This article is licensed under CC-BY 4.0 pubs.acs.org/ac Article Paper-Based Analytical Devices for Accurate Assessment of Transferrin Saturation in Diagnosed Clinical Samples from Ischemic Stroke Patients Silvia Dortez, Núria DeGregorio-Rocasolano, Mònica Millán, Teresa Gasull, Agustín G. Crevillen,* and Alberto Escarpa* Cite This: Anal. Chem. 2023, 95, 12391−12397 Read Online ACCESS Metrics & More Article Recommendations * sı Supporting Information ABSTRACT: For the first time, a paper-based analytical device (PAD) was developed for the assessment of transferrin saturation (TSAT), which is defined as the ratio between iron bound to transferrin (Tf) and the total iron-binding capacity (TIBC) of Tf. Both parameters were simultaneously measured on the same PAD using ferrozine as a chromophore and a smartphone as the color reader. To this end, Tf was first isolated from serum using anti-Tf immunomagnetic beads to ensure that only the Tf-bound iron was measured, improving the selectivity and accuracy of TSAT assessment. To demonstrate the practical utility of the device, it was validated by analyzing a certified reference material, showing excellent accuracy (Er < 4%) and good precision (RSD ≤ 6%). Finally, 18 diagnosed serum samples from ischemic stroke patients were analyzed by this approach, and the results were compared with those obtained by urea-PAGE, showing not only an excellent correlation (r = 0.93, p < 0.05) but that the PAD approach has become statistically identical to the free-interference urea-PAGE. In comparison with the slow, tedious, and non-miniaturized-PAGE, this PAD approach exhibited attractive characteristics such as low cost, disposability, and connectivity, showing great potential for future point-of-care testing, especially in developing countries and/or remote areas, where access to medical or clinical facilities is limited. T ransferrin (Tf) is a serum glycoprotein and the main iron transport source between body tissues. It can transport up to two ferric cations (Fe3+), existing three different Tf indirect methods. Regarding the direct methods, an additional stage is necessary concerning serum iron assessment, which is the addition of excess ferric iron to the serum sample to forms: apoTf (no Fe3+), monoferric-Tf, and diferric-Tf. The saturate all the free sites on Tf. Then, it is necessary to remove ratio between the amount of ferric iron bound to Tf and the the excess unbound ferric iron by using an adsorbent agent,1 maximum amount of iron that Tf can capture (total iron- but some alternatives avoid this removal step.8,9 Furthermore, binding capacity, TIBC) is defined as the percentage of TIBC can be also calculated by measuring Tf concentration, transferrin saturation (TSAT). This clinical parameter reflects considering that each mol of Tf can bind 2 mol of ferric iron.10 body iron states; therefore, it is used for diagnosing both iron Regarding indirect methods, TIBC can be calculated by the deficiency and iron overload in combination with other serum sum of serum iron concentration and unsaturated iron-binding biomarkers such as ferritin, serum soluble Tf receptor, and/or capacity (UIBC),11 which is defined as the additional amount hepcidin.1−4 Moreover, a high level of TSAT has been of ferric iron needed to fully saturate Tf (TIBC = serum iron + proposed as a biomarker of the risk for various diseases, such UIBC). as cardiovascular disease, diabetes mellitus, and cancer,1,2 and However, these methods are not reliable in all clinical cases even several clinical and preclinical studies suggest that brain because they assume that all serum iron is bound to Tf,12 damage induced by ischemia is increased by a previous systemic iron overload.5−7 TSAT is typically calculated by the ratio between serum iron Received: May 8, 2023 and TIBC (TSAT = [serum iron/TIBC] × 100). Both Accepted: July 12, 2023 parameters are routinely measured by colorimetric approaches Published: July 24, 2023 in clinical laboratories. For instance, serum iron is directly measured using ferrozine or bathophenanthroline chromo- phores.1,8 However, TIBC can be measured using direct or © 2023 The Authors. Published by American Chemical Society https://doi.org/10.1021/acs.analchem.3c01982 12391 Anal. Chem. 2023, 95, 12391−12397 Analytical Chemistry pubs.acs.org/ac Article showing differences in the TSAT value up to 35% among