THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 248, No. 2, Issue of January 25, pp. 610-618.1973

Printed in U.S.A.

Transport of Qrnithine and Citrulline across the Mitochondrial Membrane*

(Received for publication, May 26, 1972)

JAMES G. GAMBLES AND ALBERT L. LEHNINGER

From the Department of Physiological Chemistry, The Johns Hopkins University School of Medicine, Baltimore, Maryland 21,905

SUMMARY

Assay of ornithine transcarbamylase and carbamyl phos- phate synthetase I in submitochondrial fractions of rat liver mitochondria shows they are located primarily in the mito- chondrial matrix. Therefore during the urea cycle ornithine must pass from the cytoplasm, where it is formed, to the mitochondrial matrix, where it is carbamylated. The result- ing citrulline must then pass from the matrix to the cytoplasm before it can be converted to arginine. Osmotic swelling tests show that nonrespiring rat liver mitochondria do not allow the ornithine+ cation, the ornithine zwitterion, or N-ace- tylornithine to pass into the matrix, regardless of the presence of permeant anions, uncoupling agents, or valinomycin plus K+. However, when state 4 respiration is instituted with succinate as substrate, ornithine+ readily enters the mito- chondria if a permeant proton-yielding anion such as phos- phate, acetate, or bicarbonate is present. However, the permeant anions nitrate and thiocyanate, which pass the membrane without carrying protons, do not support ornithine+ entry. The respiration-energized unidirectional entry of [W]ornithine+ is inhibited by respiratory inhibitors, by uncoupling agents, and by valinomycin plus K+, but not by oligomycin. ADP also inhibits ornithine+ entry, pre- sumably by competing for respiratory energy. The driving force for entry of ornithine+ is concluded to be a negative- inside transmembrane potential produced when proton-con- ducting anions enter mitochondria to relieve the alkaline- inside pH gradient generated by electron transport. It is postulated that the L-ornithine+ cation is transported by a specific electrogenic uniport carrier. This view is supported by (a) the apparent specificity of the system, which transports ornithine+ but not the closely related arginine+‘or lysine+, (b) the stereospecificity for the L stereoisomer of ornithine+, and (c) the tissue specificity of ornithine+ transport, which occurs in liver mitochondria but not in those from heart, which cannot synthesize urea. The influx and efliux of citrulline in rat liver mitochondria does not depend on res- piratory energy or the presence of permeant anions or cations. Because penetration of citrufine occurs into liver but not

* This work was supported by Grant GM-05919 from the United States Public Health Service and by Grant GB-20519 from the National Science Foundation.

$ Recipient of a postdoctoral fellowship from the United States Public Health Service.

into heart mitochondria, it is postulated that citrulline also passes the membrane on a carrier, one specific for certain neutral amino acids.

The Krebs-Hens&it urea cycle is among those metabolic pathways requiring the participation of both mitochondrial and cytoplasmic enzymes. In mammalian liver the carbamylation of ornithine to citrulline, which is catalyzed by ornithine trans- carbamylase, takes place exclusively in the mitochondria (l-4)) as does the formation of carbamyl phosphate from ammonia, cata- lyzed by carbamyl phosphate synthetase I (5-7). The other steps of the urea cycle take place in the cytosol (8).

This paper reports experiments showing that both ornithine transcarbamylase and carbamyl phosphate synthetase I are located in the inner or matrix compartment of rat liver mito- chondria. The characteristics of the transport processes by which ornithine enters and citrulline leaves the mitochondrial matrix have been examined. Ornithine passes t.hrough the membrane by a unidirectional process dependent upon electron transport and requiring proton-carrying anions. Citrulline passes through the membrane in either direction, without requir- ing respiratory energy. Our observations strongly indicate that each of these processes is promoted by a substrate-specific trans- port system (for reviews see References 9-11) in the inner mem- brane. Some of the findings described in this paper have been briefly communicated (12).

EXPERIMENTAL DETAILS

Preparation of Mitochondria and Subnliiochondrial Fraciions- Mitochondria were isolated from livers of Sprague-Dawley albino rats by the procedure of Schneider (13) and were washed three times in cold 0.25 M sucrose. Rat heart mitochondria were pre- pared following Nagarse treatment (14). Protein was deter- mined by ultraviolet absorption (15, 16) ; crystallized bovine serum albumin was used as standard. Xfitochondria were sub- fractionated into an outer membrane plus soluble fraction and an inner membrane plus matrix (mitoplast) fraction by the digitonin procedure described by Schnaitman and Greenawalt (17). The mitoplast fraction was further separated into an inner membrane fraction (Lubrol-insoluble) and a matrix fraction (Lubrol-solu- ble), according to the method of Chan et al. (18).

