Modification of DNA in Relation to Diabetes: Identification of CMdC From Glycoxidation Reactions and Estimation of CMdC and CMdA in Fasting Human Urine

The involvement of proteins in glycoxidative and lipoperoxidation reactions1,2 leads to conditions such as cataracts in lenses, atherosclerosis, and other age-related diseases. Carboxymethyllysine (CML) was detected from both glycoxidation and lipid peroxidation reactions. This compound, identified originally from urine specimens, was proven to be an important biomarker of protein modification in relation to diabetes.3,4 Since the amino acid lysine, having an amine group, is involved in such reactions, it is clear that similar reactions are possible in other biomolecules, including nucleobases with free amine groups therein. The authors’ recent experiments indicated the presence of carboxymethyl-2’-deoxyadenosine (CMdA) from glycoxidation reactions involving adenine bases.5,6 CMdA compound was also shown to be present in the specimens of human urine, calf thymus, and human serum DNA investigated.7 Another reported work8 in this field claiming detection of CMdA and carboxymethyl-2’-deoxycytidine (CMdC) from in vitro reactions of DNA precursors was with diazoacetate.

Numerous defense systems protect cellular macromolecules against oxidation. However, there is still a high rate of damage to DNA,9,10 proteins,11 and lipids,12 leading to many types of disease and pain. Some of these free radical-induced reaction intermediates are not only potent chemical modifiers, but some are also deadly carcinogenic materials that may lead to cancer in the host systems. The steady-state level of oxidatively modified nucleobases in genomic and mitochondrial DNA in rats, and the release of these damaged products in human and rodent urine,13–15 have been determined.

8-Hydroxydeoxyguanosine (8-OH-dG), a reactive oxygen species (ROS)-induced modification of a purine residue in DNA, is a sensitive index of oxidative DNA damage.16 The urinary level of this molecule is now considered a biomarker of the total systemic oxidative stress in vivo. The purpose of the present study is to identify and evaluate the oxidative damage to DNA as results of glycoxidative modifications of deoxycytidine (dC) and deoxyadenosine (dA) assessed by measuring the excretion of CMdC and CMdA compounds in fasting human urine samples.

Experimental procedure

1. Incubations of dC with D-glucose, D-ribose, and L-ascorbic acid. A mixture of dC (24.5 mg, 0.1 mmol) with D-glucose (18 mg, 0.1 mmol) was incubated in 0.2 M phosphate buffer (1 mL, pH 7.4) (two drops of toluene were added to prevent fungal growth) at 37 °C (30 days). The reaction mixture was kept constant at pH 7.4 during the incubation period by adjusting the pH as required; 100-μL samples were collected at different time points (2, 4, 6, 8, 10, 20, and 30 days) during the incubation period and stored in the refrigerator prior to analysis. For analysis, a typical sample was adjusted to pH 7.0 and dried (<60 °C) in a rotary evaporator. Methanol (3 mL) was added to the dried residue; the mixture was sonicated, vortexed, and centrifuged; and the methanol fractions were collected. The process of extraction with methanol was repeated twice, and the combined methanol fractions were subjected to HPLC and electrospray  LC-MS/MS (Figure 1a). Similar incubations of 2’-deoxycytidine (24.5 mg, 0.1 mmol) with D-ribose (15 mg, 0.1 mmol) were also carried out and analyzed (Figure 1b).

Figure 1 – LC-MS/MS electrospray (TIC [total ion chromatograms] and ion plots) spectrum of dC with a) D-glucose, b) D-ribose, c) chloroacetic acid, and d) mass spectrum of CMdC from (c).

2. Synthesis of CMdC using dC and chloroacetic acid. Synthesis of CMdC was carried out using a mixture of dC (24.5 mg, 0.1 mmol) and chloroacetic acid (9 mg, 0.1 mmol). The sample was analyzed (Figure 1c and Figure 2a) as in experiment 1.

