*2.1. Reactants*

Lincomycin hydrochloride monohydrate (LIN), HAuCl4, sodium azide, sodium citrate, Tween-20, Triton X-100, and 4-MBA were obtained from Sigma-Aldrich (St. Louis, MO, USA). N-(3-dimethylaminopropyl)-N-ethylcarbodiimide hydrochloride (EDC) and sulfo-N-hydroxysuccinimide (NHS) were supplied from Fluka (Buchs, Switzerland). Goat antibodies against mouse immunoglobulins (GAMI) were purchased from Arista Biologicals (Allentown, PA, USA). Bovine serum albumin (BSA) was supplied from Eximio Biotec (Wuxi, China). The CdSe/ZnS QDs with an emission peak at 625 nm were obtained from Invitrogen (Catalog No A10200, Thermo Fisher Scientific, Waltham, MA, USA). All other reagents were of analytical grade.

Ultrapure water (Millipore Corporation, Burlington, MA, USA) with resistivity of 418.2 M Ω was used to prepare the AuNPs and their conjugates as well as LIN stock solutions (100 μg/mL). The LFIAs were carried out in 96-well transparent Costar 9018 polystyrene microplates provided by Corning Costar (Tewksbury, MA, USA). Amicon Ultra-0.5 mL Centrifugal Filter (100 K) was purchased from Millipore (Billerica, MA, USA).

#### *2.2. Preparation of Monoclonal Anti-LIN Antibodies*

A synthesis of the LIN–BSA conjugate and a preparation of anti-LIN antibodies were carried out in accordance with the procedure described in the study by Cao et al. [25].

#### *2.3. Synthesis and Characterization of AuNPs*

AuNPs with an average diameter of 30 nm and 40 nm were prepared according to the citrate-reduction method [26]. To obtain 30 nm AuNPs, 1 mL of 1% HAuCl4 was added to 97.5 mL of ultrapure water and heated to boiling. After that, 1.5 mL of 1% sodium citrate was added immediately to the boiling solution during vigorous stirring. The mixture was left to boil for 25 min and then cooled. The colloidal AuNPs were stored at 4 ◦C.

To obtain AuNPs with an average diameter of 40 nm, 1.5 mL of 1% sodium citrate was added to 100 mL of boiling 0.01% HAuCl4 aqueous solution under rapid agitation. The solution was then boiled for another 15 min and cooled to room temperature.

The transmission electron microscopic (TEM) images were recorded with a JEM-100C electron microscope (JEOL, Tokyo, Japan) operating at 80 kV. The AuNP preparations were applied to 300-mesh grids (Pelco International, Redding, CA, USA) coated with formvar film. The images obtained were analyzed using Image Tool software (University of Texas Health Science Center, San Antonio, TX, USA). UV–vis absorption spectra were obtained through spectrophotometer UV-2450 (Shimadzu, Kyoto, Japan).

#### *2.4. Conjugation of Antibodies to AuNPs*

Antibody–AuNPs conjugates were prepared according to the previously described technique [27]. Anti-LIN antibodies were dialyzed against a Tris-HCl bu ffer (10 mM, pH 8.5), and added to AuNPs at a concentration of 10 μg/mL (OD520 = 1). The mixture was incubated for 45 min while stirring at room temperature. BSA in the final concentration of 0.25% was further added to this preparation, followed by stirring for 15 min. The excess reagents were removed by centrifugation at 9500× *g* for 15 min, followed by resuspension of the antibody–AuNPs pellet in Tris bu ffer (10 mM, pH 8.5) with 1% BSA, 1% sucrose, and 0.1% sodium azide (TBSA).

#### *2.5. Conjugation of Antibodies with QDs*

Anti-LIN antibodies were dialyzed against a borate bu ffer (50 mM, pH 8.7). The molar ratio of QDs to anti-LIN antibodies during synthesis was 1:2. Antibodies (300 μL, 0.2 mg/mL), QDs (25 μL, 8 μM), and freshly prepared EDC and NHS solutions (50 μL, 0.8 mM each) were mixed. After incubation for 90 min in a dark place at room temperature, the resulting mixture was purified by centrifugation at 10,000× *g* for 15 min using Amicon Ultra 100 kDa tubes (Billerica, MA, USA).

The centrifugation was repeated four times, and, finally, 14 μL of QDs with a concentration of 4.26 mg/mL was obtained.

#### *2.6. Synthesis of the Raman Reporter Bioprobe*

To a solution of 40 nm diameter AuNPs (10 mL), 10 μL of 1 mM 4-MBA in ethanol was added [28]. The mixture was incubated for 3 h, followed by centrifugation at 5000× *g* for 15 min. The resulting pellet was resuspended in water.

