Vol. 176, No. 3, 1991 BlOCHEMlCALANDBIOPHYSICALRESEARCH COMMUNICATIONS

May 15, 1991 Pages 1262-1268

A 31P-NNR STUDY ON THE ENERGY STATE OF RAT LIVER IN AN EXPERIMENTAL MODEL OF

CHRONIC DIETARY IRON OVERLOAD

Daniela Ceccarelli, Guerrino Predieri *, Umberto Muscatello and Albert0 Masini

Istituti di Patologia general@ e *Igiene dell'universitadi Modena,

41100 Modena, Italy

Received March 12, 1991

"P-NMR spectroscopy of rat liver perchloric acid extracts was utilized to assess the hepatic energy state in an experimental model of chronic dietary iron overload. Oral administration of iron for a period of 65 days that induces a steady ten-fold increase in hepatic iron concentration causes a significant

decrease in the hepatic ATP level not associated with appreciable modifications of ADP and Pi levels. The phosphorylation ratio appearson the average decreased.

The values of the energy state parameters revert to the normal if the concentration of iron in the liver is reversed below the critical level upon withdrawal of iron treatment after 45 days for a period of 20 days. The implication of these energy modifications for the pathogenesis of cell damage in the siderosis is discussed. Q 1991 Academic Press, Inc.

Excess iron deposition in the liver is a feature characteristic of several

disorders including inherited alterations of iron metabolism, thalassemia major,

sideroblastic anemia, alcoholic cirrhosis, African dietary iron overload,

porphyriacutaneatarda andtransfusional iron overload (1,2). Despite convincing

clinical evidence for liver injury as a result of excess iron, the biochemical

mechanisms of the hepatocellular damage are still poorly understood (1,2). The

most favoured hypothesis is that pathological accumulation of iron in the liver

initiates lipid peroxidation of various cellular organelles that in turn would

result in alterations of physical and enzymatic properties of the membrane

components and finally in cell damage (3,4). As a matter of fact, lipid

peroxidation has been found to occur in mitochondrial, microsomal, lysosomal and

plasma cell membranes of rat liver in experimental models of chronic iron

overload (5-10). However, some lines of evidence indicate that in these models

the peroxidation of lipid components does not induce per se gross physical

alterations of the membrane structure, but rather it triggers and/or amplifies

0006-291X/91 $1.50 Copyright 0 I991 by Academic Press, Inc. All rights of reproduction in any form reserved. 1262

Vol. 176, No. 3, 1991 BIOCHEMICALAND BIOPHYSICAL RESEARCH COMMUNICATIONS

a series of biochemical events that in a long term can alter the functioning of

the cell. Consistent with this conclusion is the observation that no changes

occur in the fluidity of the mitochondria and plasma membranes upon chronic iron

overload (11).

More recently, studies on mitochondria isolated from the liver of rats made

siderotic by oral administration of iron, have shown that the presence in the

hepatocyte of iron in a concentration exceeding a threshold value, causes a

disturbance in the mitochondrial energy state most probably related to the

formation of oxygen reactive species (5-9,12,13). These studies however were

performed in an experimental model, i.e. isolated mitochondria, that does not

allow to decide whether similar alteration of the energy metabolism does occur

also in the whole tissue in vivo.

The aim of the present study was to obtain information as to the energy

state of the liver tissue in the presence of dietary iron overload. The

parameters of the energy state were estimated by "P-NMR spectroscopy on freeze-

clamped liver. This technique allows the measurement of energy parameters in the

same sample, thus abolishing the internal variability that is unavoidable by the

use of other methods.

MATERIALS AND METHODS

Female Wistar albino rats (100-120 g body weight) were purchased from Nossan (Corezzano, Milano, Italy). Rats weremade siderotic by feeding a standard diet purchased from Dr. Piccioni (Brescia, Italy) supplemented with 2,5% (w/w) carbonyl iron.

Animals were killed by decapitation after an overnight starvation period. Liver mitochondria were prepared in 0.25 M sucrose according to a standard procedure (14). The protein content of the final mitochondrial suspension was determined by biuret method with bovine serum albumin as the standard.

