Nucleic acid oxidation

We measured levels of RNA and DNA oxidation in the gill tissue of 21 Baltic Sea and 20 Icelandic A. islandica (Table ​(Table1).1). Ratios of 8-oxo-7,8-dihydroguanosine to guanosine and 8-oxo-7,8-2′-deoxyguanosine to deoxyguanosine were measured using high-performance liquid chromatography coupled to electrochemical detection (HPLC-ECD) and used as proxies for total RNA and DNA oxidation, respectively (Hofer et al. 2006). Gill tissue was ground in liquid nitrogen and homogenized on slush ice using a glass-glass Duall homogenizer in 0.6 ml (1:10 w:v) buffer (3 M guanidine thiocyanate, 0.2 % w/v N-lauroylsarcosinate, 20 mM Tris, pH 7.5, 10 mM deferoxamine mesylate (DFOM)). After transferring the solution into phase-lock gel (PLG) tubes, an equal volume of phenol/chloroform/isoamyl alcohol (25:24:1, pH 6.7) was added and the samples were vortexed for 10 s at 0 °C to completely release nucleic acids. After centrifugation, the aqueous phase was transferred into a new PLG tube and mixed with an equal volume of chloroform/isoamyl alcohol (24:1). Samples were hand-shaken and centrifuged, and the aqueous phase was collected. Nucleic acids were precipitated from the aqueous phase by addition of an equal volume of isopropanol and stored in a −80 °C freezer overnight. After centrifugation, the pellet containing nucleic acids was washed with 70 % v/v ethanol, centrifuged, dried, dissolved in 150 μl of water containing 30 μM DFOM, and hydrolyzed using 9 U nuclease P₁ and 4 U alkaline phosphatase in buffer (30 mM sodium acetate, 20 μM ZnCl₂, pH 5.3) at 50 °C for 60 min. All centrifugations were for 10 min and at 10,000×g at 0 °C.

After filtration, samples were analyzed using high-performance liquid chromatography coupled to electrochemical and UV detection (HPLC-ECD/UV). The HPLC system consisted of an EAS Model 582 pump (ESA Inc., Chelmsford, MA, USA), a chilled (4 °C) model 542 autosampler (ESA), a temperature-controlled (35 °C) column oven holding two Delta-Pak (150 × 3.9 mm inner diameter, 5 μm) C-18 reversed-phase columns (Waters, Milford, MA, USA), a SpectraSYSTEM UV1000 detector (Thermo Electron Corp., San Jose, CA, USA) set at 290 nm and an electrochemical detector (Coulochem III, ESA) with a 5011A Analytical Cell (2 nA; screen electrode, +200 mV; analytical electrode, +300 mV). Chromatograms were recorded using EZChrom Elite (Scientific Software Inc., Pleasanton, CA, USA).

Lipid peroxidation

F₂-isoprostanes (F₂-IsoPs) were measured in the gill tissue of 15 Baltic Sea and 25 Icelandic animals as markers of lipid peroxidation (Table ​(Table1).1). F₂-IsoPs were quantified using stable isotope dilution followed by gas chromatography/negative-ion chemical ionization/mass spectrometry detection as previously described (Morrow and Roberts 1999). Briefly, 100–200 mg tissue was homogenized in ice-cold Folch solution (chloroform/methanol 2:1) containing 5 mg/100 ml butylated hydroxytoluene and lipids were extracted. Extracted lipids were then chemically hydrolysed with 15 % KOH. After acidification with 1 M HCL to pH 3, isotope-labeled 8-iso-prostaglandin F_2α-d₄ was added as an internal standard. Following cleaning using C-18 and silica Sep-Pak cartridges, the eluted compounds were dried under N₂, converted to pentafluorobenzyl esters, and purified using thin-layer chromatography. Purified F₂-IsoPs were converted to trimethylsilyl ether derivatives and dissolved in undecane for quantification by GC/MS. We performed negative ion chemical ionization MS using Agilent 6890 GC and Model 5975 MSD instruments with selected ions monitored for the 8-iso-prostaglandin F_2α-d₄ internal standard (m/z 573) and F₂-IsoPs (m/z 569).

