Immunochemical device and lateral-flow immunoassay method for the determination of pharmaceutical residues and contaminants in breast milk

FIELD AND BACKGROUND

[0001] The present invention relates to lateral-flow immunoassays (LFIA) for detection of pharmaceuticals in breast milk. It is also related to fat and protein depletion of milk samples subject to drug residue detection.

[0002] Lateral-flow immunoassays (also known as dipstick assays) are used in pregnancy tests, (roadside) drug tests and in beverage testing. The handled matrices in these applications are aqueous. Breast milk is rich in fat and proteins in contrast to commonly analyzed matrices. Therefore, conventionally configured lateral-flow immunoassays are not applicable.

BRIEF SUMMARY

[0003] According to the suggested method, a breast milk sample is collected with a pre buffered polymer sponge, the collected sample is pretreated by filtration through a graphene oxide layer and the sought for analyte is detected by a competitive LFIA comprising antibody-coated gold nanoparticles as a marker on a nitrocellulose strip that underwent plasma treatment and non-contact spotting in the control and test zones. According to the invention, a sample pre-treatment step has been added to the assay. An integrated sample pad comprising a graphene oxide layer is suggested. Physiological fluids such as breast milk, blood plasma, serum, saliva or urine, can be applied directly to the cartridge. Signal read-out is performed with a hand-held reader or a smartphone with a camera and barcode reader app. A lower detection limit for diclofenac (as an exemplary analyte) in breast milk of 0.08 pg/l and a recovery of about 92 % was demonstrated. The suggested assay is simple, sensitive, fast, and inexpensive.

BRIEF DESCRIPTION OF THE DRAWINGS

[0004] Figs. 1-3 show consumables used for sample collection and filtration. Fig. 4 shows different breast milk samples with different fat content. Figs. 5 and 6 show filtrates of breast milk obtained on different types of filter materials: Fig. 5A shows breast milk before and after filtration through a filter with an integrated graphene oxide layer, Fig. 5B shows breast milk after filtration through the filter analogue without graphene oxide layer, Fig. 5C shows breast milk with high levels of fat and protein after filtration through the filter with the integrated graphene oxide layer, Fig. 5D shows breast milk with low levels of fat and protein after filtration through a filter with an integrated graphene oxide layer. Fig. 6 shows breast milk with different levels of fat and protein after filtration through the filter with an integrated graphene oxide layer. The graph in Fig. 7 shows the UV-visible absorption spectrum of different types of breast milk samples after filtration with the suggested graphene oxide-based filter (SI = Sample 1R2 (low level of fat), S2 = Sample 1R (high level of fat), S3 = Sample 1R (high level of fat) in dilution 1:50, S4 = Sample 1R2 (low level of fat) in dilution 1:50). Fig. 8A and Fig. 8B show an embodiment of a lateral-flow immunoassay device that consists of a test strip and a sample collector. Figs. 9-15 are chromatograms obtained by LC-MS/MS (liquid chromatography-tandem mass spectrometry), indicating the presence of diclofenac (DCF) in all (spiked) samples (Fig. 9 = Sample 1R1, Fig. 10 = Sample 4.1R1, Fig. 11 =

Sample 2R1, Fig. 12 = Sample 1R2, Fig. 13 = Sample 4.2R1, Fig. 14 = Sample 4.2R2, Fig. 15 = Sample 61R1). The samples were filtered through a filter with an integrated graphene oxide layer as suggested, then diluted in Milli-Q water for adjustment to the relevant range of determination of the LC-MS/MS method. Fig. 16a illustrates a top view of the test strip and Fig. 16b a sectional view of the test strip. Fig. 17 illustrates schematically the formation of graphene oxide from graphite by oxidation. Fig. 18a shows the proposed structure of graphene oxide (GO). Fig. 18b shows the proposed structure of the graphene oxide (GO) layer. Fig. 18c is a photograph of graphene oxide 'paper'. Fig. 18d is a photo of a filter with an integrated layer of graphene oxide 'paper'. Fig. 18.1a is a photo of graphene oxide powder. Fig. 18.1b is a photo showing a filter with an integrated graphene oxide powder-based filter between filter layers in the sample pad. Fig. 19 shows photographs of the nitrocellulose (NC) membrane before plasma treatment. Figs. 191 and Fig. 1911 are microscope images of spots obtained by non-contact spotting using a sciFLEXARRAYER S3 (Scienion, Berlin, Germany) spotter to create the test and control line, which are shown in Fig. 19III. Here, protein molecules migrated laterally in the membrane and are focused in a large area. As a result, the image of the spots and the lines are blurred. Fig. 20 shows photographs of the NC membrane after plasma treatment. Fig. 201 and Fig. 2011 are microscope images of spots obtained by non-contact spotting using the sciFLEXARRAYER S3 spotter to create the test and control line, which are shown in Fig. 20III. Plasma treatment increased the binding capacity of the membrane and the proteins are focused in a narrower area resulting in very sharp lines. Fig.

