asymptomatic volunteers (1) the effects of acute copper ingestion on gastric and intestinal ..... drinking water, an investigation into possible health effects.
Digestive Diseases and Sciences, Vol. 46, No. 9 (September 2001), pp. 1909 –1914 (© 2001)
Effect of Acute Copper Exposure on Gastrointestinal Permeability in Healthy Volunteers MARTIN GOTTELAND, PhD, MAGDALENA ARAYA, MD, PhD, FERNANDO PIZARRO, MaSc, and MANUEL OLIVARES, MD
In vitro studies indicate that treating cells with copper results in alteration of tight junction permeability. In humans, ingestion of a single bolus of up to 10 mg Cu/L (as copper sulfate) causes nausea and vomiting in approximately 20 and 5% of the volunteers, respectively. To understand better the gastric and intestinal effects of copper, in this study we evaluated in asymptomatic volunteers (1) the effects of acute copper ingestion on gastric and intestinal permeability and (2) whether the appearance of gastrointestinal symptoms is associated with changes in mucosal permeability. Thirty-one asymptomatic subjects were assessed in a randomized, double-blind, crossover study that included two permeability tests, one after ingesting 200 ml distilled water and the other after ingesting 200 ml water containing 10 mg Cu/L (as copper sulfate). Fifteen minutes after ingestion subjects drank a second solution containing 40 g sucrose, 7.5 g lactulose, and 2 g mannitol, and urine was collected for 5 hr. Sugar concentrations were determined by gas chromatography. Symptoms during the trials were recorded in self-administered questionnaires. Ingestion of the 10 mg/L copper solution significantly increased gastric permeability to sucrose [20.8 (11.8 –73.4) vs 28.4 (16.6 –113.9) mg, respectively; P ⫽ 0.0064] but did not change intestinal permeability to lactulose/mannitol [0.87 (0.53–2.06) vs 1.17 (0.58 –2.39)%, respectively; P ⫽ 0.18]. Gastrointestinal symptoms were reported during both the basal and the experimental conditions, but after copper ingestion they increased to 22.6% of the subjects and were significantly more intense than under basal conditions (P ⫽ 0.047). However, changes in permeability were not related to the appearance of symptoms. These results indicate that acute oral exposure to 10 mg Cu/L exerts an effect on gastric but not intestinal mucosa, reducing the gastric mucosal barrier capacity, independently of the appearance of gastrointestinal symptoms. KEY WORDS: copper; gastric permeability; intestinal permeability; sucrose; mannitol; lactulose; emetic; human.
Ingestion of large amounts of copper salts induces toxic effects that range from nausea, vomiting, diarrhea, and abdominal pain at lower doses to systemic Manuscript received July 25, 2000; accepted December 18, 2000. From the Laboratory of Microminerals, Institute of Nutrition and Food Technology (INTA), University of Chile, Santiago, Chile. This work was supported in part by the International Copper Association and Bristol Myers/Squibb/Mead Johnson Unrestricted Grant Program. Address for reprint requests: Dr. Martin Gotteland, INTA, University of Chile, Macul 5540, Santiago, Chile.