some purchased from Spinreact (Girona, Spain). Anti-Tf antibody commercial methods.13 For example, they fail during intra- (ab66952) was purchased from Abcam (Cambridge, UK). venous or oral iron supplementation treatments14,15 because in Anonymized serum samples of the multicenter, randomized, this situation the level of non-Tf-bound iron is extremely high double-blind, placebo-controlled TANDEM-17 (Thrombolysis (up to 10 μM when TSAT > 70%).16 An interesting alternative and Deferoxamine in Middle Cerebral Artery occlusion) study is the direct determination of TSAT by urea-PAGE (it allows were used to evaluate TSAT by PAD and compared with the the separation and detection apoTf, monoferric-Tf, and urea-PAGE method (as explained in DeGregorio-Rocasolano diferric-Tf) because it has proven to be more accurate and et al.18). The serum samples used were specifically those reliable than the aforementioned methods.15,17,18 However, the collected at the Hospital Universitari Germans Trias I Pujol PAGE technique is slow, tedious, and lacks automation and (HUGTP) in untreated patients. All samples were stored at miniaturization, which hamper its use in decentralized −20 °C until use. scenarios. The TANDEM-1 study was approved by the Spanish Drug In this sense, the development of new bioanalytical Agency (eudraCT 2007-0006731-31) and by local Ethics approaches that allow for the accurate determination of Committees including the HUGTP Ethics Committee and was TSAT and with potential use and development of future registered in clinicalTrials.gov as NCT00777140. point-of-care tests (POCTs) is highly needed. Indeed, low-cost A stock solution of 2 M acetate buffer pH 4.8 was prepared POCTs are an interesting and relevant alternative for by dissolving appropriate amounts of acetic acid and sodium determining TSAT, especially to facilitate the diagnosis of acetate in Milli-Q water. A stock solution of 2 M Tris buffer iron deficiency anemia in developing countries3 and/or when a pH 7.6 was prepared in Milli-Q water. A stock solution of 0.1 fast decision must be taken (such as the evaluation of the M phosphate buffer saline (PBS) pH 7.6 was prepared by growth and extent of stroke damage).18 To the best of our dissolving appropriate amounts of sodium dihydrogen knowledge, there is currently no POCT for TSAT assessment. phosphate, disodium hydrogen phosphate, sodium chloride, POCT is a top priority for the World Health Organization, and potassium chloride in Milli-Q water. The solution of especially, in resource-limited settings.19 Among POCT hydroxylamine was prepared by dissolving appropriate technologies, the combination of smartphone-based and amounts in 2 M acetate buffer of pH 4.8 to a concentration paper-based assays is attracting enormous attention from the of 100 mg mL−1. The solution of ferrozine was prepared by scientific community because the smartphone provides dissolving appropriate amounts in 2 M acetate buffer of pH 4.8 straightforward colorimetric detection, portability, and con- to a concentration of 49 mg mL−1. Iron (III) standard nectivity, and paper provides extremely low-cost analytical solutions were daily prepared by dissolving appropriate platforms, which are characteristics highly demanded by amounts in 100 mg mL−1 hydroxylamine solution (prepared developing countries.20,21 in 2 M acetate buffer of pH 4.8) or in 2 M Tris buffer of pH In the designing and developing of a novel POCT, paper- 7.6 to a concentration of 1 mg mL−1. based analytical devices (PADs) offer several advantages such All reagents and solvents were of analytical grade. All as inexpensive production, portability, operational simplicity, solutions were prepared in Milli-Q water (Merck Millipore, miniaturization, biocompatibility, minimal reagent consump- Darmstadt, Germany). tion, disposability, and capillary flow. 22−24 There are Whatman Chromatography Paper 1 CHR was purchased interesting examples of (μ)PADs with smartphone-based from Merck (Darmstadt, Germany). An Invitrogen Dyna- detection for serum iron25−27 and serum ferritin quantification MagTM-2 Magnet was purchased from Thermo Fisher (lateral flow assay)28 but, to our best knowledge, there is no Scientific (USA). The waterproof marker pen Lumocolor article reporting on TSAT assessment using a PAD approach. permanent CD/DVD/BD 310 was purchased from Staedtler In this work, we propose a new