Tryptophan Metabolism in Rheumatic Disease By ROBERT S. PINALS Five tryptophan metabolites were measured in the urine after an oral tryptophan load in patients with rheumatoid arthritis, scleroderma, systemic lupus and a variety of other diseases. An excessive excretion of kynurenine and 3-hydroxykynurenine was a common but nonspecific occurrence in rheumatic disease. The excretion did not correlate well with various clinical parameters. Inhibition of pyridoxal-dependent hepatic enzymes may be the cause of the abnormality but the mechanism is unknown. Cinque metabolitos de tryptophano esseva mesurate in le urina post cargas oral de tryptophano administrate a patientes con arthritis rheumatoidee, scleroderma, lupus generalisate, e un varietate de altere morbos. Un excretion excessive de kynurenina e 3-hydroxykynurenina esseva un occurrentia commun sed non specific in casos de morbo rheumatic. Le excretion revelava nulle nette correlation con ulle del parametros clinic studiate. Inhibition de enzymas hepatic que depende de pyridoxal es possibilemente le causa del anormalitate mentionate, sed le mechanism0 non es clar. P REVIOUS STUDIES in patients with rheumatic diseases have demonstrated disturbances in the principal pathway of tryptophan metabolism (fig. 1).In scleroderma, Price et all found marked increases in excretion of the kynurenines and lesser increases in other metabolites. Excessive amounts of 3-hydroxyanthranilic acid, not measured by Price et al., were found to be excreted in rheumatoid arthritis by McMillan,' Spiera3 and Bett,4 who also demonstrated increased excretion of kynurenine. In the present study, five metabolites in the kynurenine pathway were measured after tryptophan loading in patients with rheumatoid arthritis and other rheumatic diseases, in patients with a variety of other illnesses and in normal controls. AND MATERIALS METHODS Studies were done on 45 patients with rheumatoid arthritis,5 five with systemic lupus, four with scleroderma, 16 healthy controls and 45 patients with a variety of other diseases including, among others, cancer, pneumonia, diabetes, myocardial infarction, cirrhosis, hepatitis, emphysema, dermatitis and orthopedic conditions. Both the group with rheumatoid arthritis and the group with other diseases had 22 males and 23 females and were of comparable mean age ( 5 2 and 50, respectively). The patients with scleroderma (mean age 52) and systemic lupus (mean age 39) were all females. The healthy controls (mean age 3 7 ) had an equal sex distribution. A loadinq dose of 2.5 Gm. of L-tryptophan was given to each subject and urine collected for a 24-hour period with toluene as a preservative. No attempt was made to regulate the diet. Salicylates were discontinued 2 days before the collection. Kynurenic acid ( KA ), xanthurenic acid ( XA ), kynurenine ( Kyn), 3- From the Department of Medicine, Massachusetts General Hospital and Harmrd M e d k d School, Boston, Mass. This work was supported by grants A-3564, 2A-5067, and AM-04501 from the United States Public Health Service. This is publication No. 368 of the Robert W. Lovett Memorial Unit for the Study of Diseases Causing Deformities. 662 ARTHRITISA N D RHEUMATISM,VOL. 7, No. 6 (DECEMBER), 1964 < TRYPTOPHAN METABOLISM IN RHEUMATIC DISEASE 663 ~ H P H INDOLEACETIC ACKI TRYPTOPHAN SEROTONIN I a fryptaphan pyrralase 0 s 0 3 H INDICAN FORMYL KYNURENINE OH ~ O C H 2 f H C O O H kynureninase ANTHRANILIC ACID \ fransaminose -r m* "'1 KYNURENINE KYNURENIC ACID OH I - H Y DROXYANTHRANILIC ACID 3-HYDROXYKYNURENINE X A i k E N I C PICID 0;;; - ocooHqCoNH 1 b O4 HOC COOH NICOTINIC ACID CH3 PYR I DONE ALIPHATIC ACIDS Fig. 1.-A diagram of tryptophan metabolism in abbreviated form. The main pathway, in a quantitative sense, is shown with heavy arrows. hydroxykynurenine ( HKyn) and 3-hydroxyanthranilic acid ( HAA ) were determined by the method of Coppini, Benassi and Montorsi.22 A sample containing 0.1 to 0.2 ml. of unmodified urine was dried on Whatman # 1 paper and two-dimensional chromatography performed, using butanol-acetic acid-water (4:1:5)in the first phase and water in the second. Areas containing the previously mentioned metabolites were located under ultraviolet light, cut out and eluted. Measurement was accomplished by reaction with Ehrlich's reagent (Kyn), diazotization ( HKyn, HAA, XA) and ultraviolet absorption (KA). Recovery of. standards added to urine was from 87 to 97 per cent (average 92 per cent) for the various compounds. Two papers were run for each sample and the two areas for a particular compound eluted together when quantities were small. RESULTS The range of values for 24-hour excretion of various tryptophan metabolites after a loading dose is shown in' figures 2-6. Mean values are shown in table 1. A wide range is apparent in all groups. Serial experiments on the same individual in several instances yielded similar results. Thus it would appear that each individual has a characteristic response to tryptophan load- 664 ROBERT S. PINALS .. $>loo- % k 100- 90- lu h . .. . .... .... 