21 (insert) shows a photo of LFIA strips (gold nanoparticles as labels) obtained with varying concentrations of diclofenac (DCF). Fig. 21 (graph) shows the interpolated calibration curve of the assay, signals read with the Chembio Diagnostics LFIA reader. The curve results from plotting intensity of the T-line vs. logarithmic value of DCF concentration. Error bars are standard deviations of three independent measurements. Fig. 22 (insert) shows a photo of LFIA strips (latex particles as labels) obtained with varying concentrations of DCF. Fig. 22 (graph) shows the interpolated calibration curve of the assay, signals read with the Chembio Diagnostics LFIA reader. The curve results from plotting intensity of the T-line vs. logarithmic value of DCF concentration. Error bars are standard deviations of three independent measurements. Fig. 23 shows an embodiment of a multiplex LFIA strip in a cassette comprising three T-lines for different antibiotics (classes: macrolides, fluoroquinolones, penicillins). Fig. 24 shows constituents of a kit for performing an LFIA to analyze pharmaceuticals’ residues in breast milk. It comprises a polymer sponge on a handle for breast milk sample collection, a filter with an integrated graphene oxide layer together with a piston, needed to prepare the layer, and a sample vial for collecting the filtrate. Another constituent is the test strip in its housing (cassette) here shown as: dual cassette for parallel analysis of 2 samples for 1 compound or 1 sample for 2 compounds (left); cassette with 3 T- zones to allow for the analysis of 1 sample for 3 compounds (right).

[0005] In particular, Fig. 1 shows a polymer sponge as used for breast milk sample collection. The upper photograph shows the sponge as obtainable from https://www.porex.com/en-US/markets/in-vitro-diagnostics/sam ple- collection#SalivaUrineCollection. The lower scheme indicates the size of the sponge and its handle.

[0006] Fig. 2 shows a disposable sample vial for collecting the filtrate (on the left) https://www.porex.com/en-US/markets/in-vitro-diagnostics/sam ple-preparation, a filter insert as prepared from a syringe (center) and the piston of the syringe, used to press the filter layers into the lab-made filter holder.

[0007] Fig. 3 shows a filter assembly as prepared according to Example 4.

[0008] Fig. 4 shows a row of six different breast milk samples which illustrates different fat content of breast milk.

[0009] Fig. 5 shows filtrates obtained after filtration of breast milk. In particular, Fig. 5A shows a sample before (above) and after (below) filtration through a graphene oxide layer.

Fig. 5B shows a sample filtrated through the glass microfiber filter layers, omitting graphene oxide. Fig. 5C shows a sample after filtration through the filter of Example 3 (polycarbonate Nucleopore filter). The fat and/or protein load is still too high for LFIA. Fig. 5D shows a sample after filtration over single-layer graphene oxide flakes according to Example 4.

[0010] Fig. 6 shows 5 filtrates from breast milk samples with different fat content after filtration through the graphene oxide filter film according to Example 2. Two milliliters of breast milk were passed through fresh filter layer assemblies as described. The collected volume varies from 200 mΐ to 1 ml depending on the original fat content.

[0011] Fig. 7 shows UV-vis spectra of different samples (SI = Sample 1R2, S2 = Sample 1R, S3 = Sample 1R in dilution 1:50, S4 = Sample 1R2 in dilution 1:50), indicating still high protein content (as indicated by an absorption band at wavelength 280 nm).