approach to measure TSAT, (Nuremberg, Germany). Tesa 4024 clear packing tape and which is based on a colorimetric PAD with smartphone Samtian F40II Lightbox with LED were purchased from reading using anti-transferrin immunomagnetic beads (anti-Tf- Amazon (Spain). MBs). The approach was validated using a certified reference Instrumentation. AutoCAD 2018 (Autodesk, Student material (human serum), and then it was applied to the Version) was used to design the PAD, and they were drawn analysis of diagnosed serum samples from ischemic stroke on a sheet of filter paper (Whatman 1 CHR) using a desktop patients. These results were compared with those obtained by cutting plotter (Silhouette Cameo 3, Silhouette). a well-established free-interference method (urea-PAGE). Photos were taken using a Realme X2 mobile and a lightbox (dimensions 44 × 44 × 44 cm, including 84 brightness LED ■ MATERIALS AND METHODS Reagents and Materials. Iron (III) chloride hexahydrate, light beads inside). Subsequent image analysis was performed with ImageJ software. Procedures. PAD Design and Fabrication. Whatman 1 ferrozine (3-(2-pyridyl)-5,6-diphenyl-1,2,4-triazine-p,p′-sul- CHR was used as the filter paper because it is hydrophilic, fonic acid monosodium salt hydrate), hydroxylamine hydro- homogeneous, reproducible, biocompatible, and cheap, among chloride, acetic acid glacial, and sodium acetate anhydrous others.29 were purchased from Merck (Darmstadt, Germany). Tris was To draw the pattern of the PAD, AutoCAD software was purchased from Roche (Basel, Switzerland). Potassium used. The pattern was transferred to a piece of filter paper chloride and disodium hydrogen phosphate were purchased using a cutter plot, in which the blade was replaced by a from Scharlau (Barcelona, Spain). Sodium dihydrogen waterproof marker pen. Then, the waterproof marker pen ink phosphate was purchased from PanReac (Barcelona, Spain). penetrated the paper to form the hydrophobic walls on the Pierce Protein G Magnetic Beads and sodium chloride were filter paper. Once the PADs were drawn on the filter paper, acquired from Thermo Fisher Scientific (USA). Human serum they were individually cut using the cutter plot (the blade was (certified reference material, Spintrol H normal 1002121) was reinstalled), and it was checked whether the hydrophobic walls 12392 https://doi.org/10.1021/acs.analchem.3c01982 Anal. Chem. 2023, 95, 12391−12397 Analytical Chemistry pubs.acs.org/ac Article Figure 1. Strategy for the determination of TSAT using anti-Tf-MBs and the PAD-smartphone sensor. were well-created with the waterproof marker pen on both (i) Immunoreaction: 5 μL of anti-Tf-MBs was resuspended sides of the device. Next, the backside of the printing surface in 150 μL of serum sample (which contained Tf) and was covered with clear packing tape to prevent the solution this vial was incubated at 25 °C and 950 rpm for 45 min from leaking out underneath the PAD.27 (Eppendorf ThermoMixer C, Hamburg, Germany). The device consisted of six circular zones (detection (ii) Cleaning: the vial was placed on the magnet holding reservoirs, 10 mm diameter) at the ends of six channels (10 block for 2 min, and then, the supernatant was removed. mm length and 5 mm width), in a radial configuration where Two washing steps were carried out with 100 μL of 0.1 only the circular zones were used, assuring that all reservoirs M PBS solution of pH 7.4. are at the same distance from the point of light. This fact is very important since in this application, two different (iii) Reduction of Fe3+ to Fe2+: Tf-anti-Tf-MBs were parameters (Tf-bound iron and TIBC) were measured resuspended in 15 μL of 100 mg mL−1 hydroxylamine (reduction agent), prepared in 2 M acetate buffer of pH simultaneously, which allows the accurate assessment of the 4.8. This solution was incubated for 30 min in a TSAT as a ratio of both. Furthermore, there was another thermoshaker at 25 °C and 950 rpm, releasing iron from detection reservoir at the junction of the six channels. The Tf and reducing Fe3+ to Fe2+. Finally, the vial was placed selected device was selected for the described novel application on the magnet holding block for 2 min, and 5 μL of considering that it is one of the most ubiquitous designs in the supernatant was transferred onto the reservoir of the literature, the aspect that would facilitate other researchers to PAD. adopt