- Email: [email protected]
Studies of urinary cystine precipitation in vitro: ontogeny of cystine nephrolithiasis and identification of meso-2,3-dimercaptosuccinic acid as a potential therapy for cystinuria
Molecular Genetics and Metabolism 80 (2003) 419–425 www.elsevier.com/locate/ymgme
Studies of urinary cystine precipitation in vitro: ontogeny of cystine nephrolithiasis and identification of meso-2,3-dimercaptosuccinic acid as a potential therapy for cystinuria P. Parvex,a,b,* R. Rozen,b A. Dziarmaga,b and P. Goodyerb b
a Department of Nephrology, Geneva ChildrenÕs Hospital, Geneva, Switzerland Department of Pediatrics, McGill University, Montreal ChildrenÕs Hospital, Montreal, PQ, Canada
Received 21 August 2003; accepted 23 August 2003
Abstract Children with fully recessive (Type I/I) cystinuria have a high risk of stone formation in the first decade of life. To assess the tendency for cystine to precipitate in individual urine samples, we developed an in vitro assay in which radiolabelled cystine (4 mM) was dissolved in urine at 37 C after alkalization to pH 10. Samples were then brought to pH 5, cooled, and centrifuged. The % decrease in supernatant cpm was used as a measure of cystine precipitation (CP). CP varied widely among normal children (74% 34) whereas variability of repeated determinations on a single adult individual was modest (64% 3.3). The assay was used to compare various potential therapies for cystinuria. Precipitation of exogenous cystine from normal urine was strongly inhibited by addition of D -penicillamine (CP: 8% 3) or dimercaptosuccinic acid (DMSA) (CP: 5% 1), at urinary concentrations attained by standard oral doses of each drug. Mercaptopropionylglycine (MPG) was moderately effective (CP: 43% 9), whereas captopril was a weak inhibitor (CP: 63% 12). Precipitation of endogenous cystine (2191 lmol/L) from a cystinuric patient showed that DMSA and D -penicillamine were again highly effective compared to the other agents. In addition DMSA and penicillamine added to the same patientÕs urine reduced the free cystine by 50% (as measured by automated amino acid analyzer) whereas MPG and captopril had no effect. In conclusion, DMSA is comparable to D -penicillamine as an in vitro inhibitor. 2003 Elsevier Inc. All rights reserved. Keywords: Cystinuria; Nephrolithiasis; Kidney stones; DMSA; Penicillamine
Introduction Cystinuria is an autosomal-recessive disorder of amino acid transport affecting the renal tubules and gastrointestinal tract. The primary clinical manifestation is nephrolithiasis, with subsequent risk of renal insufficiency due to urinary tract obstruction. Cystinuria accounts for 3% of nephrolithiasis in humans; the incidence of cystinuria worldwide is about 1:7000 [1,2]. Under normal conditions, epithelial cells of the renal proximal tubules reabsorb 99% of filtered cystine. However, proximal tubular reabsorption of cystine is severely compromised in patients who inherit two mutations of either: (A) the SLC3A1 (rBAT) gene on * Corresponding author. Fax: +41-2-2382-4621. E-mail address: [email protected] (P. Parvex).
1096-7192/$ - see front matter 2003 Elsevier Inc. All rights reserved. doi:10.1016/j.ymgme.2003.08.025
chromosome 2p (Type I, fully recessive cystinuria) encoding a subunit of the luminal cystine transporter or (B) the SLC7A9 gene on chromosome 19q (non-Type I, incompletely dominant cystinuria) encoding the transport channel itself [3]. In the collecting duct, cystine concentration exceeds the limit of solubility at acid pH, allowing formation of characteristic hexagonal crystals. Dent and Senior [4,5] analyzed the limits of cystine solubility and established a graph relating urine cystine concentration to pH. At pH 5.0, cystine remains in solution up to a concentration of 1200 lmol/L. We consider here, a cohort of patients with homozygous Type I cystinuria who were followed longitudinally from birth to determine whether onset of nephrolithiasis correlates with changes in urinary cystine level [6,7]. An in vitro assay was developed to assess the tendency for cystine to precipitate in individual urine
420
P. Parvex et al. / Molecular Genetics and Metabolism 80 (2003) 419–425
from normal children at different ages. Finally, we examined the comparative efficacy of various therapeutic agents to inhibit urinary cystine precipitation in vitro; drugs commonly used in treatment of cystinuria (D penicillamine, mercaptopropionylglycine (MPG), and captopril) were compared to a potential novel therapeutic agent for cystinuria, dimercaptosuccinic acid (DMSA) [8–10].
Materials and methods Patients Eight patients with Type I/I cystinuria were identified in the neonatal period by the Quebec Genetics Network Neonatal Screening Program, and followed from birth until age 11–15. These Type I/I patients excreted >2000 lmol cystine/g creatinine while both parents excreted cystine in the normal range (<100 lmol/g creatinine). Two mutations of the rBAT gene were identified in each proband. The children were followed prospectively from birth through adolescence with annual clinic visits and renal ultrasonography; urinary cystine excretion was measured 1–3 times per year. The age of appearance of the first stone was recorded; the mean cystinuria excretion level before and after 6 years of age was calculated for each child in lmol/g creatinine and the mean urinary cystine concentration in lmol/L. In vitro cystine precipitation assay Urine samples from normal children aged 1–18 were collected and frozen at )22 C for analysis. Urine was thawed and 30-ml aliquots were dialyzed in Spectra/Por (MWCO ¼ 12–14 kDa) dialysis tubing overnight against 1 L of 0.9 g/dL NaCl. Cystine was added to each urine sample (4000 lmol/L) and adjusted to pH 10 with NaOH to promote dissolution with shaking at 37 C. Radiolabelled cystine (0.05 lCi) was added, mixed and 4 aliquots of the supernatant were counted in a liquid scintillation counter. The samples were then brought to pH 5 with HCl and left overnight at 4 C to favor cystine precipitation. Samples were spun for 10 min at 5000 rpm and the supernatant was recounted in quadruplicate. The mean percent decrease in supernatant cpm was calculated for each sample and compared to a single adult control urine sample run with each assay. Results were expressed as percentage of control. To assess intraassay variability, we assayed frozen urine samples from the same adult control on 11 separate days. Percent cystine precipitation was 63% 10 (SEM). When cystine precipitation was assayed on fresh urine samples from the same individual on eight separate days, intra-assay variability was similar (67% 7 SEM).
