Biocompatible Microextraction Devices for Simple and Green Analysis of Complex Systems

For many decades, fast and reliable analysis of complex matrices, such as food, biofluids, or environmental samples, has been a challenge to the analytical chemistry community. In spite of the significant progress achieved so far in terms of analytical instrumentation and data deconvolution software, the pretreatment of complex samples still represents a key step in the analytical workflow that critically impacts the overall quality of results acquired. Microextraction, with its multifaceted modes and configurations, has played an essential role in enabling simpler pretreatment of challenging complex matrices to facilitate instrumental analysis. In this article, the development and evolution of biocompatible solid-phase microextraction (bio-SPME) are discussed, with special emphasis on extraction phases suitable for liquid chromatography and direct mass spectrometry applications. Some of the unique applications enabled by bio-SPME devices over the years are also described.

As a sample preparation technique, solid-phase microextraction (SPME) has evolved tremendously since its inception in 1990 (1,2). Initially designed for thermal desorption and gas-chromatography (GC) applications, SPME revolutionized the philosophy of sample preparation and extraction, as it enabled the simultaneous extraction and preconcentration of analytes from a given matrix (3).  In the 1990s, especially after the commercialization of SPME devices by Supelco (now MilliporeSigma, the Life Science business of Merck KGaA, Darmstadt, Germany), their use as sample preparation tools in the field of aroma and fragrances determination, and the extraction of GC-amenable organics from noncomplex water samples increased significantly (4).  Thermal desorption of SPME devices was the optimal solution for GC applications, because it was compatible with the GC-injector port without extensive modification of the existing hardware (apart from the inner glass liner that required narrower internal diameter for SPME applications) (5).

In terms of extraction phase chemistry, a common polymer used as a stationary phase into GC columns was selected for its good sorption properties and its thermal stability: polydimethylsiloxane (PDMS). Consequently, PDMS-based SPME coatings were first commercialized, and are still to date the most commonly used extraction phases for SPME. To expand the use of the technique for the analysis of nonvolatile analytes via liquid chromatography (LC), desorption strategies using solvent systems with high affinity for the analytes of interest were implemented. The desorption solvent containing the analytes can then be injected into the LC system directly or after preconcentration or reconstitution, if necessary. Considering that molecular mass transfer in the liquid phase is slower in comparison to the gas phase, quantitative solvent desorption generally takes longer than thermal desorption. Because  solvent desorption is typically performed off-line, agitation can be used to speed up the process. Two important factors to take into consideration when performing SPME via LC are that the desorption solution should enable quantitative desorption of the analytes (and consequently avoid carryover), and the final extract should be compatible with the mobile phase composition. Often, to meet these two requirements and also to preconcentrate the extracts, the desorption solvent system can be evaporated and reconstituted with an appropriate solvent system.

In terms of extraction modes, SPME for GC applications could be performed in either headspace (HS) or direct immersion (DI), based on the volatility of the target analytes. For complex matrices, HS-SPME was usually preferred to avoid exposing the SPME device directly to the sample. Performing DI-SPME in complex matrices can likely lead to the attachment of matrix constituents to the extraction phase surface, affecting the extraction efficiency of the device and subsequently reducing its lifetime. When DI-SPME was necessary, many researchers opted for pretreating the matrix with methods including, but not limited to, dilution, centrifugation, and filtration. Although these sample pretreatment strategies were effective, most often they defeated the scope of the simple and one-step extraction process that SPME is able to provide.

Expanding the applicability of SPME to LC-based approaches posed a challenge to the technology: LC-amenable analytes are semi- or non-volatile, therefore direct immersion SPME (DI-SPME) is mandatory. Given that extensive sample pretreatment is not practical, and can potentially induce analyte loss and lack of reproducibility, alternative extraction devices were urgently needed.

In light of these factors, significant research efforts were devoted to the development of “biocompatible” or “matrix compatible” SPME extraction phases, with both descriptions referring to the SPME coating’s anti-fouling characteristics (6–8). It is also worth mentioning that biocompatible SPME devices are manufactured with materials that are non-toxic and non-injurious to a living system, thus enabling the applicability of the technique also for in-vivo sampling (9–12).

The manufacture of biocompatible SPME devices must take into account various aspects for optimal extraction performance.

