Estimation of human age using N-glycan profiles from bloodstains

Blood samples were collected from 526 blood donors from different parts of Croatia (Bjelovar, Dubrovnik, Osijek, Pula, Sisak, Split, Šibenik, Zabok, Zagreb). Of the 526 participants 402 were men (age 35 (18–63)) and 124 were women (age 29 (18–77)). Bloodstains were collected during volunteer blood donation. Capillary blood was collected on Whatman Schleicher & Schuell bloodstain cards from 2004 to 2007 and stored at +4°C. Sample collection was conducted according to the Declaration of Helsinki Ethical Principles for Medical Research Involving Human Subjects and was approved by the ethics committees of the University Hospital Centre Zagreb and the Faculty of Pharmacy and Biochemistry at the University of Zagreb.

Blood was drawn from one individual and put on Whatman Schleicher & Schuell bloodstain cards and a kitchen cloth. Bloodstains from bloodstain cards were subsequently exposed to different conditions (humidity, different temperatures and UV radiation). Humidity testing was conducted in a water incubation chamber at +37°C for 6 days. The influence of different temperatures was determined by exposing the bloodstains to either +4 °C in the fridge or +37°C and +65°C in the oven for 6 days, while the influence of UV radiation was measured by placing bloodstains under a UV lamp (60 W, 254 nm) for 20 min or leaving them in a room with a UV lamp (15 W, 253.7 nm) for 2 h. Bloodstains on the kitchen cloth were stored at room conditions, and bloodstain cards with blood were used as reference samples. Each experiment was done in tetraplicate.

From each bloodstain card, a circle with a diameter ≈ 6 mm was cut and transferred onto a microtiter plate. Each following step was done in a 96-well microtiter plate to achieve the best throughput of sample preparation. After adding 10 μL of ultra-pure water into each well with a sample of blood stain cards, samples were first denatured by addition of 20 μL of 2 % sodium dodecyl sulfate (SDS; w/v; Invitrogen, Carlsbad, CA, USA) and by incubation at 65 °C for 10 min. Subsequently, 10 μL of 4 % Igepal-CA630 (Sigma-Aldrich, St. Louis, MO, USA) and 1.25 mU of PNGase F (ProZyme, Hayward, CA, USA) in 10 μL of 5× phosphate-buffered saline (PBS) were added to the samples. The samples were incubated overnight at 37°C for N-glycan release. The released N-glycans were labeled with 2-aminobenzamide (2-AB). The labeling mixture was freshly prepared by dissolving 2-AB (Sigma-Aldrich, St. Louis, MO, USA) in a dimethyl sulfoxide (DMSO; Sigma-Aldrich, St. Louis, MO, USA) and glacial acetic acid (Merck, Darmstadt, Germany) mixture (85:15, v/v) to a final concentration of 48 mg/mL. A volume of 25 μL of labeling mixture was added to each N-glycan sample in the 96-well plate. Also, 25 μL of freshly prepared reducing agent solution [106.96 mg/ml 2-picoline borane (Sigma-Aldrich, St. Louis, MO, USA) in DMSO] was added and the plate was sealed using adhesive tape. Mixing was achieved by shaking for 10 min, followed by 2 h incubation at 65°C. The liquid component of the samples (approximately 100 μL, without paper) was transferred into a new 96-well plate and was brought to 80 % acetonitrile (ACN; v/v) by adding 400 μL of ACN (J.T. Baker, Phillipsburg, NJ, USA). Free labeling and reducing agents were removed from the samples using microcrystalline cellulose. An amount of 200 μL of 0.1-g/mL suspension of microcrystalline cellulose (Merck, Darmstadt, Germany) in water was applied to each well of a 0.45-μm GHP filter plate (Pall Corporation, Ann Arbor, MI, USA). Solvent was removed using a vacuum manifold (Millipore Corporation, Billerica, MA, USA). All wells were prewashed five times using 200 μL of water, followed by equilibration using three washes of 200 μL of acetonitrile/water (80:20, v/v). The samples were loaded in the wells of the GHP filter plate and the wells were subsequently washed seven times using 200 μL of acetonitrile/water (80:20, v/v). Glycans were eluted twice with 100 μL of water and combined eluates were stored at −20°C until usage.

Fluorescently labeled N-glycans were separated by hydrophilic interaction chromatography on a Waters Acquity UPLC instrument (Milford, MA, USA) consisting of a quaternary solvent manager, sample manager and a FLR fluorescence detector set with excitation and emission wavelengths of 330 and 420 nm, respectively. The instrument was under the control of Empower 2 software, build 2145 (Waters, Milford, MA, USA). Labeled N-glycans were separated on a Waters BEH Glycan chromatography column (150 × 2.1 mm i.d., 1.7-μm BEH particles, with 100-mM ammonium formate, pH 4.4, as solvent A and acetonitrile as solvent B). The separation method used a linear gradient of 75–53 % acetonitrile (v/v) at a flow rate of 0.56 ml/min in a 25-min analytical run. Samples were maintained at +5 °C before injection, and the separation temperature was +25 °C. The chromatograms were all separated in the same manner into 38 chromatographic regions that enabled reliable quantification. The amount of glycans in each peak was expressed as percentage of total integrated area.

To obtain normally distributed variables for 38 glycan structures, a log transformation was performed on the glycan variables. The predictive model of chronological age was built using a multivariate regression approach implemented in the “stats” package for the R programming language. The maximal model included linear and quadratic terms for each of the 38 glycan peaks, giving 76 parameters in total. Feature selection was performed using a backward elimination procedure with Akaike information criterion used as optimization criteria. Reported goodness of fit (coefficient of determination), prediction errors and regression coefficients were estimated as median values of 100 iterations of repeated, random sub-sampling validation, with two thirds (347 samples) of the sample set as a “training set” and one third (174 samples) as a "validation set."