Protein oxidation

We measured the ratio of protein carbonyls to total protein content as a proxy for protein oxidation in 15 Baltic Sea and 25 Icelandic animals (Table ​(Table1).1). Soluble and insoluble protein carbonyls were quantified by staining gill tissue with fluorescein-5-thiosemicarbazide (FTC), and FTC-stained protein carbonyls were compared against total cellular proteins stained with Coomassie Brilliant Blue (Chaudhuri et al. 2006). To extract proteins, 150–200 mg gill tissue was homogenized with 350 μl potassium phosphate buffer (20 mM KPO₄, 0.5 mM MgCl₂, 1 mM EDTA, pH 6, 1 tablet/10 ml cOmplete Mini Protease Inhibitor (Roche, Mannheim, Germany)) using zirconium silicate beads in a Retsch MM 301 homogenizer (2 min with 30/s). The lysate was centrifuged in a Beckmann TL-100 ultracentrifuge (1 h, 100,000×g at 4 °C), and the supernatant was transferred to a small reaction tube and kept on ice. Pellets were resuspended in 200 μl potassium phosphate buffer (20 mM KPO₄, 0.5 mM MgCl₂, 1 mM EDTA, 2 % SDS, 0.5 % IGEPAL, 0.5 % sodium deoxycholate, 1 tablet/10 ml cOmplete Mini Protease Inhibitor, 1 mM dithiothreitol, pH 6). Resuspended pellets were sonicated (Bandelin: Sonopuls HD200, 10 s on ice) and ultracentrifuged (10 min, 100,000×g at room temperature (RT)). Protein concentrations were measured using Pierce BCA assays (Thermo Scientific, Rockford, IL, USA).

We then measured protein carbonyl concentrations and total protein content using FTC and Coomassie brilliant blue staining. First, 500 μg protein was added to 500 μl KPO₄ buffer containing 0.3 M guanidine and 1 mM FTC label and incubated at 37 °C for 2 h. After incubation, we precipitated proteins by adding 500 μl 20 % chilled trichloroacetic acid in KPO₄ buffer and incubating mixtures on ice for 15 min. After centrifugation (15 min, 16,000×g at 4 °C), supernatants were removed and pellets were repeatedly washed in 1:1 ethanol/ethyl acetate. Tubes were kept on ice at all times to prevent protein dissolving. After the last wash, excess ethanol/ethyl acetate was removed and protein pellets were dissolved in 50 μl of 8 M Urea in KPO₄ buffer and incubated at 37 °C for 30 min. Dissolved pellets were repeatedly sonicated and centrifuged (2 min, 16,000×g at room temperature). Protein content was again measured using the BCA assay. An equal amount of protein (6.66 μg) from each sample was separated using sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE). Fluorescent protein images were captured with a Typhoon 9400 (GE Healthcare, Munich, Germany) using an excitation wavelength of 488 nm and an emission filter at 520 nm with a PMT value of 460. Gels were then washed with water and stained with Bio-Safe Coomassie (BioRad, Munich, Germany) to visualize total protein concentration. Before capturing the Coomassie image with a visible spectrum camera (Molecular Imager ChemiDoc XRS, Bio-Rad, Munich, Germany), the gel was washed with water until all residual stain had disappeared. Fluorescence and Coomassie images were analyzed using ImageJ 1.45S, and data were expressed in ratios of florescent to Coomassie-stained protein (F/C).

Protein stability

We measured protein resistance to unfolding with and without urea stress in 15 Baltic Sea and 25 Icelandic animals (Table ​(Table1).1). Denatured proteins were stained with 4,4′-dianilino-1,1′-binaphthyl-5,5′-disulfonic acid dipotassium salt (BisANS) and compared with total protein content as described by Pierce et al. (2006). To extract cytosolic proteins, 150–200 mg gill tissue was homogenized in 350 μl labeling buffer (50 mM Tris-HCl, 10 mM MgSO₄, pH 7.4, 1 tablet/10 ml cOmplete Mini Protease Inhibitor) using zirconium silicate beads in a Retsch MM 301 homogenizer (2 min with 30/s). Lysates were then ultracentrifuged in a Beckmann TL-100 ultracentrifuge (1 h, 100,000×g at 4 °C). Protein concentrations of supernatants were determined using Pierce BCA assays, and extracts were diluted to 1 mg/ml. Control extracts were stained immediately, while stress-treated extracts were incubated on ice for 1 h with 1 M urea. To stain extracts, 100 μM BisANS was added and samples were incubated for 1 h on ice under direct exposure to a 115-V 0.16-A handheld longwave UV lamp (365 nm, UVL-56, UVP, Cambridge, UK). Following photoincorporation, samples were subjected to SDS-PAGE. SDS-PAGE gels were then illuminated with 365 nm UV light and BisANS fluorescence was measured using the Molecular Imager ChemiDoc XRS Imaging System with Quantity One 4.6.8 software (BioRad, Munich, Germany). We measured total protein content using Coomasie brilliant blue as described above.

RNA and DNA oxidation

DNA and RNA oxidation increased significantly with age in both populations (Fig. 1a, b; Table ​Table1),1), and oxidation accumulated significantly faster in the Baltic Sea than in the Icelandic population (p_{DNA, slopes} < 0.0005, t = 3.6, d.f. = 37; p_{RNA, slopes} < 0.0001, t = 5.37, d.f. = 37; Fig. 1a, b). When scaled to relative observed lifespan (BS: 36 years = 100 %, IC: 226 years = 100 %), damage accumulated at the same pace in both populations (Fig. 1c, d, p_{relDNA, slopes} = 0.4, t = 0.26, d.f. = 37; p_{relRNA, slopes} = 0.41, t = 0.22, d.f. = 37). Scaled RNA oxidation was generally higher in the Icelandic population than in the Baltic Sea population, as indicated by a significantly higher intercept (p < 0.0001, F(1,38) = 22.76). Linear regression equations are listed in supplementary Table S1.