First, the outer surface of the SPME extraction phase represents the boundary phase that lies between the bulk of the matrix and the inner sorbent material. Interactions between the material and the sample matrix occur chiefly on such surface. The performance of a polymeric material for biocompatible SPME devices must have a good ability to prevent attachment of macromolecules (such as proteins and other biomolecules), and should permit smaller molecules to permeate its surface to reach the sorbent material in a reasonable time.

Second, pure polymers with antifouling properties do not always guarantee adequate extraction efficiency. Therefore, sorptive materials need to be incorporated to enhance the extraction performance. Most of these sorbents, however, are not biocompatible, so their surface must be surrounded by the antifouling polymer at the interface with the sample matrix.

Consequently, the first biocompatible extraction phase used for SPME-LC applications consisted of polyacrylonitrile (PAN), an antifouling polymer that also works as a binder to immobilize sorbents such as C18 functionalized silica particles (6). PAN and acrylonitrile-based copolymers are hydrophilic polymers broadly used in the biomedical field as membrane materials for dialysis, ultrafiltration, enzyme-immobilization, and pervaporation, due to their anti-biofouling properties and chemical stability.

The applicability of biocompatible SPME devices for biofluids and tissue analysis is highly dependent on the ability of the biocompatible polymer to prevent attachment of proteins that can affect the mass transfer of smaller organic molecules into the sorbent and act as anchors for the attachment of cells (for example, blood cells) (6). PAN, like other hydrophilic polymers, prevents the adhesion of fouling agents through the formation of a physical barrier known as hydration layer (13,14). The hydration layer is formed by hydrogen bonding between the functional groups on the device surface and water molecules in the sample matrix. The applicability of PAN for LC-based SPME devices also relies on its good binding ability toward sorptive particles to create a homogeneous slurry that can be applied as very thin layers. This feature facilitates the fabrication of devices that can be applied for in vivo and tissue analysis with improved mass transfer across the thin coating layers. Moreover, the good chemical stability of PAN toward most organic solvent facilitates solvent desorption without damaging or swelling the extraction phase, even if long desorption times are required. These unique properties make PAN-based extraction devices a very convenient solution for complex biofluids and tissue analysis that minimizes the effect of matrix interferences. PAN-based SPME devices are currently commercially available from MilliporeSigma, the Life Science business of Merck KGaA, Darmstadt, Germany (Figure 1).

Typical steps in the workflow of SPME-LC analysis of complex biospecimens by PAN-based SPME devices are as follows:

• Preconditioning. This step is generally needed to activate the sorbent particles prior to extraction, and it is performed with a solution of water and organic solvent (commonly 1:1 (v:v) MeOH:H₂O for 15-30 min).

• Rinsing. Prior to extraction in complex biomatrices, it is critical to quickly rinse the SPME device in pure water (30 s). This helps to remove residual organic solvent after the preconditioning step, which may induce protein precipitation on the device surface during extraction.

• Extraction. PAN-based biocompatible coatings can be directly exposed to untreated biofluids and tissues. Depending on the objective, the extraction time may be tuned toward maximum extraction recovery where sensitivity is pertinent or minimized for faster throughput.

• Post-extraction rinsing. Prior to desorption, it is a useful practice to quickly rinse the SPME device to remove any matrix component that can potentially be left loosely attached on its surface; this will further prevent matrix contamination. However, care must be taken not to compromise the overall amount of extracted analytes.

• Solvent desorption. During solvent desorption, the analytes need to be desorbed in a solvent system strong enough to reverse the interaction of analytes with the extraction phase. In SPME-LC, the desorption solvent ideally should match the initial composition of the mobile phase to avoid solvent mismatch and poor chromatography. When this is not achievable, evaporation of the desorption solution and reconstitution with proper solvent combinations are recommended.

• Cleanup (optional, if the devices are being re-used). Cleanup can be performed to prepare the SPME for the next cycle of extraction in the case of extraction from very complex matrices or to make sure all the analytes extracted are fully desorbed.

PAN-based SPME extraction phases have enabled a cascade of applications including in-vivo metabolomics in the brain, liver, lungs, and in various biofluids (15–22). Moreover, the easy applicability of PAN-based extraction phases onto supports of different geometries permitted the development of multiple microextraction tools, compatible with various sampling needs (such as recessed SPME [23,24], single-use samplers coated on plastic supports [25,26]) and to direct coupling to mass spectrometry (transmission mode SPME [27–29], nanospray [30], and coated blade spray [31–35], being commercialized by Restek Corporation).