Pharmacologic agents inhibiting cystine precipitation To analyze various potential inhibitors of urinary cystine precipitation, drugs were added to undialyzed aliquots of a single adult control urine containing radiolabelled cystine 4000 lmol/L. We compared the drugs at concentrations equivalent to urinary drug levels predicted from published pharmacokinetic data on each agent after standard doses. • D -Penicillamine (C4 H11 NO2 S): 42% of a standard 1 g intravenous dose is excreted in the urine over 24 h; 25–50% of a similar oral dose is excreted in the urine over 24 h [11,12]. Thus, for a standard oral dose of 1 g/day (for a 70-kg adult) we estimated that the mean urine penicillamine concentration would be 0.42 1 g in 1.5 L of urine/day. Mean urinary concentration ¼ 280 lg/ml (2 mM). • a-Mercaptopropionylglycine (MPG) (C5 H9 NO3 S): 25% of a standard 1 g daily oral dose (for a 70-kg adult) of MPG appears in the urine within 24 h [13]. Mean urinary MPG concentration ¼ 160 lg/ml (0.98 mM). • Captopril (C9 H15 NO3 S): 67% of a maximal 150 mg/ day (for a 70-kg adult) dose is excreted in the urine [14,15]. Mean urinary concentration ¼ 67lg/ml (0.3 mM). • Meso-2,3-dimercaptosuccinic acid (DMSA) (C4 H6 O4 S2 ): 18% of the standard oral dose of 30 mg/kg (used for lead poisoning in children) is excreted in the urine over 24 h [16,17]. Mean urinary DMSA concentration ¼ 250 lg/ml (1.3 mM). We also compared each drug at equimolar concentrations equivalent to the predicted molarity of D -penicillamine normally achieved in urine (2 mM) after standard oral doses. The ability of these drugs to inhibit endogenous cystine precipitation from urine of a Type I/I cystinuria patient was compared (at predicted urinary drug concentrations following standard oral doses adjusted for body weight). Additionally, the drugs were added to aliquots of urine from the same cystinuria patient and free urinary endogenous cystine levels were quantified on an automated amino acid analyzer.
Results Age-related factors influencing cystine precipitation While nephrolithiasis often begins in the first decade of life, children with cystinuria seem to be relatively protected in the early years. We wondered whether the risk of nephrolithiasis might be explained by changes in cystine concentration (lmol/L) or excretion level (lmol/ g creatinine) during this period. To examine this possibility, we studied the onset of stone formation in a group
P. Parvex et al. / Molecular Genetics and Metabolism 80 (2003) 419–425
of eight Type I/I cystinuria patients followed prospectively from birth. During the first 6 years of life, none of the patients developed nephrolithiasis detectable by annual ultrasonography or suggested by cli"nical symptoms. However, between 6 and 10 years, four children (50%) formed their first stone (mean age 9.3 years) (Table 1). The level of cystine excretion in children who formed a stone was not higher than in children without stones (Fig. 1A). Conceivably, infants take in larger fluid volumes than older children, achieving greater dilution of the filtered cystine. If so, then mean urinary cystine concentration might be lower in the first years of life, accounting for a lower risk of stone formation. However, the mean urinary cystine concentration was not significantly greater at age 6–10 (2465 lmol/L) than at age 0–6 year (2168 lmol/L)(NS) (Fig. 1B). In normal children, glomerular filtration rate (GFR) increases with age. We reasoned that older children might have to contend with an increased filtered load of cystine, excreting higher levels in the urine. However, as seen in Table 1, older (6–10 years) children actually excrete a lower level of cystine (2559 lmol/g creatinine) than young (0–6 years) children (4260 lmol/g creatinine) (p < 0:003) (Fig. 1C).
421
To determine whether cystine precipitates less readily from urine of younger versus older children, we measured in vitro precipitability (% CP) of 4 mM cystine in urine from normal children aged 0–6 (N: 13) versus 6–10 years (N: 13). Percentage CP was significantly higher in the older age group (94%) than in the younger children (76%) (p < 0:03) (Fig. 2).
Fig. 2. In vitro precipitation of cystine (4 mM) added to urine from non-cystinuric subjects: 0–6 years (76% 6.8); 6–10 years (94% 3.4). Cystine precipitation is significantly lower in the younger group (p < 0:03).
Table 1 Genotype, age at time of first stone and mean urine cystine concentration and excretion for young (0–6 years) and older (6–10) age groups Patient
1 2 3 4 5 6 7 8 Mean
Followed until age (years)
15 13 11 12 12 19 12 11
Genotype cystinuria
I/I I/I I/I I/I I/I I/I I/I I/If
Age at first stone
8.6 8.9 10 —
9.8 — — —
9.3
U [cystine] concentration (lmol/L)
U cystine excretion (lmol/g creat)
Age 0–6
Age 6–10
Age 0–6
Age 6–10
NA 3037 1870 1225 2593 3820 1475 3235
1842 2298 1845 2001 1967 2708 2565 2118
NA 2226 4369 4957 3325 4110 5103 5726
2301 1549 2750 2976 2517 1825 3256 3295
2465
2168
4260
2559
The mean age at onset of nephrolithiasis in 4 of these 8 patients was 9.3 years. NA, not available.
Fig. 1. (A) Comparison of mean of urinary cystine concentration (lmol/g creat) in stone-formers versus non-stone-formers. (B) Mean urinary cystine concentration (lmol/L) in Type I/I cystinuria patients (n ¼ 8) at age 0–6 years and 6–10 years. Urinary cystine concentration did not change with age (NS). (C) Mean urinary cystine excretion (lmol/g creat) in Type I/I cystinuria patients at age 0–6 years and 6–10 years. Cystine excretion was significantly higher in the first years of life (**p < 0:003).