organ failure, shock, and death at high doses (1–3). Because of its bitter metallic taste, copper ingestion in significant amounts is infrequent and most reports refer to accidental ingestion or deliberate suicide attempts (4, 5). In a clinical study on apparently healthy adult volunteers, we recently found that nausea and vomiting were the first and most frequent symptoms reported when fasted volunteers drank 200 ml of distilled water containing graded amounts of copper (as copper sulfate), ranging from 0 to 12 mg
Digestive Diseases and Sciences, Vol. 46, No. 9 (September 2001) 0163-2116/01/0900-1909$19.50/0 © 2001 Plenum Publishing Corporation
1909
GOTTELAND ET AL
Cu/L (6). In contrast, diarrhea and abdominal pain were rare within the concentration range tested. Except for emesis, other potential effects of copper on the gastrointestinal tract are far from clear (7). Copper seems to be a cytoprotector for gastric mucosa when it is bound to amino acids or peptides (8), but copper salts have also been shown to inhibit the glucosamine synthase activity involved in gastric mucus production (9), and consequently its presence might lead to decreased mucosal protection. The finding that acute copper ingestion may cause diarrhea has been generally interpreted as an effect on the intestinal mucosa (10). In accordance with these observations, it has recently been described that copper treatment alters the intestinal tight junction permeability of intestinal Caco-2 cells in vitro (11). However, it is not known whether such an effect may also be observed in the human gastrointestinal mucosa. Gastrointestinal permeability may be easily and noninvasively evaluated in both children and adults by using specific markers such as sucrose for gastric permeability and lactulose/mannitol for intestinal permeability. Its evaluation is widely used as an indicator of changes occurring in the human gastric and/or intestinal mucosa after a variety of stresses (12, 13). To shed light on the early effects of copper on the gastric and intestinal mucosa, in this study we assessed, in apparently healthy volunteers, the changes in gastrointestinal permeability after the ingestion of a single bolus of distilled-deionized water containing 10 mg Cu/L using the sucrose/lactulose/mannitol permeability test. We also evaluated whether the changes in permeability correlated with the appearance of gastrointestinal symptoms. SUBJECTS AND METHODS Subjects and Experimental Design. A randomized, double-blind, crossover protocol was used. Candidates to volunteer in the study answered a form that allowed the exclusion of subjects with gastrointestinal pathologies, renal disorders, diabetes, obesity, sucrose intolerance, or alcoholism and those who were on antiinflammatory drugs during the week prior to the study. Thirty-one asymptomatic subjects [15 males, 16 females; age, 32.0 ⫾ 9.7 years (range, 20 – 63 years)] participated in the protocol. The project was approved by the Committee of Ethics on Human Research, INTA, University of Chile. Subjects signed a written informed consent before the beginning of the study. After an overnight fast and after confirming that they were asymptomatic on that day, subjects were requested to drink 200 ml of deionized water containing either plain water or a 10 mg Cu/L (as copper sulfate) solution (i.e., a total dose of 2 mg of copper). Fifteen minutes after drink-
1910
ing this solution, subjects ingested 200 ml of distilled deionized water containing 40 g sucrose, 7.5 g lactulose, and 2 g mannitol. They were under direct supervision for a total of 5 hr 15 min, during which symptom appearance was recorded and urine was collected; then they left to assume their daily duties. Symptomatology. Appearance of symptoms was evaluated during the test and after 24 hr, using a chart including four outcomes (nausea, vomiting, diarrhea, abdominal pain) and seven “distracters”, i.e., unrelated symptoms used to distract the study subject from the outcomes (backache, feeling of well-being, headache, cough, dizziness, palpitations, and fatigue). Subjects recorded their perception of outcome intensity by grading them from 0 to 3 (absence of symptoms, mild, moderate, and severe, respectively). An index was calculated by adding the intensities recorded for each symptom. Increased bowel sounds and salivation, which were included in the list at the beginning of the study, were finally excluded because a long fasting period may influence the appearance of bowel sounds and the bitter metallic-taste of copper may induce salivation. Permeability Tests. Two gastrointestinal permeability tests were carried out at 2-week intervals, one after ingesting deionized distilled water (basal condition) and the other after ingesting copper-containing water (experimental condition). Urine was collected for 5 hr in a plastic container with 10 ml of 10% thymerosal, to avoid bacterial proliferation, and kept at 40°C. The volume passed was measured and aliquots were frozen at ⫺30°C until processed. Determination of sugar concentrations in urine samples was carried out as described previously (14, 15): control samples of urine with known amounts of added sucrose, lactulose, and mannitol were prepared and analyzed in parallel. Cellobiose and ␣-CH3-glucose (Sigma Chemical Co., St Louis, MO) were used as internal standards. Sugars were derivatized as described previously and dissolved in hexane. Twomicroliter samples were injected in the split mode at a column temperature of 200°C in a Varian 3600 gas chromatograph equipped with a split/splitless injector and a flame ionization detector (Varian Instruments, San Fernando, CA). Run-to-run variation of these measurements was ⬍10%. Results are expressed as the urinary excretion of each sugar (mg) and the lactulose/mannitol ratio. Statistical Analysis. Results are expressed as median ⫾ percentiles 10 –90. Statistical analyses were carried out by Wilcoxon matched-pairs test, Fisher’s exact evaluation, and Kruskal–Wallis analysis of variance.