Although PAN-based extraction phases are well suited for solvent desorption, the lack of thermal stability above 120–160 °C does not make them suitable for SPME-GC applications due to the high temperatures needed for effective thermal desorption (36).

In light of this, and to expand the applicability of DI-SPME-GC in complex matrices, a novel biocompatible (or matrix compatible) extraction phase was developed (8). This extraction phase was optimized based on commercially available SPME fibers for GC applications; it was noticed that pure PDMS extraction phases endured direct immersion into complex and untreated food matrices for longer series of extractions without noticeable coating fouling while maintaining good extraction efficiency. PDMS is well known for being a hydrophobic biocompatible polymer that prevents the formation of hydrogen bonds, thus avoiding the attachment of water and biomolecules alike (13). However, PDMS extraction efficiency is limited by its hydrophobicity. Thus, to obtain extraction phases able to provide a broader extraction range, commercial SPME devices were manufactured using PDMS as a binder for sorbents such as divinyl benzene (DVB), Carboxen (Car), and a mixture of these (4). The incorporation of sorbent particles, however, affected the outer morphology of the extraction phase compared to pure PDMS devices, making them uneven and rough. This issue was found to be detrimental when SPME devices were used for the analysis of complex matrices via DI, as residues of matrix constituents accumulate on the interstices of the extraction phase surface, and then get carbonized during thermal desorption. Subsequently, this leads to fouling buildup that would reduce the device’s extraction efficiency and affect its reusability. To overcome this issue, the design of the new PDMS/DVB/PDMS extraction phase included a thin and smooth layer of pure PDMS (~10 µm) to protect conventional commercial SPME devices such as a DVB/PDMS fibers (8). The new design, presented in Figure 1, enabled direct immersion in very complex matrices such as foodstuffs, without the need for extensive sample pretreatment. This extraction phase demonstrated its efficacy especially for fruit and vegetable analysis; the significant presence of carbohydrates in these matrices affected conventional SPME device performance. In fact, carbohydrate residues on the surface of the coating carbonize during thermal desorption, damaging the extraction phase irreversibly, and creating artifacts that will populate the chromatogram, potentially masking targeted analytes. The development of this new extraction phase enabled several applications in diverse food matrices for both targeted and untargeted analysis, including in vivo applications in fruits (37–41). In addition, the ability to add rinsing and washing steps in the analytical workflow was generally found to prolong the coating lifetime. For example, in the case of matrices with high water and carbohydrates content, a post-extraction rinsing in pure water (5–20 sec) was found effective to guarantee coating cleanness and to avoid the occurrence of artifacts due to thermal conversion of sugars into the GC injector (41). For the same matrix types, post-desorption washing in water:methanol 1:1 (v:v) also showed efficacy in removing any matrix residue on the extraction phase surface. For food matrices with high-fat content, different rinsing and washing strategies must be developed to remove oily residues from the SPME device surface to prevent extensive contamination of the GC injector. Complex matrices, such as avocado, soy milk, and dried seaweed, require a mixture of acetone and water, at different ratios, to be used for both rinsing and washing solutions (37,39,40). It is important to mention that special attention should be paid when performing the rinsing step, especially if solvents other than water are used. The rinsing time in these cases must be kept as short as possible, to minimize analyte losses. However, this phenomenon does not apply to washing procedures performed after desorption process. It is also critical to select solvents that do not affect the structural integrity of the fiber; chlorinated solvents and hydrocarbon-based solvents are known to swell PDMS.

Since their inception, PAN- and PDMS-based biocompatible coatings have facilitated the analysis of complex matrices by SPME, providing unique analytical solutions for both targeted and untargeted analysis of food, environmental, and biosamples. However, these extraction phases are specific to different separation platforms: LC in the case of PAN-based devices and GC for PDMS-devices. Therefore, sampling of complex matrices for extraction of both LC- and GC-amenable analytes could be further improved by a biocompatible SPME extraction phase compatible with both thermal and solvent desorption mechanisms.

