihuman IgG was from Proteintech (SA00001-11); HRP-conjugated streptavidin (AB 7403) and HRP-conjugated goat anti-rabbit IgG (A0208) were purchased from Abcam.
Western and lectin blot. Equal amounts (2 μg) of protein were run on 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) under reducing conditions and then transferred to polyvinylidene difluoride (PVDF) membranes (Millipore Corp.). Membranes were blocked for 2 h with 5% bovine serum albumin (BSA) in TBS-T (pH 7.5, 10 mM Tris-HCl, 0.1% Tween 20, 150 mM NaCl), and incubated with the appropriate primary Abs or lectin overnight. After washing, the blots were incubated with the corresponding secondary Abs. The proteins were visualized using an ECL system (Amersham). Density analysis was performed using Quantity One software.    38 . According to the above results, appropriate amount of AOL was added drop-wise to colloidal gold solution about 10 min. The mixture was stored for 30 min at RT. After that, 10% BSA was added drop-wise to the mixture about 10 min. The mixture was stored for 30 min at RT and centrifuged at 3,500 rpm for 10 min. The supernatant was reserved and centrifuged at 12,000 rpm for 20 min. The supernatant was removed. The colloid gold labeled AOL was used for Dot-ELISA or ICS.
Preparation of AOL-conjugated colloidal gold ICS and detection of serum samples. ICS was constructed according to the previously described method with sample application pad, AOL-conjugated colloidal gold pad, nitrocellulose membrane and the absorption pad 38 . The AOL-colloidal gold conjugates probe was added onto the glass fiber and dried at 37 °C for 2 h. SpG3 and rat anti-AOL-IgG were drawn onto the NC membrane as two discrete zones, one for control (C) and another for test (T). For each test, 100 μL of every 1: 100 diluted serum samples (1 μL serum + 99 μL PBS) was added onto the sample application pad, allowing the sample to migrate upward. After about 5 min, an image containing the color signal was showed on the strip and the results were judged by the color of the T and C lines. If both T and C lines turned red, the sample was considered as positive. If the C line turned red but the T line was no color, it was considered as negative.
ESI-MS. MS analysis was performed using an LTQ-XL linear ion trap electrospray ionization mass spectrometer (Thermo Scientific, USA) coupled with a HPLC system, as described previously 42 . The sample loaded onto a SepPak C18 SPE column and N-glycans was collected.