Figure 2 – HPLC analysis of a) CMdC synthesis using dC and chloroacetic acid, b) fasting human urine sample, and c) mixing experiment of CMdC and fasting human urine sample.

3. Synthesis of CMdA using dA and chloroacetic acid. Synthesis of CMdA was achieved by incubating deoxyadenosine (dA) and chloroacetic acid (Figure 3a). Isolation and identification of carboxymethyl-2’-deoxyadensine (CMdA) was carried out as described.5,6

Figure 3 – HPLC analysis of a) CMdA synthesis using dA and chloroacetic acid, b) fasting human urine sample, and c) mixing experiment of CMdA and fasting human urine sample.

4. Identification of CMdC in urine sample. In a typical experiment, 1 mL of fasting urine sample was adjusted to pH 7.0, evaporated in a rotary evaporator, and extracted three times with methanol. The residue was analyzed by HPLC (Figure 2) and LC-MS/MS (Figure 4) as in experiment 1.

Figure 4 – LC-MS/MS electrospray (TIC and ion plots) spectrum of CMdC and CMdA in fasting human urine sample.

5. Estimation of creatinine, CMdA, and CMdC in fasting human urine samples. Fasting urine samples were obtained from volunteer donors and frozen at –20 °C until use. Urine creatinine concentration was estimated by a colorimetric method based on a Fisher Scientific (Thermo Fisher Scientific, Waltham, MA) kit using Pointe Scientific (Canton, MI) Creatinine Reagents and procedure. The concentration of CMdC and CMdA analyzed by HPLC were expressed as ratios of urine creatinine concentrations17,18 (Figure 5).

Figure 5 – CMdC and CMdA concentration by HPLC in fasting human urine samples expressed as a ratio of urine creatinine concentration by age.


UV-VIS spectrophotometry analysis

A GENESIS 10S VIS UV-VIS spectrophotometer (Thermo Fisher Scientific) was employed for the quantitative determination of creatinine in fasting human urine samples using 40 mM picric acid as outlined in the Fisher Scientific procedure. The wavelength was set at 510 nm for the quantitation experiment.

HPLC analysis

The HPLC system was from Shimadzu Corp. (Columbia, MD) and consisted of the following components: Shimadzu low-pressure gradient pump (LC-20AT) with on-line degasser (DGU-20A5), autosampler (SIL-20AC), column oven (CTO-20A), and system controller (CBM-20Alite). Also included were a reversed-phase column (Thermo HYPERCARB 5 μm 150 × 4.6 mm) and guard column (HYPERCARB 5 μm 150 × 4.6 mm) attached to the column head. In addition, a Shimadzu UV-VIS detector (SPD-20AV) was employed at a wavelength of 254 nm.

A water/acetonitrile solvent mixture using gradient analysis was employed for HPLC analysis. Solvent A was 100% water and solvent B was 100% acetonitrile. Flow rate was 1 mL/min.

Mass spectrometry analysis

LC-MS, triple quadrupole mass spectrometry (electrospray, MS-MS multiple reaction monitoring method) was used for detection and identification of CMdC and CMdA from synthetic and urine sample specimens.


The reaction scheme for the detection of CMdC from glycoxidation reactions and fasting human urine samples is shown in Figure 6. Incubations of dC with D-glucose and D-ribose in separate experiments under physiological conditions of temp (37 °C) at pH 7.4 resulted in the formation of CMdC identified by LC-MS/MS electrospray spectroscopy (Figure 1d). The identity of CMdC was confirmed by its synthesis from 2’-deoxycytidine and chloroacetic acid.

Figure 6 – Reaction scheme for the detection of CMdC from glycoxidation reactions and fasting human urine sample.

Detection of the presence of CMdC in human urine samples was achieved by LC-MS/MS (Figure 4) and HPLC analysis (Figure 2) of the methanol extracts of the neutral pH-adjusted, dried samples. The presence of the ion M+H+ 286 (CMdC) mass spectrometry confirmed the presence of CMdC in the samples examined (Figure 1d). CMdC showed a molecular ion (M+H+ = 286) and a fragment at m/e 170, indicating breakdown of the parent ion into two groups at the N–C bond between the heterocycle and the 2’-ribose part of the molecule.

