Validation of facile test for urinary TBARS, indicator of whole body lipid oxidation products including malondialdehyde

³Scientific Advisors, Botanic Oil Innovations, Inc., 1540 South River Street, Spooner, Wisconsin, 54801. www.botanicoil.com

Introduction
Oxidative stress (OS) has been linked to 200 human diseases and the list appears to be growing.¹ OS is also a major factor in aging, necrosis and apoptosis. Some very significant pathologies related to OS, either primarily or secondarily, include heart disease, some cancers, diabetes, and neurological problems such as Alzheimer's Disease (AD), Parkinson's Disease (PD), and Huntington's Disease (HD).² Recognizing the growth in aging populations of Western Civilizations and the related increases in health care costs, medical science must develop measures of OS and learn more about the amelioration of the ill-effects of OS.

Because of the great importance of the new field of OS medicine, establishing clarity of terms and concepts is crucial. In 2001, W. A. Pryor edited a volume³ on the assessment of OS but specified the term "oxidative stress status" and included scores of publications on "bio-assays for oxidative stress status" or "BOSS." However, Pryor⁴ identifies a basic problem with tests attempting to measure "free radicals," that "radicals generally have only a fleeting existence in vivo, and 'footprints' of these elusive species must be discovered-hence, BOSS." Indeed, the chemical used to detect free radicals, malondialdehyde (MDA), is neither a radical nor the most prominent secondary lipid oxidation product in urine.⁵ Adding to the confusion, MDA is also, and more correctly, called malonaldehyde (MA), a logical conclusion of its chemical structure.

The thiobarbituric acid (TBA) method has been used by food technologists since the 1950s to estimate rancidity in foods. Tappel⁶ was one of the earliest researchers to demonstrate that lipid oxidation damaged cells. Later, Addis and coworkers^7,8 demonstrated numerous oxidation products in foods, postprandial lipoproteins and circulating lipoproteins including low-density lipoprotein (LDL). Over the same time period, the beneficial effects of vitamin E and other antioxidants, dietary factors which lower in vivo levels of free radicals and their secondary, mostly aldehydic, lipid oxidation products, was noted.

Many such studies have employed the TBA method. In 1984, Csallany et al., developed a method, by size-exclusion high performance liquid chromatography (HPLC), to determine the actual levels of free MA in tissues and foods.⁹ MA is a reactive compound and tends to bind to lysine, albumin, DNA and other cell components.⁵ The findings demonstrated clearly that TBA reacts with many compounds in addition to MA (or MDA).⁹ Generally, a four- to five-fold overestimation of MA occurred if the TBA test was used to determine it. Therefore, the authors suggested that studies using the TBA test report results in terms of thiobarbituric reactive substances (TBARS) and not in terms of MA.

Therefore the TBARS test (urinary MA test) has good news/bad news aspects: TBARS are not specific enough to satisfy the desire of chemists for chemical clarity but is general enough to quantify levels of a broad spectrum of lipid oxidation products so that clinicians and health-conscience individuals will find it useful. Indeed, TBARS analysis has the distinct benefit of simultaneously measuring several potential lipid oxidation products, thereby providing a comprehensive view of overall body lipid oxidation.

Materials and Methods

Design
A previous study reported in "Check Up" commercial literature employed a traditional TBA method (Yagi, 1987)¹⁰ to the analysis of solutions and compared the results obtained to TBARS data using the rapid "Check Up" methodology on a series of urine samples. In the present study, we added known quantities of tetramethoxypropane (TMP), a precursor to MA, to the "Check Up" urine samples plus solutions A and B. Both the older version of the test, using glass ampules, and the newer version, using plastic ampules for solutions A and B, were evaluated. We also attempted to construct a standard curve of quantity of TMP vs. absorbance at 532 nm. Data of subjective (1 to 5) color scores used by the kit were also compared to the test kit TBARS (MA) data.

Two matched (+/- 0.001 absorbance units) Varian quartz cuvettes were employed throughout the study.

Experiment 1
Glass ampules A and B were used. Urine was collected and transferred to the 10-mL vial. Duplicate samples of control and 100 µL TMP were evaluated. Ampules A and B were broken open after using a file and the bottom of each vial added to the 10 mL urine sample along with 100 µL TMP, in the case of the treated sample. For the control sample 100 µL distilled water were added to keep the control and treated samples isovolumetric. The solutions were mixed by inversion 10 times. After five minutes, the samples were evaluated subjectively on the five-point scale, opened, and transferred as needed to a cuvette for spectrophotometric analysis.