422
P. Parvex et al. / Molecular Genetics and Metabolism 80 (2003) 419–425
Comparative effects of anti-cystinuric drugs on cystine precipitation in vitro In older cystinuric patients, when recurrent nephrolithiasis begins, conservative measures to prevent stone formation may fail; in this case, several agents (D -penicillamine, MPG, and captopril) have been used to convert urinary cystine to more soluble compounds. In general these thiol-containing reagents attack the internal disulfide bond of cystine, forming mixed disulfides,
which are more soluble in the urine (Table 2). We used our in vitro cystine precipitation assay to compare these monothiols with DMSA, a dithiol which as been shown to form a similar mixed disulfide with cystine [18–24] (Table 2). Drugs were added to our in vitro assay at concentrations corresponding to predicted urinary levels following standard oral doses of each agent. Fig. 3A indicates the effect of each agent on precipitation of 4 mM cystine from normal urine. DMSA (5% 1) and D -penicillamine (8% 3) were the strongest inhibitors of
Table 2 (A–C) Monothiols combine with half-cystine to form more soluble mixed disulfide with cysteine and (D) the dithiol, DMSA, combines with two halfcystine molecules to form a mixed disulfide
P. Parvex et al. / Molecular Genetics and Metabolism 80 (2003) 419–425
CP. MPG (43% 9) was of intermediate efficacy and captopril (63% 12) was a weak inhibitor of CP. The four drugs were also compared at concentrations equimolar to the predicted concentration of urinary D -penicillamine (2 mM) following a standard oral dose (Fig. 3B). D -penicillamine (11% 2) and DMSA (10% 5) were equally effective in decreasing CP. Even at an equimolar dose, captopril (53% 22) and MPG (40% 10) was less efficient than the former agents. Furthermore, the dose of captopril required to achieve concentrations of 2 mM in urine would be about six times the maximal tolerable dose and would produce unacceptable hypotension. Effects of anti-cystinuric drugs on endogenous urinary cystine To assess the effect of these drugs on endogenous urinary cystine, we added radiolabelled cystine to urine of a Type I/I cystinuria patient (urine cystine ¼ 2000 lmol/L). After acidification and cooling, CP was suppressed by DMSA (9% of control), D -penicillamine (12% of control), MPG (18% of control) and, to a lesser extent, by captopril (50% of control) (Fig. 4). We also tested the capacity of each drug to reduce endogenous free cystine (measured on an automated
423
Fig. 4. Comparative efficacy of various drugs on inhibition of endogenous cystine precipitation from urine of a cystinuric patient. The drugs were compared at achievable urinary concentration after standard oral doses adapted to the patientÕs weight.
amino acid analyzer) in urine from the same Type I/I cystinuria patient. At urinary concentrations predicted after standard oral doses, the free cystine peak (2337 lmol/g creatinine) was sharply reduced by addition of DMSA (to 1165 lmol/g creatinine) or D -penicillamine (to 1209 lmol/g creatinine). On the other hand, captopril and MPG had little effect on the free cystine peak (to 2257 and 2488 lmol/g creatinine, respectively) (Fig. 5).
Discussion
Fig. 3. (A) Comparison of efficacy of cystine precipitation inhibitors in a normal subject, at achievable urinary concentration for each drug after standard oral doses. (B) Comparison of efficacy of cystine inhibitors in normal urine at equimolar concentration (2 mM): DMSA and D -penicillamine are strong inhibitors whereas MPG and captopril are weaker inhibitors.
Factors leading to stone formation in cystinuria are poorly understood, but high urinary cystine concentrations predisposing to cystine precipitation at acid pH are certainly a prerequisite. Among patients from a specific genetic subtype of cystinuria, the frequency of nephrolithiasis does not correlate well with urinary cystine level. However, children with homozygous Type I/I cystinuria excrete higher levels of cystine and have more stones than children with the milder mixed form (Type I/non-Type I) of cystinuria [5,6]. Dent et al. showed that high fluid intake (to dilute urinary cystine) and urinary alkalinization modifies cystine precipitation in vitro and prevents nephrolithiasis in clinical practice [3,4]. Interestingly, our observations suggest that relative resistance to nephrolithiasis in very young cystinuria patients is not explained by agerelated changes in the filtered load of cystine or in urinary dilution. However, our in vitro studies suggest that cystine is less likely to precipitate from urine of very young (age 0–6 years) children than in older (age 6–10 years) children even though urine samples are dialyzed to standardize the concentration of salts and pH. One likely explanation is that infant urine contains especially high levels of non-dialyzable factors (>12 kDa), which inhibit cystine precipitation. When endogenous inhibitors of cystine precipitation begin to fail and recurrent nephrolithiasis begins in older chil-
424
P. Parvex et al. / Molecular Genetics and Metabolism 80 (2003) 419–425
Fig. 5. (A) Effect of various therapeutic agents on free cystine level (measured on an automated amino acid analyzer) in urine from a cystinuric patient. Normal cystine level 6 100 lmol/L. (B) Free cystine peak (FC) counted with an aminoacid analyzer in urine from a cystinuric patient after addition of different drugs. FCP is decrease by D -penicillamine and DMSA.
dren, exogenous agents must be considered to prevent cystine stone formation. Both D -penicillamine and MPG are clinically effective and are thought to prevent cystine stones by forming relatively soluble mixed disulfides with the cysteine moiety [1].