0. 80- 0.. 70- h 5 60- $ 50- $ 40- i 30- 9 h 20- % 3 I .. ........ .......... ."..... ..... ........... ........... ...... ....... 100' .---* NORMAL RHEU MATOlD ARTHRITIS SCLERODERMA SYSTEMIC LUPUS OTHER DISEASES Fig. 2.-Excretion of 3-hydroxykynurenine (HKyn) in mg./24 hr. after a tryptophan load in various disease groups. Dots at the bottom of each column in this and subsequent figures represent failure to detect the compound. For 3-hydroxykynurenine, the lower limit of detectability varies from 3 to 6 mg./24 hr., depending upon the total urine volume and volume used in chromatography, -100 " - 52 '"901 0, 80- $ 70- $ g 5040- 302010- .... . ... .. .. ..... . .. . .... " 60- 3 3 . " . .. 00 NORMAL .. .... .. . I , WII RHEUMATOID ARTHRITIS " SCLERODERMA SYSTEMIC LUPUS ..Hw. n OTHER DISEASES Fig. 3.-Excretion of kynurenine (Kyn) in mg./24 hr. after a tryptophan load in various disease groups. Lower limit of detectability 2 to 4 mg./24 hr. ing, not greatly influenced by ordinary variations ini diet and activity. Only two healthy controls, both females, had HKyn spots; in one of these there was a family history of arthritis. One was studied a t several points during the menstrual cycle. Only small variations were seen, similar to those reported previously7; highest excretion of Kyn and HKyn occurred premenstrually, with little change in the other metabolites. 665 TRYPTOPHAN METABOLISM IN RHEUMATIC DISEASE . . m . . a8 m n ' .. NORMAL RHEUMATOID ARTHRITIS jCLERODERMA 0 ... . SYSTEMIC LUPUS Fig. 4.-Excretion of kynurenic acid (KA) in mg./24 hr. after a tryptophan load in various disease groups. Lower limit of detectability 1.5 to 3 mg./24 hr. . .. .. n .. "- 0' - NORMAL ! i f RHEUMATOID ARTHRITIS SCLERODERMA SYSTEMIC LUPUS -.. .. U I . . u OTHER DISEASES Fig. 5.-Excretion of 3-hydroxyanthranilic acid (HAA) in mg./24 hr. after a tryptophan load in various disease groups, Lower limit of detectability 0.8 to 1.6 mg./24 hr. The rheumatoid arthritis group had a significantly increased excretion over the normal group only for HKyn and Kyn ( p < .01). Although the mean HAA excretion was not significantly elevated, 24 per cent of the patients with rheumatoid arthritis had greater excretion than any of the normals. Included in the rheumatoid group were six patients with rheumatoid spondylitis; excretion of Kyn and HKyn was high in two and low in four. TWOpatients with psoriasis and arthritis had high values; one with psoriasis alone had low values. An attempt was made to correlate clinical findings with ex- ROBERT S. PINALS .. ____ * . . . .... .... ..... -.. .... ... ."..... ".. 0.. ..... NORMAL RHEUMATOID ARTHRITIS . .. ..... .... ...... " .......... SC LERODERMA SYSTEMIC LUPUS OTHER DISEASES Fig. 6.-Excretion of xanthurenic acid ( X A ) in mg./24 hr. after a tryptophan load in various disease groups. Lower limit of detectability 0.5 to 1mg./24 hr. cretion of each metabolite. As in other studies,3J no significant relationship was found for most parameters, including age, duration of disease, latex fixation test, presence of nodules, previous treatment and disease activity in terms of joidt swelling and tenderness. However, there was some correlation between HKyn-Kyn excretion and sedimentation rate ( Rourke-Emstene); those with values below 0.60 mm./min. had a lower HKyn ( p < .01) and Kyn ( p < .05) excretion. In scleroderma, excretions tended to be higher than id rheumatoid arthritis. The patients with systemic lupus all had chronic low-grade disease activity and, therefore, were not representative of the illness in general. The relatively low values for Kyn and high values for HAA are difficult to explain and conclusions from the study of such a small group are best avoided. The group with other diseases differed significantly from the rheumatoid arthritis group only in having lower HKyn excretion ( p < . O l ) . Among those with high HKyn values were two patients each with recent myocardial infarcTable 1.