Anal. Chem. 1999, 71, 2092-2096
Method for Synthesis and Screening of Large Groups of Molecularly Imprinted Polymers Francesca Lanza and Bo 1 rje Sellergren*
Department of Inorganic Chemistry and Analytical Chemistry, Johannes Gutenberg University Mainz, Duesbergweg 10-14, 55099 Mainz, Germany
A technique for the synthesis of molecularly imprinted polymers (MIPs) in small scale (∼55 mg) coupled with direct in situ processing and batch rebinding evaluation is reported. The primary assessment is based on quantification by HPLC or UV absorbance measurement of the amount of template released from the polymer in a given solvent. This method allows a rapid screening of the parameters of importance to reach a desired level of binding affinity capacity and selectivity for a given target molecule. This was demonstrated for the triazine herbicide terbutylazine, where an initial screening was performed for the type of functional monomer used in the MIP preparation. Thus among the six functional monomers tested, methyl methacrylate, 4-vinylpyridine, and N-vinyl-r-pyrrolidone led to rapid and quantitative extraction whereas methacrylic acid and (trifluoromethyl)acrylic acid led to polymers that retained the template the most. After having established useful functional monomers, a secondary screening for selectivity was performed. In this, nonimprinted blank polymers were prepared and a normal batch rebinding evaluation was performed. The polymer showing the highest selectivity was the one prepared using methacrylic acid as functional monomer. This polymer was shown to strongly retain chlorotriazines including atrazine when a normal-scale batch of the polymer was evaluated in chromatography. Molecular imprinting is becoming an established technique for the preparation of polymeric materials with recognition properties for small molecules.1-6 During the last years, applications of the materials as affinity phases in solid-phase extractions,7 as recognition elements in sensors,8 as stationary phases for preparative purifications9 or separations of enantiomers,10 as catalysts,11 or as adsorbents for therapeutic use12 are being actively pursued. For a general use of the technology, the class of imprintable compounds needs to be extended and the existing recognition (1) Bartsch, R. A., Maeda, M., Eds. Molecular and Ionic Recognition with Imprinted Polymers; ACS Symposium Series 703; American Chemical Society: Washington DC, 1998. (2) Wulff, G. Angew. Chem., Int. Ed. Engl. 1995, 34, 1812-32. (3) Shea, K. J. Trends Polym. Sci. 1994, 2, 166-173. (4) Mayes, A. G.; Mosbach, K. Trends Anal. Chem. 1997, 16, 321-332. (5) Sellergren, B. Trends Anal. Chem. 1997, 16, 310-320. (6) Steinke, J.; Sherrington, D.; Dunkin, I. Adv. Polym. Sci. 1995, 123, 80125.
2092 Analytical Chemistry, Vol. 71, No. 11, June 1, 1999
elements need to be improved in order to meet the requirements in the given application. A key in this development is the identification and optimization of the main factors affecting the material structure and molecular recognition properties. These factors can be the type and concentration of functional monomer, cross-linking monomer, the polymerization temperature, pressure, or solvent of polymerization. This can be achieved by scaling down the molecularly imprinted polymer (MIP) synthesis and in situ processing and evaluation of the materials. This allows rapid screening for the recognition properties of large numbers of materials (mini-MIPs). We here report