However, these agents frequently have significant side effects (fever, rash, proteinuria, hepatotoxicity, etc.) and there is a real need for agents with higher benefit/risk ratio. Captopril has been tried and decreased cystine/ creatinine ratio by 10%, but this minimal effect was not
P. Parvex et al. / Molecular Genetics and Metabolism 80 (2003) 419–425
clinically significant [19–21]. In our in vitro studies, D penicillamine was more effective than MPG and captopril in preventing cystine precipitation. Interestingly, DMSA was at least as effective or superior to penicillamine in our cystine precipitation assay. In addition, we observed that DMSA was as effective as D -penicillamine in reducing the endogenous free cystine peak in urine from a cystinuric subject (as measured on an automated amino acid analyzer). MPG is thought to be clinically efficacious in preventing cystine stone formation and was able to reduce cystine precipitation in our in vitro assay. However, MPG was somewhat less potent than either DMSA or D -penicillamine. We suggest, therefore, that MPG may be less effective than these latter agents in clinical practice, but comparative clinical trials have not been performed. It should be noted that measurement of free urinary cystine on an automated amino acid analyzer may be a useful way to adjust penicillamine or DMSA dose during treatment of cystinuria patients; free cystine levels of <1200 lmol/L would be a rational goal. However, under conditions used for sample preparation, an MPG–cystine complex was not evident and no effect on the free cystine peak was observed. This method is not useful in monitoring MPG therapy. Based on its performance in our in vitro assay, it is tempting to consider DMSA as a potential therapeutic agent for cystinuria. DMSA is widely used intravenously at low dose for nuclear scintigraphy in humans and was apparently safe at doses of 25–30 mg/kg/day during 1month oral treatment for lead poisoning in children [23,24]. Maiorino et al. [16] showed that DMSA (10 mg/ kg), given orally to 10 fasting adult subjects, appeared in the urine in disulfide linkage with L -cysteine. In monkeys (Macaca mulatta), 18% of orally absorbed DMSA is excreted in the urine (65% in the feces and 2% in expired air) [18]. These reports suggest that oral DMSA is rapidly excreted in urine at concentrations sufficient to inhibit cystine precipitation. Clinical efficacy and safety profile during longer treatment periods would certainly require careful evaluation, but the in vitro effects of DMSA on cystine precipitation suggest that it may be a promising new therapeutic agent for treatment of cystinuria.
References [1] M. Palacin, P. Goodyer, V. Nunes, P. Gasparini, Cystinuria, in: C. Scriver, A. Beaudet, W. Sly, D. Valle (Eds.), The Metabolic and Molecular Bases of Inherited Disease, vol. III, 2001, pp. 4909– 4932. [2] C. Pak, Etiology and treatment of urolithiasis, Am. J. Kidney Dis. 18 (1991) 624–637.
425
[3] P. Goodyer, M. Boutros, R. Rozen, The molecular basis of cystinuria: an update, Exp. Nephrol. 8 (2000) 123–127. [4] C.E. Dent, B. Senior, J.M. Walshe, The pathogenesis of cystinuria. Polarographic studies of the metabolism of sulfur-containing aminoacids, J. Clin. Invest. 33 (1954) 1216. [5] C.E. Dent, B. Senior, Studies on the treatment of cystinuria, Brit. J. Urol. 27 (1955) 332–377. [6] P. Goodyer, C. Clow, T. Reade, C. Girardin, Prospective analysis and classification of patients with cystinuria identified in a newborn screening program, J. Pediatr. 122 (1992) 568–572. [7] P. Goodyer, I. Sadie, P. On, G. Elks, R. Rozen, Cystinuria subtype and the risk of nephrolithiasis, Kidney Int. 54 (1998) 56–61. [8] G.K. Chow, S. Streem, Medical treatment of cystinuria: results of contemporary clinical practice, J. Urol. 156 (1996) 1576–1578. [9] J. Martens, T. Denneberg, A. Lindell, O. Textorius, Sulfur amino acid metabolism in cystinuria: a biochemical and clinical study of patients, Kidney Int. 37 (1989) 143–149. [10] E.T. Backer, A. van den Ende, W.P. Geus, Determination of urinary cystine for monitoring anticystinuric therapy, Clin. Chim. Acta 165 (1987) 351–358. [11] P. Neter, B. Bannwarth, P. Pere, A. Nicolas, Clinical pharmacokinetics of D -penicillamine, Clin. Pharmacokinet. 13 (1978) 317– 333. [12] T.F. Furlong, S. Posen, D -Penicillamine and the transport of L cystine by rat and human renal cortical brush-border membrane vesicles, Am. J. Physiol. (1990) F321–F327. [13] Y.C. Charles, C. Fuller, K.J.E. Sakhaee, V.A. Adams, Management of cystine nephrolithiasis with alpha-mercaptopropionylglycine, J. Urol. 136 (1986) 1003–1008. [14] S.H. Kubo, J.R. Cody, Clinical pharmacokinetics of angiotensin converting enzyme inhibitors: a review, Clin. Pharmacokinet. 10 (1985) 377–391. [15] M. Perazella, K.B. Gregory, Successful treatment of cystinuria with captopril, Am. J. Kidney Dis. 21 (1993) 504–507. [16] R. Maiorini, D. Bruce, V. Aposhian, Determination and metabolism of dithiol chelating agents, Toxicol. Appl. Pharmacol. 97 (1989) 338–349. [17] E. Liebelt, M. Shannon, W.J. Graef, Efficacy of oral meso-2,3dimercaptosuccinic acid therapy for low-level childhood plumbism, Pediatr. Pharmacol. Ther. 124 (1994) 313–317. [18] J.D. Cremin, M.L. Luck, N.K. Laughlin, D.R. Smith, Efficacy of succimer chelation for reducing brain lead in a primate model of human lead exposure, Toxicol. Appl. Pharmacol. 161 (1999) 283– 293. [19] J. Sloand, J.L. Izzo, Captopril reduces urinary cystine excretion in cystinuria, Arch. Intern. Med. 147 (1987) 1409–1412. [20] T.D. Cohen, S.B. Streem, P. Hall, Clinical effect of captopril on the formation and growth of cystine calculi, J. Urol. 154 (1994) 164–166. [21] S. Streem, P. Hall, Effect of captopril on urinary cystine excretion in homozygous cystinuria, J. Urol. 142 (1989) 1522–1524. [22] J.B. Besunder, D.M. Super, R. Anderson, Comparison of dimercaptosuccinic acid and calcium disodium ethylenediaminetetraacetic acid versus ethylenediaminetetraacetic acid in children with lead poisoning, J. Pediatr. 130 (1996) 966–971. [23] J.J. Chisolm, Safety and efficacy of meso-2,3-dimercaptosuccinic acid (DMSA) in children with elevated blood lead concentration, Clin. Toxicol. 38 (4) (2000) 365–375. [24] J.R. Walter, P.D. Dietrich, J.H. Ware, D.W. Dockery, M. Salganik, J. Radcliffe, R. Jones, B. Ragan, J.J. Chisolm, G.G. Rhodas, The effect of chelation therapy with succimer on neuropsychological development in children exposed to lead, N. Engl. J. Med. 344 (2001) 1421–1471.