Experiment 2
Due to the difficulty observed in mixing the glass ampule (reagents A and B) contents with the urine, and the further observation that some of the reagents in the glass ampules, in preliminary studies, had shown some loss of strength that translated to less intense color, it was decided to use the test kit comprised of the plastic ampules which were thought to be easier to use in transferring solutions A and B and because this kit was newer and contained fresher chemicals. Also, in experiment 2, 1 mL of glacial acetic acid (10%) was added to the mixing vial prior to urine addition. This was done to aid in the hydrolysis of TMP to MA⁹. The experiment was conducted as in experiment 1 except for the acetic acid addition, the use of the plastic ampules, and the use of TMP at the levels of 0, 50-, 100-, 150-, 200-, and 250-µL. Distilled water was used to maintain the solutions isovolumetric.

Experiment 3
Although experiment 2 demonstrated that adding TMP increased color development, thus validating the assay, in agreement with experiment 1, the results were too variable to be satisfactory for the development of a standard curve. Difficulties were encountered in accomplishing a quantitative transfer of the contents of ampules A and B. Therefore in experiment 3, two other changes were made. First, we used distilled water to aid in making quantitative the transfer of solutions A and B from their respective ampules. This was done first by cutting off the top of the ampule with a scissors instead of manually rotating the top of the vial until it opened. This did not work properly consistently. Then distilled water was added with the aid of a Pasteur pipette until the ampule was full. Compressing the ampule then quantitatively transferred its contents into the urine. The second major change was the use of a 50°C water bath for 10 minutes to accelerate color development.

Results and Discussion

The results of experiment 1 are presented in table 1. The results showed that adding TMP to the sample of urine increased color developed over the controls.

Table 1. Results of addition of tetramethoxypropane (TMP) to urine in the analysis of lipid oxidation products by the thiobarbituric acid test.

Treatment	Added TMP, µL	Absorbance, 532 nm	Subjective color
Control	0	0.402	1.5
TMP	100	0.759	3.0

In the case of experiment 1, the control urine was fairly high and then increased substantially with the addition of 100 µL TMP. This clearly demonstrates that the quick test is sensitive to increased MA levels as TMP hydrolyzes to MA during the five-minute incubation period. It is virtually impossible to isolate MA, especially in the presence of biological materials, given the reactive nature of MA. Therefore, TMP or TEP (tetraethoxypropane) are used as precursors to produce MA in timely fashion in assays for free radical activity.

The results of experiment 2 are presented in table 2. To be clear regarding the design of this experiment, 250 µL distilled water were added to the control and as increasing amounts of TMP were added, decreasing amounts of distilled water were added so that the total volume added always equaled 250 µL. One mL of glacial acetic acid was added to 10 mL urine. In addition the volume of each ampule added approximately 1 mL of volume to each sample. In general, the results showed an increase in absorbance and subjective color score with increased addition of TMP but the relationship was not totally satisfactory. The addition of the first 50 µL TMP did not increase color, in fact some decrease was observed. It is possible that the transfer of solutions A and B into the urine-containing vial was not quantitative resulting in a lower absorbance and subjective color than would have been noted if a quantitative transfer had been accomplished.

Table 2. Results of addition of increasing levels of tetramethoxypropane (TMP) to urine in the analysis of lipid oxidation products by the thiobarbituric acid test.

Treatment	Added TMP, µL	Absorbance, 532 nm	Subjective color
Control	0	0.588	2.0
TMP-1	50	0.541	2.0
TMP-2	100	0.725	3.0
TMP-3	150	0.843	3.0
TMP-4	200	0.860	3.5
TMP-5	250	1.201	3.5

The results of experiment 3 are shown in table 3. In this experiment, quantitative transfer of reagents from ampules A and B was accomplished by adding distilled water to the full volume of each ampule. Mathematical correction of the final absorbance and subjective color was done to adjust the data for the dilutions that occurred.

Table 3. Results of addition of increasing levels of tetramethoxypropane (TMP) to urine in the analysis of lipid oxidation products by the thiobarbituric acid test with the quantitative transfer of reagents A and B.

Treatment	Added TMP, µL	Absorbance, 532 nm	Subjective color
Control	0	1.382	2.5
TMP-1	50	1.822	3.0
TMP-2	100	2.012	3.5
TMP-3	150	2.208	3.5
TMP-4	200	2.631	4.0
TMP-5	250	2.886	4.5

Figure 1. Graph of absorbance vs. TMP (MA) added to urine. (Plot of the data of table 3.)

Figure 2. Graph of absorbance vs. subjective color score of urine. (Plot of the data of table 3.)

