Radical-based biomimetic chemistry has been applied to building-up libraries necessary for biomarker development.
The involvement of free radicals in life sciences has constantly increased with time and has been connected to several physiological and pathological processes. This subject embraces diverse scientific areas, spanning from physical, biological and bioorganic chemistry to biology and medicine, with applications to the amelioration of quality of life, health and aging. Multidisciplinary skills are required for the full investigation of the many facets of radical processes in the biological environment and chemical knowledge plays a crucial role in unveiling basic processes and mechanisms. We developed a chemical biology approach able to connect free radical chemical reactivity with biological processes, providing information on the mechanistic pathways and products. The core of this approach is the design of biomimetic models to study biomolecule behavior (lipids, nucleic acids and proteins) in aqueous systems, obtaining insights of the reaction pathways as well as building up molecular libraries of the free radical reaction products. This context can be successfully used for biomarker discovery and examples are provided with two classes of compounds: mono-trans isomers of cholesteryl esters, which are synthesized and used as references for detection in human plasma, and purine 5′,8-cyclo-2′-deoxyribonucleosides, prepared and used as reference in the protocol for detection of such lesions in DNA samples, after ionizing radiations or obtained from different health conditions.
The reactivity of free radicals revealed its enormous importance for many biological events, including aging and inflammation. Nowadays, it is more and more evident that the clarification of each chemical step involved in this reactivity is needed, in order to understand the underlying mechanisms and envisage effective strategies for the control of free radicals and repair of the damage. The contribution from chemical studies is fundamental, but the direct study in the biological environment can be difficult, since the superimposition of different processes complicates and perturbs the examination of the results and the related conclusions. Therefore, the strategy of modeling free radical reactions under biologically related conditions has become a fundamental step in the research of chemical mechanisms in biology.
In the last decade our group developed models of free radical processes under biomimetic conditions. In particular we envisaged biologically relevant transformations of unsaturated fatty acids, nucleosides, and sulfur-containing amino acids and put them in the track to be evaluated and validated as biomarkers of health status.1-4
Our general approach consists of three modules:
We chose two classes of relevant biomarkers to accredit this approach: cholesteryl esters and purine 5′,8-cyclo-2′-deoxyribonucleosides.
1. Synthesis of Mono-trans Isomers of Cholesteryl Esters
2. Isolation of Cholesteryl Esters Fraction from Human Serum
3. Characterization of Mono-trans Cholesteryl Esters by Raman Spectroscopy
4. Derivatization to Fatty Acid Methyl Esters (FAME) and Analysis by Gas Chromatography
5. Synthesis of (5’R)- and (5’S)-5′,8-cyclo-2′-deoxyadenosine
6. Synthesis of (5’R)- and (5’S)-5′,8-cyclo-2′-deoxyguanosine
7. Synthesis of Isotopic Labeled Purine (5’R)- and (5’S)-5′,8-cyclonucleosides
8. Gamma Radiolysis of DNA Aqueous Solutions
9. Enzymatic Digestion of DNA
10. DNA Sample Desalinization Prior to Analysis
11. LC-MS/MS Quantitative Analysis
The isomerization process has been described in particular for cholesteryl esters affording the mono-trans isomers of linoleic and arachidonic acids shown in Figure 1 as the first products of this attack, which can occur under conditions of free radical stress in the biological environment.5
In Figure 3 the chemical mechanism involved in the cis-trans double bond isomerization is shown. The radical source is indicated in a generic way to afford S-centered radicals. In the described protocols the radical source is UV light that is able to break homolitically the S-H bond present in the thiol molecule RSH.
Figure 4 summarizes the three steps protocol for the synthesis of the modified lipid class and their detection in human plasma: the synthesis represents a biomimetic free radical process and also provides a one-pot convenient entry to the geometrical isomers, without any contamination by positional isomers followed by purification and isolation protocols.
Several analytical methodologies can be applied for a high sensitive detection of the trans isomer content and characterization of the mono-trans cholesteryl ester library. In particular, Raman spectroscopy can be carried out directly on the cholesteryl ester fraction without derivatization (see Figure 2 and Figure 9).
The second example concerns purine 5′,8-cyclo-2′-deoxyribonucleosides, which are DNA lesions created by the attack of free radicals to the position C5′ of the sugar moiety and subsequent formation of a covalent bond between the sugar and the base moieties. Four structures can be produced, that is, 5′,8-cyclo-2′-deoxyadenosine and 5′,8-cyclo-2′-deoxyguanosine, both existing in the 5’R and 5’S diastereomeric forms (Figure 5). In Figure 6 the reaction mechanism is shown, that involves photolysis of 8-bromopurine derivatives 5 to give the corresponding C8 radical 6, which intramolecularly abstracts a hydrogen atom from the C5′ position selectively affording the 2′-deoxyadenosin-5′-yl radical 7. Radical 7 undergoes cyclization with a rate constant in the range 105-106 s-1,2 followed by oxidation of the heteroaromatic radical 8 to give the final products 9. The reaction of hydroxyl radicals, generated by radiolysis of water, with 2′-deoxyadenosine and 2′-deoxyguanosine (10) was found to occur ca. 10% by hydrogen abstraction from C5′ position.7
Our chemical biology approach for the libraries of purine 5′,8-cyclo-2′-deoxyribonucleosides (including labeled compounds) is illustrated in Figure 7, with the identification of these lesions in oligonucleotides as well as in DNA samples obtained from various sources such as, for example, treated under ionizing radiation conditions, as mimic of radical stress conditions.
