Jouarw OPBKWCIENCE ANDBIOEN~INEJWNQ Vol. 90, No. 6, 665-668. 2000
An Artificial Plastic Receptor That Discriminates Axial Asymmetry EI-ICHIRO FUKUSAKI,* AKIRA SAIGO, SHIN-ICHIRO KAJIYAMA, AND AK10 KOBAYASHI* Department of Biotechnology,
Graduate School of Engineering, Osaka University, 2-1 Yamadaoka, Suita 5654871,
Japan
Received 27 July 2QOO/Accepted13 October 2000
An &ti6cial plastic receptor that can discriminate axial asymmetry of optically active binaphthyldiamine derivativea was prepared by the ‘molecular imprinting’ techuique. A light-radical polymerization in the presence of an axially asymmetric compound, (R)-2,2’-his-methylc8rbonylanko-l,l’-blnaphtbyl (binaphthyldiamine his-a&amide (BINADA-ac)) as a template molecule with methacrylic acid (MA) as a functional monomer, and ethyleneglycol dimethacrylate (EGDMA) as a crosslinker, at 4’C. The obtained polymer exhibited a superior enantioselectivity to the (RJenantiomer by HPLC analyses. This phxstic receptor should recognke the template molecule with its shape and character of its functional groups. It should be useful in the development of chiral stationary phases for the optical resolutions of axially asymmetric compounds. mey words:
molecular imprinting, axial asymmetry, radical polymerization,
In the field of molecular recognition, the desired aim is the creation of a novel ‘host’ showing extremely high selectivity towards only the target ‘guest’ molecule. However, there are few general strategies by which one can freely design such highly guest-selective host molecules. The ‘molecular imprinting’ technique is a procedure that shows potential in creating such molecular recognition systems with high selectivity. In the ‘molecular’ imprinting technique, also known as a template polymerization, synthetic polymers are prepared with specific recognition sites for the target molecule in a tailor-made fashion. That is illustrated in Fig. 1. The procedure involves the following three steps: (i) self-assembly to yield a complex between the target molecule (template) and polymerizable monomers carrying certain functional groups (functional monomer) capable of interacting with the target molecule; (ii) polymerization of the monomers to freeze the optimal position for binding the template molecule; and (iii) subsequent removal of the template from the macro porous polymer leaving cavities that are complementary to the template in their size, shape, and arrangement of functional groups. Such ‘tailor-made’ binding sites for the template molecule would be generated through above steps. This technique can be used for preparation of sensors or stationary phases for separations (l-6). However, the ‘molecular imprinting’ technique is most useful to create stationary phases for the optical resolution of racemic compounds. Amino acids, nucleotides, and several drugs containing chiral centers have been targeted for optical resolution by ‘molecular imprinting’ (3). To date, however, there is no successful report of the optical resolution of an axial-asymmetric compound. In this current study, we report the preparation and investigation of an axial-asymmetric compound-imprinted polymer. We chose an optically active binaphthyldiamine-derivative, a typical axial-asymmetric compound as a model substrate, as binaphthyl compounds are useful as chiral catalytic elements in chiral synthesis (7, 8).
receptor]
MATERIALS AND METHODS Materials
Functional monomers, Methacrylic acid (MA) and styrene (Sty) were purchased from Tokyo Chemical Co., Tokyo. Cross linkers, ethylene glycol dimethacrylate (EGDMA) and divinylbenzen (DBV) were also purchased from Tokyo Chemical Co., Tokyo. 2,2’Azobisbutyronitrile (AIBN), chloroform, acetonitrile, methanol, acetyl chloride, propanoyl chloride, n-butanoyl chloride and benzoyl chloride were also purchased from Wako Co., Osaka. (R)-2,2’-Diamino-1, l’-binaphthy1 (binaphthyldiamine: (Z?)-BINADA), (5’)-2,2’diamino-1, I’-binaphthyl (binaphthyldiamine: (S)-BINADA), 4-vinylpyridine (CVPy) Cvinylbenzoic acid (CVBA) were purchased from Aldrich Chemical Co., USA. A template molecule, (Z?)-2,2’-bis-methylcarbonylamino-l , I’-binaphthy1 (binaphthyldiamine-bis-acetamide: (R)-BINADA-ac) were prepared from (R)-BINADA and acetyl chloride in usual manner. Analogues of template, binaphthyldiarnine-bis-n-propanoylamide ((R>BINADA-pr), binaphthy&amine-bis-n-butanoylamide ((&BINADA-bu), binaphthyldiamine-bis-benzoylamide ((Z?)-BINADA-bz), and ((R)-BINADAbinaphthyldiamine-bis-naphthoylamide nap) were prepared from (R)-BINADA and propanoyl Functionalmonomer
Crosslinkingagent
Self-assembly
Target molecule (template)
Polymerization I
* Corresponding author.