[0012] Fig. 8A shows the drop-wise application of the filtrate to the LFIA’s sample port. Usually, 1 drop, corresponding to about 50 mΐ of the sample, has been sufficient to perform the assay. Fig. 8B illustrates the images and design of the lateral flow immunoassay device with positive (left) and negative (right) results.

[0013] The table below shows details of samples, Figs. 9-15 are chromatograms obtained by LC-MS/MS (liquid chromatography-tandem mass spectrometry), indicating the presence of diclofenac (DCF) in all (spiked) samples. Fig. 9 is the chromatogram obtained with sample 1R1, diluted 1:10 according to the table below. Fig. 10 corresponds to sample 4.1R1, 1:50; Fig. 11 corresponds to sample 2R1, 1:10; Fig. 12 corresponds to sample 1R2, 1:20; Fig. 13 corresponds to sample 4.2R1, 1:10; Fig. 14 corresponds to sample 4.2R2, 1:20; and Fig. 15 corresponds to sample 6.1R1, 1:2.

DETAILED DESCRIPTION OF THE INVENTION

[0014] According to an embodiment, a sample pad for a lateral-flow immunoassay (LFIA) is suggested. The sample pad is adapted for separation of a drug residue from the major matrix constituents in a body fluid such as breast milk, blood, plasma, saliva and urine. It comprises two inert filter layers and a graphene oxide layer, which is stacked between the two inert filter layers.

[0015] Advantageously the resulting filter stack allows to safely remove, without loss of the analyte, fat and/or protein load of a sample. Therefore, the sample pad enables even an untrained user to safely apply a freshly obtained sample of, e.g., a body fluid onto an LFIA device without any sample pretreatment.

[0016] According to an embodiment, the two inert layers comprise a micro fibrous glass, comprising pores of a size of 0.5-2 pm, particularly 1.0- 1.2 pm.

[0017] Glass microfibers are widely used for filtration purposes due to their inertness and good flow properties (bubble point).

[0018] According to an embodiment the graphene oxide layer is selected from the list consisting of: a graphene oxide paper consisting of multiple graphene oxide nanosheets, and a graphene oxide powder, comprising flakes consisting of multiple graphene oxide nanosheets.

[0019] According to the invention, graphene oxide has proven to be a good adsorbent for fats and proteins (albumin and immunoglobulins) present, e.g., in breast milk, without altering the concentration of the model drug diclofenac (DCF).

[0020] According to an embodiment, a use of a sample pad comprising a graphene oxide layer for treatment of a sample comprising a physiological fluid immediately before analysis by removal of fat and/or protein from the sample is suggested, wherein the physiological fluid is selected from the list, consisting of: a breast milk, a plasma, a serum, a saliva, a urine, and full blood.

[0021] According to an embodiment, a volume of the sample is 10-100 pi, preferably 25- 75 mΐ, in particular about 50 mΐ (i.e. one drop), wherein a fat load of the LFIA sample as high as 10%, typically 4.4%, does not hamper the detection of an analyte, e.g. a drug such as DCF, in the sample, e.g. a breast milk.

[0022] Advantageously, up to 250 mΐ of the original sample may be put onto the suggested sample pad. [0023] According to an embodiment, thickness of the one graphene oxide layer is between 15-20 pm by AFM (graphene oxide nanosheets), and a graphene oxide powder, comprising flakes consisting of multiple graphene oxide nanosheets between 0.5-0.8 mm.

[0024] Advantageously such a thin layer does not delay the steady stream of liquid along the LFIA strip but is effective in removing accompanying fat and protein from the applied sample.

[0025] According to an embodiment, a lateral-flow immunoassay for detection of an analyte is suggested, wherein the analyte comprises a drug and/or drug residue in a physiological fluid, and the lateral-flow immunoassay comprises filtration of the sample through a graphene oxide layer, the graphene oxide layer being arranged with the sample pad of a lateral-flow immunoassay strip.

[0026] Advantageously such LFIA allows for easy and “on the side” (bedside, roadside) detection of analytes in physiological fluids by untrained personnel, particularly by a lay(wo)man.

[0027] According to an embodiment, the graphene oxide layer is selected from the list consisting of: a graphene oxide paper consisting of multiple graphene oxide nanosheets, and a graphene oxide powder, comprising flakes consisting of multiple graphene oxide nanosheets.