Fluorinated polymers constitute a unique class of materials with high chemical resistance and thermal stability. This class of polymers is known to be chemically inert or relatively unreactive. Polytetrafluoroethylene (PTFE), also known by its trade name Teflon, is the first fluoropolymer to be discovered in 1938 and exhibits exceptional ability to repel water, oils, adhesives, and so on (42). Moreover, it is a well-established biocompatible material, often used for the production of medical devices. One major disadvantage in the use of PTFE is that it is not soluble and does not swell in most solvents, thus machining techniques are commonly used to process it. To overcome this limitation, amorphous fluoropolymers such as PTFE-AF were developed. PTFE-AF is a copolymer of tetrafluoroethylene and 2,2-bis(trifluoromethyl)-4,5-difluoro-1,3-dioxole, and exhibits improved mechanical stability and high solubility in fluorinated solvents. Moreover, the fluorinated backbone of PTFE-AF provides similar biocompatibility and stability as the PTFE polymer (42). These characteristics make PTFE-AF an excellent candidate for the manufacturing of biocompatible SPME devices suitable for both solvent and thermal desorption. The first report of receptor-doped fluorous films for SPME was reported in 2014 (43), followed by the fabrication of PTFE-AF-based SPME fiber that incorporated hydrophilic-lipophilic balance particles (HLB) in 2017 (44). This HLB-PTFE-AF extraction phase was specifically designed to serve as a multipurpose sampling tool for complex matrices. Although the PTFE-AF guaranteed compatibility to LC and GC desorption techniques, the HLB particles provided broader extraction coverage, and improved recovery for more polar analytes. The compatibility with different chromatographic platforms together with the collection of a broader range of analytes make this extraction phase well suited for untargeted analysis. This new biocompatible extraction phase was tested for the extraction of a broad range of LC and GC amenable analytes in biofluids such as whole blood, saliva, serum, and urine, and in Concord grape juice, a food matrix particularly challenging for its high content in sugars and pigments. When repetitive DI-SPME extraction/desorption cycles were performed prior to GC and LC analysis from the matrices mentioned above, good performance was achieved up to at least 50 consecutive cycles for both solvent and thermal desorption techniques. Moreover, it was assessed that the chemistry of this new extraction phase and in particular the inertness of the PTFE-AF material, drastically minimizes the impact of the matrix on the overall analytical process (45).

Conclusion

In summary, the introduction of biocompatible extraction phases has significantly expanded the applicability of the SPME technology and enabled convenient analysis of complex matrices with minimum or no sample pretreatment. This results in numerous advantages in terms of the throughput of the analytical routine and minimization of laboratory waste production. Additionally, the unique properties of these SPME devices together with their miniaturized geometry offer exceptional sampling opportunities applicable to in-vivo analysis.