Open in a separate window

Fig. 1

Nucleic acid damage in gill tissue of Baltic Sea (circles) and Icelandic (triangles) animals. a DNA and b RNA oxidation in both populations over age. Significance of the slopes, i.e., increase of oxidation with increasing age is displayed within the graphs. c DNA and d RNA oxidation calculated over the relative lifespan of both populations

Lipid peroxidation and protein oxidation

Both F₂IsoPs and protein carbonyl content remained stable over all investigated ages (Figs. 2 and 3a, b; supplementary Table S1). Mean levels of F₂IsoPs across ages were higher in Icelandic individuals than Baltic Sea individuals (p < 0.0001, t = 4.55, d.f. = 37). Oxidation levels of soluble and insoluble proteins did not differ between populations (soluble proteins: p = 0.19, t = 1.35, d.f. = 38; insoluble proteins: p = 0.99, t = 0.02, d.f. = 17).

Open in a separate window

Fig. 2

Lipid peroxidation in gill tissue of Baltic Sea and Icelandic A. islandica. Lipid peroxidation over age quantified as F₂-isoprostanes relative to total wet weight in a the Baltic Sea (circles) and b the Icelandic (triangles) population (scales are different for each population)

Open in a separate window

Fig. 3

Protein damage in gill tissue of Baltic Sea (circles) and Icelandic (triangles) A. islandica. Relative quantification of protein carbonyls (in FTC fluorescence) to total protein content (Coomassie absorbance) in a soluble and b insoluble proteins. c Protein resistance with age to 1 M urea unfolding stress measured as BisANS-coupled fluorescent units relative to total protein content in Icelandic samples. d Protein resistance to 1 M urea stress of Baltic Sea animals. e Mean levels of protein resistance to urea stress of all measured BS and IC individuals ± SEM. Equal letters indicate nonsignificant differences among populations and treatments, whereas different letters indicate significances at the p ≤ 0.05 level

Nucleic acid oxidation—a true aging-marker?

Nucleic acid oxidation in the clams increases according to relative age within each population instead of merely accumulating over chronological time. Oxidized nucleic acids accumulated significantly faster over time in the shorter-lived Baltic Sea than in the Icelandic population (Fig. 1). But when standardized to population lifespan, oxidation of each nucleic acid accumulated at the same rate in both populations. Nucleic acid oxidation can be compared between populations because both populations are members of the same species, and the only difference between populations is lifespan. Here, the use of 226 years as maximal lifespan for Icelandic A. islandica may be more realistic than the previously reported 507 years because individuals with ages around 200 years were found in previous studies, but only one individual with an age >375 years has been found so far (Schöne et al. 2005; Butler et al. 2013). One caveat to our nucleic acid observations is that it is possible that only individuals with high oxidative damage reach old age. A potential mechanism for oxidative damage promoting survival is that damage-inducing ROS can serve as a beneficial antimicrobial defense mechanism. We consider this unlikely because bivalves are expected to experience low extrinsic mortality once they reach adulthood (Philipp and Abele 2009). Instead, we consider nucleic acid oxidation to be a true marker of biological age because it accumulates in both long-lived and short-lived A. islandica at the same rate over biological age.

Further evidence for a correlation between nucleic acid oxidation and biological age comes from comparisons across species. DNA oxidation levels in A. islandica increase at rates of 6.73 and 9.25 % per year in gill tissue of Icelandic and Baltic Sea individuals, respectively. Nucleic acid oxidation rate in adult human muscle tissue is 3.34 % per year (Mecocci et al. 1999) which is of the same order of magnitude as in A. islandica. Likewise, human lifespan is of the same order of magnitude as observed A. islandica lifespan (122 and 36–226 years, respectively). In contrast, DNA oxidation increases much faster in short-lived rodents: 21–370 % over a rodent’s 18-month lifespan in several tissues (Hamilton et al. 2001). Because the biologies of humans, rodents, and clams differ, our results stress the advantage of intraspecific damage-marker comparison between populations with different maximum lifespans. While nucleic acid oxidation is a genuine marker for primary aging, its causes are not yet completely understood.

We thank Prof. Philip Rosenstiel and the Institute of Clinical Molecular Biology in Kiel as well as the Excellence Clusters Future Ocean and Inflammation of Interfaces and the Max Planck International Research Network on Aging for financial support. We are also grateful to H. P. Halldorson at the Suðurnes University Research Centre, Sandgerði, Iceland, and the GEOMAR Helmholtz Centre for Ocean Research, Kiel, Germany, for the support and possibility to collect Arctica islandica in the Baltic Sea and around Iceland. Additionally we want to thank the anonymous reviewer who gave inspiring comments that helped to improve our manuscript.