Table 3 and Figures 1 and 2, data graphed from Table 3, illustrate that, as experiments 1 and 2 demonstrated, there is a strong trend for color subjective score and absorbance at 532 nm to increase as increasing quantities of TMP were added to the urine sample. This validates the fact that the test kit is able to detect and quantify, in a semi quantitative manner, the malonaldehyde in urine (as well as several other oxidation products).

Potential inaccuracies

Several potential inaccuracies are inherent in any urinary analysis; in particular a rapid in home test, and it is useful to discuss some of the major potential sources of inexactness in data collection. One obvious source of variation, not related to lipid oxidation, is urine concentration, a factor usually normalized by quantification of creatinine in the urine11. Creatinine determination is not practical under the circumstances of an "in-home" urinary malonaldehyde test. Therefore, we recommend that subjects be firmly directed to maintain excellent hydration by drinking several glasses of water each day. The visible yellow color of urine can be used to adequately assess hydration; a well-hydrated individual will produce a very light yellow color to almost colorless urine.

Another potential source of variation is the transfer of solutions A and B into the urine. This may occasionally lead to less color development following the five-minute incubation period. Clinicians should be alert to this possibility.

A third potential problem would be the production of urinary malonaldehyde via metabolic activity related to intense exercise11. Subjects should be admonished to minimize exercise on the days of testing, unless physical activity is being studied.

Recommended use of test

The urinary malonaldehyde test is only semi-quantitative. However, skillful use of the test can overcome any potential inaccuracies outlined in the foregoing section. In a typical case, the patient will usually product intensely color (pink or pinkish-red) at the start of the treatment. Using a multiple intervention paradigm designed to promote a healthy lifestyle, a gradual drop in color development, indicating excretion of lower levels of lipid oxidation products such as MA, should be observable with the "Check-up" Free Radical Test Kit.

Conclusion

The urinary malonaldehyde test is valid as a semi-quantitative approximation of the detrimental lipid oxidation reactions occurring in the body. Although not studied in this investigation, several changes in life-style should reduce urinary malonaldehyde levels. These include the cessation of smoking, the reduction in the consumption of fried foods, especially deep-fried foods, the increased intake of fat- and water-soluble vitamins and antioxidants, the increased intake of phytochemicals, many of which are antioxidants, and the increased intake of ribose¹¹, a cellular energy enhancer that has been shown to reduce malonaldehyde in exercising humans.

The deviations from the test kit protocol used in this validation study are not necessary for the proper determination of urinary MA in the clinical situation.

References

1. Free Radical Biology and Medicine, 40, 341-347, 2006.

2. Blass, J. P., and McDowell, F. H. Eds. "Oxidative/Energy Metabolism in Neurodegenerative Disorders." Annals of the New York Academy of Sciences, 893, 1999.

3. Pryor, W. A. Ed. "Bio-assays for Oxidative Stress Status (BOSS)." Elsevier Science, Amsterdam, The Netherlands, 2001.

4. Pryor, W. A. Ed. Oxidative stress status: OSS, BOSS, and "Wild Bill" Donovan. "Bio-assays for Oxidative Stress Status (BOSS)." Elsevier Science, Amsterdam, The Netherlands, 1-2, 2001.

5. Draper, H. H., Csallany, A. S., and Hadley, M. Urinary aldehydes as indicators of lipid peroxidation in vivo. Pryor, W. A. Ed. "Bio-assays for Oxidative Stress Status (BOSS)." Elsevier Science, Amsterdam, The Netherlands, 184-190, 2001.

6. Tappel, A. L. Lipid peroxidation damage to cell components. Fed. Proc. 32, 1870-1874, 1973.

7. Addis, P.B., Emanuel, H.A., Bergmann, S.D. and Zavoral, J.H. Capillary GC quantification of cholesterol oxidation products in plasma lipoproteins of fasted humans. Free Radical Biology and Medicine. 7:179-182, 1989.

8. Emanuel, H.A., Hassel, C.A., Addis, P.B., Bergmann, S.P. and Zavoral, J.H. Plasma cholesterol oxidation products (oxysterols) in human subjects fed a meal rich in oxysterols. Journal of Food Science 56:843-847, 1991.

9. Csallany, A. S., Guan, M. D., Manwaring, J. D., and Addis, P. B. Free malonaldehyde determination in tissues by high-performance liquid chromatography. Analytical Biochemistry, 142: 277-283, 1984.

10. Yagi, K. Lipid peroxides and human diseases. Chemistry of Lipids, 45: 337-351, 1987.

11. Seifert, J. G., Subudhi, A. W., Fu, M. Riska, K. L., John, J. C., St.Cyr, J. A. and Shecterle, L. M. Benefits of ribose supplementation on oxidative stress during and following hypoxic exercise. (Submitted)