the described methodology without any significant changes. For TIBC assessment, besides previous steps, a new one Ferrozine was used as a colorimetric agent because it is (iron saturation) was carried out between (ii) and (iii) (see Figure S1): commonly used for the colorimetric iron assay.30−32 The (ii, b) Iron saturation: Tf retained by anti-Tf-MBs was optimal conditions of this chromogenic reagent were chosen saturated with iron (III) using 10 μL of 10 μg mL−1 iron (III) based on our previous article.27 0.5 μL of 49 mg mL−1 standard solution in 2 M Tris buffer pH 7.6 and incubated for ferrozine in 2 M acetate buffer of pH 4.8 was added to each 15 min, at 25 °C and 950 rpm. After that, the vial was placed detection reservoir of PAD, and it was dried completely in an on the magnet holding block for 2 min, and then, the oven at 60 °C for 2 min. Then, the PAD was ready to use. supernatant was removed. Two washing steps were carried out Immunopurification Step for Tf-Bound Iron Isolation. with 100 μL of 0.1 M PBS solution of pH 7.4. Anti-Tf-MBs were used for Tf isolation in serum samples and TSAT Colorimetric Assay Using PAD. The PAD contains were prepared according to the literature.33,34 This step seven detection reservoirs: three were used for Tf-bound iron allowed us to isolate Tf-bound iron from the rest of the iron measurement, the other three for TIBC assessment, and the (III) species present in serum, ensuring that we only measured last for the blank (100 mg mL−1 hydroxylamine in 2 M acetate iron bound to Tf in the following steps.27 buffer of pH 4.8). 12393 https://doi.org/10.1021/acs.analchem.3c01982 Anal. Chem. 2023, 95, 12391−12397 Analytical Chemistry pubs.acs.org/ac Article Figure 2. Color intensity (green channel) using the immunopurification step and the smartphone-PAD sensor: (A) At different concentrations of 10 μL of Fe3+ in 2 M Tris buffer of pH 7.6 during 15 min for Tf saturation. (B) At different Tf saturation times using 10 μL of 10 μg mL−1 Fe3+ in 2 M Tris buffer of pH 7.6. (C) At different volumes of 10 μg mL−1 Fe3+ in 2 M Tris buffer pH 7.6 solution during 15 min (n = 3). For other experimental conditions see the Immunopurification step for the Tf-bound iron isolation section. The assay was as follows: 5 μL of each sample was added to Then, the assessment of TSAT was carried out by a parallel the PAD and then, the paper turned purple after 10 min of and simultaneous colorimetric measurement on the same PAD reaction between Fe2+ and ferrozine. The reaction was carried of two parameters: (i) the Tf-bound iron was present out at room temperature. The intensity of the purple color was physiologically in the sample and (ii) TIBC (after the directly proportional to the iron concentration. saturation of Tf with Fe3+). The methodology for Tf-bound Images of the PAD were taken by a smartphone (Realme iron assessment was previously developed by our group,27 so X2). The whole imaging process was performed inside a we first develop a method for TIBC assessment, and then both professional photography box, where the smartphone was on methodologies were combined for designing a novel approach top. In addition, the PAD was placed 22 cm below the for TSAT assessment (see eq 1) using unique diagnosed serum smartphone, keeping the same reservoir−smartphone camera samples from ischemic stroke patients. distance.27 TIBC is defined as the maximum amount of iron (III) that The images were analyzed by ImageJ software in the red− saturates all serum Tf. Our methodology for TIBC assessment blue−green color format. The green channel was selected consists of four steps: (i) Tf isolation from serum sample using because it provides the maximum intensity due to it being the anti-Tf-MBs; (ii) saturation of the isolated Tf with a Fe3+ complementary color to the purple of the colored product of solution; (iii) release of Fe3+ from isolated Tf and reduction to the reaction of ferrozine with iron.27 Next, the image was Fe2+; and (iv) determination of Fe2+ using the colorimetric inverted (pure white and black backgrounds were considered PAD. to be of 0- and 255-pixel intensity, respectively), and the The concentration of Fe3+ solution for Tf saturation (TIBC detection zones for each reservoir of the PAD were selected assessment) was first studied. Solutions of 5, 10, and 20 μg individually (circular diameter of 60 pixels) to