RESULTS Urinary sucrose excretion during the basal and experimental conditions is shown in Figure 1: a significant increase of 36.5% in gastric permeability to sucrose was observed after drinking water containing 10 mg of copper [28.4 (16.6 –113.9) vs 20.8 (11.8 – 73.4) mg, respectively; P ⫽ 0.0064]. Copper ingestion did not change urinary lactulose excretion [12.5 (5.6 –24.6) vs 11.4 (5.2–20.3) mg, respectively; P ⫽ 0.64] and decreased, but not significantly, mannitol excretion [172.7 (102.4 –235.0) vs Digestive Diseases and Sciences, Vol. 46, No. 9 (September 2001)
ACUTE COPPER EXPOSURE AND GASTROINTESTINAL PERMEABILITY
Fig 1. Urinary excretion of sucrose as a marker of gastric permeability in healthy volunteers after ingestion of distilled water (control) or distilled water that contained 10 mg/L copper (copper). P ⫽ 0.0064, Wilcoxon matched-pairs test.
157.3 (75.1–286.7) mg, respectively; P ⫽ 0.096]. Consequently, as shown in Figure 2, intestinal permeability reflected by the lactulose/mannitol ratio was not significantly altered by copper ingestion in comparison with basal conditions [1.17 (0.58 –2.39) vs 0.87 (0.53–2.06)%, respectively; P ⫽ 0.18]. Table 1 shows the index of frequency/intensity of each symptom; taking into account the confounding symptoms, the appearance of outcomes was significantly higher after copper ingestion. Of 31 subjects who participated in the study, 10 (32.3%) presented at least one of four outcomes. No diarrhea was observed during the study, and abdominal pain was described by three subjects after copper ingestion, but also by three subjects in the basal condition. This suggests that this is not a good discriminating symptom in this type of study, probably due to the long fasting period (overnight plus 5.5 hr for the permeability test). Taking into account only vomiting and nausea, seven volunteers (22.6%) were symptomatic. Nausea was described by five volunteers during the copper test and one during the basal test. Two volunteers vomited during the first 15 min following copper ingestion. One of them also presented vomiting during the basal test; however, this event occurred 120 min after water ingestion and coincided with the Digestive Diseases and Sciences, Vol. 46, No. 9 (September 2001)
Fig 2. Urinary lactulose/mannitol ratio as a marker of intestinal permeability in healthy volunteers after ingestion of distilled water (control) or distilled water that contained 10 mg/L copper (copper). P ⫽ 0.18, Wilcoxon matched-pairs test.
appearance of intense headache, in a subject who reported suffering migraines. As shown in Table 2, no differences in urinary excretion of sucrose, lactulose, and mannitol or in lactulose/mannitol rate were observed between symptomatic and asymptomatic subjects. DISCUSSION Results of this study show that the integrity of gastrointestinal mucosa is indeed affected by the acute ingestion of 10 mg Cu/L (as copper sulfate). The effect on the stomach was clear, but it did not reach statistical significance in the intestine. Copper is an essential trace element implicated in the function of several proteins indispensable in cellular respiration, free radical defense, connective tissue synthesis, and others (16). On the other hand, copper also acts as a strong prooxidative compound (17) and may exert toxic effects at high amounts; however, at which doses the first adverse effects appear, and their scope, are unclear. Current recommended dietary intakes are based on few and incomplete studies, and the range of safe tolerable oral intake is still unclear (18). Therefore, to determine adequate recommendations it is important to establish the precise doses and effects of this metal in humans (19 –21).