Studies of urinary cystine precipitation in vitro: ontogeny of cystine nephrolithiasis and identification of meso-2,3-dimercaptosuccinic acid as a potential therapy for cystinuria P. Parvex,a,b,* R. Rozen,b A. Dziarmaga,b and P. Goodyerb b
a Department of Nephrology, Geneva ChildrenÕs Hospital, Geneva, Switzerland Department of Pediatrics, McGill University, Montreal ChildrenÕs Hospital, Montreal, PQ, Canada
Received 21 August 2003; accepted 23 August 2003
Abstract Children with fully recessive (Type I/I) cystinuria have a high risk of stone formation in the first decade of life. To assess the tendency for cystine to precipitate in individual urine samples, we developed an in vitro assay in which radiolabelled cystine (4 mM) was dissolved in urine at 37 C after alkalization to pH 10. Samples were then brought to pH 5, cooled, and centrifuged. The % decrease in supernatant cpm was used as a measure of cystine precipitation (CP). CP varied widely among normal children (74% 34) whereas variability of repeated determinations on a single adult individual was modest (64% 3.3). The assay was used to compare various potential therapies for cystinuria. Precipitation of exogenous cystine from normal urine was strongly inhibited by addition of D -penicillamine (CP: 8% 3) or dimercaptosuccinic acid (DMSA) (CP: 5% 1), at urinary concentrations attained by standard oral doses of each drug. Mercaptopropionylglycine (MPG) was moderately effective (CP: 43% 9), whereas captopril was a weak inhibitor (CP: 63% 12). Precipitation of endogenous cystine (2191 lmol/L) from a cystinuric patient showed that DMSA and D -penicillamine were again highly effective compared to the other agents. In addition DMSA and penicillamine added to the same patientÕs urine reduced the free cystine by 50% (as measured by automated amino acid analyzer) whereas MPG and captopril had no effect. In conclusion, DMSA is comparable to D -penicillamine as an in vitro inhibitor. 2003 Elsevier Inc. All rights reserved. Keywords: Cystinuria; Nephrolithiasis; Kidney stones; DMSA; Penicillamine
Introduction Cystinuria is an autosomal-recessive disorder of amino acid transport affecting the renal tubules and gastrointestinal tract. The primary clinical manifestation is nephrolithiasis, with subsequent risk of renal insufficiency due to urinary tract obstruction. Cystinuria accounts for 3% of nephrolithiasis in humans; the incidence of cystinuria worldwide is about 1:7000 [1,2]. Under normal conditions, epithelial cells of the renal proximal tubules reabsorb 99% of filtered cystine. However, proximal tubular reabsorption of cystine is severely compromised in patients who inherit two mutations of either: (A) the SLC3A1 (rBAT) gene on * Corresponding author. Fax: +41-2-2382-4621. E-mail address: [email protected] (P. Parvex).
1096-7192/$ - see front matter 2003 Elsevier Inc. All rights reserved. doi:10.1016/j.ymgme.2003.08.025
chromosome 2p (Type I, fully recessive cystinuria) encoding a subunit of the luminal cystine transporter or (B) the SLC7A9 gene on chromosome 19q (non-Type I, incompletely dominant cystinuria) encoding the transport channel itself [3]. In the collecting duct, cystine concentration exceeds the limit of solubility at acid pH, allowing formation of characteristic hexagonal crystals. Dent and Senior [4,5] analyzed the limits of cystine solubility and established a graph relating urine cystine concentration to pH. At pH 5.0, cystine remains in solution up to a concentration of 1200 lmol/L. We consider here, a cohort of patients with homozygous Type I cystinuria who were followed longitudinally from birth to determine whether onset of nephrolithiasis correlates with changes in urinary cystine level [6,7]. An in vitro assay was developed to assess the tendency for cystine to precipitate in individual urine
420
P. Parvex et al. / Molecular Genetics and Metabolism 80 (2003) 419–425
from normal children at different ages. Finally, we examined the comparative efficacy of various therapeutic agents to inhibit urinary cystine precipitation in vitro; drugs commonly used in treatment of cystinuria (D penicillamine, mercaptopropionylglycine (MPG), and captopril) were compared to a potential novel therapeutic agent for cystinuria, dimercaptosuccinic acid (DMSA) [8–10].
Materials and methods Patients Eight patients with Type I/I cystinuria were identified in the neonatal period by the Quebec Genetics Network Neonatal Screening Program, and followed from birth until age 11–15. These Type I/I patients excreted >2000 lmol cystine/g creatinine while both parents excreted cystine in the normal range (<100 lmol/g creatinine). Two mutations of the rBAT gene were identified in each proband. The children were followed prospectively from birth through adolescence with annual clinic visits and renal ultrasonography; urinary cystine excretion was measured 1–3 times per year. The age of appearance of the first stone was recorded; the mean cystinuria excretion level before and after 6 years of age was calculated for each child in lmol/g creatinine and the mean urinary cystine concentration in lmol/L. In vitro cystine precipitation assay Urine samples from normal children aged 1–18 were collected and frozen at )22 C for analysis. Urine was thawed and 30-ml aliquots were dialyzed in Spectra/Por (MWCO ¼ 12–14 kDa) dialysis tubing overnight against 1 L of 0.9 g/dL NaCl. Cystine was added to each urine sample (4000 lmol/L) and adjusted to pH 10 with NaOH to promote dissolution with shaking at 37 C. Radiolabelled cystine (0.05 lCi) was added, mixed and 4 aliquots of the supernatant were counted in a liquid scintillation counter. The samples were then brought to pH 5 with HCl and left overnight at 4 C to favor cystine precipitation. Samples were spun for 10 min at 5000 rpm and the supernatant was recounted in quadruplicate. The mean percent decrease in supernatant cpm was calculated for each sample and compared to a single adult control urine sample run with each assay. Results were expressed as percentage of control. To assess intraassay variability, we assayed frozen urine samples from the same adult control on 11 separate days. Percent cystine precipitation was 63% 10 (SEM). When cystine precipitation was assayed on fresh urine samples from the same individual on eight separate days, intra-assay variability was similar (67% 7 SEM).