[0028] Advantageously these materials are commercially available or can be easily prepared, e.g. from graphite by oxidation ( cf Fig. 17).

[0029] According to an embodiment, the drug and the drug residue are selected from the list consisting of: carbamazepine (CBZ), caffeine (CAF), cholesterol, diclofenac (DCF), isolithocholic acid (ILA), creatinine, cotinine, atenolol, bisoprolol, metoprolol, carvedilol, celiprolol, esmolol, labetalol, levobunolol, metipranolol, diazepam, dihydrocodeine, doxepine, ibuprofen, methadone, metoprolol, morphine, diazepam, oxazepam, primidone, gentamicine, sotalol and tonalide, cocaine, benzoylecgonine, THC, MDMA, herbicides (glyphosate, atrazine, metolachlor), pesticides (carbofuran, carbaryl), imidacloprid, fenitrothion, clenbuterol, a phthalate ester, a poly- or perfluorinated alkyl substance (PFAS), a bisphenol, an antibiotic penicillin analogue, sulfonamide, tetracycline analogue, fluoroquinolone analogue, macrolide analogue, b-lactam, an alkaloid, and others (e.g. florfenicol, chloramphenicol, lincomycin). [0030] According to an embodiment, a lateral-flow immunoassay device is suggested which comprises a sample pad, wherein the sample pad comprises a graphene oxide layer.

[0031] Advantageously the device comprises a housing, having a bottom and a top cover, wherein the top cover has at least two openings which are arranged over a thin layer or over a membrane being shaped as a strip, i.e. a membrane strip.

[0032] According to an embodiment, the sample pad is pressed by a housing cover onto a membrane strip, which can be made from glass fiber, rayon, polyester, nylon, cellulose, spun polyethylene, or other suitable materials with size from 5 - 8 mm x 18 - 20 mm, and thickness 1 - 3 mm, preferably 1.5 - 2 mm ( cf Fig. 16b). Said sample pad is the place where the sample is applied (added) which will then flow to the NC membrane and be analyzed by flowing along it to the absorbent pad.

[0033] Advantageously, the housing of the device comprising the cover lid and the bottom are clipped or glued together. Thus, a rim surrounding a first opening in the cover lid may be permanently pressed against a membrane stack placed beneath it between the cover lid and the bottom of the housing such as to secure fluidic contact between the different layers, i.e. the lowermost membrane strip and the placed thereon stacked filter layers, comprising a graphene oxide layer. The size of the sample pad comprising the graphene oxide is selected to accommodate within the pores of the membrane stack enough sample for drug detection by LFIA.

[0034] A preferred embodiment of an LFIA device is illustrated in Fig. 8B. The device includes a housing and cover (Fig. 8A) having an inlet port with variable size which extends from the exterior surface of the cover to the interior of the housing for receiving a sample containing the one or more selected analytes to be determined. The inlet port allows the sample to be introduced to a sample receiving device, which is attached to the interior surface of the cover as shown in Fig. 8A. The sample pad may be formed of cotton, glass fiber, rayon, polyester, nylon, cellulose, spun polyethylene, or other suitable materials.

[0035] According to an embodiment the lateral-flow immunoassay is designed as a multiplex assay. Accordingly, the nitrocellulose strip comprises in addition to the control line (C-line) at least two, preferably three, more preferably even four different test lines (T-lines), wherein each T-line comprises a different antigen. Typically, the different T-lines are arranged one after another along the strip (i.e. in flow direction of the liquid in the strip). [0036] Advantageously, applying such design of the test strip, groups of different analytes can be detected at low cost, e.g. if polyclonal antisera containing IgG with altogether broad specificity, are used, and hence allowing to perform a multiplex lateral-flow immunoassay. Remarkably, the graphene oxide used for sample filtration did not change the concentration of antibiotics as different as macrolides, fluoroquinolones, and penicillins (cf. Fig. 23).