calculate the mL−1 Fe3+ in 2 M Tris buffer pH 7.6 solution were selected to mean intensity of the selected area, subtracting the blank value. saturate a solution of 2.03 g L−1 Tf (after immunopurification). Finally, TSAT was calculated by dividing the concentration The results showed that the highest color intensity was of Tf-bound iron by TIBC (μg mL−1). obtained using 10 μg mL−1 Fe3+, so it was chosen as the best [Tf bound iron] iron concentration to saturate Tf for the rest of the TSAT (%) = × 100 experiments (see Figure 2A). The next study consisted of [TIBC] (1) optimizing the reaction time to saturate Tf with 10 μg mL−1 Calibration Curves. External calibration linear plots were Fe3+ in 2 M Tris buffer pH 7.6 solution. Reaction times of 15, constructed using Fe3+ standard solutions. 10 μL of the Fe3+ 30, and 45 min were studied. No significant differences were standard solution was mixed with 10 μL of 100 mg mL−1 obtained between the different times, so 15 min was chosen hydroxylamine solution (in 2 M acetate buffer pH 4.8) in a (see Figure 2B). Finally, the added volume of 10 μg mL−1 Fe3+ vial. Then, it was kept under agitation for 15 min at 25 °C and solution was also optimized. Volumes of 10, 50, and 100 μL 950 rpm using a thermoshaker. In this process, Fe3+ was were assayed. As for Tf saturation time, no significant reduced to Fe2+. Next, 5 μL the previous mixture was added differences were obtained between the different volumes. For onto the detection zone of the PAD, where the paper turned this reason, 10 μL was selected as the optimum volume for Tf purple after 2 min of the reaction between Fe2+ and ferrozine saturation (see Figure 2C). (at room temperature). A blank (100 mg mL−1 hydroxylamine Analytical Performance. Then, the analytical perform- in 2 M acetate buffer of pH 4.8) was also prepared and ance of the PAD was carefully evaluated. Linear calibration analyzed in the same way. Finally, the images were taken and plots were constructed by analyzing Fe3+ standard solutions processed in the same way as in the previous section (TSAT from 5 to 40 μg mL−1 with three different PADs (see Figure colorimetric assay using PAD). S2). As Figure S2 inset shows, the intensity of the color formed ■ RESULTS AND DISCUSSION Analytical Design and Optimization of the PAD- depended on the iron concentration. A complete calibration plot was constructed using a single PAD. The calibration plot for iron was linear in the range of 5−40 μg mL−1, showing an Based Approach for TSAT Assessment. Figure 1 shows the excellent correlation coefficient (r = 0.997). The calibration analytical strategy for the assessment of the clinical parameter slope was 0.77 ± 0.03 px mL μg −1 and the intercept was 3.8 ± TSAT. First, Tf was isolated from the serum sample using anti- 0.6 px (n = 3). In addition, the reproducibility of calibration Tf-MBs to ensure that only iron bound to Tf is measured. slopes using different PADs was very good (RSD = 3%). 12394 https://doi.org/10.1021/acs.analchem.3c01982 Anal. Chem. 2023, 95, 12391−12397 Analytical Chemistry pubs.acs.org/ac Article However, the limit of detection (LOD) was 2 μg mL−1 (calculated as 3sa/b, where sa is the standard deviation of the intercept), which is not enough to detect iron in human serum (normal range from 0.7 to 1.7 μg mL−1).35 However, this is not a problem when serum samples were analyzed because an immunopurification step was implemented for the isolation of Tf using anti-Tf-MBs,33,34 yielding a 10 times preconcentration (Tf present in 150 μL serum is concentrated in a 15 μL solution during the immunopurification). In this sense, the LOD of the overall method was reduced from 2 to 0.2 μg mL−1. Moreover, this step ensures that only Fe3+ bound to Tf is measured, avoiding the interference of non-Tf-bound iron Figure 3. Stability of colorimetric PAD (in plastic zipper bags at room and improving the accuracy of TSAT evaluation. temperature and protected from light). Experimental conditions: 5 μL of 500 μg mL−1 Fe3+ standard solution prepared in 100 mg mL−1 Table 1 lists the results obtained for the determination of Tf- hydroxylamine solution (2 M acetate buffer pH 4.8) and 2 min bound iron and TIBC in a certified reference material (human reaction time (n = 3). Table 1. Analysis of Certified Reference Material by PADa certified value parameter (μg mL−1) PAD value (μg mL−1) Er (%) RSD x̅ ± s (%) serum iron 1.09 1.05 ± 0.01 1 4 TIBC 2.85 2.8 ± 0.2 6 2 a Values are expressed as mean value ± SD (n = 3). serum) (n = 3). In addition, both parameters were measured on the same PAD, demonstrating its multiplexing capability and highlighting their practical use (see Figure S3). Based on these results, our method showed excellent accuracy and Figure 4. Correlation between TSAT values obtained by PAD and precision. urea-PAGE (n = 3). Experimental condition: see the section On studying these data thoroughly, the quantification values Immunopurification Step for Tf-Bound Iron Isolation. for serum iron and TIBC were excellent and reproducible, although always below the certified value. This might be because our method only measured the iron bound to Tf, 0.93). More importantly, both methods did not show while the reference values were obtained by measuring all Fe3+ significant differences at the 95% level since the values in the serum (serum iron). However, we must bear in mind obtained for slope (0.94 ± 0.09) and the intercept (2 ± 3), that this work aims to evaluate TSAT, so the Fe3+ not bound to included 1 and 0, respectively, revealing an excellent accuracy Tf must be not considered. This event would improve the since the urea-PAGE method is a free-interference method. accuracy of the TSAT assessment. If TSAT is calculated for the On the other hand, the urea-PAGE method separates and certified reference material (using eq 1, these data are not detects the different forms of Tf (apoTf, monoferric-Tf, and provided in the certificate), the value is 38.2%, which is slightly diferric-Tf), so it measures the TSAT directly without higher than that given by our method (37.6%). considering the iron not bound to Tf, but it takes 18 h. Our The storage stability of the PADs was also evaluated for approach takes less than 2 h and showed not only an excellent practical purposes. To this end, several PADs were prepared correlation but a high agreement in terms of accuracy with the and stored in plastic zipper bags at room temperature and free-interference urea-PAGE even at high TSAT (see Table S1, protected from light. The color intensity generated in these S17, and S18 samples), meaning that our strategy, based on the devices (three different PADs) was measured over time. Figure immunopurification of Tf and the colorimetric PAD, only 3 represents the percentage variation of the color intensity over measures the Tf-bound iron. This contrasts with results time concerning the color intensity obtained on the first day. provided by other colorimetric methodologies that rely on the After 40 days, there were no significant differences in color measurement of total iron serum (bound and not bound to intensity compared with the first day (test t, p < 0.05, two Tf). sides, n = 3), so the PADs are stable for at least 40 days under Moreover, current methods for TSAT assessment employ the selected storage conditions. automatic benchtop equipment,7,11 which is expensive and Assessment of TSAT in Serum Samples from an located in clinical laboratories or medical centers. To our best Ischemic Stroke Patient. To demonstrate the applicability of knowledge, there is no PAD or POCT for TSAT assessment the proposed PAD in clinical samples, TSAT was measured in reported in the literature, so our PAD approach is the pioneer 18 anonymized serum samples from ischemic stroke patients of in this application and, interestingly, a potential candidate for a the TANDEM-1 study. The samples were analyzed following future POCT. Indeed, the use of a processed sample (serum) our approach and the obtained TSAT values were compared and the multistep procedure of our approach does not meet with those obtained by the urea-PAGE method18 (see Table the user-friendly feature required in POCT devices yet;24 S1). In addition, Figure 4 shows the correlation plot between however, it meets other important features such as being both sets of data, exhibiting a good correlation coefficient (r = affordable, sensitive, specific, and deliverable to end-users. 