Pharmacologic agents inhibiting cystine precipitation To analyze various potential inhibitors of urinary cystine precipitation, drugs were added to undialyzed aliquots of a single adult control urine containing radiolabelled cystine 4000 lmol/L. We compared the drugs at concentrations equivalent to urinary drug levels predicted from published pharmacokinetic data on each agent after standard doses. • D -Penicillamine (C4 H11 NO2 S): 42% of a standard 1 g intravenous dose is excreted in the urine over 24 h; 25–50% of a similar oral dose is excreted in the urine over 24 h [11,12]. Thus, for a standard oral dose of 1 g/day (for a 70-kg adult) we estimated that the mean urine penicillamine concentration would be 0.42 1 g in 1.5 L of urine/day. Mean urinary concentration ¼ 280 lg/ml (2 mM). • a-Mercaptopropionylglycine (MPG) (C5 H9 NO3 S): 25% of a standard 1 g daily oral dose (for a 70-kg adult) of MPG appears in the urine within 24 h [13]. Mean urinary MPG concentration ¼ 160 lg/ml (0.98 mM). • Captopril (C9 H15 NO3 S): 67% of a maximal 150 mg/ day (for a 70-kg adult) dose is excreted in the urine [14,15]. Mean urinary concentration ¼ 67lg/ml (0.3 mM). • Meso-2,3-dimercaptosuccinic acid (DMSA) (C4 H6 O4 S2 ): 18% of the standard oral dose of 30 mg/kg (used for lead poisoning in children) is excreted in the urine over 24 h [16,17]. Mean urinary DMSA concentration ¼ 250 lg/ml (1.3 mM). We also compared each drug at equimolar concentrations equivalent to the predicted molarity of D -penicillamine normally achieved in urine (2 mM) after standard oral doses. The ability of these drugs to inhibit endogenous cystine precipitation from urine of a Type I/I cystinuria patient was compared (at predicted urinary drug concentrations following standard oral doses adjusted for body weight). Additionally, the drugs were added to aliquots of urine from the same cystinuria patient and free urinary endogenous cystine levels were quantified on an automated amino acid analyzer.
Results Age-related factors influencing cystine precipitation While nephrolithiasis often begins in the first decade of life, children with cystinuria seem to be relatively protected in the early years. We wondered whether the risk of nephrolithiasis might be explained by changes in cystine concentration (lmol/L) or excretion level (lmol/ g creatinine) during this period. To examine this possibility, we studied the onset of stone formation in a group
P. Parvex et al. / Molecular Genetics and Metabolism 80 (2003) 419–425
of eight Type I/I cystinuria patients followed prospectively from birth. During the first 6 years of life, none of the patients developed nephrolithiasis detectable by annual ultrasonography or suggested by cli"nical symptoms. However, between 6 and 10 years, four children (50%) formed their first stone (mean age 9.3 years) (Table 1). The level of cystine excretion in children who formed a stone was not higher than in children without stones (Fig. 1A). Conceivably, infants take in larger fluid volumes than older children, achieving greater dilution of the filtered cystine. If so, then mean urinary cystine concentration might be lower in the first years of life, accounting for a lower risk of stone formation. However, the mean urinary cystine concentration was not significantly greater at age 6–10 (2465 lmol/L) than at age 0–6 year (2168 lmol/L)(NS) (Fig. 1B). In normal children, glomerular filtration rate (GFR) increases with age. We reasoned that older children might have to contend with an increased filtered load of cystine, excreting higher levels in the urine. However, as seen in Table 1, older (6–10 years) children actually excrete a lower level of cystine (2559 lmol/g creatinine) than young (0–6 years) children (4260 lmol/g creatinine) (p < 0:003) (Fig. 1C).
421
To determine whether cystine precipitates less readily from urine of younger versus older children, we measured in vitro precipitability (% CP) of 4 mM cystine in urine from normal children aged 0–6 (N: 13) versus 6–10 years (N: 13). Percentage CP was significantly higher in the older age group (94%) than in the younger children (76%) (p < 0:03) (Fig. 2).
Fig. 2. In vitro precipitation of cystine (4 mM) added to urine from non-cystinuric subjects: 0–6 years (76% 6.8); 6–10 years (94% 3.4). Cystine precipitation is significantly lower in the younger group (p < 0:03).
Table 1 Genotype, age at time of first stone and mean urine cystine concentration and excretion for young (0–6 years) and older (6–10) age groups Patient
1 2 3 4 5 6 7 8 Mean
Followed until age (years)
15 13 11 12 12 19 12 11
Genotype cystinuria
I/I I/I I/I I/I I/I I/I I/I I/If
Age at first stone
8.6 8.9 10 —
9.8 — — —
9.3
U [cystine] concentration (lmol/L)
U cystine excretion (lmol/g creat)
Age 0–6
Age 6–10
Age 0–6
Age 6–10
NA 3037 1870 1225 2593 3820 1475 3235
1842 2298 1845 2001 1967 2708 2565 2118
NA 2226 4369 4957 3325 4110 5103 5726
2301 1549 2750 2976 2517 1825 3256 3295
2465
2168
4260
2559
The mean age at onset of nephrolithiasis in 4 of these 8 patients was 9.3 years. NA, not available.
Fig. 1. (A) Comparison of mean of urinary cystine concentration (lmol/g creat) in stone-formers versus non-stone-formers. (B) Mean urinary cystine concentration (lmol/L) in Type I/I cystinuria patients (n ¼ 8) at age 0–6 years and 6–10 years. Urinary cystine concentration did not change with age (NS). (C) Mean urinary cystine excretion (lmol/g creat) in Type I/I cystinuria patients at age 0–6 years and 6–10 years. Cystine excretion was significantly higher in the first years of life (**p < 0:003).
