Methylated purines in urinary stones.
Limited attention has been paid to the possible role of metabolites of methylxanthines (caffeine, theophylline, and theobromine) in the pathogenesis of urolithiasis, although their average daily output in urine, depending on dietary intake, is ~500 mg/day (4), which equals the output of UA. The main metabolites--methyluric acids--have properties similar to those of UA, including low solubility in aqueous solutions. Detection of their presence in stones by use of traditional analytical techniques (x-ray diffraction, infrared spectroscopy, and semiquantitative wet tests) has not been reported. We have recently developed a gradient HPLC method for separation of 16 purines, including 10 methyl derivatives of Xan and UA, in urinary calculi (5). We present here results of analysis of purine derivatives in urinary calculi by this method.
Materials and Methods
We examined urinary stones from 48 adult patients (31 men and 17 women) of Caucasian descent [mean (SD) age of 49 (15) years] who underwent surgery at the urology ward in Szczecin. If the stones were heterogeneous, i.e., their respective parts or layers differed in color or structure from each other, separate samples were taken from each part, e.g., nucleus and outer layer. In total, 65 samples were analyzed.
Urinary stone samples were dissolved in 0.1 mol/L NaOH as described previously (6). The extracts were diluted in 50 mmol/L K[H.sub.2]P[O.sub.4] and analyzed by HPLC (5). Using this method, we separated 16 compounds: UA; Xan; hypoxanthine (Hyp); 2,8-DHA; allopurinol; oxypurinol; 1-, 3-, 7-, and 9-methyluric acid (1-, 3-, 7-, and 9-MUA); 1,3-, 1,7-, and 3,7-dimethyluric acid (1,3-, 1,7-, and 3,7-DMU); and 1-, 3-, and 7-methylxanthine (1-, 3-, and 7-MX). Limits of detection for individual compounds ranged from 6 to 35 [micro]g/g of stone weight, and imprecision (CV) was 0.5%-2.4%. Other components of the stones were quantified by wet methods (7).
When multiple samples were taken from a stone, the arithmetic mean was calculated for the content of each compound. Normality of distributions was checked with the Shapiro-Wilk test. Correlations were measured with Spearman rank correlation coefficient ([r.sub.s]).
The purine compound found as the main stone constituent was UA; its content exceeded 50% in 9 (19%) stones. Of the 48 stones analyzed, 14 (29%) contained no UA, 14 (29%) contained detectable UA that constituted <1% of stone mass, 8 (17%) contained 1%-25% UA, 3 (6%) contained 25%-50% UA, 1 (2%) contained 50%-75% UA, and 8 (17%) contained >75% UA.
In all UA-containing calculi, we found 9 other purines in addition to UA: Xan; Hyp; 2,8-DHA;1-, 3-, and 7-MUA; 1,3-DMU; and 3-and 7-MX (Table 1 and Fig. 1). Allopurinol, oxypurinol, 9-MUA, 1,7- and 3,7-DMU, and 1-MX were not detected in any sample. Other main constituents of the calculi were calcium oxalates in 26 (54%) stones, calcium phosphates in 6 (13%), struvite (magnesium ammonium phosphate) in 6 (13%), and cystine in 1 stone (2%).
The distributions of UA and other purines in the stones were far from gaussian; therefore, the results were not expressed as mean (SD). The median was not a useful parameter either because most of the compounds were detected in less than one-half of the stones.
Several purines were detected in stones containing at least 4% UA (1-MUA was detectable even when UA constituted 1% of stone mass) but were never found in stones containing no UA. There were strong positive correlations (P <0.0001) between the UA content and each purine (Table 1).
The correlation plot of UA and 1-MUA in the stones is shown in Fig. 2. For other purines, correlation plots also suggested a linear regression, with the exception of 3-MX, for which the highest concentrations were observed in stones containing 10%-20% UA.
The strong linear correlation of UA with admixtures of other purines justifies expressing their quantities as purine/UA quotients, calculated as the ratio of masses of any purine to UA in a stone. The quotients could not be calculated for stones containing no UA; for calculi containing little UA and accompanying purines, purine/UA quotients would be very imprecise. That is why only 19 quotients calculated for stones containing >4% UA were analyzed further (Table 1). The distributions of quotients were gaussian with the exception of the 3-MX/UA and Xan/UA quotients. The mean of the quotients (MQ) is an optimal estimate of the regression coefficient in the equation formulating dependence between content of UA and other purines: predicted purine content = MQ x (UA content).
[FIGURE 1 OMITTED]
All of the UA calculi examined contained N-methyl derivatives of UA (1-MUA, 3-MUA, 7-MUA, and 1,3-DMU) and Xan (3-and 7-MX). Methylated purines were not present only as the result of contamination of the stone surface by urine; they were often present in larger amounts in the stone nuclei than in the outer layers. The maximum total content of methylated purines in the calculi examined was 1.7% of stone mass. This amount is relatively small but considerably greater than the sum of endogenous purines (Xan, Hyp, and 2,8-DHA), which amounted to 0.16% of the weight.