12395 https://doi.org/10.1021/acs.analchem.3c01982 Anal. Chem. 2023, 95, 12391−12397 Analytical Chemistry pubs.acs.org/ac Article Moreover, our PAD can be used by any skilled health 28805 Alcala de Henares, Madrid, Spain; orcid.org/ professional and even blood samples may be analyzed instead 0000-0002-1999-5605 of serum, thanks to the use of anti-Tf-MBs (not assayed). Núria DeGregorio-Rocasolano − Cellular and Molecular Neurobiology Research Group, Department of Neurosciences, ■ CONCLUSIONS A new approach to accurately assess TSAT was successfully Germans Trias I Pujol Research Institute (IGTP), 08916 Badalona, Barcelona, Spain developed by measuring the Tf-bound iron and TIBC on a Mònica Millán − Department of Neurociences, Germans Trias single PAD using a smartphone as a color reader. Thanks to its I Pujol University Hospital, Universitat Autònoma de multiplexing capabilities, Tf-bound iron and TIBC assays were Barcelona, 08916 Badalona, Barcelona, Spain simultaneously measured on the same PAD, eliminating the Teresa Gasull − Cellular and Molecular Neurobiology inter-PAD variability and the variability due to changes in Research Group, Department of Neurosciences, Germans external factors such as light intensity, temperature, humidity, Trias I Pujol Research Institute (IGTP), 08916 Badalona, and smartphone model, among others, that can affect the Barcelona, Spain; orcid.org/0000-0002-9321-1741 analysis. Complete contact information is available at: Importantly, the use of anti-Tf-MBs allowed us to measure https://pubs.acs.org/10.1021/acs.analchem.3c01982 only the Tf-bound iron, avoiding inaccuracies in TSAT evaluation, even at high values due to the presence of non- Author Contributions Tf-bound iron. The manuscript was written through the contributions of all Our approach was validated by analyzing a certified the authors. All the authors have approved the final version of reference material, showing excellent accuracy (recoveries > the manuscript. 96%) and good precision (RSD ≤ 6%). Moreover, it was Funding successfully applied to the analysis of serum from ischemic This work has been financially supported by the TRANS- stroke patients, showing a very good correlation and excellent NANOAVANSENS program from the Community of Madrid accuracy with a reference method (urea-PAGE). Despite the (P2018/NMT-4349) (A.E.), by the grant PID2020- analysis time (about 1:30 h.), it was a distinctive feature 118154GB-I00 funded by MCIN/AEI/ 10.13039/ compared to the urea-PAGE, which takes 18 h. 501100011033 (A.E.), by the RICORS RD21/0006/0024 Even more exciting are the opportunities our approach and 2021SGR00925 (T.G.), and by the Spanish Ministry of opens up. It has several interesting features such as low cost, Economy and Competitiveness (CTQ2017-86441-C2-1-R, disposability, and connectivity to be the potential basis of a FPI fellowship (S. D.)). future POCT for TSAT assessment, making it very attractive Notes for use in developing countries or remote areas, where access The authors declare no competing financial interest. to medical or clinical facilities is restricted. ■ ASSOCIATED CONTENT ■ ACKNOWLEDGMENTS S.D. acknowledges the FPI Fellowship from the Spanish * sı Supporting Information Ministry of Economy and Competitiveness. The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acs.analchem.3c01982. Scheme of the analytical reactions used for determi- ■ ABBREVIATIONS Anti-Tf-MBs anti-transferrin immunomagnetic beads nation of TSAT; image of a PAD used for the calibration LOD limit of detection curve (with the corresponding calibration plot) and MBs immunomagnetic beads multiplexing analysis; and table with TSAT values of μPAD microfluidic paper-based analytical device serum samples from ischemic stroke patients obtained PAD paper-based analytical device by a PAD and urea-PAGE (PDF) PBS phosphate buffer saline POCT point-of-care testing ■ AUTHOR INFORMATION Corresponding Authors px Tf TIBC pixel transferrin total iron-binding capacity Agustín G. Crevillen − Department of Analytical Sciences, TSAT transferrin saturation percentage Faculty of Sciences, Universidad Nacional de Educacion a UIBC unsaturated iron-binding capacity Distancia (UNED), 28040 Madrid, Spain; orcid.org/ 0000-0002-4470-6502; Email: agustingcrevillen@ ccia.uned.es ■ REFERENCES (1) Elsayed, M. E.; Sharif, M. U.; Stack, A. G. Adv. Clin. Chem. 2016, Alberto Escarpa − Department of Analytical Chemistry, 75, 71−97. Physical Chemistry and Chemical Engineering and Chemical (2) Szoke, D.; Panteghini, M. Clin. Chim. Acta 2012, 413, 1184− 1189. Research Institute “Andrés M. Del Río” (IQAR), University of (3) Lopez, A.; Cacoub, P.; Macdougall, I. C.; Peyrin-Biroulet, L. Alcala, 28805 Alcala de Henares, Madrid, Spain; Lancet 2016, 387, 907−916. orcid.org/0000-0002-7302-0948; (4) Camaschella, C. Blood Rev. 2017, 31, 225−233. Email: alberto.escarpa@uah.es (5) Castellanos, M.; Puig, N.; Carbonell, T.; Castillo, J.; Martinez, J. M.; Rama, R.; Dávalos, A. Brain Res. 2002, 952, 1−6. Authors (6) García-Yébenes, I.; Sobrado, M.; Moraga, A.; Zarruk, J. G.; Silvia Dortez − Department of Analytical Chemistry, Physical Romera, V. G.; Pradillo, J. M.; Perez De La Ossa, N.; Moro, M. A.; Chemistry and Chemical Engineering, University of Alcala, Dávalos, A.; Lizasoain, I. Neurochem. Int. 2012, 61, 1364−1369. 