422
P. Parvex et al. / Molecular Genetics and Metabolism 80 (2003) 419–425
Comparative effects of anti-cystinuric drugs on cystine precipitation in vitro In older cystinuric patients, when recurrent nephrolithiasis begins, conservative measures to prevent stone formation may fail; in this case, several agents (D -penicillamine, MPG, and captopril) have been used to convert urinary cystine to more soluble compounds. In general these thiol-containing reagents attack the internal disulfide bond of cystine, forming mixed disulfides,
which are more soluble in the urine (Table 2). We used our in vitro cystine precipitation assay to compare these monothiols with DMSA, a dithiol which as been shown to form a similar mixed disulfide with cystine [18–24] (Table 2). Drugs were added to our in vitro assay at concentrations corresponding to predicted urinary levels following standard oral doses of each agent. Fig. 3A indicates the effect of each agent on precipitation of 4 mM cystine from normal urine. DMSA (5% 1) and D -penicillamine (8% 3) were the strongest inhibitors of
Table 2 (A–C) Monothiols combine with half-cystine to form more soluble mixed disulfide with cysteine and (D) the dithiol, DMSA, combines with two halfcystine molecules to form a mixed disulfide
P. Parvex et al. / Molecular Genetics and Metabolism 80 (2003) 419–425
CP. MPG (43% 9) was of intermediate efficacy and captopril (63% 12) was a weak inhibitor of CP. The four drugs were also compared at concentrations equimolar to the predicted concentration of urinary D -penicillamine (2 mM) following a standard oral dose (Fig. 3B). D -penicillamine (11% 2) and DMSA (10% 5) were equally effective in decreasing CP. Even at an equimolar dose, captopril (53% 22) and MPG (40% 10) was less efficient than the former agents. Furthermore, the dose of captopril required to achieve concentrations of 2 mM in urine would be about six times the maximal tolerable dose and would produce unacceptable hypotension. Effects of anti-cystinuric drugs on endogenous urinary cystine To assess the effect of these drugs on endogenous urinary cystine, we added radiolabelled cystine to urine of a Type I/I cystinuria patient (urine cystine ¼ 2000 lmol/L). After acidification and cooling, CP was suppressed by DMSA (9% of control), D -penicillamine (12% of control), MPG (18% of control) and, to a lesser extent, by captopril (50% of control) (Fig. 4). We also tested the capacity of each drug to reduce endogenous free cystine (measured on an automated
423
Fig. 4. Comparative efficacy of various drugs on inhibition of endogenous cystine precipitation from urine of a cystinuric patient. The drugs were compared at achievable urinary concentration after standard oral doses adapted to the patientÕs weight.
amino acid analyzer) in urine from the same Type I/I cystinuria patient. At urinary concentrations predicted after standard oral doses, the free cystine peak (2337 lmol/g creatinine) was sharply reduced by addition of DMSA (to 1165 lmol/g creatinine) or D -penicillamine (to 1209 lmol/g creatinine). On the other hand, captopril and MPG had little effect on the free cystine peak (to 2257 and 2488 lmol/g creatinine, respectively) (Fig. 5).
Discussion
Fig. 3. (A) Comparison of efficacy of cystine precipitation inhibitors in a normal subject, at achievable urinary concentration for each drug after standard oral doses. (B) Comparison of efficacy of cystine inhibitors in normal urine at equimolar concentration (2 mM): DMSA and D -penicillamine are strong inhibitors whereas MPG and captopril are weaker inhibitors.
Factors leading to stone formation in cystinuria are poorly understood, but high urinary cystine concentrations predisposing to cystine precipitation at acid pH are certainly a prerequisite. Among patients from a specific genetic subtype of cystinuria, the frequency of nephrolithiasis does not correlate well with urinary cystine level. However, children with homozygous Type I/I cystinuria excrete higher levels of cystine and have more stones than children with the milder mixed form (Type I/non-Type I) of cystinuria [5,6]. Dent et al. showed that high fluid intake (to dilute urinary cystine) and urinary alkalinization modifies cystine precipitation in vitro and prevents nephrolithiasis in clinical practice [3,4]. Interestingly, our observations suggest that relative resistance to nephrolithiasis in very young cystinuria patients is not explained by agerelated changes in the filtered load of cystine or in urinary dilution. However, our in vitro studies suggest that cystine is less likely to precipitate from urine of very young (age 0–6 years) children than in older (age 6–10 years) children even though urine samples are dialyzed to standardize the concentration of salts and pH. One likely explanation is that infant urine contains especially high levels of non-dialyzable factors (>12 kDa), which inhibit cystine precipitation. When endogenous inhibitors of cystine precipitation begin to fail and recurrent nephrolithiasis begins in older chil-
424
P. Parvex et al. / Molecular Genetics and Metabolism 80 (2003) 419–425
Fig. 5. (A) Effect of various therapeutic agents on free cystine level (measured on an automated amino acid analyzer) in urine from a cystinuric patient. Normal cystine level 6 100 lmol/L. (B) Free cystine peak (FC) counted with an aminoacid analyzer in urine from a cystinuric patient after addition of different drugs. FCP is decrease by D -penicillamine and DMSA.
dren, exogenous agents must be considered to prevent cystine stone formation. Both D -penicillamine and MPG are clinically effective and are thought to prevent cystine stones by forming relatively soluble mixed disulfides with the cysteine moiety [1].
