Abstract
Corticosterone in water-ethanol solution can eject “solvated electrons” (eaq−) when excited into the singlet state by monochromatic UV-light (λ=254 nm). As a consequence of this process free radicals and H+ ions were also generated. Hence, the objectives of this study were to determine the quantum yield, Q, at different corticosterone concentrations, and elucidate the fate of the generated free radicals and the involved reaction mechanisms. Because of the formation of associates, which consume a part of the emitted eaq−, the Q decrease with increase of cortisone concentration. Additionally the H+ ions scavenge and convert a part of the ejected eaq− into H-atoms. In comparsion with progesterone, the Q of corticosterone is much higher. Evidently, this effect is due to the two OH groups of corticosterone, which act as intense emission centres for eaq−. Thereby, the generated free radicals from corticosterone lead to formation of metabolites, which were analyzed by combination of liquid-chromatography with mass spectrometry (LC/MS) method. Two of them were identified: 5α-pregnan-3α, 21-diol-11, 20-dione and 20β-dihydroxycortisone. Both have the same mass number of 348.230. To explain the involved, rather complicated processes, a probable reaction mechanism is suggested.
Hormones are generated by highly specified endocrine glands at an extremely low concentration. They control the harmonic course of specific biological functions in the organism.
Recently it was found that hormones in polar media, when excited into the singlet state, eject solvated electrons (e−aq) (1, 2). Additionally, hormones can also consume e−aq (reaction rate constant, k≥109-1010 l/mol/s) as well as transfer e−aq (acting as electron mediators) to other hormone types or to biological systems directly, or via the brain, making communication between them feasible (3).
Previous studies of aqueous organic and inorganic substances showed that the e−aq yield strongly depends on the pH of the medium, substrate concentration and absorbed radiation energy (4). The same effects were also found to be valid for corticosterone (5). Thereby with the increase of corticosterone concentration, the emitted e−aq- yield decreases because of associated formation (unstable complexes) at more than 10−7 mol/l corticosterone, which consume a part of the generated e−aq. The hormone molecules eject e−aq, resulting in the formation of free radicals. Both e−aq and the free radicals lead to generation of metabolites with specific biological properties, depending on the current milieu. Furthermore, it was established that hormones can be re-generated by electron transfer from an efficient electron donor, such as vitamin C (1, 2, 5). This effect depends on several factors, such as substrate concentration, pH of the medium and the reaction rate constants of the involved processes (5).
Based on current knowledge, it was of special interest to determine the quantum yield (Q) of the ejected e−aq in dependence of the absorbed UV dose and furthermore to identify the metabolites generated as subsequent products of the free radicals formed by electron emission. Therefore, the objectives of the present investigations were to learn more about the reaction mechanisms of the electron emission from corticosterone and consequently about metabolite generation.
Materials and Methods
Corticosterone of highest purity available (≥92%; Sigma-Aldrich, Vienna, Austria) was used as obtained. Since corticosterone is not completely soluble in water, a solvent mixture of triple-distilled water/ethanol (40/60) was applied. To prevent oxidation of corticosterone during the solvation process, the solvent mixture was first saturated in the irradiation vessel with high purity argon at 37°C for 20 min. Then the substrate was added under intense stirring. A 4π-geometry double-wall irradiation vessel (6) in combination with a low-pressure mercury lamp (HNS 12, OSRAM, 12 W, OSRAM GmbH, Vienna, Austria) with incorporated VYCOR filter for elimination of the 185 nm line was used for irradiation purposes (6,7). The intensity of the monochromatic UV light (λ=254 nm; 4.85 eV/hv) was determined by monochloric acetic acid actinometer (8).
Samples containing different substrate concentration and having been treated with different UV doses (hv/l) were analysed immediately afterwards. The e−aq emitted by corticosterone at given UV doses and corticosterone-concentrations were scavenged by chlorethanol (1×10−2 mol/l). The resulting chloride ions were determined spectrophotometrically (9).