To prepare the AuNPs–MBA–Ab bioprobe, Au–MBA conjugate and anti-LIN antibodies were adjusted to pH 8.9 with 0.1 M K2CO3. Anti-LIN antibodies at a concentration of 10 μg/mL were added to 2 mL of Au–MBA and incubated for 3.5 h at room temperature. Then, 50 μL of 10% BSA was added and incubated overnight at 4 ◦C. After that, the mixture was centrifuged at 9000× *g*, for 10 min. The pellet was resuspended in an equal volume of water and stored at 4 ◦C.

#### *2.7. Preparation of Test Strips*

The schemes of three LFIA formats are shown in Figure 1. Test strips were assembled using MdiEasypack membrane sets (Advanced Microdevices, Ambala Cantt, India) comprising the following elements: a plastic support, a CNPC nitrocellulose working membrane with a pore size of 15 μm, a PT–R7 conjugate fiberglass pad (in case of conventional AuNP- and QD-based LFIAs), a GFB-R4 sample pad, and an AP045 absorbent pad. The control line was formed by applying 0.5 mg/mL GAMI in a K-phosphate bu ffer (PBS, 50 mM, pH 7.4, with 0.1 M NaCl) by an Iso-Flow automatic dispenser (Imagene Technology, Hanover, NH, USA). To form a test line, LIN–BSA conjugate (0.5 mg/mL—for AuNPs-based, 0.15 mg/mL—for QD-based, and 0.2 mg/mL—for SERS-based LFIAs, in PBS) was applied. After that, the test strips were dried at 37 ◦C for 2 h. For AuNP-based LFIA, the antibody-AuNPs conjugate in TBSA containing 0.05% Tween-20 was applied to the conjugate pad and dried at room temperature overnight. For QD-based LFIA, 1 μL of antibody–QDs conjugate (0.09 μM) in a borate bu ffer (BB, 0.05 M with 1% BSA, 0.1% sucrose, and 0.1% sodium azide, 0.05% Tween-20) was applied to the interface of the sample pad and nitrocellulose membrane and dried at room temperature overnight. Finally, the assembled multimembrane composites were cut into individual test strips 3 mm wide using an automatic guillotine cutter (Index Cutter-1, A-Point Technologies, Gibbstown, NJ, USA).

### *2.8. LFIA Procedures*

#### 2.8.1. Colorimetric and Fluorescent LFIAs

Solutions of LIN (1 μg/mL–1 pg/mL) in PBST (100 μL) were dripped onto the microplate wells. The test strips were vertically placed into the well and left to react for 15 min. The color intensity (in the case of AuNP-based LFIA) of the formed bands was scanned by the CanoScanLiDE 90 (Canon, Tokyo, Japan). The fluorescence intensity (for QD-based LFIA) was recorded under UV light excitation. The obtained images were then digitized using the TotalLab program (Nonlinear Dynamics, Newcastle upon Tyne, UK).

**Figure 1.** Schemes of lateral flow immunoassay (LFIA) formats developed in the study: conventional colorimetric gold nanoparticle (AuNP)-based LFIA (**a**); fluorescent quantum dot (QD)-based LFIA (**b**); AuNP-based LFIA with surface-enhanced Raman scattering (SERS) detection (**c**).

### 2.8.2. SERS-Based LFIA

To perform the SERS-based LFIA, 2 μL of AuNPs–MBA–Ab bioprobe was pipetted onto the sample pad approximately 1 cm below the nitrocellulose membrane, and 100 μL of LIN (100,000–1 × 10−<sup>8</sup> ng/mL) in PBS containing 0.05% Triton X-100 (PBST) was added into the microplate wells. The test strips were vertically inserted into the wells and left to react for 15 min. Then, the Raman spectra from 10 points along the middle of the test line were collected using a DXR Raman microscope (Thermo Fisher Scientific, Madison, WI, USA). The SERS settings were selected with the identical registering technique [29,30]. All spectra were obtained under the same conditions: The excitation source was tuned at 780 nm and laser power of 20 mW; the exposure time was 10 s. A 10× objective lens (NA = 0.25) was used to focus a laser spot on the surface of the test strip.