a scaled-down version of the previous monolith procedure13,14 allowing rapid in situ extraction and rebinding evaluation. The approach is similar to the one described recently by Takeuchi et. al. for the combinatorial synthesis of molecularly imprinted polymers.15 It distinguishes itself, however, with respect (7) (a) Sellergren, B. Anal. Chem. 1994, 66, 1578. (b) Andersson, L. I.; Paprica, A.; Arvidsson, T. Chromatographia 1997, 46, 57-62. (c) Martin, P.; Wilson, I. D.; Morgan, D. E.; Jones, G. R.; Jones, K. Anal. Commun. 1997, 34, 45-47. (d) Rashid, B. A.; Briggs, R. J.; Hay, J. N.; Stevenson, D. Anal. Commun. 1997, 34, 303-305. (e) Muldoon, M. T.; Stanker, L. H. Anal. Chem. 1997, 69, 803-808. (f) Matsui, J.; Okada, M.; Tsuruoka, M.; Takeuchi, T. Anal. Commun. 1997, 34, 85-87. (g) Mullett, W. M.; Lai, E. P. C. Anal. Chem. 1998, 70, 3636-3641. (h) Walshe, M.; Howarth, J.; Kelly, M. T.; O’Kennedy, R.; Smyth, M. R. J. Pharm. Biomed. Anal. 1997, 16, 319-325. (i) Zander, Å.; Findlay, P.; Renner, T.; Sellergren, B.; Swietlow, A. Anal. Chem. 1998, 70, 3304-3314. (8) (a) Piletsky, S. A.; Parhometz, Y. P.; Lavryk, N. V.; Panasyuk, T. L.; El’skaya, A. V. Sens. Actuators, B 1994, 19, 629-31. (b) Kriz, D.; Ramstro ¨m, O.; Mosbach, K. Anal. Chem. 1997, 69, 345A-349A. (c) Dickert, F. L.; Forth, P.; Lieberzeit, P.; Tortschanoff, M. Fresenius' J. Anal. Chem. 1998, 360, 759-762. (d) Turkewitsch, P.; Wandelt, B.; Darling, G. D.; Powell, W. S. Anal. Chem. 1998, 70, 2025-2030. (e) Lulka, M. F.; Chambers, J. P.; Valdes, E. R.; Thompson, R. G.; Valdes, J. J. Anal. Lett. 1997, 30, 23012313. (f) Piletsky, S. A.; Piletskaya, E. V.; El’skaya, A. V.; Levi, R.; Yano, K.; Karube, I. Anal. Lett. 1997, 30, 445-455. (9) Joshi, V. P.; Karode, S. K.; Kulkarni, M. G.; Mashelkar, R. A. Chem. Eng. Sci. 1998, 53, 2271-2284. (10) (a) Kempe, M.; Mosbach, K. J. Chromatogr., A 1995, 694, 3-13. (b) Hosoya, K.; Shirasu, Y.; Kimata, K.; Tanaka, N. Anal. Chem. 1998, 70, 943-945. (c) Sajonz, P.; Kele, M.; Zhong, G.; Sellergren, B.; Guiochon, G. J. Chromatogr., A 1998, 810, 1-17. (d) Armstrong, D. W.; Schneiderheinze, J. M.; Hwang, Y.-S.; Sellergren, B. Anal. Chem. 1998, 70, 3304-3314. (11) (a) Davis, M. E.; Katz, A.; Ahmad, W. R. Chem. Mater. 1996, 8, 18201839. (b) Wulff, G.; Gross, T.; Scho ¨nfeld, R. Angew. Chem., Int. Ed. Engl. 1997, 36, 1962-1964. (12) (a) Sreenivasan, K. Angew. Makromol. Chem 1997, 246, 65-69. (b) Asanuma, H.; Kakazu, M.; Shibata, M.; Hishiya, T.; Komiyama, M. Chem. Commun. 1997, 1971-1972. (c) Sellergren, B.; Wieschemeyer, J.; Boos, K.-S.; Seidel, D. Chem. Mater., 1998, 10, 4037-4046. (13) Sellergren, B.; Shea, K. J. J. Chromatogr., A 1993, 635, 31-49. (14) Dauwe, C.; Sellergren, B. J. Chromatogr., A 1996, 753, 191-200. (15) Takeuchi, T.; Fukuma, D.; Matsui, J. Anal. Chem. 1999, 71, 285-290. 10.1021/ac981446p CCC: $18.00
© 1999 American Chemical Society Published on Web 04/30/1999
Scheme 1
to the method for screening of the recognition properties. The system, which we developed independently from Takeuchi et. al., is based on two screening steps. In the first, the release of template, in the same solvent (porogen) as used in the polymerization step is analyzed, and in the second, the selectivity of the mini-MIPs selected in the first screen. This may speed up the synthesis of new MIPs with high affinity and selectivity for the template. EXPERIMENTAL SECTION Materials. Triazine samples of technical grade were generously provided by Novartis (Basel, Switzerland). Methacrylic acid (MAA), methyl methacrylate (MMA), hydroxyethyl methacrylate (HEMA), N-vinyl-R-pyrrolidone (NVP), ethylene glycol dimethacrylate (EDMA), (trifluoromethyl)acrylic acid (TFM), and dichloromethane were purchased from Sigma-Aldrich. 