1911
GOTTELAND ET AL TABLE 1. INDEX OF THE FREQUENCY/INTENSITY OF THE OUTCOMES AND DISTRACTING SYMPTOMS DURING THE PERMEABILITY TEST CARRIED OUT UNDER THE BASAL AND EXPERIMENTAL CONDITIONS (FISHER EXACT TEST, P ⫽ 0.047) Symptom Outcomes Nausea Vomiting Abdominal pain Diarrhea Total “Distracters” Backache Feeling of well-being Headache Cough Dizziness Palpitations Fatigue Total
Basal trial, water
Experimental trial, 10 mg Cu/L
3 (3)* 0 3 (3) 0 6 (6)
13 (6) 4 (2) 3 (3) 0 20 (7)
0 0 5 (3) 0 3 (2) 0 1 (1) 9 (4)
0 0 4 (4) 0 0 1 (1) 3 (3) 8 (6)
*Numbers in parentheses indicate the number of patients feeling the symptom.
The fate of copper ingested with water on an empty stomach is not well known. Preliminary data (unpublished) suggest that at the low gastric pH most copper salts are dissolved and therefore become readily available to interact with the mucin layers, both with degraded soluble mucins such as gastroferrin (naturally present in gastric juice) and with the adherent mucous layer that participates in defining the microenvironment for the epithelial cells (24). The binding rate to soluble mucins probably depends on the volume of ingested water and its copper concentration. Several models of gastrointestinal damage suggest that copper bound to amino acids displays a protec-
tive effect on the gastrointestinal mucosa (8); in this study copper was delivered in water and after a long fasting period. This probably favors copper dissolution in the stomach and avoids copper binding to dietary proteins, which would decrease its potential effect. In an interesting study, Ferruza et al. recently showed that incubation of Caco-2 intestinal cells with Cu– histidine did not affect the integrity of the cell monolayers (11). In contrast, luminal treatment of Caco-2 monolayers with 10 –100 M CuCl2 resulted in a time- and concentration-dependent alteration of the tight junction permeability, in association with a perturbation of the cellular cytoskeleton which was reversible after 24 hr in a medium without copper (11). There are obvious differences between the in vivo and the in vitro studies. However, it is tempting to speculate that in our study the conditions in the stomach may resemble those of the luminal treatment of Caco 2 cells. At the gastric level we observed an increase in sucrose permeability, which is thought to go across the gastric epithelium by the paracellular route (12). We would like to speculate that this effect of copper might occur early after water ingestion, before its binding to soluble mucins. On the other hand, alterations of the intestinal tight junction permeability by copper were not observed in our study, as indicated by the unchanged permeability to lactulose, a dissacharide absorbed by the paracellular pathway (13). Whether the presence of proteins in the intestinal lumen may be responsible for this lack of effect cannot be answered by the results of this study. It is intriguing that for mannitol there was a trend toward decreased permeability. This monosaccharide is
TABLE 2. URINARY EXCRETION (MG) OF SUCROSE, LACTULOSE, AND MANNITOL AND LACTULOSE/MANNITOL RATIO (%) IN ASYMPTOMATIC AND SYMPTOMATIC APPARENTLY HEALTHY VOLUNTEERS, UNDER BASAL CONDITIONS OR AFTER INGESTION OF A 10 mg Cu/L SOLUTION (AS COPPER SULFATE)* Urinary excretion of Sucrose Basal Copper Lactulose Basal Copper Mannitol Basal Copper L/M Basal Copper
Asymptomatic subjects (n ⫽ 24)
Symptomatic subjects (n ⫽ 7)
33.6 [10.2–73.4] 49.2 [11.3–131.0]
15.2 [11.8–73.5] 41.9 [17.7–73.5]
0.20
0.82
15.3 [7.5–22.4] 15.3 [4.7–27.9]
10.7 [3.7–20.3] 12.5 [8.4–14.6]
0.71
0.50
194 [66–370] 164 [99–270]
194 [109–280] 173 [102–235]
0.025
0.95
1.24 [0.67–2.39] 1.32 [0.58–2.39]
0.58 [0.42–1.40] 1.22 [0.49–2.60]
1.17
0.32
F
P
*Results are expressed as means [percentiles 10 –90] and are compared by Kruskall– Wallis ANOVA.
