Rapid31P NMR test of liver function

MAGNETIC RESONANCE IN MEDICINE 8,220-223 (1988) zyxw zyxw zyxwvut COMMUNICATIONS Rapid 31PNMR Test of Liver Function WILLIAM J. THOMA* AND a M I L UGURBIL zyxwv zyxwvut zyxwvut zyxwvuts zyxwvuts Departments of Biochemistry and Radiology and the Gray Freshwater Biological Institute, University of Minnesota, P.O. Box 100, Navarre. Minnesota 55392 Received January 19, 1988; revised July 14, 1988 The "P NMR spectrum of perfused rat liver was found to be dependent on the exogenous carbon availableto the tissue. When pyruvate was supplied to liver initially perfused with lactate, Pidecreased,phosphoenol pyruvate and phosphomonoesters increased,and nucieotide pools remained the same. It is proposed that these changes can be used to evaluate liver function. o 1988 Academic prw,Inc. zyxwvu Successful use of nuclear magnetic resonance spectroscopyin diagnoses and noninvasive evaluation of organ competence requires the existence of easily recognizable differences between the NMR spectra of normal and pathological tissues. While dramatic changes due to the appearance of new resonances have been noted in 31PNMR spectra of tumors ( 1 , 2) and the decline of high-energy phosphate compounds in ischemic tissue is well documented, it has become increasingly clear that the type and the steady-state concentrations of numerous NMR detectable compounds may remain unaltered in diseased tissue. Consequently, protocols which induce metabolic stress have been employed in conjunction with NMR measurement of detectable rametabolites such as the phosphocreatine (PCr) to inorganic orthophosphate (Pi) tios in muscle ( 3 , 4 ) . Protocols proposed for evaluation of liver function by NMR include response to fructose (5- 7), the measurement of ratios of various metabolites, and metabolism of 13C-labeledprecursors such as pyruvate or glucose. In this communication, we present data that show that the 31PNMR spectrum of liver is strongly dependent on the carbon substrate presented to this tissue and suggest that the metabolism of pyruvate as studied by 3LP NMR may provide a rapid means of evaluating liver function. Figure 1 shows the effect of 5 mMpyruvate on the isolated perfused rat liver. Livers were prepared as described elsewhere (8, 9 ) . After the preparation had been established using lactate as the carbon source 170 ml of perfusate containing 5 m M pyruvate was presented to the liver ( N = 4) .l Spectrum (a) illustrates the 146.1-MHz 31P spectrum of the liver while receiving lactate. The spectrum is the result of 200 FIDs * To whom correspondenceshould be addressed at Department of Radiology,University of Iowa Hospitals, University of Iowa, Iowa City, Iowa 52242. ' Nrepresentsthe number ofdifferent livers from different animals. zyxwvutsrqp 0740-3194/88 $3.00 Copyright Q 1988 by Academic Press, Inc. All rights of reproduction in any form reserved. 220 COMMUNICATIONS zyxwv zy 22 1 b zyx zyxwvu zyxw zyxwvut zyxwvutsrqp a I I 5 0 I1 -5 1 -10 - F - 7 - -15 -20 PPM FIG. I. The effect of 5 m M pyruvate on the 3'P NMR spectra of isolated perfused rat liver. The spectra were the result of the summation of 200 FIDs recorded at 146.I MHz resulting from 80' pulses recorded every 0.8 s. Spectra were treated with a 12-Hz exponential filter before Fourier transformation. Spectrum (a) shows the perfused liver with lactate as the carbon source. The changes in the 3'P spectrum caused by exposure to 170 ml of 5 m M pyruvate in a recirculating perfusate circuit can be observed in spectra ( b ), (c), (d), and (e). These spectra were acquired in 2.3-min blocks; the spectra displayed were taken at 7min intervals. Within 30 min of exposure to pyruvate, the spectra return to near baseline metabolite levels. 222 zyxwvuts zyxwv zyxw zyxwvu zyxwv COMMUNICATIONS accumulated with 80" pulses with an 0.8-s delay between pulses. Under these conditions there was partial saturation of inorganic phosphate, monophosphodiesters, and phosphodiesters; therefore, the acquisition parameters may not have allowed full sensitivity to metabolic changes. The broad resonance of the liver was suppressed as described previously (8).The addition of 5 m A4 pyruvate to the perfusate caused the spectral changes observed in spectra (b), (c), (d), and (e) taken at 7-min intervals. In response to pyruvate exposure, there was an increase in liver phosphomonoester, as well as phosphoenol pyruvate (PEP), and a decrease in Pi.The time course of metabolic change was highly