In Figure 8 a typical photoreactor equipment is shown. The device allows for irradiation of compounds dissolved in the appropriate solvent. It consists of: i) the reaction chamber equipped with the inlet for the inert gas (on the bottom) and two inlets for the gas exit and for the reagents addition; ii) an internal chamber containing the appropriate mercury lamp connected to a cooling system and electrical power, which is inserted into the reaction chamber through a glass joint.
Figure 1. Mono-trans isomers of cholesteryl linoleate and arachidonate.
Figure 2. Cholesteryl esters in human serum and direct analysis for mono-trans isomers by Raman spectroscopy.
Figure 3. The addition-elimination process that leads to the cis-trans isomerization of double bonds by thiyl radicals.
Figure 4. Three steps protocol for development of mono-trans cholesteryl esters as biomarker of free radical stress.
Figure 5. The four purine 5′,8-cyclo-2′-deoxyribonucleoside diastereoisomers. Click here to view larger figure.
Figure 6. Generation of C5′ radicals, either by photolysis of 5 or by reaction of 10 with HO• radicals, and the mechanism of purine 5′,8-cyclo-2′ deoxyribonucleoside formation. Click here to view larger figure.
Figure 7. Protocol for identification of some radical-based DNA lesions; mtDNA: mitochondrial DNA, nDNA: nuclear DNA.
Figure 8. Photochemical reactor.
Figure 9. Representative Raman spectrum of plasma cholesteryl ester. In the inset the comparison of a specific region with cholesteryl linoleate and cholesteryl mono-trans linoleic acid isomers.
Figure 10. Representative GC analysis of FAME obtained from plasma cholesteryl esters.
Figure 11. Representative HPLC run containing 2′-deoxyadenosine and 2’deoxyguanosine together with their oxidative and purine 5′,8-cyclo-2′-deoxyribo nucleosides. Click here to view larger figure.
The conversion of naturally occurring cis unsaturated fatty acids to geometrical trans isomers is a transformation connected with the production of radical stress in the biological environment. Cell membrane lipids, which contain fatty acids, are a relevant biological target for radical stress and we first studied the endogenous cis-trans phospholipids isomerization in cell cultures, animals and humans assessing analytical protocols in each case.8-10 We demonstrated that this transformation can occur by a variety of S-containing compounds, including thiols, thioethers and disulfides, which under different radical stress conditions are able to generate thiyl radicals, i.e. the isomerizing agent (Figure 3). The example shown in this article focuses on the class of cholesteryl esters, which represent a well-known fraction of plasma lipids, strictly involved in lipoprotein metabolism. The ester linkage between fatty acids and cholesterol is biosynthesized by the transfer of fatty acids from the position 2 of the glycerol moiety of phosphatidylcholine to cholesterol, a step catalyzed by the enzyme lecithin cholesterol acyl transferase (LCAT). Therefore, plasma cholesteryl esters are strictly connected with membrane lipid turnover, and contain relatively high proportions of the polyunsaturated fatty acids (PUFA) typically present in phosphatidylcholines, i.e. linoleic and arachidonic acids. Lipoprotein formation is involved in cardiovascular and metabolic diseases. The reactivity of natural cholesteryl esters with free radicals can occur at the double bonds of linoleate and arachidonate residues, which can be transformed in the corresponding trans geometrical isomers (see Figure 1 for the structures). Characterization of the trans cholesteryl ester content in biological samples is interesting for biomarker development. An indirect methodology consists of the transformation of cholesteryl esters isolated from plasma to the corresponding fatty acid methyl esters (FAME) and separation by gas chromatographic protocols. In this case, the calibration of the standard references of cis and trans fatty acid methyl esters is performed, in order to allow the quantitation of the trans content in the samples. Based on the analytical studies carried out on the cholesteryl ester library, we proposed to apply also a method based on Raman spectroscopy, that can be carried out directly on the cholesteryl ester fraction isolated from plasma, without further derivatization to the corresponding FAME (see Figures 2 and 9). It is worth noting that up to now no successful methods are described to separate cis and trans isomers of fatty acid containing lipids by HPLC, as instead described by cholesteryl ester hydroperoxides. So far, the indirect gas chromatographic method is still the best available method so far. By this method the first quantitative evaluation of the mono-trans content derived from cholesteryl esters isolated from plasma of healthy subjects was provided. Using Flame Ionization Detector (FID) the limit of detectability is satisfactory (ppb) and nanomolar quantities of the compounds have been detected.5. With different detection systems this limit can be even lowered. The effect of ionizing radiation on cholesteryl esters is matter of further studies, whether a linear response is obtained relative to the applied dose.
