Carboxy water-soluble QDs-COOH (ZnCdSe/ZnS, core/shell) with a size of 12 nm were obtained from Wuhan Jiayuan Quantum Dots Corporation, Ltd. (Wuhan, China). Staphylococcal protein-A (SPA) was purchased from Beijing Solarbio Science & Technology Corporation, Ltd. Fetal calf serum (FCS) was bought from Tian hang Biological Technology Co., Ltd. (Zhejiang, China). The following substances were purchased from Sigma-Aldrich Chemical Co. (St. Louis, MO): 25-OH VD antigen, bovine serum albumin (BSA), staphylococcal protein-A (SPA), and N-(3-dimethylaminopropyl)-Nʹ-ethylcarbodiimide hydrochloride (EDC). 25-OH VD monoclonal antibody was purchased from Bioventix Biotechnology, United Kingdom. Vitamin D and 25-OH VD standards were purchased from Hanzun Biological Co., LTD. 25-OH VD derivative were purchased from Xiamen Tongrenxin Gong Department. The nitrocellulose membrane, sample pad, backing card, absorbent pad and the filter membrane were from Millipore, US. The commercial Chemiluminescence Immunoassay (CLIA) kit for 25-OH VD quantitative analysis was from Immunodiagnostic Systems Holdings PLC (IDS), United Kingdom. Millipore’s Milli-Q filtration system was used to create ultrapure water. One-hundred clinical samples were collected from patients who visited the outpatient clinic or physical examination center of Zhengzhou Central Hospital in 2023. Each patient underwent a health interview and was asked to sign an informed consent form.

The synthesis of QDs-mAb conjugates was carried out as described in an earlier study (Zhou et al., 2021). In this study, improved synthesis conditions were used for the 25-OH VD probes. A reaction vessel was filled with 12.5 μL of the QDs-COOH (ZnCdSe/ZnS) (8 μM) and 30 μL of the EDC (4.0 mg/mL), and the mixture was incubated for 30 min at room temperature 25°C. EDC solution was then used to activate the QDs-COOH. Next, 30 μL of 25-OH VD mAb (4.0 mg/mL) in borate-buffered saline (BBS, 10 mM, pH = 7.6) was added to the reaction. A shaking incubator was used to allow the mixture to continue to react for an additional 2 hours at 25°C. At the end of the reaction, the reaction was centrifuged at 8,000 rpm for 3 min to remove any agglomerates and supernatant was discarded. The sample was concentrated and purified five times using ultrafiltration tubes, and the final product was redissolved in the appropriate target coupling buffer. QDs-mAb conjugates were saved in a new brown microcentrifuge tube and stored at 4°C for future use.

The QDs-FICA consists of five main parts: sample pad, conjugate pad, nitrocellulose membrane, absorbent pad, and backing card. The 25-OH VD-BSA conjugate and SPA solution were immobilized on NC membranes as the test T) and control C) lines using a BioDot XYZ3050 dispensing platform (Irvine, CA, United States). The NC membranes was then dried in an electric blast oven at 40°C for 4 hours. The distance between the T and C lines was roughly 5 mm. Next, the test strips were assembled in the order of the sample pad, conjugate pad, NC membrane, and absorption pad. They were then packaged in a plastic cassette after being sliced into 2.8 mm-wide test strips using a BioDot CM 4000 Guillotine Cutter. They were then stored with desiccant and sealed in aluminum foil at 4°C.

The inputs of EDC, 25-OH VD mAb, and pH value play a crucial role during the conjugation process. These conditions were optimized by an orthogonal experiment, which is described in Table 1. The fluorescence intensity on the test paper and the peaks of the fluorescence spectral curves were recorded using an MD-980 Multi-channel Fluorescent Immunoassay Analyzer (Micro detection Corporation, Ltd., Nanjing, China). The measurements were used to determine the best set of conjugation conditions.

