Sol-Gel Entrapped Levonorgestrel Antibodies: Activity and Structural Changes as a Function of Different Polymer Formats

Abstract

The paper describes development of a sol-gel based immunoaffinity method for the steroid hormone levonorgestrel (LNG) and the effects of changes in the sol-gel matrix format on the activity of the entrapped antibodies (Abs) and on matrix structure. The best sol-gel format for Ab entrapment was found to be a tetramethoxysilane (TMOS) based matrix at a TMOS:water ratio of 1:8, containing 10% polyethylene glycol (PEG) of MW 0.4 kDa. Addition of higher percentages of PEG or a higher MW PEG did not improve activity. No activity was obtained with a TMOS:water ratio of 1:12, most likely because of the very dense polymer that resulted from these polymerization conditions. Only minor differences in the non-specific binding were obtained with the various formats. TMOS was found to be more effective than tetrakis (2-hydroxyethyl)orthosilicate (THEOS) for entrapment of anti-levonorgestrel (LNG) Abs. However, aging the THEOS-based sol-gel for a few weeks at 4 °C stabilized the entrapped Abs and increased its binding capacity. Confocal fluorescent microscopy with fluorescein isothiocyanate (FITC) labeled immunoglobulines (IgGs) entrapped in the sol-gel matrix showed that the entrapped Abs were distributed homogenously within the gel. Scanning electron microscopy (SEM) images have shown the diverse structures of the various sol-gel formats and precursors.

Keywords: THEOS; TMOS; antibodies; immunoaffinity purification; levonorgestrel; polyethylene glycol; sol-gel.

Figure 1

Effects of PEG on the structure and activity of entrapped Abs in 1:8-TMOS-based sol-gel. (A) LNG (100 ng) was applied on 1-day-aged sol-gel columns containing 0, 10 or 20% of 0.4-kDa or 10-kDa PEG. Amount of eluted LNG was determined by ELISA. Vertical bars labeled ‘Empty’ represent columns that were not doped with Abs, and therefore represent non-specific binding. Bars labeled ‘Ab’ represent the total binding capacity and bars labeled ‘Net’ represent the specific binding values (Ab minus empty). Each bar represents the mean ± S.E.M of n experiments, where: for 0% PEG, n = 2; for 10 or 20% of 0.4-kDa PEG, n = 6; for 10 or 20% of 10-kDa PEG, n = 4. Means with the same letter did not differ significantly at p < 0.05; (B) SEM image of 1:8 TMOS-based sol-gel containing 0% PEG; (C) SEM image of 1:8 TMOS-based sol-gel containing 10% of 0.4-kDa PEG. Magnification, ×50,000; scale bar represents 500 nm.

Figure 2

Effects of various sol-gel formats (characterized by differing TMOS:HCl ratios) on their structure and on the activity of entrapped Abs. (A) LNG (100 ng) was applied on 1-day-aged sol-gel columns prepared with 10 or 20% of 0.4-kDa or 10-kDa PEG, at four different TMOS:HCl ratios (1:4, 1:6, 1:8 and 1:12). Amounts of eluted LNG were determined by ELISA. Data are presented as net binding, i.e., binding of LNG to Abs-doped columns minus the non-specific binding of LNG to 'empty' columns. Each vertical bar represents the mean ± S.E.M of n experiments, where: for TMOS:HCl = 1:8, n = 6; and for 1:4, 1:6, and 1:12, n = 2. An asterisk indicates significant differences at p < 0.05. (B), (C), and (D) are SEM images of the sol-gels prepared at various TMOS:HCl ratios: 1:4, 1:8, and 1:12, respectively, all containing 10% of 0.4-kDa PEG. Magnification, ×50,000; scale bar represents 500 nm.

Figure 3

Structures of the sol-gel precursors: (A) TMOS; (B) THEOS.

Figure 4

Effects of different monomers on the structure of the sol-gel and on the activity of entrapped Abs. (A) LNG (100 ng) was applied on 1-day-aged sol-gel columns prepared with TMOS or THEOS as precursor, at a precursor:HCl ratio of 1:8, with 10% of 0.4 kDa PEG. Amounts of eluted LNG were determined by ELISA. Each vertical bar represents the mean ± S.E.M of two experiments. Means with the same letter did not differ significantly at p < 0.05; (B) and (C) are SEM images of freshly prepared sol-gels based on 1:8 THEOS and 1:8 TMOS, respectively, containing 10% of 0.4-kDa PEG. All other details are as described in the legend to Figure 2. Magnification × 50,000; scale bar represents 500 nm.

Figure 5

Effects of aging on matrix structure and on Abs activity. (A) LNG (100 ng) was loaded on columns containing sol-gels based on TMOS at a TMOS:HCl ratio of 1:8 containing 10% of 0.4-kDs PEG, or on THEOS at a THEOS:HCl ratio of 1:8 without PEG, as precursors. The columns were stored for periods of 1 day (Week 0) up to 4 weeks at 4 °C. Amounts of eluted LNG were determined by ELISA. Binding capacity represent the ratio (expressed as a percentage) between the amounts of LNG recovered from a TMOS or THEOS column at each time point and that from a freshly prepared TMOS sol-gel column (Week 0, designated as 100%). Each vertical bar represents the mean ± S.E.M of two experiments. Statistical analysis compared the binding capacity of doped TMOS- or THEOS-based sol-gel columns separately as a function of aging (indicated by letters) and of the binding capacity of TMOS vs. THEOS at each time point (indicated by an asterisk). Means with the same letter did not differ significantly at p < 0.05. Means marked with an asterisk (*) differed significantly at p < 0.05. (B) and (C), respectively, are SEM images of freshly prepared and 4-week-old TMOS-based sol-gels with a TMOS:HCl ratio of 1:8 containing 10% of 0.4-kDa PEG. Magnification, × 50,000; scale bar represents 500 nm.

Figure 6

IgG distribution within a TMOS-based sol-gel prepared with a TMOS:HCl ratio of 1:8 containing 10% of 0.4-kDa PEG. A. Confocal microscopy image of ‘empty’ xerogel (i.e., sol-gel without IgGs). B. Images of xerogels containing FITC-labeled IgGs; xerogels were mounted on microscope slides and scanned with a fluorescent confocal microscope (a) phase image (without using laser beam at the tested wavelength); (b) fluorescence level of a doped sol-gel sample; and (c) superimposition of the phase and fluorescent images in (a) and (b). C. Confocal microscopy images of (I) empty, and (II) doped xerogels containing FITC-labeled IgGs and based on TMOS at a TMOS:HCl ration of 1:8 containing 10% of 0.4-kDa PEG. Images A and B were taken with a magnification of × 60; those in C were taken with a × 60 magnification and a × 2 zoom at a depth of 15 μm. The scale bars in (A) and (B) represent 50 μm; and in (C) represents 20 μm.