As a second example, we chose modified nucleosides which can be produced by free radical damage of DNA. Hydroxyl radicals (HO• ) are known to be the most harmful Reactive Oxygen Species (ROS) for their ability to cause chemical modifications to DNA. Single or multiple lesions may occur on DNA, that in eukaryotic cells is located in nucleus and mitochondria. Identification and measurement of the main classes of oxidative generated damages to DNA require the appropriate molecular libraries in order to set up the analytical protocols. We focused our interest on the smallest tandem lesions, which are purine 5′,8-cyclo-2′-deoxyribonucleosides, having an additional covalent bond between the base and the sugar moieties created by the free radical attack. The compounds are 5′,8-cyclo-2′-deoxyadenosine and 5′,8-cyclo-2′-deoxyguanosine existing in the 5’R and 5’S diastereomeric forms (Figure 5). Their potential to become free radical stress marker is matter of fundamental research.2 Indeed, when DNA is exposed to HO• radical hydrogen abstraction from C5′ position of the sugar is one of the possible events leading to the formation of these tandem lesions. Purine 5′,8-cyclonucleosides can be measured as sum of diastereomers by HPLC-MS/MS in enzymatically digested γ-irradiated DNA samples varying from 1 to 12 lesions /106 nucleosides/Gy going form absence of oxygen to physiological level of oxygen in tissues, the diastereomeric ratio 5’R/5’S being ~4 and ~3 for 5′,8-cdAdo and 5′,8-cdGuo, respectively (Figure 11).11 It is worth noting that the relationship between radiation dose and 5′,8-cdAdo and 5′,8-cdAdo lesions detected in cellular DNA is far from being understood. The single experiment based on 2kGy irradiation reported in the experimental cannot be considered conclusive.11 Further experiments of this kind and analytical quantitation of the four lesions are necessary for defining such relationship. The detection of these lesions and the more popular oxidative transformations (such as 8-oxo-2′-deoxyguanosine, 8-oxodGuo) are matter of intense investigations, evidencing the importance of both lesions during oxidative metabolism.6,13 The use of HPLC-MS/MS (triple quadrupole) has a detection limit close to 30fmol for all four lesions. Last improvements are claimed to reach detection limits of attomol levels by the instrument producers. Based on the very recent literature,6 analytical procedures have to include the proper cleanup of the sample and enrichment in order to meet the detection limits of the MS/MS/MS (ion trap) or of the MS/MS (triple quadrupole) used in our case.
Bio-inspired synthetic procedures of compounds 1-4 were developed starting from 8-bromopurine derivatives under or photolysis.7,12 These procedures involve a radical cascade reaction that mimics the DNA damage mechanism of formation of 5′,8-cdAdo and 5′,8-cdGuo lesions. From biological perspectives, it was found that these lesions accumulate with aging in a tissue-specific manner (liver>kidney>brain), providing evidence that DNA repair mechanisms are inadequate to preserve the genetic material from these lesions.13 Indeed, nucleotide excision repair (NER) is the only pathway currently identified for the repair of these lesions.2
The two classes of compounds shown in Figures 1 and 5 are not commercially available at the moment, however by the synthetic strategies described in the literature it would not be difficult to prepare these compounds for commercial use.
The multidisciplinary approach provided by chemical biology studies not only has an enormous value in the identification of novel mechanisms occurring in the biological environment, but also gives a fundamental contribution to biomarker discovery and diagnostics, ultimately bringing novelty in health care and prevention strategies.14 The chemical contribution is needed for a successful development of molecular medicine, creating integrated platforms and panels for metabolic profiling which are expected to allow for an optimal rationalization of intervention design, either therapeutic and nutritional, reducing uncertainties and failures when they can be predictable.
The authors have nothing to disclose.
Financial support from the Ministero dell’Istruzione, dell’Universitá della Ricerca (PRIN-2009K3RH7N_002) and Marie Curie Intra-European Fellowship (CYCLOGUO-298555) as well as the sponsorship of COST Action CM0603 on ‘Free Radicals in Chemical Biology and COST Action CM1201 on “Biomimetic Radical Chemistry” are gratefully acknowledged.