Hepatic iron concentration was determined by an atomic absorption Perkin- Elmer spectrophotometer (mod. 306) as in Ref.(7)

Phosphorus containing compounds were determined by "P-NMR spectroscopy using a Varian XL-200 Fourier transform spectrometer operating at 80.98 MHz. Livers were freeze-clamped with aluminum clamps pre-cooled in liquid nitrogen from ether anesthetized animals, the liver still being perfused by the animal's blood supply. The intercurrent time between the break of organ perfusion and the rapid freezing was not longer than 5 sec. Frozen tissue was ground in liquid

nitrogen and homogenized with 3 volumes of 7% HClO, at 0"; after centrifugation at 1000 xg the extract was neutralized with 10% KOH and re-centrifuged. The

supernatants were passed through a column of Chelex 100 (Calbiochem 400 Mesh) to remove paramagnetic contributions of iron and alkaline earths (15) and then lyophilized. The dry samples were kept at -18'C and resuspended in D:O immediately before NMR analysis. The usual spectrometer conditions were: 60' flip angle, sweep-width 8000 Hz, line width 3 Hz, acquisition time 2 s, transient

1263

Vol. 176, No. 3, 1991 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

accumulation 2500: the spectra were 'H decoupled. Chemical shift data have been reported relatively to 85% inorganic orthophosphate (ppm = 0). Methylene

diphosphonic acid (MDP, 1 M in $0) was used as primary standard for measuring absolute concentrations of ATP, ADP and Pi. Peak assignment was performed by

comparison with a standard mixture containing a known amount of phosphorylated compounds; quantitation of ATP, ADP and P, was performed by interpolating the ratio between the peak areaof the phosphorus containing compound and the primary standard (MDP) with the calibration curve of the same compound.

The ATP synthesis of isolated mitochondria wasdetermined under respiratory State 3 conditions as follows: aliquots of 50 ul were collected at the indicated times and immediately added to 1 ml of boiling water. After 2 min they were rapidly cooled and the ATP content was determined by bioluminescence photometry, employing purified firefly luciferase (16). Luminescence was detected by an LKB 1250 luminometer.

The results represented the mean of 6-7 different experiments f S.D. The Student's t-test was used to compare group means.

Carbonyl iron was purchased from Fluka (Fluka AG, Bushs, CH-9479,

Switzerland).

RESULTS

Feeding rats a diet containing 2.5% (w/w) carbonyl iron results in marked

and progressive increase in the hepatic iron concentration which reaches a

maximum value at 40 days and then remains mainly constant for a further period

of 30 days examined (13).

Table I shows that the content of iron in the tissue after 65 days of

treatment is 1640 + 180 pg/g wet tissue, a value which is one order of magnitude

higher than the control value of 178 + 20 ug/g wet tissue. The withdrawal of the

iron supplemented diet after 45 days of treatment, for a period of 20 days,

results in a significant decrease of the hepatic iron concentration to a value

of 1170 t 170 ug Fe/g wet tissue.

TABLE I. Effect of iron treatment and withdrawal on the hepatic tissue iron concentration in an experimental model of chronic

dietary siderosis

Experimental conditions fig Fe/g wet tissue

Control

Fe (65 days)

Fe (45 days) withdrawn (20 days)

178 f 20

1640 ?: 180

1170 + 170'

yalues obtained from S-10 animals, are expressed as means ? S.D. P > 0.05 compared with the value of rats treated with Fe for 65 days. Iron

concentration was determined by atomic absorption spectroscopy as described in the Methods.

1264

Vol. 176, No. 3, 1991 BIOCHEMICALANDBIOPHYSICAL RESEARCH COMMUNICATIONS

The effect of iron treatment on the phosphorylated metabolites, as measured

by "P-NMR spectroscopy on freeze clamped livers, is presented in Table II. The

hepatic ATP concentration is significantly decreased after 65 days of iron

treatement in comparison to control. Indeed, the ATP value of 1.485 ? 0.084 of

control drops to 1,113 ?r 0.203 after 65 days of treatment with a 25% decrease.

It is worthy to be noted that the withdrawal of iron treatment for a period of

20 days completely restores the hepatic ATP content. On the contrary, no

appreciable modification in the hepatic ADP and Pi content in respect to the

control is observable following iron treatment, In the same Table, the

phosphorylation ratios of the livers from control, 65 days iron treated or 20

days standard diet fed after 45 days iron treated rats, as calculated from the

phosphorylated metabolites, are reported. The phosphorylation ratio of the

control rat livers is 0.488 + 0.144 mM-', a value in agreement with previous

reports (17,18). The iron treatment for 65 days brings about an average decrease

in the phosphorylation ratio of about 30% which is completely reverted to normal

values when the iron treatment is interrupted. However, this apparent

modification does not achieve statistical significance.