1912
Digestive Diseases and Sciences, Vol. 46, No. 9 (September 2001)
ACUTE COPPER EXPOSURE AND GASTROINTESTINAL PERMEABILITY
transported mainly by the transcellular pathway, and a decrease in mannitol permeability is usually interpreted as a diminution of the absorptive area (13). The proportion of individuals that presented gastrointestinal symptoms after ingesting the 10 mg/L copper solution (22.6%) is similar to those observed in previous studies (6, 26); also, as in these studies, only nausea and/or vomiting (without diarrhea) were observed at the concentration tested. The basic function of vomiting is the elimination of noxious substances from the digestive system before their intestinal absorption. The primary sensory system that detects their presence is constituted by chemoreceptors present in the gastrointestinal mucosa, the most sensitive region being the antropyloric area (26). The emesis phenomenon is regulated peripherally by 5-HT4 receptors localized in the gastrointestinal mucosa and centrally by 5-HT2, dopamine, and opiate receptors present in the central nervous system or in the blood– brain barrier, the vagus nerve being implicated in mediating the responses activated from the stomach (27). The second objective of this study was to evaluate whether the gastrointestinal permeability of symptomatic subjects differs from that of asymptomatic individuals. Results show that the presence of gastrointestinal symptomatology is not associated with alterations of gastric or intestinal permeability, suggesting that the mechanisms implicated in mucosal permeability alterations are independent of those implicated in nausea and vomiting. In conclusion, this study shows that in healthy volunteers acute exposure to 10 mg/L copper alters their gastric mucosal barrier function and that about 20% of them develop gastrointestinal symptomatology, mainly nausea and/or vomiting, but that these symptoms are not associated with specific changes in gastrointestinal permeability. REFERENCES 1. Eife R, Weiss M, Barros V, Sigmund B, Goriup U, Komb D, Wolf W, Kittel J, Schramel P, Reiter K: Chronic poisoning by copper in tap water: I. Copper intoxications with predominantly gastrointestinal symptoms. Eur J Med Res 4:219 –223, 1999 2. Eife R, Weiss M, Muller-Hocker M, Lang T, Barros V, Sigmund B, Thanner F, Welling P, Lange H, Wolf W, Rodeck B, Kittel J, Schramel P, Reiter K: Chronic poisoning by copper in tap water: II. Copper intoxications with predominantly systemic symptoms. Eur J Med Res 4:224 –228, 1999 3. Knobeloch L, Ziarnik M, Howard J, Theis B, Farmer D, Anderson H, Proctor M: Gastrointestinal upsets associated with ingestion of copper contaminated water. Environ Health Perspect 102:958 –961, 1994 4. Gill JS, Bhagar CI: Acute copper poisoning from drinking lime Digestive Diseases and Sciences, Vol. 46, No. 9 (September 2001)
5. 6.
7.
8.
9.
10. 11.
12.
13.
14.
15.
16. 17.
18.
19.
20.
21. 22.