HPLC analysis, as shown in Figure 2, clearly indicated the presence of CMdC in the fasting human urine sample. Mixing experiments with synthesized CMdC confirmed the identity of the CMdC present. At this point, creatinine estimation on the urine samples was carried out, and the concentrations of CMdC and CMdA were calculated from the HPLC experiments performed on each sample. The concentrations of CMdC and CMdA in fasting human samples were expressed as ratios of urine creatinine concentration (Figure 5).

Role of carboxymethylation in protein chemistry

In protein chemistry, carboxymethylation is a major protein secondary transformation reaction producing established biomarkers such as CML4 and CEL (carboxyethyllysine).1 These biomarkers are widely used to assess damage due to diabetes and other age-related diseases. The authors’ recent investigations implicated carboxymethylation as a consequence of DNA modification also, and they were able to identify CMdA in the specimens of fasting human urine, calf thymus, and human serum DNA examined.7 This finding led the authors to believe that DNA molecules are also modified by glycoxidation reactions and possibly play a role in complications of diseases such as diabetes, mutations of DNA, and cancer.

The present investigation involves glycoxidative modification of deoxycytidine and the identification and estimation of carboxymethylated dA in the specimens of human urine examined. The identities of the CMdC compound from glycoxidative reactions of 2’-deoxycytidine with D-glucose and D-ribose were investigated using LC-MS/MS and HPLC. The identity of CMdC compound was confirmed by synthesizing CMdC using 2’-deoxycytidine and chloroacetic acid raw materials. The presence of the glycoxidatively produced and in vitro identified carboxymethyl derivative (CMdC) in urine sample, from human subjects, was identified and confirmed by HPLC and LC-MS/MS. These experiments proved conclusively that CMdC is produced in human systems. CMdC is the second reported urinary metabolite with CMdA reported earlier, demonstrating that glycoxidation in DNA is a consequence of complications in diabetes, mutations, and other age-related disease processes.

HPLC analysis of fasting urine samples and mixing experiments using authentic specimens clearly identified CMdC (Figure 2) and CMdA (Figure 3) compounds. Quantitative estimation of these metabolites with reference to creatinine contents of the respective urine specimens showed increasing amounts of the CMdC and CMdA contents, indicating a biomarker-type relationship of CMdC and CMdA with age.