[0037] According to another embodiment, a method for detection of drug residues in a body fluid is proposed. The method comprises the following steps:

Collecting a sample of typically up to 2 ml with a polymer sponge by soaking the polymer sponge in the body fluid, wherein the polymer sponge has been treated beforehand in a 25 mM borate buffer solution, wherein the buffer solution comprises 0.01 M EDTA, 0.05 % Triton X-100, 0.01 % NaN3, and 0.05 % Tween 20, and dried;

Filtrating the collected sample through a graphene oxide layer, which is provided either as “Graphene Oxide Film - Super Paper” (www.ac smaterial . com/graphene-oxide-film- super-paper- 1061.html) or as “Single Layer Graphene Oxide Flake”, H Method ( w w w . ac smaterial . com/single-layer- graphene-oxide-h-method- 1001.html or www .cheaptubes .com/product/single-layer- graphene-oxide- 1 -20um/) between two glass microfiber filters (e.g., 1.2 pm pore size, Whatman 1822-100 on top, and 1.0 pm pore size, Whatman 1821-100, at the bottom of a filter sandwich); wherein a thickness of the graphene oxide is 15-20 pm for obtaining a filtrate, wherein the graphene oxide powder, comprises flakes which consist of multiple graphene oxide nanosheets of a lateral size between 0.5-0.8 mm;

Performing a competitive lateral-flow immunoassay with about 50 mΐ (typically 1 drop) of the filtrate as obtained in the previous step; and

Determining an amount of the residue of the drug in the sample using a calibration curve.

[0038] According to an embodiment, the sponge during the collecting step is typically soaked into the body fluid for 1-5 min.

[0039] Advantageously the short time allows for easily handling by a lay(wo)man without any additional equipment (e.g. clock or timer) and quick completion of the test. [0040] According to an embodiment the body fluid is breast milk and the drug whose residual concentration is detected is selected from diclofenac and an antibiotic, the antibiotic being selected from the classes macrolides, fluoroquinolones, and penicillins.

[0041] The detrimental effect of diclofenac and of antibiotics on infants is well known. Therefore, preventing intoxication is highly desirable.

[0042] The terms “graphene,” “graphene oxide,” “graphene oxide nanosheet,” and “graphene nanosheet” describe two-dimensional carbon structures and are used interchangeably throughout the present specification.

[0043] Each embodiment described above may be combined with any other embodiment or embodiments unless clearly indicated to the contrary.

[0044] A full and enabling disclosure of the present invention, including the best mode thereof, to one of ordinary skill in the art, is set forth more particularly in the remainder of the description, including reference to the accompanying figures, which form a part hereof, and in which we show, by way of illustration, specific embodiments and features of the invention. It is to be understood that other embodiments may be utilized and structural or logical changes may be made without departing from the scope of the present invention. The following detailed description, therefore, is not to be taken in a limiting sense, and the scope of the present invention is defined by the appended claims.

[0045] Methods used for the determination of pharmaceutical compounds (drugs) in breast milk are used in research labs or specialized clinical labs. A well-established instrumental method is high-performance liquid chromatography (HPLC), especially in combination with mass spectrometry (HPLC-MS). However, its major drawback is the requirement of expensive equipment and trained personnel. The method is not adequately sensitive and rugged and therefore requires intensive sample preparation which makes it costly and time-consuming. Therefore, it is not suitable for the screening of a large number of samples and utterly unsuitable for use under domestic conditions by a lay(wo)man, in a physician’s examination room, or in a pharmacist’s lab.

[0046] LFIA, i.e. pregnancy test-like assays, are most valuable as home, bedside or roadside assay systems due to the non- sophisticated procedure, low cost, evaluation by the naked eye, easy operation, and their small size. Yet, milk and especially, breast milk, is a highly complex matrix with significant amounts of lipids and proteins; thus, the determination of trace residues of pharmaceuticals and their metabolites requires extensive sample extraction and preparation prior to analysis. Such prevents its analysis by the lay(wo)man, moreover if a more precise value (and not merely a yes/no answer) has to be obtained.

[0047] In LFIA, a liquid sample or its extract containing the sought for analyte moves along a strip of a membrane thereby passing various zones where molecules (“antigens”) have been attached that exert more or less specific interactions with the analyte. Thus, a typical LFIA format comprises a surface layer to carry the sample from the sample pad via the conjugate release pad along the strip encountering the detection zone up to the absorbent pad. Current membrane strips often comprise nitrocellulose but may be fabricated as well from nylon, polyether sulfone, polyethylene or even fused silica.