MATERIALS | |||
Cholesteryl linoleate ≥98% | Sigma-Aldrich | C0289-100 mg | |
Cholesteryl arachidonate≥95% | Sigma-Aldrich | C8753-25mg | |
2-mercaptoethanol | Sigma-Aldrich | M6250-100 ml | |
2-propanol | Sigma-Aldrich | 34965-1L | |
Methanol 215 SpS | Romil | H409-2,5 L | |
Ethanol | Sigma-Aldrich | 02860-2.5L | |
Chloroform SpS | Romil | H135 2,5 L | |
n-Hexane 95% SpS | Romil | H389 2,5 L | |
Acetonitrile 230 SpS | Romil | H047 2,5 L | |
Dichloromethane SpS | Romil | H2022,5 L | |
Carbon tetrachloride | Sigma-Aldrich | 107344-1L | |
Sodium iodide | Sigma-Aldrich | 383112-100G | |
Sodium hydrogen carbonate | Carlo Erba | 478536-500 g | |
Diethyl ether | Sigma-Aldrich | 309966-1L | |
NaCl | Sigma-Aldrich | S7653-5KG | |
NaOH solid | Sigma-Aldrich | 221465-25G | |
NH4OH sol. 28%-30% | Sigma-Aldrich | 221228-1L-A | |
Acetic acid | Sigma-Aldrich | 320099-500ML | |
Ammonium Cerium(IV)sulfate dihydride | Sigma-Aldrich | 221759-100G | |
Ammonium Molybdate tetrahydrate | Sigma-Aldrich | A7302-100G | |
Sulfuric Acid 95%-98% | Sigma-Aldrich | 320501-1L | |
Silver Nitrate | Sigma-Aldrich | 209139-25G | |
Sodium sulfate anhydrous | Sigma-Aldrich | 238597-500G | |
Nuclease P-I from penicillium citrinum | Sigma-Aldrich | N8630-1VL | |
Phosphodiesterase II type I-sa | Sigma-Aldrich | P9041-10UN | |
Erythro-9-(2-hydroxy-3-nonyl)adenine, hc | Sigma-Aldrich | E114-25MG | |
Phosphatase alkaline type VII-t from*bov | Sigma-Aldrich | P6774-1KU | |
Phosphodiesterase I type VI | Sigma-Aldrich | P3134-100MG | |
Deoxyribonuclease II type IV from*porcin | Sigma-Aldrich | D4138-20KU | |
Trizma(r) base, biotechnology performanc ce | Sigma-Aldrich | T6066-100G | |
EDTA | Sigma-Aldrich | E1644-100G | |
Succinic acid bioxtra | Sigma-Aldrich | S3674-250G | |
Calcium chloride | Sigma-Aldrich | C5670-100G | |
Formic acid, 98 % | Sigma-Aldrich | 06440-100ML | |
Amicon Ultra-0.5 Centrifugal Filter Unit with Ultracel-3 membrane | Millipore | UFC500324 | |
8-Bromo-2′-deoxyguanosine | Berry & Associates | PR3290-1 g | |
8-Bromo-2′-deoxyadenosine | Berry & Associates | PR3300-1 g | |
Sodium iodide | Sigma-Aldrich | 383112-100G | |
Sodium hydrogen carbonate | Carlo Erba | 478536-500 g | |
2′-deoxyguanosine:H2O (U-15N5, 96-98%) | Cambridge Isotope Laboratories, Inc | CILNLM-3899-CA-0.1 | |
2′-deoxyadenosine (U-15N5, 98%) 95%+ CHEMICAL PURITY | Cambridge Isotope Laboratories, Inc | CILNLM-3895-0.1 | |
Nitrous oxide (N2O) | Air Liquide | ||
Deoxyribonucleic acid from calf thymus | Sigma Aldrich | D4522-5MG | |
EQUIPMENT | |||
60Co-Gammacell | AECL- Canada | 220 | |
Immersion well reaction medium pressure 125 watts | Photochemical reactors ltd | Model 3010 | |
Evaporating flask 250 ml | Heidolph | P/N NS 29/32 514-72000-00 | |
Luna 5 μm C18(2) 100 Å, LC Column 250 x 4.6 mm | Phenomenex | 00A-4252-E0 | |
Alltima C8 Column 250 x 10 mm 5 μm | Grace | 88081 | Semipreparative |
SecurityGuard Kit | Phenomenex | KJ0-4282 | Analytical holder kit and accessories |
Holder for 10.0 mm ID cartridges | Phenomenex | AJ0-7220 | Semipreparative holder |
10.0 mm ID cartridges | Phenomenex | AJ0-7221 | |
High-performance liquid chromatography (HPLC) | Agilent | 1100 | |
LC/MS/MS | Applied Biosystems | 4000QTRAP System | |
Tandem mass ESI spectrometer | (Bruker Daltonics) | Esquire 3000 plus | |
Vial 2-4 ml | SUPELCO | Cod 27516 | |
Vial 4 ml | SUPELCO | Cod 27517 |