The colorimetric paper-based device using lipase inhibition assay was developed for rapid and visual detection of orlistat in weight loss supplements. The paper device with five circular detection zones was simply fabricated from filter paper No 1 using a low-cost paper craft puncher with a design of a flower-like shape. The enzymatic reaction on the detection zone employed a small volume of a substrate, a-naphthyl acetate, a sample, and a lipase enzyme. After incubation, the Fast Blue B solution was used as a chromogenic reagent. The decrease in purple color can be observed by the naked eye, in the presence of orlistat. Under optimized conditions, the paper device showed satisfactory sensitivity and selectivity. The device was applied for rapid screening of orlistat in weight loss supplements and the results agreed with those obtained from TLC analysis.

paper-based device, orlistat, lipase inhibition, weight loss supplements

Overweight and obesity are recognized as the most common metabolic disease worldwide. Anti-obesity drugs and alternative treatments such as herbal slimming formulations and weight loss supplements may be used as part of a weight-control program. One of the approved weight management medications by the Food and Drug Administration (FDA) in many countries is orlistat, a semisynthetic derivative of lipstatin. It prevents the absorption of fats from the human diet by acting as a potent and specific inhibitor of intestinal lipase. The use of orlistat seems to have beneficial effects on blood pressure and carbohydrate metabolism. However, several mild-to-moderate gastrointestinal adverse effects, such as oily stools, diarrhea, abdominal pain, and fecal spotting, and a few cases of serious hepatic adverse effects have been reported (European Medicines Agency 2012). The absorption of certain fat-soluble vitamins (such as A, D, and E) and many drugs (such as warfarin, amiodarone, ciclosporin, and thyroxine as well as fat-soluble vitamins) can be affected by orlistat, affecting their bioavailability and effectiveness (Liu et al. 2020)

Owing to its potential use for the treatment of overweight and obesity, orlistat is reported to be one of the undeclared, unapproved pharmaceutical ingredients found in weight loss supplements. Being unaware of taking the drug ingredient in dietary supplements, overuse, interactions with other medications, or with underlying health conditions could cause serious adverse health effects (Calahan et al. 2016; Tucker et al. 2018). The detection of drug adulteration in dietary supplements is therefore important. Various analytical approaches for the detection of synthetic adulterants in herbal medicine based on chromatographic and spectroscopic techniques such as high-performance liquid chromatography (HPLC) with ultraviolet (UV) and diode array detectors (DAD), ion mobility spectrometry (IMS), high-performance, thin-layer chromatography (HPTLC), liquid chromatography-tandem mass spectrometry (LC-MS/MS), nuclear magnetic resonance (NMR) analysis, and hyphenated-mass spectrometric techniques are still widely used (Pratiwi et al. 2021). These methods are sensitive and selective but require expensive lab instruments and the expertise of an analyst. HPLC and TLC have been reported for the determination of orlistat in pharmaceutical formulations and herbal products (Ahamad et al. 2016; Joshi et al. 2017; Fathy et al. 2019).

Recently, the paper-based analytical device has received considerable attention as a tool for drug analysis. This device offers several advantages, such as simplicity, low cost, rapid analysis, low consumption of reagents and samples, and portable analytical tools capable of on-site drug screening (Novian et al. 2020; Pratiwi et al. 2021). A paper-based device has been developed for the determination of adulterated drugs, dexamethasone and prednisolone in traditional medicine, and sibutramine in slimming products (Primpray et al. 2019; Karamahito et al. 2021). The most widely used method for detecting substances on paper is the colorimetric method based on rapid chemical and enzymatic-based reactions (Garnier and Then 2013). Color visualizations of orlistat on TLC plates with universal chemical staining reagents such as phosphomolybdic acid and p-anisaldehyde and the enzymatic colorimetric detection of orlistat based on its potent and specific lipase inhibitory activity were reported (Hassan 2012; Bayineni et al. 2014; Tang et al. 2016; Fathy et al. 2019). In this study, a low-cost colorimetric paper-based device for rapid screening of orlistat based on the enzymatic inhibitory reaction was developed. The color change upon the enzyme-substrate reaction with and without orlistat on the paper device is observed and captured using a scanner and analyzed by the image analysis software.