[0048] The technical object of the described embodiments is therefore to provide an efficient method for sample pretreatment, particularly for fat and protein depletion of breast milk samples. Further, it is an object to provide reliable selection criteria for a membrane material suitable as LFIA strips and a method for plasma pretreatment, especially of nitrocellulose. Furthermore, it is an object of the invention to provide a system, comprising an LFIA kit consisting of a filter, which is adapted to a dropper for taking a fixed amount of the filtrated sample, and a plastic housing for the strip, wherein the housing carries a QR code. The QR code is adapted for directing a user to a webpage with a download link for a smartphone app, wherein the app will read the test lines and return the quantitative results to the user for interpretation, which allows reproducible read-out of LFIA results by the very smartphone.

[0049] A barcode reader that is adapted to read a barcode on the test/test packaging is utilized to retrieve calibration curve data held in the portal. For traceability of results, a user log-in is suggested. Users are required to log in to the app and their results after logging in are linked to their own account.

[0050] According to an embodiment the nitrocellulose strip can comprise up to 4 different test lines (T-lines), i.e. can be designed as a multiplex assay. Thus, advantageously, up to 4 different tests can be run simultaneously and recorded. For example, four antibiotics from different groups (e.g. macrolides, fluoroquinolones, and penicillins) can be detected in the sample simultaneously using one single strip. Thus, for instance, on one strip, 3 compounds (azithromycin conjugate, ciprofloxacin conjugate and amoxicillin conjugate) were spotted onto the NC membrane as test lines (see Fig. 23), and a goat anti-rabbit immunoglobulin as the control line. Rabbit anti-macrolides IgG, rabbit anti- fluoroquinolones IgG, rabbit anti-penicillin IgG were labeled with gold nanoparticles or latex particles and placed onto the conjugate pad. Advantageously, not merely one particular analyte (e.g. an antibiotic) but whole groups of antibiotics can be detected simultaneously from a single sample, if polyclonal sera, containing IgG that make up for group selectivity, for example, the macrolides group, are used.

[0051] A mandatory timer (e.g. 10 min) that must elapse before users are able to record a test result (a skip feature available during development), will be included to aid usability and reduce the chance of false negatives and false positives. Reading the results with the use of the smartphone app helps to obtain correct diagnostic values in a home environment.

[0052] The App (i.e. mobile application, also referred to as a mobile app, is a computer program or software application designed to run on a mobile device such as a phone, tablet, or a “smart” watch) contains a database of drugs and dietary supplements that may affect breastfeeding. It includes information on the levels of such substances in breast milk and infant blood, and possible adverse effects in the nursing infant. The App helps to track the nursing progress. It may further contain supporting information with photos and video clips, basic information and requirements around breastfeeding (how to breastfeed, how to collect the breastmilk and how to store it, diet during the breastfeeding, etc.). Furthermore, scanning the QR code successfully allows the database curator to track the website and app activity through appropriate software tools.

[0053] The selection and modification of the membranes in the absorbent pad and the membranes of the strips in the lateral-flow device have been optimized in a way that it can cope with the breast milk matrix. The viscosity of the breastmilk is an important property, which greatly influences the flow rate. As we observed, membranes with a slower flow rate can distinguish the difference in fluid viscosity more effectively than membranes with a faster flow rate. The membranes produced the following capillary rise times for a 4 cm travel distance along the membrane: MDI (www.mdimembrane.com) CNPC 8 pm: 220 sec, CNPC 12 pm: 110 sec, CNPH 150 pm: 90 -150 sec. For a travel distance of 12 mm on the LFIA device, the time difference between running times for breastmilk and deionized water is around 5 sec on CNPC 8 pm, 8 sec on CNPC 12 pm and 13 sec on 150 CNPH. CNPC 8 pm membrane is too slow. The type CNPH 150 pm without plasma treatment gave a large image for the T line (with BSA conjugate spotted on the test line, cf Fig. 19) because the amount of detector particles is focused in a large area. A suitable NC membrane is CNPC 12 pm in combination with an absorbent pad comprising AP 110 (MDI) which results in an increase of the flow rate of 15 sec for the complete length of the strip. Moreover, by this, the sensitivity of the assay could be increased which allows for more reliable results. Nitrocellulose (NC) membranes bind proteins electrostatically through interaction of the strong dipole of the nitrate ester with the strong dipole of the peptide bonds of the protein. Plasma treatment in ammonia and ammonia-hydrogen mixtures for membrane CNPC 12 mhi (Fig. 19) has been employed to modify the surface of nitrocellulose membranes. The plasma treatment showed to be effective to increase the binding capacity of the NC membrane by enhancing the total amount of grafted amino groups, since H atoms produced in the plasma discharge could transform grafted nitro groups into amino groups by a reduction process.