The 25-OH VD stock solution was diluted with 0.01 mol/L PBS buffer solution to achieve final concentrations of 0, 5, 10, 20, 30, 40, 60, 80, and 100 ng/mL. Then, 100 μL of each standard solution was incubated for 3 min with 3 μL of standard immunoprobe before being applied to the spiking wells of the test strips. The T/C ratios of the strips were measured using a gold standard reader after 15 min. Each concentration was tested five times. The sensitivity, IC₅₀ value, and linear quantification range of 25-OH VD immunochromatographic strips were determined by plotting the standard curve with the logarithm of 25-OH VD concentration as the horizontal coordinate and the T/C value as the vertical coordinate. The sensitivity was computed by subtracting the mean value for the negative samples from their triple standard deviation.

The 25-OH VD stock solution was diluted with 0.01 mol/L PBS buffer solution to the series standard concentration of 0, 12.5, 5, 7.5, 15, 25, 50, 60, 80, 100, 125, 130, 140, and 150 ng/mL. The FI_T/FI_C ratio of the test strip was detected using the fluorescence instrument, and each concentration was measured in parallel five times. The FI_T/FI_C ratio of the test strip with a concentration of 0 ng/mL of added 25-OH VD was denoted as B0 and the FIT/FIc ratios of the other added concentration was denoted as B. The competition inhibition curve of 25-OH VD strips with PBS as the sample matrix was plotted by logarithmic mapping of the B/B0 value of the 25-OH VD concentration, and the IC₅₀ value and linear quantitative range of the strips were determined at this time.

To evaluate the specificity of 25-OH VD colloidal gold quantitative test strips, six common 25-OH VD structural analogs, namely, 25-OH VD₂, 25-OH VD_3, 125-OH₂VD₃, 125-OH₂VD₂, VD2, and VD3, were selected for cross-reaction experiments. A series of standards concentration with final concentrations of 0, 5, 10, 20, 30, 40, 60, 80, and 100 ng/L were prepared in the stock solution, and the above solutions were detected with 25-OH VD colloidal gold strips. Each concentration was measured three times in parallel, and the standard curve was drawn according to the T/C ratio measured by the colloidal gold reader. The IC₅₀ of each competitor was obtained, and the cross-reaction rate of each analogue with 25-OH VD was calculated according to the following formula: (Cr) %= (25-OH VD IC₅₀) (cross analogue IC₅₀) 100%.

The test was repeated twice with the same dilution of the interference test samples; then, the samples of low, medium, and high concentration levels were used as the base samples. We divided each base sample into five portions, one of which was added to the sample dilution, without interfering substances, as the control sample. The other four portions were added in equal volumes, with different concentrations of interfering substances, as the analysis sample. The measurement was repeated three times to determine the average value of each sample to calculate the interference rate, using the following formula: Interference rate = (average concentration of analyzed sample average concentration of base sample)/average concentration of base sample × 100%. The interference effect between the addition of interferon and the absence of interferon was calculated, and ≤10% bias was used as the judgment standard.

The 25-OH VD standard was added into 0.01 mol/mL PBS buffer solution until the final concentration was 25, 50, and 100 ng/mL. The same batch of test strips was used to detect the three concentrations of low, medium, and high levels; measurement was performed every 3 days and repeated three times for each sample. To evaluate the accuracy and precision of the test strips, the recovery rate of each concentration and the difference between batches of the test strips was calculated.

Stability experiments were performed by randomly selecting 100 test strips from the same batch and placing them in the oven at 60.1°C. The experiments were designed according to the Arrhenius formula, the time intervals for which 60.1°C needs to be tested are the Day 0, Day 1, Day 2, Day 4 and Day 8 (about 0 days, 40 days, 80 days, 160 days, and 320 days at 25°C). Three standard antibody dilutions of 1:500, 1:2,000, and 1:5,000 were measured and three parallels were set up for each sample. The average FI_T/FI_C values were calculated. The QDs solution with a concentration of 0.5 mg/mL was configured and directly scribed to the T-line position of the NC membrane using a membrane scribe, assembled into a test card and then placed in the tester for 100 consecutive readings to compare the average lifetimes of the QDs and mAb QDs.