-Mean Normal Rheumatoid arthritis Scleroderma Systemic lupus Other diseases Values (mg./24 hr.) for Excretion of Tryptophan Metabolites in Various Groups HKyn Kyn XA KA HAA 2.63 31.2' 85.2' 46.4' 9.72 5.64 33.2* 87.2" 8.70 23.9 4.00 5.04 4.80 5.41 3.03 13.0 14.8 28.5f 7.40 14.1 3.07 5.80 8.52" 10.2* 4.82 uSienificantly elevated over normal at p < .01. +Significantly elevated over normal at p < .05. Undetectable quantities were considered as zero for this calculation. TRYPTOPHAN METABOLISM IN RHEUMATIC DISEASE 667 tion and carcinoma, and one each with megaloblastic anemia, acute leukemia, pneumococcal pneumonia and sarcoma of bone. The latter had the highest excretions of HKyn, Kyn, KA and XA in all groups studied. DISCUSSION The present study confirms the findings of Price et a1.l in scleroderma and indicates that a similar abnormality in the kynurenine pathway exists in rheumatoid arthritis. Bett found excessive Kyn excretion in rheumatoid arthritis4s8and could reverse the abnormality, as could Price, by the administration of pyridoxine. We have also found this in preliminary studies. However, the pattern of urinary metabolites differs from that seen in pyridoxine deficiency or antagonism with desoxypyridoxine in that XA excretion is not significantly e l e ~ a t e dThe . ~ observations suggest that one or more pyridoxaldependent hepatic enzymes, most particularly kynureninase, may be partially inhibited in certain disease states. The elevated output of HAA in rheumatoid a r t h r i t i ~ may ~ - ~ have a different basis. No data on the output of this metabolite after pyridoxine administration are available. Bett4 found that the difference in excretion of HAA between the control and rheumatoid groups diminished after tryptophan loading, in contrast to Kyn where loadidg greatly exaggerated the difference. This would make enzyme inhibition in the pathway at a point beyond HAA an unlikely explanation. Spiera3 has speculated about mechanisms for increased HAA excretion. Increased diversion of tryptophan into the knurenine pathway, perhaps through accelerated protein turnover, is a reasonable possibility. Derangements in this pathway may be found in non-rheumatic conditions as shown' by the present study and many others. Various hematologic and malignant diseases are at times associated with increased excretion of one or more metabolites.9-l3 Abnormalities have also been reported in schizophrenia,14J5porphyria,15 hepatitis,17 asthma,lRgastrointestinal disorders18 and in some apparently healthy individuals with a family history of one or another i l l n e s ~ . *Little ~ ~ ~ information is available on the relationship of the biochemical abnormality to the clinical state, in terms of remissions and exacerbations in individual patients and relative severity of disease in different patients. Reversal with pyridoxine has been variable. Noteworthy in most studies involving tryptophan loading has been the wide scattering of values for excretion of metabolites, both in healthy and diseased groups. This may be a reflection of numerous actual or potential variable factors such as absorption, renal clearance, contribution of endogenous protein turnover to the tryptophan pool and relative activities of the several enzymes participatidg in the intermediary metabolism of tryptophan. These latter may be influenced in turn by the availability of pyridoxal and possibly other members of the vitamin B complex, by hormonal factors and by inhibitors postulated in various disease states. Known hormonal influences include the induction of tryptophan pyrrolase by corticosteroids,20 the idhibition of kynurenine transaminase by estrogens21 and the alteration of tryptophan metabolism in pregnancy,6 simulating pyridoxine deficiency but 668 ROBERT S . PINALS inkcompletely reversed by pyrid~xine.