(1) (2)The metabolites arising from the e−aq emission by corticosterone were analysed using liquid chromatography/mass spectroscopy (LC/MS). Luna pentafluorophenyl 150 mm × 2.1 mm ID (Phenomenex Inc., Torrance, CA, USA) column operated at 45°C and with a column flow of 500 μl/min was applied. Substances were eluted using a gradient composed of 25×10−3 mol/l ammonium acetate (Sigma-Aldrich) in water as mobile phase A and acetonitrile (Sigma-Aldrich) as mobile phase B. The separation column was mounted into an UltiMate 3000 nano RSLC HPLC system (ThermoFisher, Germering, Germany). Samples were injected using the user-defined protocol with mobile phase A as transport liquid. The HPLC program is shown in Table I.
The spectrophotometric measurements, prior to mass detection were performed at three wavelengths: 214 nm, 244 nm and 280 nm.
The MS detection was performed using positive electrospray ionization on a maXis Impact q-ToF mass spectrometer (Bruker, Bremen, Germany). Capillary voltage was set to 4.5 kV, dry gas was nitrogen at 8.0 l/min, nebulizer gas was set to 3.5 bar, ion-transfer capillary temperature was 270°C, and the mass range from m/z 50 to 2500 was scanned with 4 Hz for MS and 8Hz for MS/MS. Automatic MS/MS analysis was performed on the top three ions with intensities higher than 3,000 counts. Active exclusion for 60 s or after three spectra was used and single-charged ions were included for MS/MS analysis. Lock mass of m/z 1221.9906 was used for internal calibration and measurement corrections. Argon was used as collision gas and automatic MS/MS settings were applied.
Results and Discussion
Emission of e−aq. As indicated above, most UV-induced corticosterone metabolites originate from free radicals formed by the e−aq emission process. As previously stated, for certain hormones, the Q(e−aq) decreases with increasing substrate concentration (2). This effect is ascribed to associate formation (instable complex), which consumes a part of the ejected e−aq (reaction rate constant, k≥1010 l/mol/s). In the present case, H+ ions, which are simultaneously generated by the OH-groups of corticosterone, also react with the electrons (see below). These effects were studied using various corticosterone concentrations and different UV doses. The results are presented in Table II.
By increasing the corticosterone concentration, Q(e−aq) is strongly reduced. Obviously, the processes of e−aq ejection and consumption of e−aq by associates, and the reaction with H+ ions and free radicals in the medium, are competing.
With the progression of corticosterone photolysis, the yield of metabolites increases correspondingly. These metabolites absorb certain parts of the UV-light, and likewise emit e−aq. This process can be followed by studying the Q(e−aq) yield in dependence of the UV dose absorbed.
The results obtained for 5×10−5 mol/l corticosterone are shown in Figure 1.
In Figure 1, the course of curve A shows a very rapid decrease of Q(e−aq) with increase of the absorbed UV dose. The graph passes a maximum at a dose of 1.5×1021 hv/l, which corresponds to the yield of e−aq emitted by the primary formed metabolites. By further increasing the UV dose applied, a second, but smaller maximum of e−aq appears, reflecting the e−aq emitted by the secondary metabolites. In accordance with previous experimental data (5), the obtained results indicate that e−aq emission can mostly be attributed to the OH-groups of the corticosterone molecule, whereas the basic molecular skeleton remains more or less unchanged. Curve B of Figure 1 shows corticosterone photolysis as a function of absorbed UV dose, having a rather similar course to that of curve A, which confirms the statement above.
Comparing the structure formulas of progesterone and corticosterone (Figure 2), it is obvious that both have the same basic molecular skeleton, namely, –C=O groups at the 3rd and 20th positions. However, corticosterone exhibits two OH-groups at positions 11 and 21 additionally. As is well-known, when electronically excited, OH-substituted groups in organic molecules (phenol, methoxybenzenes etc.) act as centres for e−aq emission (4). For illustration, some Q(e−aq) data are presented in Table III for progesterone and corticosterone.