For construction of the quantitative standard curve for 25-OH VD detection, the extract was mixed with 25-OH VD to make standard solutions (0, 0.01, 0.05, 0.1, 0.5, 1.0, 2.5, 5.0, 10, and 50 ng/mL). 25-OH VD standard solution detection was performed using fluorescent test strips. The cut-off value, i.e., the lowest 25-OH VD concentration that caused T line fluorescent band invisibility, was used to evaluate the qualitative performance of QDs-FICA. The T line fluorescence intensity dropped as the 25-OH VD concentration increased, disappearing under a handheld UV light. The MD-800 multiple immunochromatographic test strip analyzer measured the T line, C line, and FI ratio (FI_T/FI_C) fluorescence intensity three times for each standard test for quantitative detection.

The QDs were synthesized by encapsulating ZnCdSe/ZnS using a microemulsion method as previously reported (Shen et al., 2015). Agarose gel electrophoresis (AGE) and sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) experiments indicated that anti-25-OH VD-mAb-QDs have a large molecular weight, which slows down their migration in agarose gels (Figure 1A). When anti-25-OH VD-mAb were coupled to QDs, the excitation and emission wavelengths remained unchanged at about 605 nm, but the fluorescence intensity was weaker than that of QDs, which may be attributed to the effect of the antibody on fluorescence detection or the occurrence of fluorescence burst during the coupling process (Figure 1B). The particle size distribution and potential variation of the nanoparticles were measured by dynamic light scattering (DLS), as shown in (Figures 1C,D), The mAb-QDs conjugates reached a size of 160.82 nm, and the polymer dispersity index (PDI) value of mAb-QDs conjugates was 0.372. In comparison, the QDs were only 19.27 nm in size, and the PDI value of the QDs was 0.187. As a result, the size of the hydrated QDs was significantly smaller than that of the mAb-QDs. Furthermore, Figures 1E,F show the zeta potential (ζ potential) values of the anti-25-OH VD-mAb-QDs conjugates and QDs. While the potential of QDs was detected by DLS to be −15.7mV, the potential of anti-25-OH VD-mAb-QDs was −2.9 mV. Transmission electron microscopy (TEM) images showed (Supplementary Figure S1) that the prepared anti-25-OH VD-mAb-QDs had a regular spherical shape with a relatively uniform particle size distribution. High-resolution TEM images of individual particles showed that a large number of oil-soluble ZnCdSe/ZnS QDs were tightly embedded in the polymer. The above several sets of data indicate that there is effective coupling between the QDs and anti-25-OH VD-mAb, confirming successful synthesis of the immunofluorescence probe.

QDs-FICA is a straightforward and visually appealing approach for validating the bioactivity of mAb-QDs. A schematic illustration for the detection of 25-OH VD using the QDs-FICA-based platform is shown in Figure 2. Briefly, QDs were used to replace colloidal gold particles as signal markers. QDs were coupled with the corresponding antibody of 25-OH VD to be tested, and then sprayed on the binding pad. Immunochromatographic test strips were made by assembling the sample pad, the binding pad, the reaction membrane, and the absorbent pad. The FICA technique revealed that QDs could not selectively interact with the T line (25-OH VD-BSA) and C line (SPA). However, anti-25-OH VD-mAb -QDs underwent strong interactions with the C and T lines when exposed to ultraviolet light from a portable UV lamp with an excitation wavelength of 365 nm.

Because of the pH, EDC and anti-25-OH VD mAbs can have a significant effect on the stability and biological activity of the conjugate. Therefore, coupling conditions were optimized in this study. As shown in Figure 3A, when the pH was 7.6, the highest values of both FI_T and FI_C were obtained. When the molar ratio of QDs to EDC is 1:2000, FI_T and FI_C are saturated (Figure 3B). The excessive amount of EDC would lead to a large amount of conjugate aggregation, which is unfavorable for mutual coupling. In addition, the results showed that FI_T and FI_C were saturated when the molar ratio of QDs to the anti-25-OH VD mAb was 1:8 (Figure 3C). The optimal coupling conditions were as follows: a pH of 7.6, an EDC ratio of 1:2000, and a 25-OH VD:mAb ratio was 1:8.