A low-cost colorimetric paper-based device for the detection of orlistat based on a lipase inhibition assay was developed. In this study, a-naphthyl acetate was used as the enzyme substrate because it is inexpensive and can be catalyzed by lipase to yield a-naphthol, which can subsequently react with FBB to give a distinct purple-colored diazonium dye (Fig. 1) (Hassan 2012). In the presence of orlistat, a potent lipase inhibitor, the reduction of purple color intensity was observed by the naked eye and with image analysis software.

Figure 1. 

Reaction of lipase with a-naphthyl acetate and the subsequent formation of the purple color with Fast Blue B.

The effect of the paper type on the observation of color intensity was studied. In general, the same grades of filter and chromatography paper produced comparable results in terms of signal intensity and uniformity (Evans et al. 2014). The filter paper grades W1, W3 and W4 were utilized in the optimization study because they were less expensive than the chromatography paper. Each paper grade has unique characteristics, especially the thickness. The thickest filter paper is W3 while W1 and W4 are characterized by similar thicknesses and qualitative grades.

The results showed that the thicker paper (W3) yielded poorer color intensity; better intensity was obtained with thinner papers (W1 and W4) (Fig. 2). According to Evans et al. 2014, a reason for this is that the colorimetric detection method is mainly sensitive to the compounds present on the surface of the paper device. Due to being thicker and more opaque, the paper device made from W3 yielded poor color intensity. W1 showed better color intensity and uniformity than W4, possibly due to the smaller pores resulting in slower wicking speed, which increased reaction efficiency and resulted in a more noticeable color observation. Therefore, W1, the most popular and the least expensive filter paper was employed for the following studies.

Figure 2. 

Color intensity of the paper device made from filter paper grades W1, W3 and W4.

The intensity of the purple color in the detection zone was affected by different concentrations of lipase enzyme (2.5, 5 and 10 mg/mL), a-naphthyl acetate (0.75, 1.0 and 1.5 mg/mL), and FBB (0.5 and 1.0 mg/mL). Good color intensity and uniformity were observed with the use of enzyme concentration at 5 and 10 mg/mL, substrate concentration at 1.0 and 1.5 mg/mL, and FBB solution at 0.5 and 1.0 mg/mL (Fig. 3a–c). The cases which showed good color intensity were further investigated with the presence of orlistat at 0.02 mg/mL to observe the change of purple color reduction on the paper device as compared to the case used without orlistat. The results showed that the reduction in purple color in the presence of orlistat was not that obvious when a high concentration of lipase at 10 mg/mL (L10) was employed (Fig. 4). The case with orlistat at low concentrations but using the excess amount of lipase and substrate still produced good purple color intensity. Thus, the optimal conditions for visual detection of orlistat by an enzymatic colorimetric paper-based device were 5 mg/mL lipase, 1.5 mg/mL a-naphthyl acetate, and 1.0 mg/mL FBB solution. A scheme of the procedure for the determination of orlistat using the paper device is shown in Fig. 5.

Figure 3. 

(a) Effect of lipase concentrations (2.5, 5 and 10 mg/mL) with 3 µL of 1.5 mg/mL a-naphthyl acetate and 45 µL of 1.0 mg/mL FBB. (b) Effect of a-naphthyl acetate (NA) concentrations (0.75, 1.0 and 1.5 mg/mL) with 3 µL of 5 mg/mL lipase and 45 µL of 1.0 mg/mL FBB. (c) Effect of FBB concentration (0.5 and 1.0 mg/mL) with 3 µL of 1.5 mg/mL a-naphthyl acetate and 3 µL of 5.0 mg/mL lipase.

Figure 4. 

Color intensity and the relative percentage change (rel. DI%) of color intensity with and without orlistat at various conditions on the paper device.

Figure 5. 

The schematic diagram of paper-based lipase inhibition assay for orlistat detection.

The performance of the paper-based device was examined against orlistat at various concentrations ranging from 0.00008–1.2 mg/mL. The intensity of the purple color at the detection zone decreased with increasing orlistat concentrations (Fig. 6). As observed by the naked eye, there was little difference in the reduction of purple color intensity at orlistat concentrations from 0.02 to 1.2 mg/mL. The limit of detection (LOD) by visual detection was estimated to be around 0.005 mg/mL. The dose-dependent inhibition responses of orlistat and the semi-log of the data were constructed to determine LOD given by IC₅₀ (Jin et al 2020). The calculated LOD (IC₅₀) fitted by the AAT Bioquest website program was found to be 0.0018 ± 0.0003 mg/mL (0.0036 ± 0.0006 mM).