reproducible: standard error of the peak intensities was less than f5% ( N = 4). Thirty minutes after the initiation of pyruvate exposure, the spectrum returned to the prepyruvate state. Enzymatic analysis of the perfusate showed increased glucose levels subsequent to the addition of pyruvate. The decrease in the Pi levels while maintaining constant nucleotide levels noted upon addition of pyruvate to liver perfusate is consistent with similar changes observed in perfused cardiac muscle in response to analogous manipulations of the exogenous carbon substrate ( 10-12). In cardiac muscle, this observation was ascribed to activation ofpyruvate dehydrogenase and consequent changes in steady-state levels of TCA cycle intermediates and high-energy phosphates ( 11, 12). Unless exogenous pyruvate is supplied to the tissue, carbon substrate entry into the TCA cycle may be a rate-limiting step in mitochondria1 respiration. Pyruvate activates pyruvate dehydrogenase ( 13);consequently it is a substrate that is rapidly utilized by mitochondria. Given this property and the alterations caused in the 31PNMR spectra, exposure to pyruvate can serve as a test of mitochondrial competence in liver as well as cardiac muscle. This test has the following advantages over the others proposed: (1) unlike fructose exposure, pyruvate does not cause the deamination of liver adenine nucleotides and subsequent decrease in the nucleotide pool ( 14); ( 2 ) the use of 31Pinstead of 13Callows a quicker test due to better sensitivity of the nuclei and avoids any of the problems associated with proton decoupling ( 1 5 ) ;and ( 3 ) the metabolism of pyruvate is not as complex as glucose. Notwithstanding the question of the glucose paradox ( 1 6 ) ,glucose metabolism is primarily divided among giycolysis, pentose shunt, and glycogen synthesis. Pyruvate, on the other hand, is directly utilized by mitochondria and is converted to glucose by gluconeogenesis. The time course of the metabolism of pyruvate to phosphoenol pyruvate and ultimately to glucose is dependent on the state of the mitochondria. Therefore, this time course could be used to evaluate liver function. For example, we have noted that the clearance of 5 mMpyruvate from the perfused rat liver after 30 min of warm ischemia and subsequent recovery was depressed compared to the nonischemic organ ( N = 2). In summary, it is concluded that the intensities of phosphorylated metabolites observed in liver 31PNMR spectra are strongly dependent on exogenous carbon substrate and that this property can be employed as a simple test of tissue bioenergetics and function. zyxwv zyxwvuts zyxwvuts zyxwvu ACKNOWLEDGMENTS This work was supported by NIH Grants HL33600, HL32427, and 1K04HL01241. zyxwvu zyxwvutsrqponm zyxwvutsrq zyxwvuts zyxwv zyxw zyxwvuts zyxwvuts zyxwvu COM MU NlCATlONS 223 REFERENCES I . R.A . G m t l A M , R. A. MEYER,B. S. SZWERC~LD, ANDT.R. BROWN,J. B i d . (‘hem. 262.35 (1987). 2. J. M. MARIS,A. E. EVANS,A. c. MCLAUGHLIN, G . J. D’ANGIO,L. BOLINGER. H. MANOS, A N D B. CHANCE, N. E~ig/.J. Mcd. 312, 1500(1985). 3. G. RADDA,Science 233,640 ( 1986). 4. B. CHANCE, B. J. CLARKE, S. NIOKA,H. SUBRAMANIAN, J. M. MARIS,Z. ARGOV, A N D M. BODE, Cardio/ogy72(SupplI V ) , I03 (1985). 5 . R. D. OBERHAENSLI, G. J. GALI.OWAY, D. J. TAYLOR. P. J. BORE,AND G. K. RADDA,Brit. J. Radio/. 59,695 (1986). 6 . A. R. GRIVEGNEE. C. SEGEBARTH. P.R. INYTEN, A N D J. A. DEN HOLLANDFR.Radiology P 165,67 (1987).[Abstract No. 1331 7. P. VOCK, J. COTTING, R.LAREBECK. F. TERRIER. J. REICHEN,A N D D. HENTSCHEL,Radiology P 165, 346(1987).[Abstract No. 11561 8. W. J. TIIOMA, L. M. HENDERSON,AND K. UGURBIL. J. Mugn. Reson. 61, 141 (1985). 9. W. J. THOMA AND K. UGURRIL, Biochim. Biophys. Actu893,225 (1987). 10. K. UGURBIL, M. PETEIN, R. MAIDAN.s. MICHURSKI,A N D A. H. L. FROM,Biuchi.mistry 25, 100 (1 986). 11. A. H. L. FROM,M. A. PETEIN. s. MICHURSKI. s. D. ZIMMER.A N D K. UGURBll.. I.%IBSLert.206,257 ( 1986). 12. K. IJGUKBII., P. B. ~ N G S L E Y-HICKMAN,E. Y . SAKO,S. ZIMMER. P. MOHANAKRISIiNAN, P.M. L. ROBITAII.LE,W. J . THOMA, A. JOHNSON.J. E. FOKER. A N D A. H. L. FROM, Pruc. N . Y. Acad. Sci. 508,265 (1987). 13. s. c. DENNIS,A. PADMA. M. s. DEBUYSFRE,AND M. s.OLSON. Rioc/iernistry 17. 1252 (1978). 14. P.H.MAENPAA, K. 0.RA~VO, A N D M. P. KEKOMAKI. Science 161,1253 (1968). 15. M. AVISON, H. P. HEI.HERINGTON. AND R. G . SHULMAN, Annu. Rev. Biuphjs. Chem. 15,377(1986). 16. J . KArz, M. KlJWAJIMA, D. W. FOSTER,A N D J. D. MCGARRY,TIBS I I, 136 (1986).