  1. Rodríguez-García, J.; Requena, J.R. et al. Increased concentrations of serum pentosidine in rheumatoid arthritis. Clin. Chem.  1998, 44(2), 250–5.
  2. Januszewski, A.S.; Alderson, N.L. et al. Role of lipids in chemical modification of proteins and development of complications in diabetes. Biochem. Soc. Trans.  2003, 31, 1413–16.
  3. Petrovic, R.; Futas, J. et al. Rapid and simple method for the determination of epsilon-(carboxymethyl)lysine and nepsilon-(carboxyethyl)lysine in urine using gas chromatography/mass spectrometry. Biomed. Chromatogr.  2005, 19(9), 649–54.
  4. Ahmed, M.U.; Thorpe, S.R. et al. Identification of N-carboxymethyllysine as a degradation product of fructose lysine in glycated proteins. J. Biol. Chem. 1986, 261, 4889–94.
  5. Ahmed, M.; Lila, Z.A. et al. Investigations on DNA/RNA damage in diabetes involving glycoxidation reactions. In: 1890 Research: Sustainable Solutions for the 21st Century; 15th Biennial Research Symposium, Mar 28–Apr 1, 2009, Atlanta, GA; p 216.
  6. Lila, Z.A.; Mahtab, R. et al. Glycoxidative modification of DNA–RNA in relation to diabetes. In: The 1890 Land Grant System; AED/ARD Land Grant Conference, Jun 8–11, 2008, Memphis, TN, p 37.
  7. Lila, Z.A.; Allen, P. et al. Modifications of DNA in relation to diabetes: identification of carboxymethyl-2’-deoxyadenosine from glycoxidation reactions, calf thymus DNA, human urine and DNA;
  8. Wang, J.; Wang, Y. Synthesis and characterization of oligodeoxyribonucleotides containing a site-specifically incorporated N6-carboxymethyl-2’-deoxyadenosine or N4-carboxymethyl-2’-deoxycytidine. Nucl. Acids Res.  2010, 38(19), 6774–84.
  9. Al-Aubaidy, H.A.; Jelinek, H.F. Oxidative DNA damage and obesity in type 2 diabetes mellitus. Eur. J. Endocrinol. 2011, 164(6), 899–904.
  10. Synold, T.; Wuenschell, G.E. Advanced glycation end products of DNA: quantification of N2-(1-carboxyethyl)-2'-deoxyguanosine in biological samples by liquid chromatography electrospray ionization tandem mass spectrometry. Chem. Res. Toxicol.  2008, 21(11), 2148–55.
  11. Oliver, C.N.; Bong-Wham, A. et al. Age-related changes in oxidized proteins. J. Biol. Chem.  1987, 262, 5488–91.
  12. Negre-Salvayre, A.; Coatrieux, C. et al. Advanced lipid peroxidation end products in oxidative damage to proteins. Potential role in diseases and therapeutic prospects for the inhibitors. Br. J. Pharmacol.  2008, 153(1):6–20.
  13. Cathcart, R.; Schwiers, E. et al. Thymine glycol and thymidine glycol in human and rat urine: a possible assay for oxidative DNA damage. Proc. Natl. Acad. Sci. USA  1984, 81, 5633–7.
  14. Lila, Z.A.; Manzano, K. et al. Glycoxidative Modification of DNA (dC) and RNA (C) in Relation to Diabetes: Estimation of Carboxymethyl- 2’-Deoxyadenosine (CMdA) and Carboxymethyl- 2’-Deoxycytidine (CMdC) in Fasting Human Urine; Proceedings of the Network of Minority Research Investigators, 10th Annual Workshop held April 19–20, 2012, Bethesda, MD.
  15. Shigenaga, M.K.; Gimeno, C.J. et al. Urinary 8-hydroxydeoxy-2’-deoxyguanosine as a biological marker of in vivo DNA damage. Proc. Natl. Acad. Sci. USA  1989, 86, 9697–9701.
  16. Loft, S.; Fischer-Nielsen, A. et al. 8-Hydroxydeoxy-guanosine as a urinary biomarker of oxidative DNA damage. J. Toxicol. Environ. Health  1993, 40, 391–404.
  17. Kumar, A.K.; Ramachandran, G. et al. Can urine lamivudine be used to monitor antiretroviral treatment adherence? MedGenMed  2006, 8(4), 53.
  18. Barr, D.B.; Wilder, L.C. et al. Urinary creatinine concentrations in U.S. population: implications for urinary biologic monitoring measurements. Environ. Health. Perspect.  2005, 113(2), 192–200.

The authors are with the Department of Biological and Physical Sciences, South Carolina State University, 300 College St., Orangeburg, SC 29117, U.S.A.; tel.: 803-516-8979; fax: 803-516-4685; e-mail:, with the exception of R. Idris, who is with the Department of Family and Consumer Sciences, South Carolina State University. The authors thank the 1890 Evans-Allen Research Program, South Carolina State University (SCSU), Project Export MUSC-SCSU (subcontract from NIH grant #5 P60 MD 000243); SCAMP (SCSU) Program for financial help; and SCSU for providing the facilities for this work. They also thank Dr. William E. Cotham of the Mass Spectrometry Laboratory, University of South Carolina, Columbia, for performing the LC-MS, triple-quadrupole mass spectrometry analysis experiments.