#### **3. Results and Discussion**

#### *3.1. Synthesis and Characterization of Signal Markers*

AuNPs were used as a reporter label in conventional and SERS-based LFIAs. AuNPs of a diameter close to 30 nm were reported to be optimal for traditional immunochromatography [31], whereas larger particles are preferable in SERS-based LFIAs. The optimal size of AuNPs for the preparation of a SERS-active probe was previously found to be no more than 50 nm [32]. Therefore, to achieve a desirable sensitivity and high reproducibility in SERS-based LFIAs, AuNPs with an average size of 40 nm were preferred. To prepare AuNPs with diameters of 30 and 40 nm, a simple method of sodium citrate-associated reduction of chloroauric acid was applied. According to this method, the size of the resulting AuNPs is varied by adding di fferent amounts of the reducing agent: to obtain larger particles, a smaller volume of reducing agen<sup>t</sup> is required. The selection of 4-MBA as a Raman reporter molecule stems from the widespread use of thiol-containing aromatic molecules because of their ability to conjugate directly to the gold surface and provide surface carboxyl groups for biomolecule binding [33]. To optimize the composition of the AuNPs–MBA–Ab bioprobe, the amount of added 4-MBA was varied in the range of 10–50 μL per 10 mL of AuNPs. It was demonstrated that an excess of 4-MBA could cause the aggregation of AuNPs, as evidenced by a red-shifted absorbance peak in UV–vis spectra and a color change of the AuNPs–MBA–Ab probe (data not shown). Therefore, 10 μL of 1 mM MBA was proven to be su fficient for preparation of a stable AuNPs–MBA conjugate.

The size and shape of AuNPs were estimated by TEM and UV–vis spectroscopy (Figure 2). The as-prepared AuNPs showed localized surface plasmon resonance at 523 and 527 nm for AuNPs of 30 and 40 nm, respectively. After the conjugation process, the slight redshift of the maximum peak of AuNPs–MBA–Ab was observed, which indicates a successful conjugation of Au–MBA and anti-LIN antibodies. The TEM images revealed spherical morphology and homogeneity with a size distribution in the range of 29.5 ± 7.4 nm and 39.5 ± 5.0 nm, and a degree of ellipticity of 1.3 for two AuNPs preparations (Figure 2b,c). According to the manufacturer, the size of carboxyl quantum dots varies from 15 to 20 nm.

**Figure 2.** (**a**) UV–vis spectra of AuNPs of 30 nm (**a**) and 40 nm (**b**) diameter, and AuNPs functionalized with 4-mercaptobenzoic acid (4-MBA) and coupled with anti-lincomycin (LIN) monoclonal antibodies (AuNPs–MBA–Ab) (**c**); (**b**) microphotograph of AuNPs for colorimetric LFIA. The average diameter is 29.5 ± 7.4 nm and the degree of polydispersity is 1.3; (**c**) microphotograph of AuNP–MBA for SERS-based LFIA. The average diameter is 39.5 ± 5.0 nm and the degree of polydispersity is 1.3.

### *3.2. AuNP-Based LFIA*

Given the need to detect a low molecular weight compound in this study, a direct competitive LFIA format was performed (Figure 1). In this assay, free LIN present in the sample competed with the immobilized LIN–BSA conjugate in regard to binding with specific anti-LIN antibodies. The binding sites of the specific anti-LIN antibodies labeled with di fferent nanodispersed markers were first occupied with the target analyte. And thereafter the excess labeled antibodies were captured by the LIN-BSA conjugate, which in turn was detected by employing di fferent detection techniques. Thus, the signal intensity on the test line of the strip was inversely proportional to the concentration of LIN in the sample. The preliminary characterization of the immune properties of monoclonal anti-LIN antibodies used by ELISA confirmed their high a ffinity (Figure S1) and allowed for the development of LFIAs.

The scheme of the conventional AuNP-based LFIA is presented in Figure 1a. For the LFIA, LIN–BSA conjugate, and GAMI were applied to form test and control lines on the working membrane, respectively. The specific antibody-labeled AuNPs was immobilized on the fiberglass pad. The assay conditions were optimized to achieve the lowest detection limit at a high amplitude of the analytical signal. As a result, the following conditions were found to be optimal for three formats of assay: 0.5 mg/mL—for AuNP-based, 0.15 mg/mL—for QD-based, and 0.2 mg/mL—for SERS-based LFIAs (the concentration varies from 0.2 to 1 mg/mL) and 0.5 mg/mL for GAMI (the concentration varies from 0.15 to 0.5 mg/mL). The AuNPs-anti-LIN antibodies solution was then applied to the conjugate pad at the concentration corresponding to OD520 = 1 (we tested OD520 in the range from 0.5 to 2.5). The overall performance of the LFIA was explored by varying the concentration of the analyte (from 1000 to 0.001 ng/mL). Under optimal experimental conditions, the AuNP-based LFIA exhibits linearity over the range of 0.7–7.2 ng/mL with an instrumental detection limit of 0.4 ng/mL (Figure 3). The cuto ff was 10 ng/mL with the assay duration of 15 min.

**Figure 3.** Calibration curve of LIN in the AuNP-based LFIA and the digital photographs of the LFA strips after conventional AuNP-based LFIA procedure. The LIN-BSA conjugate was applied at the test line at a concentration of 0.5 mg/mL. The AuNPs-anti-LIN antibodies solution was applied at a concentration corresponding OD520 = 1. LIN concentrations are given at the bottom of the test strips. The error bars indicate the standard deviations for three measurements.