From the data in Table III, it can be concluded that the Q(e−aq) for corticosterone are several times higher compared to those for progesterone. Obviously, two pathways for the ejection of e−aq by corticosterone are possible: (i) e−aq emission resulting from the skeleton structure as in the case of progesterone; and (ii) e−aq ejection from the OH–groups, acting as emission centres for e−aq. Analogous to phenol (4), the e−aq ejection from an OH–group is accompanied by formation of H+ ions, which can scavenge a part of e−aq. This is demonstrated by the pH decrease in the medium with increasing the UV dose absorbed, as shown in Figure 3. As previously mentioned, the H+ ions can react with a part of the emitted e−aq by converting them into H atoms. For example, this is demonstrated in the case of phenol: (3) (4)
LC/MS analysis of corticosterone metabolites. The metabolites and products resulting from irradiation of a solution containing 5×10−5 mol/l corticosterone are shown in the chromatogram in Figure 4. The corresponding peak areas are given in Table IV for different absorbed UV doses. From the data presented in Table IV, it is clearly shown that: (i) corticosterone (peak 1) strongly decomposes with increasing UV dose except at the dose of 7.8×1020 hv/l; (ii) peak 2, corresponding to 5α-pregnan-3α,21-diol-11,20-dione, increases with UV dose, but start degrading at higher doses (>11.6×1020 hv/l); (iii) peak 3, corresponding to the metabolite 20β-dihydroxycorticosterone, which at first increases with UV dose, at 7.8×1020 hv/l shows a minimum yield; (IV) based on the obtained MS data, peaks 4 and 5 seem to be unidentified artefacts. The obtained results demonstrate the involvement of a rather complicated reaction mechanism, which is initiated by the emission of e−aq and thereby resulting in the production of free radicals.
Obviously, by irradiation, the carbonyl groups (−C=0) at position 3 (ring A) and position 20 (see Figure 2) are converted into OH-groups, resulting in 5α-pregnan-3α,21-diol-11,20-dione (peak 2) and 20β–dihydroxycorticosterone (peak 3) (Figure 4), respectively. The possible reaction steps for the conversion of –C=O group (position 3) into the OH-group are presented schematically below: (5) (6) Similar reaction steps are postulated for –C=O (position 20): (7) Both metabolites (peak 2 and 3) have different structure formulas (see Figure 5), but exactly the same mass 348.230.
On the other hand, OH-groups (position 11 and 21 in the corticosterone molecule) can also be re-generated after e−aq emission. The two-step reaction concerning the OH-group as a source of e−aq emission on position 11 and subsequent regeneration of OH-groups by H-addition (see Equation 4) are probably as follows: (8) Peaks 4 and 5 (see Table IV) were also registered in unirradiated samples. This possibly indicates, as previously mentioned and based on MS data obtained, that these areas yet unidentified artefacts.
The series of small peaks numbered 6 to 26 (Figure 4) very likely result from the combination of free radicals formed in the photolysis of corticosterone in the bulk of the solution.
The main points of the present studies can be summarized as follows. (i) Corticosterone in water–ethanol solution emits e−aq, when excited into singlet state. (ii) Q(e−aq) decreases with increasing corticosterone concentration because of the formation of associates and H+ ions generated simultaneously with the e−aq ejection from OH-groups (Table II). (iii) The Q(e−aq) of corticosterone is much higher compared to that of progesterone, although both hormones possess the same basic skeletal molecular structure (Figure 2). A reason for this effect may be that the corticosterone molecule possesses two additional OH-groups (positions 11 and 21; Figure 2) which act as intense ejecting centres for e−aq. (iv) Simultaneously with e−aq emission, the corticosterone free radicals generate metabolites, which were analysed by LC/MS. (v) The mass chromatogram (Figure 4) shows several peaks, representing the various metabolites. Two of them are identified: peak 2 being 5α-pregnan-3α,21-diol-11,20-dione and peak 3 being 20β-dihydroxycorticosterone (see Figure 5). (vi) For explanation of the results obtained a probable reaction mechanism is suggested. The obtained results provide a deep insight into the reaction mechanisms of corticosterone and are, therefore, of potential interest for medicine and molecular biology.
- Received March 24, 2014.
- Revision received May 28, 2014.
- Accepted May 29, 2014.
- Copyright © 2014 International Institute of Anticancer Research (Dr. John G. Delinassios), All rights reserved