FIG. 1.
665
Principle of molecular imprinting technique.
666
J. Bmsc~. BIOENO.,
FUKUSAKI ET AL. Functional monomers BINADA: R=NHz BINADA-SC:R=NHCOCHs BINADA-: RzNHCOCzHs BlNADA-bu: R=NHCOC$H, BINADA-bz: R=NHCOPh BINADA-nap: R-NHCONaph BINAOHzR=OH
(R)-emntiomer
FIG. 2.
(S)-enmtiomer
Structures of target molecule and its derivatives. Cross linkers
chloride, n-butanoyl chloride, benzoyl chloride and naphthoyl chloride respectively in usual manners. The enantiomers of template and its analogues ((S)-BINADAac, (S)-BINADA-pr, (S)-BINADA-bu, (S)-BINADA-bz, (S)-BINADA-nap) were also prepared in the same manner described above except using @-BINADA instead of (R)-BINADA. Structures of target molecule and its derivatives are indicated in Fig. 2. Molecular imprinting The typical procedures for the preparation of (R)-BINADA-imprinted polymers are as follows. For the preparation of the imprinted polymer, 0.5 mm01 of (R)-BINADA-ac as a template, 5 mmol of functional monomer, 12.5 mm01 of cross linker and 0.2mmol of AIBN as an initiator were dissolved in 50ml of chloroform in a glass ampoule. After the mixture had been degassed by injection of nitrogen gas, the ampoule was sealed under vacuum. The polymerization was initiated by UV light irradiation (365 run) at 4°C and the reaction mixture was left for 12 h. The rigid polymer obtained was ground in a mortar (ANMlOO, Nitto Kagaku Co., Osaka) and sieved to collect 30-60 micrometer polymer particles. Fine particles were removed by decantation in acetonitrile. Reference polymer was prepared in the same manner without the addition of (R)BINADA (template molecule). Evaluation
of binding ability of imprinted polymer
The high performance liquid chromatography (I-IPLC) study was carried out to evaluate the binding affinities of imprinted polymers to the template molecule. The HPLC system consisted of a pump (LC-lOAT, Shimadzu, Kyoto) and UV-detector (SPD-lOA, Shimadzu). The polymer particles obtained above were slurry-packed into stainless steel column (150mm x 4.6mm i.d.) and template molecule was washed out exhaustively with acetonitrile until stable base line was obtained. Reference polymers were prepared in the same manners. A 2.5 micro liter of sample solution containing 0.25 microgram of sample was injected and monitored at 240nm. Acetonitrile was used as the eluent at a flow rate 1 ml/min. The retention capacity factor for (R)-enantiomer (kx) was calculated by the equation k’= (fR- &J/t 0, where tR is the retention time for the (R)-enantiomer sample and to is the time to elute the void marker toluene. The capacity factor for (S)-enantiomer (k’s) was calculated in the same manners described above except using ts value, which is the retention time for the (s)enantiomer, instead of tR value. Enantioselectivity was expressed as the enantio separation factor (a), which is a ratio of the k’ value of the template to that of its enantiomer ((~=k’a/k’s). All the steps were carried out at room temperature. RESULTS AND DISCUSSION Effect of the functional
monomer
on stereoselectivity
FIG. 3.
Structures of functional monomers and cross linkers.
We first examined the effects of monomer functionality on the stereospecificity of the imprinted polymers. The structures of the functional monomers tested are shown in Fig. 3. Methacrylic acid (MA) is a commonly employed as a functional monomer containing a carboxylic acid. In addition to MA, 4-vinylbenzoic acid (CVBA), 4-vinylpyridine (4-VPy) and styrene (Sty) were investigated as functional monomers. Ethyleneglycoldimethacrylate (EGDMA) was the primary cross linker used in this study, though in the case of MA, divinylbenzene (DBB) was also used as a cross linker. Several molecular imprintings were performed with (R)-BINADA-ac as a template according to the method described in Materials and Methods. Each polymer’s capacity factor (k’) and enatioseparation factor (a) were calculated from the chromatographic results obtained with BINADA-ac as the anatyte. Results are summarized in Table 1. The highest enantioselectivity was observed when MA and EGDMA were used as the functional monomer and the cross linker respectively (a=5.5, kX=3.9, k$=0.7) (run 1). When divinylbenzene (DVB) was used as a crosslinker instead of EGDMA, a more moderate enantioselectivity was observed (a= 3.9, kX= 3.4, k’s=O.9) (run 2). When styrene (Sty), a vinyl monomer without a carboxylic group, was tested as a functional monoTABLE 1.