References

1. C.L. Arthur and J. Pawliszyn, Anal. Chem. 62(19), 2145–2148 (1990). doi:10.1021/ac00218a019
2. N. Reyes-Garcés, E. Gionfriddo, G.A. Gómez-Ríos, M.N. Alam, E. Boyaci, B. Bojko, V. Singh, J. Grandy, and J. Pawliszyn, Anal. Chem.90(1), 302–360 (2018). doi:10.1021/acs.analchem.7b04502
3. J. Pawliszyn, Handbook of Solid Phase Microextraction (Elsevier, Waltham, Massachussetts, 2012).
4. J. Pawliszyn, Development of SPME Devices and Coatings in Handbook of Solid Phase Microextraction (Elsevier, Waltham, Massachussetts, 2012), pp. 61–97
5. J. Pawliszyn, L. Kudlejova, S. Risticevic, and D. Vuckovic, Handbook of Solid Phase Microextraction (Elsevier, Waltham, Massachussetts, 2012), pp. 201–249
6. M. L. Musteata, F. M. Musteata, and J. Pawliszyn, Anal. Chem. 79(18), 6903–6911 (2007).doi:10.1021/ac070296s
7. N. H. Godage and E. Gionfriddo, TrAC Trends Anal. Chem. 111, 220–228 (2019).doi:10.1016/j.trac.2018.12.019
8. É.A. Souza Silva and J. Pawliszyn, Anal. Chem. 84(16), 6933–6938 (2012). doi:10.1021/ac301305u
9. F.M. Musteata, M. Sandoval, J.C. Ruiz-Macedo, K. Harrison, D. McKenna, and W. Millington, Anal. Chim. Acta 933, 124–133 (2016). doi:10.1016/j.aca.2016.05.053
10. F. M. Musteata, TrAC Trends Anal. Chem. 45, 154–168 (2013).
11. N. Reyes-Garcés and E. Gionfriddo, TrAC Trends Anal. Chem. 113, 172–181 (2019). doi:10.1016/j.trac.2019.01.009
12. B. Bojko, K. Gorynski, G.A. Gomez-Rios, J.M. Knaak, T. Machuca, V.N. Spetzler, E. Cudjoe, M. Hsin, M. Cypel, M. Selzner, M. Liu, S. Keshavjee, and J. Pawliszyn, Anal. Chim. Acta803, 75–81 (2013). doi:10.1016/j.aca.2013.08.031
13. J.L. Harding and M.M. Reynolds, Trends Biotechnol.32(3), 140–146 (2014). doi:10.1016/j.tibtech.2013.12.004
14. J. L. Dalsin and P.B. Messersmith, Mater. Today8(9), 38–46 (2005). doi:10.1016/S1369-7021(05)71079-8
15. S. Lendor, S.A. Hassani, E. Boyaci, V. Singh, T. Womelsdorf, and J. Pawliszyn, Anal. Chem.91(7), 4896–4905 (2019). doi:10.1021/acs.analchem.9b00995
16. A. Napylov, N. Reyes-Garces, G. Gomez-Rios, M. Olkowicz, S. Lendor, C. Monnin, B. Bojko, C. Hamani, J. Pawliszyn, and D. Vuckovic, Angewandte Chemie 59(6), 2392–2398 (2020). doi:10.1002/anie.201909430
17. E. Cudjoe, B. Bojko, I. de Lannoy, V. Saldivia, and J. Pawliszyn, Angewandte Chemie52(46), 12124–12126 (2013). doi:10.1002/anie.201304538
18. B. Bojko, K. Gorynski, G.A. Gomez-Rios, J.M. Knaak, T. Machuca, E. Cudjoe, V.N. Spetzler, M. Hsin, M. Cypel, M. Selzner, M. Liu, S. Keshjavee, and J. Pawliszyn, Lab. Invest.94, 586–594 (2014). doi:10.1038/labinvest.2014.44
19. N.T. Looby, M. Tascon, V.R. Acquaro, N. Reyes-Garcés, T. Vasiljevic, G.A. Gomez-Rios, M. Wasowicz, and J. Pawliszyn, Analyst144(12), 3721–3728 (2019). doi:10.1039/c9an00041k
20. E. Boyacı, B. Bojko, N. Reyes-Garcés, J.J. Poole, G.A. Gómez-Ríos, A. Teixeira, B. Nicol, and J. Pawliszyn, Sci. Rep. 8(1), 1167 (2018). doi:10.1038/s41598-018-19313-1
21. N. Reyes-Garcés, M. N. Alam, and J. Pawliszyn, Anal. Chim. Acta1001, 40–50 (2018).doi:10.1016/j.aca.2017.11.014
22. N. Reyes-Garcés, M. Diwan, E. Boyacl, G.A. Gómez-Ríos, B. Bojko, J.N. Nobrega, F.R. Bambico, C. Hamani, and J. Pawliszyn, Anal. Chem.91(15), 9875–9884 (2019). doi:10.1021/acs.analchem.9b01540