[FIGURE 2 OMITTED]
The methylated purines detected in the stones are excreted in urine as metabolites of ingested methylxanthines: 1,3,7-trimethylxanthine (caffeine); 1,3-dimethylxanthine (theophylline); and 3,7-dimethylxanthine (theobromine). These compounds are present in popular dietary items: coffee and some carbonated beverages (caffeine), tea (caffeine and theophylline), and cocoa and chocolate (theobromine). Theophylline and its derivative aminophylline are frequently used as drugs. A simplified diagram of the metabolism of methylxanthines in humans is shown in Fig. 3. Activities of particular enzymes show great variability in the population, being determined genetically and modified by numerous factors such as age, smoking, alcohol, drugs, and liver diseases (4,8-10).
The order of MUAs arranged according to their content in urine (1-MUA > 7-MUA > 3-MUA) (4) is consistent with their content relative to UA in stones (MQ values in Table 1). The amount of 3-and 7-MX was proportionally much less in stones than reported in urine. Neither 1-MX nor 1,7-DMU were detected in stones, although they are present in urine. The absence of 9-MUA is consistent with the assumption that MUAs in stones are metabolites of methylxanthines originating from plants, which never contain a methyl group at N-9 of the purine ring.
[FIGURE 3 OMITTED]
Urinary concentrations of methyl derivatives of UA and Xan probably never exceed the saturation point of the solution, with the exception of 1-MUA, for which the upper limit of excretion is ~800-900 mg/day (4). To our knowledge, only 1 patient with urinary stones consisting mainly of 1-MUA has been reported (unpublished data; personal communication from Dr H. Anne Simmonds, Purine Research Laboratory, London, UK). That patient drank at least 8 cups of coffee per day. This unusual case supports the hypothesis that precipitation of 1-MUA may be the cause of urolithiasis in people ingesting excessive amounts of caffeine.
Precipitation of any compound may occur when its concentration in urine exceeds its solubility (11). However, the results of the present work suggest that compounds present in urine at concentrations much below the saturation point of their solutions may coprecipitate during UA stone formation. Methylated purines were present exclusively in the stones containing UA, and their content showed a linear correlation with UA content. This finding supports the notion that in an environment of UA supersaturation, coprecipitation with UA from undersaturated solution occurs.
The results presented here enabled us to identify 3 independent factors determining the content of compounds accompanying UA in stones:
* The UA content in the stone correlates positively with the amount of purine admixtures.
* The concentrations of particular compounds in urine are consistent with their content in the stones, e.g., 1-MUA > 7-MUA > 3-MUA.
* The greater the similarity of the structure of the derivative to the main constituent of the stone (UA), the higher the admixture. For example, 1,3-DMU with 2 methyl substituents was present in amounts ~5-fold less than those of monomethyl 7-MUA. Monomethyl derivatives of Xan, lacking an oxo substituent at C-8 of the purine ring, which is present in UA and its derivatives, were found in much lower amounts than N-1-and N-7-monomethyl UA derivatives. Xan was present in stones in amounts 2- to 4-fold greater on average than Hyp. This result can be explained by the presence of an oxo substituent in 2 positions of the Xan ring (C-2 and C-6) and in only 1 position of Hyp ring (C-6; Fig. 1).
These 3 factors may influence the incorporation of admixtures of components present in solution into the growing crystal. In theory, the amount of admixture is directly proportional to its concentration in solution and to the amount of the main constituent of the crystal (12). Our results are consistent with this theory because the concentration of purines in the solid phase (or percentage of stone mass) was directly proportional to their concentration in urine and the UA content in the stone. The proportionality factor is determined by the similarity of the UA structure to the coprecipitating compound.
Processes of coprecipitation are related to the ability of crystallizing compounds to form a mixed crystal (13), which is a solid solution (14) containing ions or molecules of one compound in the crystal lattice of another. Crystals of purines are formed by molecules and classified as molecular crystals (15). Solid solutions of organic compounds are always substitutional, formed by replacing some solvent molecules with the admixture (16). The necessary condition for 2 or more organic compounds to form a solid solution is the similarity of shape and size of their molecules (13). This condition is fulfilled by purine derivatives identified in urinary stones (Fig. 1).
We propose a model of the UA stone molecular structure for which some purines may form a solid solution with UA as solvent. In the course of UA crystallization from a supersaturated solution, admixtures of other compounds are incorporated into the expanding crystal lattice. The proportion of substituted molecules is greater at higher concentrations of admixture in solution (which increases frequency of contact of its molecules with crystal surface) and when the difference between the structures of the main component and the admixture is smaller [which facilitates their incorporation into the crystal lattice (17)].