Effects of varying the functional monomer and cross linker combination
Run’
Functional monomerb
Cross linkerb
1 2 3 4 5 6” 7’
MA MA Sty 4-VPy 4-VBA MA MA
EGDMA DBV EGDMA EGDMA EGDMA EGDMA DBV
Capacity factorc
Segy$
, kR
ks
a
3.9 3.4 0.1 0.2 0.6 0.4 1.0
0.7 0.9 0.1 0.2 0.5 0.4 1.0
5.5 3.9 1.0 1.0 1.2 1.0 1.0
B Polymerization conditions: template, (R)-BINADA-ac (0.5 mmol); functional monomer (5 mmol); cross linker (12.5 mmol); AIBN (0.2 mmol); chloroform, 50 ml; 4°C; 365 nm: 12 h. b The structures of functional monomers and cross linkers are summarized in Fig. 3. c Retention capacity factor (k’= (f~ - @/to). d Enantioseparation factor (a = kk/k’,) for BINADA-ac as analyte. e Non-imprinted control without template.
VOL.
TABLE 2.
Runa (m%l) 1 2 3 4 5 6 7 8
ARTIFICIAL PLASTIC RECEPTOR
90, 2ooo
1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0
Effects of the ratio of functional monomer and cross linker Ratio $Yt?t?$ 12.5 12.5 12.5 12.5 12.5 12.5 12.5 12.5
W&$Y
0.6 0.6 2.0 2.3 3.9 3.1 3.8 1.9
(k’s)
(a)
0.3 0.3 0.6 0.7 0.7 1.0 1.2 0.9
2.0 2.0 3.2 3.3 5.5 3.1 3.0 2.1
* Polymerization conditions: template, (R)-BINADA-ac (0.5 mmol); AIBN (0.2 mmol): chloroform, 50 ml: 4°C: 365 nm: 12 h. b Retention capacity factor (k’= (ta - tO)/tO). c Enantioseparation factor (a = k’a/k’s) for BINADA-ac as analyte.
mer, not only was no enantioselectivity observed (a= 1.0) but the analyte template was barely retained (k’, =O.l, k$=O. 1) (run 3). When Cvinylpyridine (4-Vpy), an aromatic amine capable of forming the hydrogen bonds, was used as the functional monomer, no enatioselectivity was showed (a= 1.0) although the template molecule was retained somewhat (Ka =0.2, k$= 0.2) (run 4). These results suggest that carboxyl groups are necessary to recognite the template (Z?)-BINADA-ac. However when 4-vinyl benzoic acid (4-VBA), a vinyl monomer with an aromatic carboxyl group, was employed as a functional monomer, contrary to expectation, very little enantioselectivity was observed (a= 1.2) (run 5). Above result indicated that a combination of MA as a functional monomer and EGDMA as a cross linker would be the best one for a molecular imprinting of optically active BINADA-ac. We next examined the ratio of MA and EGDMA (Table 2.) The highest enantioselectivity was observed when the EGDMA/MA ratio is 2.5. Finally, this optimized imprinted polymer as prepared in the conditions described above was subjected to a substrate selectivity examination. Specticity of imprinted polymer to BINADA derivatives In order to evaluate the substrate specificity of the (R)-BINADA-ac-imprinted polymer, various BINADA derivatives were subjected to HPLC analysis as shown in Fig. 2. The tested derivatives are briefly described below: (i) propanoyl amide (BINADA-pr) and n-butanoyl amide (BINADA-bu) as aliphatic amides in which chain length differs from BINADA-ac; (ii) benzoyl
FIG. 4
Specificity of imprinted polymers to BINADA derivatives
SQa&J Run
@GDMA’MA) (k’s) 12.5 6.3 4.2 3.1 2.5 2.1 1.8 1.6
TABLE 3.