Figure 6. 

a. A representative image of the paper devices and the rel. DI% of orlistat signals at various concentrations; b. A dose-response curve; c. A semi-log plot of the data, fitted by the AAT Bioquest website program to determine IC₅₀. Data are the means (± SD) of three replicates.

Orlistat-spiked samples at varied concentrations were analyzed to assess the ability of the paper device for visual detection of orlistat in real samples. The color intensity difference (DI) between the orlistat-free sample and the orlistat-spiked sample showed that the DI increased with increasing orlistat concentration (Fig. 7a). The concentrations of orlistat higher than 0.02 mg/mL showed apparent saturation in the reduction of color intensity. This was possible since orlistat irreversibly inhibits lipase, thus, when the orlistat concentration increases, the catalytic site of the enzyme becomes saturated.

Figure 7. 

a. The color intensity difference (DI) between the orlistat-free sample and the orlistat spiked sample at various concentrations; b. A representative image of the paper devices with orlistat spiked sample at 0.002 and 0.02 mg/mL.

A four-parameter logistic regression equation was used to determine the analytical recovery of the spiked sample at a concentration of 0.02 mg/mL, a 10-fold concentration to the calculated LOD value. The percent recovery (±SD) was 96.9 ± 12.5. The color intensity at higher concentrations than 0.02 mg/mL did not decline significantly, resulting in unsuitability for quantitative analysis. However, in terms of qualitative analysis, the reduction of color intensity was clearly observed for the orlistat-spiked sample at 0.002 mg/mL and 0.02 mg/mL on the paper device (Fig. 7b). The detection performance of the device at these concentrations was determined by the intra- and inter-day precision of the orlistat spiked sample. The data showed that rel. DI% for intra-day analysis of the orlistat spiked sample at 0.002 and 0.02 mg/mL were -39.6 and -50.9, and for the inter-day were -38.2 and -50.8, respectively. %RSD of rel. DI% for intra-day precision of the samples at 0.002 and 0.02 mg/mL were 9.0 and 2.9, and for the inter-day precision were 8.9 and 3.1, respectively. The detection performance with good precision was obtained with the orlistat spiked sample at 0.02 mg/mL, suggesting that the device may be beneficial for the qualitative detection of adulterated orlistat in a real sample.

Known compounds reported for lipase inhibition and herbal extracts commonly found in a fat blocker supplement were tested on the paper device. At the detection zones, no obvious change in the color intensity was observed for caffeine and quercetin and the herbal extracts as compared to the blank (Fig. 8a). All did not inhibit the enzymatic reaction as the orlistat did on the paper device, suggesting the selectivity of the proposed method for orlistat detection.

Figure 8. 

a. A representative image of each sample analysis (with rel. DI% ± SD given in parenthesis) on the paper device; b. TLC of S2 and orlistat (OL) using 0.5% w/v phosphomolybdic reagent as visualization reagent.

The paper-based analytical device was applied for rapid detection of lipase inhibitory activity in four different weight loss supplements claimed as fat blockers (S1–S4) and the orlistat capsule. The results show a distinguished purple color intensity reduction for S2 and the orlistat capsule (Fig. 8a). Other samples did not show an apparent decrease in color intensity nor a significant decreasing rel. DI%. The results from the paper-based assay revealed that S2 could be adulterated with orlistat. The presence of orlistat in S2 was ensured by TLC analysis, showing a distinctive blue band of orlistat at the R_f value of 0.6 (Fig. 8b).

A simple, minimized reagent consumption and low-cost colorimetric paper-based device for visual detection of orlistat in weight loss supplements was developed. The color detection was based on enzymatic inhibitory reaction. The color change can be easily observed by the naked eye or with image analysis software. The device provided satisfactory sensitivity and selectivity for the qualitative analysis. Compared with the TLC analysis, the device used in the proposed method could be easily fabricated in the laboratory using widely available, affordable and disposable materials. Furthermore, the method avoided the use of unsafe organic mobile phase and required simple procedures, thus also enabling the portability for in-field analysis and applicable for rapid screening orlistat adulteration and potential lipase inhibitors in weight loss supplements.

The authors are very grateful to the Faculty of Pharmacy, Silpakorn University for financial assistance and for providing facilities to support this project. The authors would also like to extend our sincere gratitude to Mr. Anthony Phonpituck for his careful editing of the manuscript.