The model presented explains the linear correlation between the amounts of UA and admixtures of other purines in stones. Purines can incorporate into the crystal lattice formed by another purine (UA), but not into different crystal structures of stones. For verification of the presented model, it would be necessary to study the relationships between conditions of crystallization and the composition of crystals in vitro, using combination of various modern techniques (18).
Derivatives of methylxanthines were the dominant purines accompanying UA in stones in this study. Experiments by others regarding the role of various trace elements, or other organic compounds, in the processes of stone formation showed that they can have either inhibitory or promoting influences on crystal growth (19,20). There are no such data for purines. Methylated purines in urine may also inhibit or promote UA crystallization.
Extensive prospective studies concerning the influence of beverages (e.g., tea and coffee with and without caffeine) were carried out in the United States (21,22). The results showed that consumption of such beverages is associated with a lower frequency of urolithiasis, although this influence was relatively small. There was no difference between coffee containing and not containing caffeine. Unfortunately, the study did not take into consideration the composition of the stones. It is possible that the influence of methylated purines on calculi formation is limited to UA stones. Such calculi comprise as little as 5%-10% of all stones in the US population (23); therefore, the influence of methylxanthines on the occurrence of all kinds of urolithiasis could remain imperceptible.
In conclusion, the results of this study indicate that methyl derivatives of UA and Xan, being metabolites of methylxanthines originating from plants (caffeine, theophylline, and theobromine), are permanent constituents of urinary calculi containing UA. We developed a model in which admixtures of purines form substitutional solid solutions in UA. This model is consistent with the results presented but needs further experimental verification. It appears that 1-MUA may be the main constituent of urinary stones in patients with a history of excessive ingestion of methylxanthines. The frequency of such urolithiasis may be underestimated because of difficulties in diagnosis. Further studies are needed to establish the role of methylated purines in the pathogenesis of urolithiasis.
This work was supported by Grant 4 P05B 028 15 from the Polish Committee for Scientific Research.
Received January 10, 2005; accepted May 3, 2005.
Previously published online at DOI: 10.1373/clinchem.2005.048033
(1.) Carvalho M, Lulich JP, Osborne CA, Nakagawa Y. Role of urinary inhibitors of crystallization in uric acid nephrolithiasis: Dalmatian dog model. Urology 2003;62:566-70.
(2.) Moran ME. Uric acid stone disease. Front Biosci 2003;8:s1339-55.
(3.) Sahota A, Tischfield J, Kamatani N, Simmonds HA. Adenine phosphoribosyltransferase deficiency: 2,8-dihydroxyadenine lithiasis. In: Scriver CR, Beaudet AL, Sly WS, Valle D, eds. The metabolic and molecular bases of inherited disease, 8th ed. New York: McGraw-Hill, 2001:2571-84.
(4.) Ullrich D, Compagnone D, Munch B, Brandes A, Hille H, Bircher J. Urinary caffeine metabolites in man. Age-dependent changes and pattern in various clinical situations. Eur J Clin Pharmacol 1992; 43:167-72.
(5.) Safranow K, Machoy Z. Simultaneous determination of 16 purine derivatives in urinary calculi by gradient reversed-phase high-performance liquid chromatography with UV detection. J Chromatogr B Analyt Technol Biomed Life Sci 2005;819:229-35.
(6.) Safranow K, Machoy Z, Ciechanowski K. Analysis of purines in urinary calculi by high-performance liquid chromatography. Anal Biochem 2000;286:224-30.
(7.) Farrington CJ, Liddy ML, Chalmers AH. A simplified sensitive method for the analysis of renal calculi. Am J Clin Pathol 1980; 73:96-9.
(8.) Denaro CP, Wilson M, Jacob P, Benowitz NL. The effect of liver disease on urine caffeine metabolite ratios. Clin Pharmacol Ther 1996;59:624-35.
(9.) Grant DM, Tang BK, Kalow W. Variability in caffeine metabolism. Clin Pharmacol Ther 1983;33:591-602.
(10.) Miners JO, Birkett DJ. The use of caffeine as a metabolic probe for human drug metabolizing enzymes. Gen Pharmacol 1996;27: 245-9.
(11.) Mandel N. Mechanism of stone formation. Semin Nephrol 1996; 16:364-74.
(12.) Ohara M, Reid RC. Modeling crystal growth rates from solution. Englewood Cliffs: Prentice-Hall, 1973:272pp.
(13.) Kitaigorodsky AI. Mixed crystals. Berlin: Springer-Verlag, 1984: 388pp.
(14.) Putnis A, Pina CM, Astilleros JM, Fernandez-Diaz L, Prieto M. Nucleation of solid solutions crystallizing from aqueous solutions. Philos Transact Ser A Math Phys Eng Sci 2003;361:615-31.