The dosages of 25-OH VD-BSA and anti-25-OH VD mAb-QDs are particularly important for optimal sensitivity and fluorescence intensity of the FICA. The experimental results are presented in Table 1. Which shows that the fluorescence intensity of the T and C lines of the test strips was stronger when the concentration of 25-OH VD-BSA was 0.4 mg/mL and the immunoprobe dosage was 1.0 μL (group 4), The FI_T and FI_C were 642 ± 23 and 815 ± 12, respectively. The competition inhibition rate of 1.0 ng/mL positive samples was as high as 80.27 ± 0.02 (n = 3), which indicated the best detection effect relative to the rest of the groups. Therefore, the optimization results for group 4 were determined as the best quantitative analysis conditions for the test strips.

In order to achieve the best performance of the QDs-FICA platform, we further optimised the immunoreaction time of the assay, the salt ions and the ethanol content of the sample solution. As shown in Figure 4A, The FI_T and FI_C improved continuously within 45 min The fluorescence signals of T and C lines were gradually strengthened after 10 min of spiking, and the FI_T/FIc ratio tended to be 0.8 when the immunoreaction time reached 15 min. Specific experimental results are shown in Figure 4B. When the salt ion concentration is 0.2 mol/L, the T/C value of the negative sample is 1.365, and the T/C value of the positive sample is 0.028. The competitive inhibition rate reaches 92.5%. The maximum values of FI_T/FI_C for negative samples and competitive inhibition for positive samples were reached when the concentration of ethanol in the samples was 10%. When the concentration of ethanol is greater than 10%, FI_T/FI_C and the competitive inhibition rate decrease rapidly (Figure 4C). The reason is that the concentration of organic solution exceeds the capacity of protein, which will lead to the loss of protein activity.

The stability results showed that the detection efficiency of the test strips stored at 60.1°C for 4 days (160 days, 25°C) was weakened, and only 1:500 high concentration antibody standards could be measured. So it was inferred that the storage time of the test strips in this study was about 160 days at room temperature. See Supplementary Figure S2. The fluorescence lifetime measurement results show that although the fluorescence intensity of anti-25-OH VD-mAb-QDs is slightly lower than that of QDs, the fluorescence intensity of 100 consecutive measurements of the T-line remains basically unchanged. This demonstrates that the lifetime of anti-25-OH VD-mAb-QDs can satisfy normal clinical detection applications (Supplementary Figure S3).

The standard curve was shown in Figure 5, its linear regression equation is y = −0.02088 logx)+1.444 (R ² = 0.9050). The concentration of 25-OH VD shows a very good linear relationship between 5 ng/mL and 100 g/mL, according to the linear regression equation. It is concluded that the IC₅₀ of the test strip was 39.6 ± 1.33 ng/mL (n = 3). Table 2 showed the cross-reaction rates of 25-OH VD_3, 125-OH₂VD₃, and 125-OH₂VD₂ with the strip were 26%, 11%, and 25%, respectively. The cross-reaction rates of VD₃ and VD₂ were less than 1% with the test strip. Therefore, the fluorescence test strip has good specificity.

As shown in Supplementary Table S1, the interference rate is less than 8%. This indicates that the method is more resistant to strong interference. The test results for the 25-OH VD immunochromatographic strips between batches and within batches are shown in Table 3. The recovery rates of 25-OH VD addition samples at different concentrations in the test strip range from 80.48% to 93.41%, and the coefficient of variation for the test strip ranged from 1.78% to 9.96%. The recoveries of the 25-OH VD addition samples were 85.35%–100.65%, and the coefficient of variation was between 2.39% and 9.35%. The results show that the 25-OH VD QD fluorescent microsphere strip has good precision and accuracy.

We collected 100 human serum samples from Zhengzhou Central Hospital. For comparison with the chemiluminescence immunoassay method, a semi-quantitative commercial ELISA kit was purchased from DiaSorin. The results were presented in Figures 6A, B, the serum results were compared with the results for the control reagents. The regression equation was Y = 1.455X+13.89, and the correlation with DiaSorin was R ² = 0.9438. The positive samples containing a high viral load showed color very quickly (5–10 min) at the T lines; this was statistically significant, and confirms that the results of the tests are in good agreement.