4-Vinylpyridine (VPY) was purchased from Merck (Darmstadt, Germany), and the initiator azo-N,N′-bisisobutyronitrile (AIBN) was obtained from Janssen. No further purification of the monomers was performed. The UV lamp used in the photopolymerization was a high-pressure mercury vapor lamp (Philips, HPK 125 W). The glass vials (1.5 mL) with rubber septa used as polymerization reactors were purchased from Supelco. All the chromatographic evaluations were done using a Hewlett-Packard instrument (HP1050) equipped with a quaternary pump, an autosampler, a variable-wavelength detector, and an HP work station. Small-Scale MIP Synthesis (Mini-MIPs). Two mother solutions (with or without template) were prepared for the scaleddown version of the polymerization (mini-MIPs). Both were obtained by mixing EDMA 4.75 mL (25 mmol), AIBN 50 mg (0.30 mmol), and CH2Cl2 7 mL. The template mother solution also contained 286 mg (1.25 mmol) of tertbutylazine. A total of 117.5 µL of each mother solution was then dispensed into a 1.5-mL glass
vial so that the resulting polymerization mixture of the scaleddown version had the following composition: EDMA 47.5 µL (250 µmol), CH2Cl2 70 µL, AIBN 0.5 mg (3 µmol), and for the template solutions terbutylazine 2.86 mg (12.5 µmol). After addition of one of the functional momomers (MAA, TFM, MMA, HEMA, VPY, NVP; 50 µmol) to a total volume of ∼120 µL, the vial was sealed with a rubber septum. All the vials were purged with nitrogen for 5 min while cooled to 15 °C in a thermostated water bath, and then they were symmetrically placed around a UV lamp and the solutions allowed to polymerize for 12 h at 15 °C and for 13 h at 30 °C. Extraction and Rebinding Experiments. A 1-mL volume of CH2Cl2 was dispensed into each of the 12 vials containing the 6 imprinted and blank polymers. The vials were then sonicated for 1 h without heating, and the content of tertbutylazine was quantified in each extract by HPLC using an external standard. The HPLC evaluation was carried out using a C18 reversed-phase column, (Prodigy 5 µm ODS3, 125 × 4.6 mm2), MeCN/HOAc/ H2O 92.5:5:2.5 (v/v/v) as mobile phase, and a detection wavelength of 260 nm. The vials were inserted into the HP autosampler and the supernatant solution (10 µL) was directly injected onto the column. The vials were then left overnight at room temperature, sonicated again for 1 h at 40 °C, and finally sonicated for 1 h after the addition of 50 µL of HOAc, all treatments followed by HPLC analysis of terbutylazine in the extracts. The supernatant was thereafter removed from each vial, and the polymer was sonicated for 15 min in 1 mL of CH2Cl2, 1 mL of CH2Cl2 + 5% HOAc, and finally 1 mL of CH2Cl2. The rebinding experiment was then performed by addition of 1 mL of 1 mM terbutylazine in CH2Cl2 to each vial and sonication of the vials for 1 h. Thereafter the concentration of free (unbound) terbutylazine was determined by reversed-phase HPLC. Analytical Chemistry, Vol. 71, No. 11, June 1, 1999
2093
Table 1. Total Recovery of the Template, Terbutylazine, from Mini-MIPs after Different Consecutive Extraction Stepsa CH2Cl2 sonicationc
sonication at 40 °Ce
overnightd
CH2Cl2 + 5% HOAcf
polymerb
Cfree (mM)
yield (%)
Cfree (mM)
yield (%)
Cfree (mM)
yield (%)