667
Sample” BINADA-ac BINADA-pr BINADA-bu BINADA-bz BINADA-nap BINADA BINAOH
Capacity factorb
(k’d
(k’s)
(4
3.9
0.7 0.7 0.7 0.5 0.8 0.4 0.1
5.5 2.5 2.1 1.2 1.1 1.0 1.0
1.7 1.4 0.6 0.9 0.4 0.1
a The structures of the tested samplesare summarizedin Fig. 2. b Retention capacity factor (k’=&-to)&). c Enantioseparation factor (a = k’,Jk’s).
amide (BINADA-bz) and naphthanoyl amide (BINADAnap) as aromatic amides; (iii) the free amine (BINADA); and (iv) binaphthol (BINAOH). The results are summarized in Table 3. Comparing first the results of aliphatic amide derivatives (runs 2 and 3) with the template (run l), the enantioselectivity decreased, as the chain length of the amide group increased (BINADA-ac, a = 5.5; BINADA-pr, a=2.5; BINADA-bu, a=2.1). These results suggest that the enantioselectivity should be lowered by the increase in the steric hindrance around amide group. The more sterically hindered benzoyl amide (BINADA-bz) was barely recognized enantioselectively (a = 1.2, run 4). The very hindered naphthanoyl amide (BINADA-nap) was not only enantioselectively recognized extremely poorly (a=l.l, run 5), it showed little retention #X=0.9, k’s= 0.8), suggesting that the naphthalene ring of the amide group, rather than bi-naphthyl part of BINADA-nap might be inserted into the cavity generated by molecular imprinting which originally recognized bi-naphthyl part of the BINADA-ac. The free amine, BINADA, was not enantioselectively recognized and was also retained very little (run 6, a=l.O, k’a=0.4, k’s=O.4). These results suggest that the carboxylic groups derived from MA in the imprinted polymer are in an unsuitable location to form a hydrogen bonding with the free amino group of BINADA. In the case of the sterically unhindered free, BINAOH (run 7), not only was no enantioselectivity observed (a = 1.O), but it was not retained at all (k&=0.1, k’s=O.l). The above results could not indicate the clear pattern with which imprinted polymer recognizes the template guest molecule, BINADA-ac. However a possible bind-
Possible mechanism of recognition of BINADA by the imprinted polymer.
668
J. BIOSCI.BIOBNCL,
F’UKUSAKI ET AL.
ing pattern should be mentioned that each carbonyl of the guest is hydrogen bonded to an MA-derived hydroxyl group of the imprinted polymer, and the binaphthyl naphthalenes are bonded by Van der Waals interaction in a pocket of polymer backbone methylenes derived from EGDMA (Fig. 4). Conclusion A plastic receptor that enantioselectively discriminates the axial asymmetry of binaphthyldiamine (BINADA) derivatives was prepared by the ‘molecular imprinting’ technique and exhibited superior enantioselectivities with observed enantioseparation factors of up to 5.5. The plastic receptor should be useful in the development of chiral stationary phases for the optical resolution of axial asymmetric compounds. In the future, it should be possible for the prepared plastic receptor to be used as an entropy catalyst that regulates the stereochemistry in the templates coupling reaction of naphthol derivatives to yield binaphthol compounds. ACKNOWLRDCMKNTS The authors wish to thank Mr. Akira Maeda for technical assistance. The author also thank Kansai Research Institute Co., Japan for their iinancial support.
REFERENCES 1. Conuack, P. A. G. and MOS~JWII, K.: Molecular imprinting: recent developments and the road ahead. Reactive & Functional Polymer, 41, 115-124 (1999). 2. Owens, P.K.,Karlason, L., Lutz, E. S.M., and Andemon, L. I.: Molecular imprinting for bio and pharmaceutical analysis. Trends Anal. Chem., 18, 146-154 (1999). 3. Takeucki, T. and Aaginaka, J.: Separation and sensing based on molecular recog&ion using molecularly imprinted polymers. J. Chromatog. B. 728. l-20 (1999). 4. Mckert, F. L. and &y&a, b.: M&cul~ar imprinting in chemical sensing. Trends Anal. Chem., 18, 192498 (1999). 5. Yano, K. and Kambe, I.: Molecularly imprinted polymers for biosensor application. Trends Anal. Chem., 18, 199-204 (1999. 6. Wet&y, S. A., Panasynk, T. L., Piletnkaya, E. V., Nicholls, I. A., and Ulbrickt, M.: Receptor and transport properties of imprinted polymer membranes. J. Membrane Sci., 157, 263278 (1999). 7. Ohta, T., Miyake, T., seldo, N., Knmobayaebi, A., and Tak~ya, H.: Asymmetric hydrogenation of olefins with aprotic oxygen functionalities catalyzed by BINAP-Ru(I1) complexes. J. Org. Chem., 68, 357-363 (1995). 8. Knmobayasbl, H., Sayo, N., Akntagawr, S., Sakagnchi, T., and Tsnruta, H.: Industrial asymmetric synthesis by use of metal-BINAP catalysts. J. Chem. Sot. Jpn., 835-846 (1997).