Analytical grade solvents were purchased from Sintorgan (Buenos Aires, Argentina). LiChrosolv® methanol (MeOH) and dichloromethane (CH₂Cl₂) were supplied by Merck (Darmstadt, Germany). Formic acid was purchased from Baker (New Jersey, USA). Ultrapure water was generated in-house with a Barnstead Thermo Scientific™ purification system (resistance ≥18.2Ω). Analytical thin layer chromatography (TLC) was performed on silicagel 60-coated F₂₅₄ aluminium plates (Merck KGaA, Germany). Tadalafil working standard was isolated from tablets of CialisTM (20 mg) and its identity and purity was evaluated by nuclear magnetic resonance (NMR), Fourier transform infrared spectroscopy (FT-IR) and mass spectroscopy (MS) and compared to those obtained from tadalafil pharmaceutical secondary standard from Sigma Aldrich. Dimethylsulfoxide-d6 (DMSO-d6, 99.8%) (Sigma, St. Louis, MO, USA) was used as solvent in nuclear magnetic resonance (NMR) experiments.

Fourier transform infrared spectroscopy (FT-IR) experiments were performed on an Alpha FT-IR Spectrometer (Bruker, Madison WI, USA). FT-IR data were processed with a Bruker OPUS/IR 7.5.18 software and recorded over a spectral range spanning from 4,000 to 400 cm−1 with an optical resolution of 4 cm−1. NMR spectra were recorded on a Bruker Avance III 600 MHz equipped with a cryoprobe. Mono- and bidimensional NMR experiments were run using the programs from the Bruker software. Chemical shift values (δ) were expressed in parts per million (ppm). The HPLC-MS/MS system consisted of an UltiMate 3000 HPLC device coupled to a TSQ Quantum Access MAX Triple Quadrupole Mass Spectrometer with electrospray ionization (Massachusetts, USA). A Thermo Scientific™ Acclaim™ Mixed-Mode HILIC-1 (150 x 4 mm, 5 µm particle size) column was used. Microscopic analysis was performed in a Zeiss Axioscop 2 plus microscope equipped with a Moticam X Wifi camera.

As a first approach, a TLC analysis of CH₂Cl₂ and MeOH extracts of the capsule content was carried out to analyse the presence of the herbal drugs declared on the label. The TLC analysis of plant extracts normally renders chromatograms with numerous spots corresponding to a wide variety of plant metabolites. Many sources including pharmacopeias such as the US Pharmacopeia-National Formulary and the European Pharmacopoeia, and reference bibliography 14 detail the TLC techniques that can be used to analyse and identify different plant drugs. Normally, when herbal products containing many plant organs are analysed by TLC after an extraction procedure, complex chromatograms are obtained, especially those obtained with methanolic extracts. The TLC chromatograms obtained with the CH₂Cl₂ and MeOH extracts of the DS showed a small number of spots blurred by a greenish colour. The simplicity of the chromatograms was suspicious, together with the finding that the CH₂Cl₂ extract presented a major compound. However, the purification of such compound was not easy due to sample matrix complexity, which is known to affect the performance of analytical methods, e.g. co-extraction phenomena may occur 13. Knowing that the majority of the adulterated DS used as sexual enhancers contain approved synthetic PDE-5 inhibitors or their analogues and considering that most of those inhibitors contain nitrogen atoms in their structure 131516 we developed a rapid and easy purification method.

The isolation consisted of a liquid-liquid extraction (water/CH₂Cl₂) profiting from the acid-base properties of such compounds, followed by re-crystallisation from ethanol, to afford the target compound as white crystals. In our previous work, the isolation of the adulterant was accomplished by a single column chromatography process that yielded the pure compound. In this case, the sample matrix seemed less complex as shown by the TLC and HPLC profiles 8. Spectroscopic data (Figure 1 and Figure 2) allowed the identification of the major compound as tadalafil (Figure 3) 1718.