Analysis of "P-NMR spectra shows that the ratio phosphomonoesters (PME)

to Pi is significantly reduced after 65 days of iron treatment: in fact the ratio

decreases from 2.148 ? 0.187 to 1.802 + 0.106 (P < 0.005). The withdrawal of iron

from the diet induces little, if any reversal to normal values after 20 days. It

should be noted that the PME region arises from different monophosphorylated

TABLE II. Effect of Fe treatment and withdrawal on the concentration of phosphorylated metabolites and phosphorylation ratio in rat liver determined by "P-NMR

Eperimental Conditions

[ATPI [ADPI [Pi1 [ATPI 1Liml-m

Control

Fe (65 days)

Fe (45 days) withdrawn (20 days)

1.485 + .084 1.022 + .178 3.295 f .503 0.468 +- .144

1.113 ?: .202 1.058 f .232 3.532 f .384 0.328 f .162 :

1.451 k -212 1.027 + -247 3.606 2 -633 0.431 t .171

Concentrations are expressed in pmol/g intracellularwater. VaJues, obtained from 6-7 animals for each group of treatment, are expressed as means k S.D. P < 0,005. Phosphorylated metabolites analysis was performed by 3'P-NMR as described in the Methods.

1265

Vol. 176, No. 3, 1991 BIOCHEMICALAND BIOPHYSICALRESEARCH COMMUNICATIONS

C= 2.3

Fe =2.6

Fig. t Mitochondria (3 mg/ml) were incubated in a standard medium with pyruvate/malate as the respiratory substrate. The reaction was started by the addition of 0.33 mM ADP. At timed intervals indicated in the figure, samples of 50 1.11 were collected and the ATP content measured as described in the Methods.HControl; -Fe-treated.

metabolites including phosphorylated precursors of phospholipid biosynthesis,

sugar phosphates, AMP and glycolitic intermediates. Since these metabolites

cannot be resolved individually by the technique used, the interpretation of

their modification would be only matter of speculation.

Given the central role played by mitochondria in cellular energy

metabolism, the oxidative phosphorylation capability of isolated mitochondria

from 65 days iron treated rats has been measured. Fig. 1 shows that the rate of

ATP production as well as the phosphorylative efficiency, as indicated by the

measured ATP/O ratios, are not appreciably modified by the iron treatment.

DISCUSSION

The present "P-NMR study reveals at the organ level that concentrations

of iron higher than normal in the liver are associated with perturbation of the

energy state of the hepatic tissue. This energy perturbation is seen when the

iron concentration in the liver has reached a critical value (above 1500 ug Fe/g

wet tissue) which remains then unchanged all-over a long period before the onset

of structural damages. During this stage, the alteration of the energy state is

reversible if the concentration of iron falls down below the critical value.

The perturbation of the energy state of the hepatic tissue is revealed by

the use of "P-NMR spectroscopy on freeze-clamped liver, a method now widely used

for detection of the parameters relevant to assess the energy state (17). In the

experimental model studied, i.e. when the concentration of iron in the liver has

reached a maximal steady value, the level of ATP appears significantly decreased.

1266

Vol. 176, No. 3, 1991 BIOCHEMICALAND BIOPHYSICALRESEARCH COMMUNICATIONS

Furthermore, it was found that the phosphorylation ratio, which is characteristic

of the energy balance between ATP production and ATP consumption (19), is also

decreased in the experimental animals. All these parameters revert to normal

values when the concentration of iron in the liver decreases below the threshold

value.

The measurements of the phosphorylation ratio enable to calculate the

phosphoprylation potential, AGATp, by using the classical equations reported for

example by Chance et al. (19). The maximum hGATp calculated from these figures

appears to be -16.07 kCa1 in the controls and -15.86 kCa1 in the sjderotjc

animals. This implies that a minor aliquot of the energy released during

oxidation of NADH is incorporated in the ATP terminal phosphate bond in the case

of siderotic livers. By assuming that the oxidative phosphorylation mechanism is

unimpaired, an assumption suggested by evidence obtained on mitochondria isolated

from the liver of rats at the very same stage of siderosis (9,13) and here

confirmed (see Fig l),it can be concluded that a certain amount of energy

deriving from the oxidation of NADH is dissipated. This conclusion is consistent

with the results reported in the above mentioned studies on mitochondria isolated

from the siderotic livers. In fact it was found that the siderotic condition

induces a shift in the redox state of mitochondrial pyridine nucleotides toward

the oxidized form which is associated to the activation of a speficic CaC' efflux

route from mitochondria. Since the released Ca" is actively re-accumulated into

mitochondria, there is in fact a continous energy draining to maintain the re-

accumulation process (9,13). The in vivo activation of this energy dissipating

caCi cycle may account for the observation reported above that an aliquot of

energy released from the oxidation of NADH is dissipated.