~ The wide range of metabolite excretion might hide an existing relationship to disease activity or severity. Perhaps only through serial studies on the same individual during different phases of his illness could such correlations be demonstrated. The significance of abnormal tryptophan metabolism in rheumatic and other diseases is an unsettled matter. An identical defect in many disparate ilhesses would suggest that the phenomenon is a secondary one, probably not etiologic. However, present knowledge cannot justify such a conclusion. In view of the complexity of tryptophan metabolism, alternative explanation&might be found for any abnormal pattern of response to a loading dose. Further understanding of the mechanisms involved and observation of the clinical course of individual patients in relation to the biochemical abnormality and its reversal with pyridoxine would be of value. ACKNOWLEDGMENTS The author thanks Dr. Hugo Muench for his aid in statistical anlysis and Dr. Stephen Krane for his advice and guidance. REFERENCES supplementation on the urinary ex1. Price, J. M., Brown, R. R., Rukavina, J. G., Mendelson, C., and Johnson, S. A. M.: Scleroderma (acrosclerosis), 11. Tryptophan metabolism before and during treatment by chelation (EDTA). J. Invest. Dermat. 29:289, 1957. 2. McMillan, M.: The identification of a fluorescent reducing substance in the urine of patients with rheumatoid arthritis. The excretion of 3-hydroxyanthranilic acid in this and other conditions. J. Clin. Path. 13:140, 1960. 3. Spiera, H.: Excretion of a tryptophan metabolite in rheumatoid arthritis. Arth. & Rheumat. 6:364, 1963. 4. Bett, I. M.: Metabolism of tryptophan in rheumatoid arthritis. Ann. Rheumat. Dis. 21:63, 1962. 5. Ropes, M. W., Bennett, G. A., Cobb, S., Jacox, R., and Jessar, R. A.: 1958 Revision of diagnostic criteria for rheumatoid arthritis. Arth. & Rheumat. 2:16, 1959. 6. Wachstein, M., and Lobel, S.: Abnormal tryptophan metabolites in human pregnancy and their relation to deranged vitamin B6 metabolism. Proc. SOC. Exper. Biol. & Med. 86:624, 1954. 7. Brown. R. R., Thornton, M. T., and Price, J . M.: The effect of vitamin cretion of tryptophan metabolites in pregnant women. J. Clin. Invest. 40: 617, 1961. 8. Bett, I. M.: Effect of pyridoxine on tryptophan metabolism in rheumatoid arthritis. Ann. Rheumat. Dis. 21:388, 1962. 9. Price, J. M.: Disorders of tryptophan metabolism. Univ. Michigan M. Bnll. 24:461, 1958. 10. Marver, H. S.: Studies on tryptophan metabolism. I. Urinary tryptophan metabolites in hypoplastic anemias and other hematologic dismders. J. Lab. & Clin. Med. 58:425, 1961. 11. Leppanen, V. V. E., and Oka, M.: Metabolism of tryptophan in cancer of various sites. Ann. med. exper. et biol. Fenniae 41: 123, 1963. 12. Musajo, L., Benassi, C. A,, and Parpajola, A.: Excretion and isolation of kynurenine and 3-hydroxykynurenine from human pathologic urine. Clin. chim. acta 1:229, 1956. 13. Boyland, E., and Williams, D. C.: The metabolism of tryptophan. I. The metabolism of tryptophan in patients suffering from cancer of the bladder. Biochem. J. 64:578, 1956. 14. Benassi, C . A., Benassi, P., Allegri, G., and Ballarin, P.: Tryptophan metabolisin in schizophrenic patients. J. ~~ TRYPTOPHAN METABOLISM IN RHEUMATIC DISEASE Neurochem. 7:264, 1961. 15. Price, J. M., Brown, R. R., and Peters, H. A.: Tryptophan metabolism in porphyria, schizophrenia, and a variety of neurologic and psychiatric diseases. Neurology 9:456, 1959. 16. Benassi, C. A., Perissinotto, B., and Allegri, G.: The metabolism of tryptophan in patients with bladder cancer and other urological diseases. Clin. chim. acta 8:822, 1963. 17. Quagliarello, E., Tancredi, F., Saccone, C., and Piazza, M.: Interrelation between tryptophan and nicotinic acid in human viral hepatitis. Nature, London 194:976, 1962. 18. Knapp, A.: Tryptophanbelastung und Vitamin-B6-Mangel. I. Die Ausscheidung von Xanthurensaure, Kynurenin, Trigonellinamid und 5-Hydroxydolessigsaure bei verschiedenen Erkrankungen. Dentsche Gesundh. 16:941, 1961. 19. -: 669 Tryptophanbelastung und VitaminB6-Mangel. 111. Hereditare Faktoren fur den Ausfall des Tryptophanbelastungs tests. Deutsche Gesundh. 16: 1041, 1961. 20. Knox, W. E.: Two mechanisms which increase in vivo the liver tryptophan peroxidase activity: Specific enzyme adaptation and stimulation of the pituitary-adrenal system. Brit. J. Exper. Path. 32:462, 1951. 21. Mason, M., and Gullekson, E. H.: Estrogen-enzyme interactions: Inhibition and protection of kynurenine transaminase by the sulfate esters of diethylstilbesterol, estradiol, and estrone. J. Biol. Chem. 235: 1312, 1960. 22. Coppini, D., Benassi, C. A., and Montorsi, M.: Quantitative determination of tryptophan metabolites (via Kynurenine) in biologic fluids. Clin. Chem. 5:391, 1959. Robert S . Pinab, M.D., formerly Post-doctoral Fellow, United States Public Health Service; presently Director, Arthritis Unit, Lemuel Shuttuck Hospital, Boston, Mass.