610

by guest on April 21, 2020

http://ww

w.jbc.org/

Dow

nloaded from

611

Enxyme dssays-Carbamyl phosphate synthetase I (EC 2.7.25) was assayed by a modification of the procedure of Kerson and Appel (19). Bicarbonate labeled with l4C was converted in the presence of NH,+ and N-acetylglutamate into carbamyl phosphate, which was t,hen reacted with citrulline in a coupled assay with excess ornithine transcarbamylase (EC 2.1.3.3). The resulting acid-stable, nonvolatile radioactivity was deter- mined by liquid scinOillation counting. The assay was linear with time and protein concentration under the conditions em- ployed. Glut.amate dehydrogenase (EC 1.4.1.2) was assayed according to Beaufay et al. (20), malic dehydrogenase (EC 1.1.1.37) and monoamine oxidase (EC 1.4.3.4) according to Schnaitman el al. (21), and ornithine transcarbamylase by the method of Schimke (22). All membranous fractions were treated with 0.2 mg of Lubrol WX per mg of mitochondrial pro- tein prior to enzyme assays.

From these data it is clear that in the operation of the urea cycle ornithine generated by the arginase reaction in the cytosol must pass through t,he outer and inner mitochondrial membranes into the mat,rix before it can undergo transcarbamylation from carbamyl phosphate formed in the mitochondrial matrix by carbamyl phosphate synthetase I. The resulting citrulline must then pass from the mitochondrial matrix through the two mito- chondrial membranes to the cytosol, where the subsequent steps of the urea cycle take place. Throughout the remainder of this paper t.he reasonable assumption is made that the outer mem- brane is permeable to both these metabolites, as is true for most small solute molecules, and that the inner membrane is the true permeability barrier (9, 10).

Osmotic Swelling-The general method of Chappell and Crofts (23) was employed. l\Iitochondrial volume changes were ob- served by recording t’he change in absorbance of mitochondrial suspensions at 700 nm and 25” in isosmotic (250 mosM) solutions of various solutes, supplemented with 1.0 mM Tris buffer (pH 7.4), 0.5 IIIM EGTA’ to suppress changes due to movements of endogenous Ca2+ and Mgz+, and rotenone (0.5 PM) to suppress endogenous respiration. Swelling due to entry of solute and the accompanying water results in an increase in the light trans- mitted, which was recorded on strip charts with a Gilford-Beck- man DU spectrophotometer. The various ornithine salts used in these experiments were prepared by ion exchange on DEAE- Sephadex A-25 anion exchanger (Pharmacia Chemicals, Pis- cataway, N. J.).

Uptake of [‘4C]Ornifkine-Mitochondria were recovered from the test medium on 0.45 p Millipore filters. The filters were immediately dried under a lamp and counted in a toluene-based scintillation fluid. Results obtained by the filtration procedure were in good agreement with those obtained by centrifugation of the suspension followed by counting of the supernatant me- dium and pellet.

Reagents-n- and L-Ornithine, N-acetylornithine, L-lysine, and lactic dehydrogenase were obtained from Sigma Chemical Co., St. Louis, MO., malic dehydrogenase from Worthington Bio- chemicals, Freehold, S. J., sodium [14C]bicarbonate (11.4 mCi per mmole) from New England Nuclear, Boston, Mass., and nL-[1-‘4Clornithine (3i mCi per mmole) from Amersham-Searle, Arlington Heights, Illinois.

Nonpenetration of Respiration-inhibited Mitochondria by Or- nithine Salts-Since ornithine has an isoelectric pH of about 9.7, it is largely present as a cation at pH 7.4. Therefore, net movement of ornithine+ into the matrix compartment of non- respiring mitochondria can occur only if electroneutrality of the latter can be maintained. The three sets of circumstances which would meet this condition are (a) if a permeant anion enters the matrix with ornithine+ in stoichiometric symport, (b) if the ornithine+ cation exchanges with some matrix cation, such as H+ or K+, in stoichiometric antiport, of (c) if ornithine+ is first converted enzymatically into an electroneutral species. To test these possible modes of entry rat liver mitochondria were suspended in isosmotic solutions of ornithine salts (250 mosM) containing rotenone to inhibit endogenous respiration and the requirements for osmotic swelling due to entry of or- nithine examined. It was found that respiration-inhibited rat liver mitochondria completely failed to swell when suspended in isosmotic solutions of ornithine chloride (anion impermeant) or ornithine nitrate, acetate, or phosphate (anions permeant) (Table I). Thus the ornithine+ cation fails to enter respiration- inhibited mitochondria, resembling in this respect the failure of K+ and Na2+ to penetrate the membrane of nonrespiring rat liver mitochondria in the absence of ionophores (23). The failure of the mitochondria to swell in an ornithine acetate me- dium (Table I) further indicates that the neutral ewitterionic