Cfree (mM)
yield (%)
P-MAA P-TFM P-MMA P-HEMA P-VPY P-NVP
2.85 5.23 9.86 7.78 9.25 10.3
25 47 88 69 83 92
3.56 5.53 10.4 8.38 9.92 10.9
32 49 92 75 88 97
4.17 5.95 11.7 10.2 11.2 12.0
38 53 104 91 100 107
9.29 8.88 11.8 10.3 11.3 11.9
83 79 106 92 100 106
a The mini-MIPs were prepared according to the procedure described in the Experimental Section and as shown in Scheme 1. After polymerization, 1 mL of dichloromethane was added to each vial whereafter all the vials were identically treated. After each treatment, the supernatant was analyzed by reversed-phase HPLC based on duplicate injections (Figure 1). The concentration of free terbutylazine was calculated with reference to an external standard and the yield calculated by assuming a theoretical maximum concentration of 11.2 mM. b Scaled-down version of MIPs (miniMIPs) using the functional monomers indicated by abbreviation (defined in text). c The vials were sonicated without heating for 1 h. d The vials were left overnight at room temperature. e The vials were sonicated at 40 °C for 1 h. f Acetic acid (HOAc) was added to each vial to a concentration of 5% (v/v) followed by sonication for 15 min.
Normal-Scale MIP Synthesis. A previously described procedure was followed for the normal-scale version of the polymerization.14 EDMA (20 mmol, 3.8 mL), MAA (4 mmol, 0.34 mL), terbutylazine (1 mmol), and AIBN (0.24 mmol, 40 mg) were added in CH2Cl2 (5.6 mL) and the solution was then transferred into a glass tube (14 mm i.d.). The polymerization mixture was then degassed with nitrogen for 5 min while cooled to 15 °C in a thermostated water bath. All the tubes were placed at ∼10 cm distance from a UV light source and then turned at regular intervals during the first 30 min, to obtain a more even exposure. After 24 h, the tubes were crushed, the polymer was ground and sieved under water, and the particle size fraction of 25-36 µm was collected. This fraction was then repeatedly washed with 50mL aliquots of MeOH/H2O 1:1, MeOH, MeOH/HOAc 80:20 (v/ v), and MeOH and then used for the chromatographic evaluation. Stainless steel HPLC columns (125 × 4 mm2) were slurry packed using MeOH/H2O 80:20 (v/v) as pushing solvent at pressures up to 300 bar. A primary evaluation of the polymer selectivity was done using MeCN/H2O/HOAc (92.5:2.5:5, v/v/v) as the mobile phase. Thereafter, the polymers were evaluated using MeCN as mobile phase. RESULTS AND DISCUSSION To demonstrate the synthesis and screening approach, we chose the previously studied triazine model system14,16 using terbutylazine as template. The polymers were prepared by photoinitiation following the previous protocol using EDMA as a cross-linking monomer and dichloromethane as solvent (Scheme 1).14 The standard size batch was scaled down 80 times to a total volume of ∼120 µL. The mixture of cross-linker, solvent, template, initiator, and functional monomer was pipetted into 1.5-mL size sample vials designed for the HPLC autoinjector. The final compositions are given in the Experimental Section. As seen in Scheme 1, the screening was performed in two steps. In the first step, the amount of template released from the polymer was measured. Thus, six polymers with different functional monomers (16) (a) Matsui, J.; Miyoshi, Y.; Doblhoff-Dier, O.; Takeuchi, T. Anal. Chem. 1995, 67, 4404-4408. (b) Muldoon, M. T.; Stanker, L. H. J. Agric. Food Chem. 1995, 43, 1424-1427. (c) Siemann, M.; Andersson, L. I.; Mosbach, K. J. Agric. Food Chem. 1996, 44, 141-145.