For the quantification of tadalafil in the DS, the HPLC chromatographic conditions needed optimisation. Several methods using liquid chromatography have been previously described for the identification of tadalafil and related compounds 192021. Given the chemical characteristics of tadalafil, several additives, such as triethanolamine, trifluoroacetic acid and formic acid have been used to avoid tailing. As shown in figure 3, the tadalafil molecule presents several nitrogen atoms, which interact with the stationary phase resulting in an irregular peak shape. This has been partially improved by the addition of mobile phase modifiers. In our method, we replaced the typical reverse phase by one that combines both reverse-phase and hydrophilic interactions. The Acclaim® Mixed-Mode HILIC-1 column provided by Thermo Scientific was employed. The particular properties of this stationary phase are given by the covalent functionalisation of high-purity spherical silica with a silyl ligand consisting of both an alkyl chain and a diol group. The use of this stationary phase not only avoids peak tailing, thus providing an improved symmetry and sensitivity, but also prevents ionic suppression given by the addition of mobile phase additives and the detriment of the column performance.

For the optimisation of MS/MS conditions, a standard solution of 1.0 µg/mL tadalafil was first infused directly into the spectrometer at a flow rate on 10 µL/min. In a full scan spectrum, the main peak of the standard corresponded to a m/z=390.4, representing the ^M+H+. Therefore, this was chosen as the precursor ion. Secondly, the most abundant product ions were determined, and their collision energies optimised. These product ions correspond to m/z=268.0 and m/z=240.0 (Table 1). Once the tuning conditions were optimised, the method settings were established. Three scan events were set, which comprised a full scan, a single ion monitoring (SIM) and a data dependent experiment. While the SRM experiment was used for the quantification of tadalafil, the data dependant experiment was employed for the identification of tadalafil by comparing the full spectrum of the peak with that of the standard. Chromatograms and spectra obtained from the standard and sample are shown in figure 4.

The content of tadalafil in the DS was determined by HPLC-MS/MS. The mean content per capsule was 21.2 ± 2.6 mg which represents a slightly higher value than that found in approved products in Argentina, i.e., 5 or 20 mg per tablet 22. Taking these results into account, it can be concluded that, unlike pharmaceuticals, certain DS are not subjected to a complete quality control process. This entails a substantial variability in the tadalafil content among capsules. Even though the amount of tadalafil found is similar to that found in approved marketed products, the presence of this drug in DS poses a serious health risk. As stated above, most patients consuming DS are unaware of potential adulterations and many consider DS as safe natural products. The common side effects associated with PDE-5 inhibitors include flushing, headache, dyspepsia, back pain, myalgia, dizziness, and rhinitis, but these effects are self-limited. However, some other more serious drug-related problems may occur, such as visual and auditory changes, cardiovascular effects, and alterations of renal and hepatic functions 11. Adulteration with synthetic drugs and their analogues pose a pharmacological and toxicological health risk but also the presence of impurities and degradation products may be of relevance 23.

Microscopic methods are used as part of the identification tests when the DS contains either intact plant material or its powder 4. Microscopic methods can detect other herbal drugs as adulterants as well as the presence of undeclared plant parts. In order to verify the presence of the herbal drugs declared in the DS label, we analysed the capsules content by light microscopy and polarized light microscopy. The analysis showed mainly the presence of the alga Arthrospirasp. and numerous simple starch grains (Figure 5). The presence of Arthrospirawas confirmed by comparison of their powder characteristics with those found in the literature 24. Arthrospirais a genus of cyanobacteria that constitutes dietary supplements commonly known as spirulina. This alga has long been used as food, and although most alga are innocuous, a significant proportion of cyanobacterial species are harmful due to the production of cyanotoxins 25. The possible presence of these toxins in these DS reinforces the need for a better-quality control 26.

Authors wish to thank Cátedra de Farmacognosia, Facultad de Farmacia y Bioquímica, Universidad de Buenos Aires and Consejo Nacional de Investigaciones Científicas y Técnicas (CONICET) for the use of equipment and facilities.

This article does not contain any studies with human participants or animals performed by any of the authors.

Informed Consent Not applicable