EXAMPLES

[0054] Example 1 - Preparation of sample collection sponge

[0055] A polyester sponge as shown in Fig. 1 was modified with a 25 mM sodium borate buffer solution, pH 8.5 (established by adding HC1) comprising 0.01 M EDTA, 0.05 % Triton X-100. The sponge was immersed in the indicated solution for 2 minutes. Afterwards, the sponge was dried overnight at room temperature in a desiccator over silica gel or calcium sulfate (Drierite™, Sigma- Aldrich) and stored until use at room temperature in a closed box.

[0056] Example 2 - Preparation of sample pretreatment filter A

[0057] Filters for sample pretreatment, i.e. cell, protein and fat depletion, were prepared from the tip part of 10 ml BD Difco™ Discardit™ Luer-slip two-piece syringes by cutting off their front end with a sharp knife ( cf Fig. 3). A circular piece of a glass microfiber filter with 1.0 pm pore size (Whatman 1821-100) of appropriate size was placed at the bottom of the front end and pressed with the syringe piston. Then, a layer of Graphene Oxide (GO) Film - Super Paper" (www.acsmaterial.com/graphene-oxide-film-super-paper-1061.ht ml) was pressed against the first layer. On top of the GO layer another glass microfiber filter with 1.2 pm pore size (Whatman 1822-100) of appropriate size was tightly pressed with the piston.

[0058] Example 3 - Preparation of sample pretreatment filter B

[0059] Different to Example 2, between the two glass microfiber filters with 1.0 pm and 1.2 pm pore size, a polycarbonate Whatman® Nuclepore™ Track-Etched Membrane, pore size 12.0 pm, was used instead of the graphene oxide film.

[0060] Example 4 - Preparation of sample pretreatment filter C

[0061] Different to Example 2, between the two glass microfiber filters with 1.0 pm and 1.2 pm pore size, a layer of about 0.5-0.8 mm "Single Layer Graphene Oxide Flake", H Method (www.acsmaterial.com/single-layer-graphene-oxide-h-method-10 01.html) was used instead of the graphene oxide film.

[0062] Example 5 - Filtration of diclofenac- spiked breast milk samples

[0063] According to practical example 5, fat and protein depletion by filtration of original breast milk samples which were spiked with diclofenac was performed according to the table below. The samples were spiked gravimetrically with different DCF concentrations (see table). The samples were stored at -20 °C until their analysis or stored at 4 °C until their analysis within 3 days.

[0064] A woman’s diet can influence her breast milk composition, composition changes dynamically within a feeding, with time of day, over lactation, and between mothers and populations. Normally, mature human breastmilk contains 3%-5% fat, 0.8%-0.9% protein, 6.9%-7.2% carbohydrate calculated as lactose, and 0.2% mineral constituents expressed as ash. Its energy content is 60-75 kcal/100 ml. The level of fat in the breast milk was determined by the Rose-Gottlieb Method (International Dairy Federation. 1987. Milk. Determination of fat content - Rose Gottlieb gravimetric method, IDF, Brussels, Belgium). In this work, low level of fat is < 2% and high level > 5%.

[0065] The samples were spiked with different concentrations of DCF and then diluted in Milli-Q water for adjustment to the relevant range of determination of the FC-MS/MS method. A dilution of the sample as high as 1:50 did not hamper the recovery. Rather a lower dilution (1:2) appears to deteriorate the recovery for DCF concentrations, perhaps due the limit of determination (FOD) of FC-MS/MS, which is 5-10 pg/l (see sample # 7 with DCF concentration below 5 pg/l). It is recommended, with high fat (but < 10% fat) level of breast milk, to dilute the sample at least 1:2 for the LC-MS/MS method, to avoid contamination of the chromatographic system.