cordial prepared and left overnight in an old urn. Med J Aust 170:510, 1999 Yelin G, Taff ML, Sadowski GE: Copper toxicity following massive ingestion of coins. Am J Forens Med Pathol 1:78 – 85, 1987 Olivares M, Araya M, Pizarro F, Uauy R: Nausea threshold in apparently healthy individuals who drink fluids containing graded concentrations of copper. Reg Toxicol Pharmacol (in press), 2001 Niijima A, jiang ZY, Daunton NG, Fox RA: Effect of copper sulphate on the rate of the afferent discharge in the gastric branch of the vagus nerve in the rat. Neurosci Lett 80:71–74, 1987 Franco I, Doria D: Prostaglandins and nitric oxide in coppercomplex mediated protection against ethanol-induced gastric damage. Pharmacol Res 36:395–399, 1997 Fujita T, Sakuma S, Takahashi K, Bohtani Y, Nishida H, Fujimoto Y: Inhibition of rabbit gastric glucosamine synthetase activity by Cu2⫹, Zn2⫹ and Se4⫹. Res Commun Mol Pathol Pharmacol 96:203–208, 1997 Stenhammar L: Diarrhoea following contamination of drinking water with copper. Eur J Med Res 4:217–218, 1999 Ferruza S, Scarino ML, Rotilio G, Ciriolo MR, Santaroni P, Muda AO, Sambuy Y: Copper treatment alters the permeability of tight junctions in cultured human intestinal Caco-2 cells. Am J Physiol 277:G1138 –G1148, 1999 Medding JB, Sutherland LR, Byles NI, Wallace JL: Sucrose. A novel permeability marker for gastrointestinal disease. Gastroenterology 104:1619 –1626, 1993 Hollander D: The intestinal permeability barrier. A hypothesis as to its regulation and involvement in Crohn’s disease. Scand J Gastroenterol 27:721–726, 1992 Vera JF, Gotteland M, Chavez E, Vial MT, Kakarieka E, Brunser O: Sucrose permeability in children with gastric damage and Helicobacter pylori infection. J Pediatr Gastroenterol Nutr 24:506 –511, 1997 Gotteland M, Cruchet S, Araya M, Espinoza J, Brunser O: Permeabilidad intestinal en el primer an ˜o de vida. Efecto de la diarrhea. An Esp Pediatr 49:125–128, 1998 Linder MC, Hazegh-Azam M: Copper biochemistry and molecular biology. Am J Clin Nutr 63:797S– 811S, 1996 Luza SC, Speisky H: Liver copper storage and transport during development. Implications for cytotoxicity. Am J Clin Nutr 63:812S– 820S, 1996 Olivares M, Araya M, Uauy R: Copper homeostasis in infant feeding: Genetic and environmental determinants. J Pediatr Gastroenterol Nutr (in press), 2000 Olivares M, Gonzalez M, Icaza G, Uauy R: Mechanism of adaptation to copper deficit and excess in humans: Basis to assess health risk. In Proceedings of the International Workshop “Risk Assessment of Copper in the Environment,” GE Lagos, R Badilla-Ohlbaum (eds). 1997, pp 291–326 Fewtrell L, Kay D, Jones F, Baker A, Mowat A: Copper in drinking water, an investigation into possible health effects. Public Health 110:175–177, 1996 Fitzgerald DJ: Safety guidelines for copper in water. Am J Clin Nutr 67 (Suppl 5):1098S–1102S, 1998 Lagos G, Reveco F: Human exposure model for copper in drinking water in the city of Santiago, Chile. In Proceedings of the International Workshop “Risk Assessment of Copper in the Environment,” GE Lagos, R Badilla-Ohlbaum (eds). 1997, pp 291–326
1913
GOTTELAND ET AL 23. Olivares M, Pizarro F, Speisky H, Lo ¨nnerdal B, Uauy R: Copper in infant nutrition: safety of WHO provisional guideline value for copper content of drinking water. J Pediatr Gastroenterol Nutr 26:251–257, 1998 24. Powell JJ, Jugdaohsingh R, Thompson RPH: The regulation of mineral absorption in the gastrointestinal tract. Proc Nutr Soc 58:147–153, 1999 25. Pizarro F, Olivares M, Uauy R, Contreras P, Rebelo A, Gidi V:
1914
Acute gastrointestinal effects of graded levels of copper in drinking water. Environ Health Perspect 107:117–121, 1999 26. Kayashima N, Tanaka M, Iwasaki M, Hayama T: Site of emetic action of oral copper sulfate in dogs. (I) Thresholds of various portions of gastrointestinal tract to locally applied copper sulfate. Jpn J Pharmacol 28:775–781, 1978 27. Lang IM: Noxious stimulation of emesis. Dig Dis Sci 44:58S– 63S, 1999
Digestive Diseases and Sciences, Vol. 46, No. 9 (September 2001)