12396 https://doi.org/10.1021/acs.analchem.3c01982 Anal. Chem. 2023, 95, 12391−12397 Analytical Chemistry pubs.acs.org/ac Article (7) Millán, M.; Degregorio-Rocasolano, N.; Pérez de la Ossa, N.; Reverté, S.; Costa, J.; Giner, P.; Silva, Y.; Sobrino, T.; Rodríguez- Yáñez, M.; Nombela, F.; Campos, F.; Serena, J.; Vivancos, J.; Martí- Sistac, O.; Cortés, J.; Dávalos, A.; Gasull, T. Antioxidants 2021, 10, 1270. (8) Yamanishi, H.; Kimura, S.; Iyama, S.; Yamaguchi, Y.; Yanagihara, T. Clin. Chem. 1997, 43, 2413−2417. (9) Yamanishi, H.; Iyama, S.; Yamaguchi, Y.; Kanakura, Y.; Iwatani, Y. Clin. Chem. 2003, 49, 175−178. (10) Gambino, R.; Desvarieux, E.; Orth, M.; Matan, H.; Ackattupathil, T.; Lijoi, E.; Wimmer, C.; Bower, J.; Gunter, E. Clin. Chem. 1997, 43, 2408−2412. (11) Strzelak, K.; Rybkowska, N.; Wiśniewska, A.; Koncki, R. Anal. Chim. Acta 2017, 995, 43−51. (12) Frank, C.; Rienitz, O.; Jährling, R.; Schiel, D.; Zakel, S. Metallomics 2012, 4, 1239−1244. (13) Eleftheriadis, T.; Liakopoulos, V.; Antoniadi, G.; Stefanidis, I. Ren. Fail. 2010, 32, 1022−1023. (14) Scheiber-Mojdehkar, B.; Lutzky, B.; Schaufler, R.; Sturm, B.; Goldenberg, H. J. Am. Soc. Nephrol. 2004, 15, 1648−1655. (15) Kitsati, N.; Liakos, D.; Ermeidi, E.; Mantzaris, M. D.; Vasakos, S.; Kyratzopoulou, E.; Eliadis, P.; Andrikos, E.; Kokkolou, E.; Sferopoulos, G.; Mamalaki, A.; Siamopoulos, K.; Galaris, D. Haematologica 2015, 100, e80−e83. (16) Angoro, B.; Motshakeri, M.; Hemmaway, C.; Svirskis, D.; Sharma, M. Clin. Chim. Acta 2022, 531, 157−167. (17) Agarwal, R. Kidney Int. 2004, 66, 1139−1144. (18) DeGregorio-Rocasolano, N.; Martí-Sistac, O.; Ponce, J.; Castelló-Ruiz, M.; Millán, M.; Guirao, V.; García-Yébenes, I.; Salom, J. B.; Ramos-Cabrer, P.; Alborch, E.; Lizasoain, I.; Castillo, J.; Dávalos, A.; Gasull, T. Redox Biol. 2018, 15, 143−158. (19) Drain, P. K.; Hyle, E. P.; Noubary, F.; Freedberg, K. A.; Wilson, D.; Bishai, W. R.; Rodriguez, W.; Bassett, I. V. Lancet Infect. Dis. 2014, 14, 239−249. (20) Karim, K.; Lamaoui, A.; Amine, A. J. Pharm. Biomed. Anal. 2023, 225, 115207. (21) Vashist, S. K.; Luppa, P. B.; Yeo, L. Y.; Ozcan, A.; Luong, J. H. T. Trends Biotechnol. 2015, 33, 692−705. (22) Martinez, A. W.; Phillips, S. T.; Butte, M. J.; Whitesides, G. M. Angew. Chem., Int. Ed. 2007, 46, 1318−1320. (23) Noviana, E.; Ozer, T.; Carrell, C. S.; Link, J. S.; McMahon, C.; Jang, I.; Henry, C. S. Chem. Rev. 2021, 121, 11835−11885. (24) Yamada, K.; Shibata, H.; Suzuki, K.; Citterio, D. Lab Chip 2017, 17, 1206−1249. (25) Shrivas, K.; Monisha; Kant, T.; Karbhal, I.; Kurrey, R.; Sahu, B.; Sinha, D.; Patra, G. K.; Deb, M. K.; Pervez, S. Anal. Bioanal. Chem. 2020, 412, 1573−1583. (26) Serhan, M.; Jackemeyer, D.; Long, M.; Sprowls, M.; Diez Perez, I.; Maret, W.; Chen, F.; Tao, N.; Forzani, E. IEEE J. Transl. Eng. Heal. Med. 2020, 8, 2800309. (27) Dortez, S.; Crevillen, A. G.; Escarpa, A. Talanta 2023, 253, 123914. (28) Srinivasan, B.; O’Dell, D.; Finkelstein, J. L.; Lee, S.; Erickson, D.; Mehta, S. Biosens. Bioelectron. 2018, 99, 115−121. (29) Carrilho, E.; Martinez, A. W.; Whitesides, G. M. Anal. Chem. 2009, 81, 7091−7095. (30) Stookey, L. L. Anal. Chem. 1970, 42, 779−781. (31) Mori, L.; Bekkering, A.; Traini, J.; Vanderlinden, L. Clin. Chem. 1981, 27, 1441−1444. (32) Smith, G. L.; Reutovich, A. A.; Srivastava, A. K.; Reichard, R. E.; Welsh, C. H.; Melman, A.; Bou-Abdallah, F. J. Inorg. Biochem. 2021, 220, 111460. (33) Sierra, T.; Crevillen, A. G.; Escarpa, A. Biosens. Bioelectron. 2021, 179, 113098. (34) Sierra, T.; Henry, C. S.; Crevillén, A. G.; Escarpa, A. Anal. Sens. 2022, 2, No. e202100038. (35) Breuer, W.; Hershko, C.; Cabantchik, Z. I. Transfus. Sci. 2000, 23, 185−192. 12397 https://doi.org/10.1021/acs.analchem.3c01982 Anal. Chem. 2023, 95, 12391−12397