(15.) Wright JD. Molecular crystals, 2nd ed. Cambridge: Cambridge University Press, 1994:235pp.
(16.) Nyvlt J, Ulrich J. Admixtures in crystallization. Weinheim: VCH, 1995:391pp.
(17.) York P, Grant DJW. A disruption index for quantifying the solid state disorder induced by additives or impurities. I. Definition and evaluation from heat of fusion. Int J Pharm 1985;25:57-72.
(18.) Giordano F, Bettini R, Donini C, Gazzaniga A, Caira MR, Zhang GG, et al. Physical properties of parabens and their mixtures: solubility in water, thermal behavior, and crystal structures. J Pharm Sci 1999;88:1210-6.
(19.) Hofbauer J, Steffan I, Hobarth K, Vujicic G, Schwetz H, Reich G, et al. Trace elements and urinary stone formation: new aspects of the pathological mechanism of urinary stone formation. J Urol 1991;145:93-6.
(20.) Fink DA, Sours RE, Swift JA. Modulated uric acid crystal growth in the presence of acridine dyes. Chem Mater 2003;15:2718-23.
(21.) Curhan GC, Willett WC, Rimm EB, Spiegelman D, Stampfer MJ. Prospective study of beverage use and the risk of kidney stones. Am J Epidemiol 1996;143:240-7.
(22.) Curhan GC, Willett WC, Speizer FE, Stampfer MJ. Beverage use and risk for kidney stones in women. Ann Intern Med 1998;128: 534-40.
(23.) Shekarriz B, Stoller ML. Uric acid nephrolithiasis: current concepts and controversies. J Urol 2002;168:1307-14.
KRZYSZTOF SAFRANOW * and ZYGMUNT MACHOY
Department of Biochemistry and Chemistry, Pomeranian Medical University, Szczecin, Poland.
 Nonstandard abbreviations: UA, uric acid; Xan, xanthine; 2,8-DHA, 2,8-dihydroxyadenine; Hyp, hypoxanthine; 1-, 3-,7-, and 9-MUA, 1-, 3-,7-, and 9-methyluric acid; 1,3-,1,7-, and 3,7-DMU, 1,3-,1,7-, and 3,7-dimethyluric acid; and 1-, 3-, and 7-MX, 1-, 3-, and 7-methylxanthine.
* Address correspondence to this author at: Department of Biochemistry and Chemistry, Pomeranian Medical University, Powstancow Wielkopolskich 72 70-111 Szczecin, Poland. Fax 48-91-4661516; e-mail email@example.com.
Table 1. Content of purines detected in the examined urinary stones. Correlation No. (%) with UA of stones Content content in which range, ([r.sub.s]) Compound detected (a) % (b) UA 34 (71) 0-90.7 Xan 20 (42) 0-0.12 0.88 Hyp 19 (40) 0-0.04 0.87 2,8-DHA 16 (33) 0-0.012 0.84 1-MUA 24 (50) 0-1.20 0.94 3-MUA 14 (29) 0-0.05 0.80 7-MUA 22 (46) 0-0.31 0.92 1,3-DMU 18 (38) 0-0.08 0.87 3-MX 13 (27) 0-0.04 0.63 7-MX 18 (38) 0-0.16 0.85 Content relative to UA content, (c) x [10.sup.-3] g purine/g UA Mean (MQ) Compound (d) Median SD Range UA Xan 1.43 0.83 1.81 0.31-8.55 Hyp 0.35 0.37 0.17 0-0.82 2,8-DHA 0.09 0.07 0.06 0-0.19 1-MUA 9.91 9.48 3.97 2.60-18.74 3-MUA 0.23 0.23 0.19 0-0.59 7-MUA 2.38 2.65 1.03 0.20-3.63 1,3-DMU 0.52 0.48 0.25 0-0.92 3-MX 0.60 0.07 1.00 0-2.90 7-MX 0.98 0.86 0.73 0-2.69 (a) Content of purines detected in the 48 examined stones is expressed as percentage of total stone mass. (b) Correlations with UA content are calculated as Spearman rank correlation coefficients ([r.sub.s]). (c) Statistics of purine/UA quotients were calculated for each purine in 19 stones containing >4% UA. (d) Mean purine content in a stone (in mg) = MQ x UA content (in g).
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||General Clinical Chemistry|
|Author:||Safranow, Krzysztof; Machoy, Zygmunt|
|Date:||Aug 1, 2005|
|Previous Article:||Stable-isotope dilution liquid chromatography--electrospray injection tandem mass spectrometry method for fast, selective measurement of...|
|Next Article:||Maternal serum invasive trophoblast antigen and first-trimester down syndrome screening.|