These modifications of the energy state remain practically at the same

level all-over the period of treatment studied without detectable cell damages

and are completely reversible. This implies that at these levels of energy

perturbation, most of the energy-requiring processes are not hindered. However,

in view of the central role of the energy state for the living cell, it is

feasible that in a long term its perturbation renders the hepatocyte more

susceptible to metabolic damages in case of increased ATP requirements and\or

1267

Vol. 176, No. 3, 1991 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

impaired ATP production. These results thus suggest that the structural changes,

that occur in the hepatocyte in subsequent stages of siderosis, may be the result

not only of physical alterations of the membrane components due to lipid

peroxidation, but also of metabolic derangements.

ACKNOWLEDGMENTS

This work was supported by a grant from Minister0 dell'universita della Ricerca Scientifica e Tecnologica of Italy (40X), project "Patologia da Radicali Liberi e degli Equilibri Redox".

REFERENCES

1.

2.

3. 4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

15. 16.

17.

18.

19.

Jacobs, A. (1980) in Iron in Biochemistry and Medicine II (Jacobs, A. and Woorwood, M., eds.), pp 427-459, Academic Press, London. Powell, L.W. (1985) in Liver and Biliary Disease (Wright, R., Millward-

Sadler, G.H., Alberti, K.G.M.N. and Korran, S., eds.), pp 963-982, Bailliere Tindall Saunders, London. Bacon, B.R. and Britton, R.S. (1989) Chem. Biol. Interactions 70, 183-206. Fletcher, L.M., Roberts, F.D., Irving, M.G., Powell, L.W. and Halliday, J.W. (1989) Gastroenterology 97, 1011-1018. Bacon, B.R., Tavill, A.S., Brittenham, G.M., Park, C.H. and Recknagel, R.O. (1983) J. Clin. Invest. 71, 429-439. Bacon, B.R., Park, C.H., Brittenham, G.M., O'Neil, R. and Tavill, A.S.

(1985) Hepatology 8, 789-797. Masini, A., Ceccarelli-Stanzani, D., Trenti, T. and Ventura, E. (1984) Biochim. Biophys. Acta 802, 253-258.

Masini, A., Trenti, T., Ventura, E., Ceccarelli-Stanzani, D. and Muscatello, U. (1984) Biochem. Biophys. Res. Commun. 24, 462-469. Masini, A., Trenti, T., Ceccarelli-Stanzani, D. and Muscatello, U. (1986)

Ann. N.Y. Acad. Sci. 488, 517-519. Peters, T.J., O'Connel, M.J. and Ward, R.J. (1985) in Free Radicals in Liver Injury (Poli, G., Cheeseman, K.M., Dianzani, M.U. and Slater, T.F., eds.), pp 107-115, IRL Press, Oxford. Pietrangelo, A., Grandi, R., Tripodi, A., Tomasi, A., Ceccarelli, D., Ventura, E. and Masini, A. (1990) Biochem. Pharmacol. 39, 123-128. Masini, A., Trenti, T., Ceccarelli, D. and Muscatello, U. (1988) Biochem. Biophys. Res. Commun. 151, 320-326. Masini, A., Ceccarelli, D., Trenti, T., Corongiu, F.P. and Muscatello, U. (1989) Biochim. Biophys. Acta 1014, 133-140. Masini A., Ceccarelli-Stanzani D. and Muscatello U. (1984) FEBS Lett. 160, 137-140. Burt C.T., Glonec T. and Barany M. (1976) J. Biol. Chem. 251, 2584-2591. Masini, A., Ceccarelli, D. and Muscatello, U. (1983) Biochim. Biophys. Acta 724, 251-257. Iles, R.A., Stevens, A.N., Griffiths, J.R. and Morris, P.G. (1985) Biochem. J. 229, 141-151. Ling, M. and Brauer, M. (1990) Biochim. Biophys. Acta 1051, 151-158. Gyulai, L., Roth, Z., Leigh, J.S., and Chance, B. (1985) J. Biol. Chem. 260, 3947-3954.

1268