2094 Analytical Chemistry, Vol. 71, No. 11, June 1, 1999
Figure 1. Examples of the reversed-phase HPLC elution profiles obtained in the rebinding experiment using the mini-MIPs and blank polymers. The polymers were subjected to 1 mL of a solution of terbutylazine (1 mM) in CH2Cl2 for 1 h under sonication whereafter the supernatants were analyzed for free terbutylazine. The large peak corresponds to terbutylazine, and in order of increasing size of this peak, the elution profiles represent P-MAA-MIP, P-MAA-blank, and an external standard of terbutylazine (1 mM). The arrow indicate the approximate elution time for possible remaining unreacted monomers from the polymerization. The HPLC evaluation was carried out using a C18 reversed-phase column (Prodigy 5 µm ODS3, 125 × 4.6 mm2); MeCN/HOAc/H2O 92.5:5:2.5 (v/v/v) as mobile phase and at a detection wavelength of 260 nm.
were synthesized in the presence of the template. Thereafter, the polymers were washed successively with dichloromethane under sonication for 1 h, overnight with no sonication, at 40 °C for 1 h with sonication, and finally after addition of acetic acid (5%) and further sonication. The supernatants were then analyzed by reversed-phase HPLC after each of these treatments (Figure 1 and Table 1). Already after the first wash it is obvious that the polymers prepared using MAA and TFM as functional monomers show the lowest release of template followed by the material
Table 2. Batch Rebinding of the Template, Terbutylazine, to Mini-MIPs and Blank Nonimprinted Polymers in Dichlormethanea mini-MIP polymer
Cfree (mM)
nb (µmol/g)
P-MAA P-TFM
0.44 0.36
10.2 11.6
nonimprinted blank polymer KMIPc (mL/g)
Cfree (mM)
23 32
0.90 0.77
nb
(µmol/g) 1.82 4.18
separation factor
Kblankc (mL/g)
R () KMIP/Kblank)
2.0 5.4
11 5.9
a The polymers were sonicated for 15 min consecutively with 1 mL of dichloromethane, 1 mL dichloromethane containing 5% acetic acid, and finally with dichloromethane. Then 1 mL of a solution of the template (1 mM terbutylazine) in dichloromethane was added and the solution sonicated for 1 h. Thereafter the concentration of free terbutylazine in the supernatant (Cfree) was determined by reversed-phase HPLC (duplicate injections) using an external standard as reference. Only the polymers listed showed enhanced uptake of the template. Conditions otherwise as in Table 1 and as described in the Experimental Section. b The amount of adsorbed terbutylazine per gram of dry polymer, assuming a dry polymer weight of 55 mg/vial. c K ) n/Cfree.
Table 3. Capacity Factors for Various Triazines Obtained in Chromatography Using as Stationary Phase a Normal-Scale Version of P-MAA Imprinted with Tertbutylazinea imprinted polymer k′MIP
nonimprinted blank k′blank
separation factor R ) k′MIP/k′blank
triazine
10 nmol
100 nmol
10 nmol
100 nmol
10 nmol
100 nmol
terbutylazine ametryn prometryn atrazine
>25 9.7 9.9 22
18 6.7 6.3 10
1.46 2.46 2.47 1.23
1.25 2.25 2.25 1.08
>17 3.9 4.0 18
14 3.0 2.8 9.3
a The polymer synthesis and processing were as described in the Experimental Section. After conditioning of the columns (125 × 4 mm2) in MeCN/H2O/HOAc (92.5:2.5:5, v/v/v), the columns were evaluated using MeCN as mobile phase at a flow rate of 1 mL/min, a detection wavelength of 260 nm, and by 10-µL injections of two different loads of the triazines (10 or 100 nmol).