23. J.J. Poole, J.J. Grandy, M. Yu, E. Boyaci, G.A. Gómez-Ríos, N. Reyes-Garcés, B. Bojko, H. Vander Heide, and J. Pawliszyn, Anal. Chem.89(15), 8021–8026 (2017). doi:10.1021/acs.analchem.7b01382
24. S. Lendor, S.A. Hassani, E. Boyaci, V. Singh, T. Womelsdorf, and J. Pawliszyn, Anal. Chem.91(7), 4896–4905 (2019). doi:10.1021/acs.analchem.9b00995
25. N. Reyes-Garcés, B. Bojko, D. Hein, and J. Pawliszyn, Anal. Chem.87(19), 9722–30 (2015). doi:10.1021/acs.analchem.5b01849
26. T. Vasiljevic, G.A. Gómez-Ríos, and J. Pawliszyn, Anal. Chem.90(1), 952–960 (2018). doi:10.1021/acs.analchem.7b04005
27. G.A. Gómez-Ríos and J. Pawliszyn, Chem. Commun. 50(85), 12937–12940 (2014). doi:10.1039/C4CC05301J
28. G.A. Gómez-Ríos, E. Gionfriddo, J. Poole, and J. Pawliszyn, Anal. Chem. 89(13), 7240–7248 (2017). doi:10.1021/acs.analchem.7b01553
29. G.A. Gómez-Ríos, T. Vasiljevic, E. Gionfriddo, M. Yu, and J. Pawliszyn, Analyst 142(16), 2928–2935 (2017). doi:10.1039/c7an00718c
30. G.A. Gómez-Ríos, N. Reyes-Garcés, B. Bojko, and J. Pawliszyn, Anal. Chem. 88(2), 1259–1265 (2016). doi:10.1021/acs.analchem.5b03668
31. A. Khaled, G.A. Gómez-Ríos, and J. Pawliszyn, Anal. Chem. 92(8), 5937–5943 (2020). doi:10.1021/acs.analchem.0c00093
32. A. Kasperkiewicz, G.A. Gómez-Ríos, D. Hein, and J. Pawliszyn, Anal. Chem. 91(20), 13039–13046 (2019). doi:10.1021/acs.analchem.9b03225
33. G.A. Gómez-Ríos, M. Tascon, N. Reyes-Garcés, E. Boyacl, J. Poole, and J. Pawliszyn, Sci. Rep. 7(1) (2017). doi:10.1038/s41598-017-16494-z
34. M. Tascon, G.A. Gómez-Ríos, N. Reyes-Garcés, J. Poole, E. Boyacı, and J. Pawliszyn, J. Pharm. Biomed. Anal. 144(10), 10.1016/j.jpba.2017.03.009 (2017). doi:10.1016/j.jpba.2017.03.009
35. G.A. Gómez-Ríos, M. Tascon, N. Reyes-Garcés, E. Boyacı, J.J. Poole, and J. Pawliszyn, Anal. Chim. Acta Accepted manuscript (2017).
36. S. Xiao, W. Cao, B. Wang, L. Xu, and B. Chen, J. App. Polym. Sci.127(4), 3198–3203 (2013). doi:10.1002/app.37733
37. E. Gionfriddo, D. Gruszecka, X. Li, and J. Pawliszyn, Talanta211(10), (2019), 120746 (2020). doi:10.1016/j.talanta.2020.120746
38. S. Risticevic, E. A. Souza-silva, E. Gionfriddo, J.R. Deell, J. Cochran, W.S. Hopkins, and J. Pawliszyn, Sci. Rep. 10(1), 1–11 (2020). doi:10.1038/s41598-020-63817-8
39. L. Zhang, E. Gionfriddo, V. Acquaro, and J. Pawliszyn, Anal. Chim. Acta1031, 83-97 (2018).  doi:10.1016/j.aca.2018.05.066
40. S. De Grazia, E. Gionfriddo, and J. Pawliszyn, Talanta167, 754–760 (2017). doi:10.1016/j.talanta.2017.01.064
41. É.A. Souza-Silva, V. Lopez-Avila, and J. Pawliszyn, J. Chromatogr. A1313, 139–146 (2013). doi:10.1016/j.chroma.2013.07.071
42. V.F. Cardoso, D.M. Correia, C. Ribeiro, M.M. Fernandes, and S. Lanceros, Polymers10(2),1–26 (2018). doi:10.3390/polym10020161
43. D. Lu and S. Weber, J. Chromatogr. A1360, 17–22 (2014). doi:10.1016/j.chroma.2014.07.060
44. E. Gionfriddo, E. Boyaci, and J. Pawliszyn, Anal. Chem. 89(7) (2017). doi:10.1021/acs.analchem.6b04690
45. J. Pawliszyn, E. Gionfriddo, and E. Boyacı, Solid Phase Microextraction Coating. US Patent 10,545,077 B2 (2017).

Emanuela Gionfriddo is in the Department of Chemistry and Biochemistry at the School of Green Chemistry and Engineering at The University of Toledo, in Toledo, Ohio. Direct correspondence to: Emanuela.Gionfriddo@UToledo.edu