However, these agents frequently have significant side effects (fever, rash, proteinuria, hepatotoxicity, etc.) and there is a real need for agents with higher benefit/risk ratio. Captopril has been tried and decreased cystine/ creatinine ratio by 10%, but this minimal effect was not
P. Parvex et al. / Molecular Genetics and Metabolism 80 (2003) 419–425
clinically significant [19–21]. In our in vitro studies, D penicillamine was more effective than MPG and captopril in preventing cystine precipitation. Interestingly, DMSA was at least as effective or superior to penicillamine in our cystine precipitation assay. In addition, we observed that DMSA was as effective as D -penicillamine in reducing the endogenous free cystine peak in urine from a cystinuric subject (as measured on an automated amino acid analyzer). MPG is thought to be clinically efficacious in preventing cystine stone formation and was able to reduce cystine precipitation in our in vitro assay. However, MPG was somewhat less potent than either DMSA or D -penicillamine. We suggest, therefore, that MPG may be less effective than these latter agents in clinical practice, but comparative clinical trials have not been performed. It should be noted that measurement of free urinary cystine on an automated amino acid analyzer may be a useful way to adjust penicillamine or DMSA dose during treatment of cystinuria patients; free cystine levels of <1200 lmol/L would be a rational goal. However, under conditions used for sample preparation, an MPG–cystine complex was not evident and no effect on the free cystine peak was observed. This method is not useful in monitoring MPG therapy. Based on its performance in our in vitro assay, it is tempting to consider DMSA as a potential therapeutic agent for cystinuria. DMSA is widely used intravenously at low dose for nuclear scintigraphy in humans and was apparently safe at doses of 25–30 mg/kg/day during 1month oral treatment for lead poisoning in children [23,24]. Maiorino et al. [16] showed that DMSA (10 mg/ kg), given orally to 10 fasting adult subjects, appeared in the urine in disulfide linkage with L -cysteine. In monkeys (Macaca mulatta), 18% of orally absorbed DMSA is excreted in the urine (65% in the feces and 2% in expired air) [18]. These reports suggest that oral DMSA is rapidly excreted in urine at concentrations sufficient to inhibit cystine precipitation. Clinical efficacy and safety profile during longer treatment periods would certainly require careful evaluation, but the in vitro effects of DMSA on cystine precipitation suggest that it may be a promising new therapeutic agent for treatment of cystinuria.
References [1] M. Palacin, P. Goodyer, V. Nunes, P. Gasparini, Cystinuria, in: C. Scriver, A. Beaudet, W. Sly, D. Valle (Eds.), The Metabolic and Molecular Bases of Inherited Disease, vol. III, 2001, pp. 4909– 4932. [2] C. Pak, Etiology and treatment of urolithiasis, Am. J. Kidney Dis. 18 (1991) 624–637.
425
[3] P. Goodyer, M. Boutros, R. Rozen, The molecular basis of cystinuria: an update, Exp. Nephrol. 8 (2000) 123–127. [4] C.E. Dent, B. Senior, J.M. Walshe, The pathogenesis of cystinuria. Polarographic studies of the metabolism of sulfur-containing aminoacids, J. Clin. Invest. 33 (1954) 1216. [5] C.E. Dent, B. Senior, Studies on the treatment of cystinuria, Brit. J. Urol. 27 (1955) 332–377. [6] P. Goodyer, C. Clow, T. Reade, C. Girardin, Prospective analysis and classification of patients with cystinuria identified in a newborn screening program, J. Pediatr. 122 (1992) 568–572. [7] P. Goodyer, I. Sadie, P. On, G. Elks, R. Rozen, Cystinuria subtype and the risk of nephrolithiasis, Kidney Int. 54 (1998) 56–61. [8] G.K. Chow, S. Streem, Medical treatment of cystinuria: results of contemporary clinical practice, J. Urol. 156 (1996) 1576–1578. [9] J. Martens, T. Denneberg, A. Lindell, O. Textorius, Sulfur amino acid metabolism in cystinuria: a biochemical and clinical study of patients, Kidney Int. 37 (1989) 143–149. [10] E.T. Backer, A. van den Ende, W.P. Geus, Determination of urinary cystine for monitoring anticystinuric therapy, Clin. Chim. Acta 165 (1987) 351–358. [11] P. Neter, B. Bannwarth, P. Pere, A. Nicolas, Clinical pharmacokinetics of D -penicillamine, Clin. Pharmacokinet. 13 (1978) 317– 333. [12] T.F. Furlong, S. Posen, D -Penicillamine and the transport of L cystine by rat and human renal cortical brush-border membrane vesicles, Am. J. Physiol. (1990) F321–F327. [13] Y.C. Charles, C. Fuller, K.J.E. Sakhaee, V.A. Adams, Management of cystine nephrolithiasis with alpha-mercaptopropionylglycine, J. Urol. 136 (1986) 1003–1008. [14] S.H. Kubo, J.R. Cody, Clinical pharmacokinetics of angiotensin converting enzyme inhibitors: a review, Clin. Pharmacokinet. 10 (1985) 377–391. [15] M. Perazella, K.B. Gregory, Successful treatment of cystinuria with captopril, Am. J. Kidney Dis. 21 (1993) 504–507. [16] R. Maiorini, D. Bruce, V. Aposhian, Determination and metabolism of dithiol chelating agents, Toxicol. Appl. Pharmacol. 97 (1989) 338–349. [17] E. Liebelt, M. Shannon, W.J. Graef, Efficacy of oral meso-2,3dimercaptosuccinic acid therapy for low-level childhood plumbism, Pediatr. Pharmacol. Ther. 124 (1994) 313–317. [18] J.D. Cremin, M.L. Luck, N.K. Laughlin, D.R. Smith, Efficacy of succimer chelation for reducing brain lead in a primate model of human lead exposure, Toxicol. Appl. Pharmacol. 161 (1999) 283– 293. [19] J. Sloand, J.L. Izzo, Captopril reduces urinary cystine excretion in cystinuria, Arch. Intern. Med. 147 (1987) 1409–1412. [20] T.D. Cohen, S.B. Streem, P. Hall, Clinical effect of captopril on the formation and growth of cystine calculi, J. Urol. 154 (1994) 164–166. [21] S. Streem, P. Hall, Effect of captopril on urinary cystine excretion in homozygous cystinuria, J. Urol. 142 (1989) 1522–1524. [22] J.B. Besunder, D.M. Super, R. Anderson, Comparison of dimercaptosuccinic acid and calcium disodium ethylenediaminetetraacetic acid versus ethylenediaminetetraacetic acid in children with lead poisoning, J. Pediatr. 130 (1996) 966–971. [23] J.J. Chisolm, Safety and efficacy of meso-2,3-dimercaptosuccinic acid (DMSA) in children with elevated blood lead concentration, Clin. Toxicol. 38 (4) (2000) 365–375. [24] J.R. Walter, P.D. Dietrich, J.H. Ware, D.W. Dockery, M. Salganik, J. Radcliffe, R. Jones, B. Ragan, J.J. Chisolm, G.G. Rhodas, The effect of chelation therapy with succimer on neuropsychological development in children exposed to lead, N. Engl. J. Med. 344 (2001) 1421–1471.