prepared using HEMA. A larger release was seen from the polymers prepared using VPY, MMA, and NVP. From the release test, additional useful information concerning the amount of unreacted monomer or side reactions may be obtained (Figure 1). A modest increase in the recovery was obtained after an additional wash in dichloromethane, but as expected, only after adding strongly competing modifiers such as acetic acid were high overall recoveries of the template obtained. The selection criteria in the first screen are thus based on the amount of released template and/or kinetics of release in the same solvent used as porogen during polymerization. In most cases, this solvent is the medium in which the polymers exhibit the most pronounced recognition.17,18 Thus, for a particular MIP, a quantitative release of the template in this step indicates that it will not recognize the template to a significant degree and the MIP may thus be discarded. On the basis of these criteria, only the miniMIPs prepared using MAA, TFM, and possibly HEMA can be expected to recognize the template in CH2Cl2. Nevertheless, a complete secondary screen for binding and selectivity (Table 2) was performed for all the polymers. In this, the rebinding of the template to the mini-MIPs and to blank nonimprinted polymers was compared. Alternatively, the selectivity may here be assessed by comparing the binding of the template with that of an added internal standard structurally related to the template.15 Supporting the validity of the suggested screening approach, only the MAA and TFM mini-MIPs rebound the template selectively under these conditions. Also it is clear that the highest selectivity was obtained using MAA as functional monomer. This is in agreement with previous results using the chlorotriazine atrazine as template15 (17) Yu, C.; Mosbach, K. J. Org. Chem. 1997, 62, 4057-4064. (18) Spivak, D.; Gilmore, M. A.; Shea, K. J. J. Am. Chem. Soc. 1997, 119, 43884393.
and appears mainly to be due to a difference in the nonspecific binding to the P-TFM and P-MAA blank polymers (Table 2). A normal-scale version of the P-MAA MIP was assessed as stationary phase in chromatography. As seen in Table 3, a number of structurally related triazines are strongly retained on the corresponding column. Interestingly, the separation factor for atrazine in acetonitrile, R ) 18, is the highest reported so far for MIPs evaluated under similar conditions. The retention of atrazine was also significantly higher than the retention obtained using a polymer imprinted with atrazine. This shows that template analogues can be successfully used to generate binding sites showing high affinity and selectivity for a particular target compound. CONCLUSIONS The described screening protocol allows a rapid identification of the factors of importance for the generation of binding sites of high affinity and selectivity for a particular target compound. It consists of two screening steps, the first based on the amount of released template from the polymers in the porogenic solvent and the second based on the amount of template rebound to the polymers after an exhaustive extraction of the template from the polymers. In the rebinding step, the selectivity is calculated based on a reference, which can be an internal standard or the rebinding to a blank nonimprinted polymer. The system is suited for automation which should allow a rapid processing of a large number of polymers. Rebinding quantification can be based on direct measurement of, for instance, UV absorption, fluorescence emission, or radioactivity, by MS detection or by analytical chromatography. In view of the large number of parameters that influence selectivity, affinity, capacity, and kinetic properties in the rebinding to MIPs, this and other methods15 for combinatorial Analytical Chemistry, Vol. 71, No. 11, June 1, 1999
2095
synthesis and screening of the polymers are expected to significantly accelerate the development of new molecular recognition elements. ACKNOWLEDGMENT The authors acknowledge support from the European Union. This work was part of the European network project MICA (Contract no. FMRX-CT98-0173) with the participants: Damia Barcelo´, CID-CSIC, Barcelona, Spain; Werner Blau, Trinity College Dublin, Ireland; Karl-Siegfried Boos, Maximilians-Universita¨t Mu¨nchen, Germany; Kees Ensing, University of Groningen, The Netherlands; George Horvai, Technical University of Budapest,
2096 Analytical Chemistry, Vol. 71, No. 11, June 1, 1999
Hungary; Lars Karlsson, Astra Ha¨ssle AB, Mo¨lndal, Sweden; David Sherrington, University of Strathclyde, U.K. Support also came from the Graduiertenkolleg: Chemie und Physik supramolekularer Systeme, at the University of Mainz. The authors also thank Hewlett-Packard, Waldbronn, Germany for providing the HP1050 system and Novartis (Basel, Switzerland) for the gift of the triazine samples.
Received for review December 30, 1998. Accepted February 26, 1999. AC981446P