[0066] The primary mouse monoclonal anti-DCF IgG antibodies were kindly provided by Prof. Dr. Dietmar Knopp (Technische Universitat Miinchen). 40 nm gold nanoparticles (BioReady™ 40 nm Carboxyl Gold, product number AUXR40) have been used as label. They were obtained from NanoComposix (https://nanocomposix.com/)· These GNPs are functionalized with a tightly bound monolayer containing terminal carboxylic acid functional groups which can be activated through EDC/NHS chemistry. Carboxyl surfaces were used to covalently bind free amino groups of antibodies to the surface of the particles. Estapor® carboxyl-modified dyed microspheres from Millipore (product number Kl-030) have been used as label. Microspheres based on latex with the size 0.3 pm were also used. Antibodies were covalently coupled to the surface of carboxylated latex using EDC/NHS chemistry.

[0067] The nitrocellulose (NC) membrane used was Laminate Type 150 CNPH, from MDI (w w w . mdimembrane .com), treated in a plasma. Their characteristic parameters are:

PHYSICAL TESTS

Polyester Thickness (pm): 102 Membrane Thickness (pm): 103

FUNCTIONAL TESTS

Wicking Rate, 4cm Distance (seconds) (Blocked): 122

Wicking Rate, 4cm Distance (seconds) (Normal Saline): 145 Dot spreading with 10 pi Solution (seconds): 3-5

Adhesion Test (24 hours in H2O): Passed Test

DIMENSIONAL TESTS

Membrane Width (mm): 25 Dimensions of PVC (mm): 58 x 260 Color of PVC: White

Thickness of PVC (pm): 250 PVC (poly vinyl chloride) was the material for the backing on which different parts of the lateral-flow immunoassay (NC membrane, absorbent pad and sample pad) are assembled (using a suitable adhesive).

[0068] The plasma treatment set-up for the NC membrane consists mainly of a glass reactor with a non- symmetrical configuration of electrodes. Low frequency power (50-75 kHz), provided by an industrial generator, is capacitively coupled to the blade-type electrode. The NC membrane with the PVC backing (58x260 mm) is enrolled on the grounded cylinder. The real exposure time of the polymer to the discharge is calculated from the total duration of plasma discharge and the plasma width on the membrane. For a total plasma treatment duration of 1 s, the calculated real time is equal to 0.03 s. The gas is introduced through mass flow controllers (MKS Instruments, Inc.) and the pressure is monitored with an MKS capacitive gauge.

[0069] The general structure of the used LFIA strips and the geometrical properties of the pads used in the LFIA are shown in Fig. 16.

[0070] Practical results obtained for DCF detection in breast milk using the described LFIA are indicated in the table below.

[0071] To summarize, some aspects of the embodiments described herein can be briefly described as follows:

1. We claim a method for lay(wo)man analysis of breast milk from a subject;

2. We claim that pharmaceutical residues and contaminants can be sensitively determined in a breast milk sample; 3. We claim an included technical measure by which the sample is treated to facilitate the release of the analyte from a binding protein or a binding agent;

4. We claim that graphene oxide is used to reduce the antigen binding to particles used in the LFIA strip;

5. We claim a plasma treatment of the nitrocellulose membrane that provides uniform sample flow on the nitrocellulose substrate;

6. We claim the use of a QR bar code printed on or attached to either the top or the bottom housing, which gives the test strip a unique code for use and initiates download of the smartphone app for reading and result evaluation and transmission to a hosting service;

7. We claim a housing top where two or more extended wells are accommodated for processing multiple samples;

8. We claim that the subject, from which the breast milk stems, is a human, or any other lactating mammal;

9. We claim, as carrier for the antibodies, a gold nanoparticle, but it can also be a quantum dot, a core/shell polymer or polymer/silica particle, or a metal-organic framework particle;

10. We claim the use of antibodies capable of binding the analyte (pharmaceutical or pollutant molecule);

11. We claim a lateral-flow test device to be used in combination with a sample collector for breast milk;

[0072] The present invention has been explained with reference to various illustrative embodiments and examples. These embodiments and examples are not intended to restrict the scope of the invention, which is defined by the claims and their equivalents. As is apparent to one skilled in the art, the embodiments described herein can be implemented in various ways without departing from the scope of what is invented. Various features, aspects, and functions described in the embodiments can be combined with other embodiments.