PERCENT RECOVERY 0 I

IO 2p ;o 4,” 5,o y ;o “p 90 I ‘9”

OUTER MEMBRANE

INTER-

M YEENE ORNITHINE TRANSCARBAMYLASE (OTC)

RESULTS

Intramitochondrial Location of Ornithine Transcarbamylase and Carbamyl Phosphate Synthetase I-Although these enzymes have long been known to reside exclusively in the mitochondria of liver tissue (l-8)) their specific intramitochondrial location has not been reported. Data in Fig. 1 show their distribution in subfractions of rat liver mitochondria, compared with the dis- tribution of t,he marker enzymes malate dehydrogenase (matrix), monoamine oxidase (outer membrane), and cytochrome oxidase (inner membrane). Both ornithine transcarbamylase and car- bamyl phosphate synthetase I (see Reference 8) are primarily localized in the mat#rir, since their distribution follows closely that of the matrix markers glutamate and malate dehy- drogenases.

INNER MEMB+ANE

MATRIX

INNER MEMBRANE FRACTION

I CYTOCHROME OXIDASE 1 PSI

OTC

MATRIX FRACTION

1 The abbreviations used are: EGTA, ethylene glycol bis(@ aminoethyl ether)-iV,A\7’-tetraacetic acid; FCCP, carbonyl cya- nide p-trifluoromethoxyphenylhydrazone; NEM, N-ethylmale- . imide.

FIG. 1. Distribution of ornithine transcarbamylase and car- bamyl phosphate synthetase I in subfractions of rat liver mito- chondria compared with the distribution of characteristic marker enzymes. The per cent recoveries were calculated from assays of the unfractionated mitochondria. The specific activity of ornithine transcarbamylase in unfractionated mitochondria was 5,250 nmoles of ornithine transformed per mg of protein per min.

by guest on April 21, 2020

http://ww

w.jbc.org/

Dow

nloaded from

612

TABLE I Factors a$ecting entry oj ornithine salts into mitochondria

The basic system contained mitochondria (1.0 mg of protein) suspended in 2.0 ml of isosmotic ornithine salts as indicated below, 1.0 ITIM Tris-chloride buffer, pH 7.4, 0.5 &M rotenone t,o suppress endogenous respiration, and 0.5 mM EGTA in a total volume of 2.0 ml. Potassium succinate was added where shown at 5.0 mat. The swelling tests were run at 25”. The data given represent the initial rate of volume increase (first minute). NEM was added at 30 PM, gramicidin at 5 pM, and FCCP at 5 j.&M.

Anion

Experiment 1 Chloride. _ . . Nitrate. . . . . Nitrate. Nitrate. Acetate. Acetate........ Acetate........ Phosphate. .

Experiment 2 Phosphate. Phosphate. Acetate........ Acetate.. , . Chloride. . Chloride. Nitrate Nitrat.e Thiocyanate. Thiocyanate. .

Experiment 3 Phosphate. Phosphate. . . Acetate........ Acetate........

Experiment 4 Phosphate. , Phosphate. . . Phosphate.. , Phosphate.. . Phosphate.. . Phosphate. .

Experiment 5 Phosphate. .

Phosphate.

Phosphate. .

Respiratory substrate

None None None None None None None None

None Succinate None Succinate None Succinate None Succinat.e None Succinate

Succinate Succinate Succinate Succinate

Succinate Succinate Succinate Succinate Succinate Succinat.e

Succinate

Succinate

Succinate

Additions Rate of

swelling

!- -AAiao/min

None None Gramicidin FCCP None Gramicidin Gramicidin None

0.00 0.00

0.00

0.00

0.01 0.01 0.00 0.00

None None None None None None None None None None

0.02 0.35 0.02 0.42 0.00 0.00 0.00 0.00 0.02 0.01

None 0.41 NEM 0.02 None 0.46 NEM 0.46

(L-Ornithine) (n-Ornithine) (A-Acetylornit,hine) (L-Lysine) (L-Arginine) (Choline)

0.19 0.02 0.01 0.02 0.01 0.00

(L-Ornithine, rat liver)

(L-Ornithine, rat heart)

(L-Ornithine, beef heart)

0.23

0.02

0.01

species of ornithino does not enter mit.ochondria, in contrast to the ueutral NH, molecule, which can enter readily (23). Mi- tochondria suspended in 250 InOSM N-acetylornithine also failed to swell, an observation which excludes the possibility t,hat the ornithiuef cation is first acetyla,ted to an electrically neutral species by an enzymatic reaction in the cytosol to facilitate its transport.

The impermeability of respiration-inhibited mitochondria to ornithine salts is most strikingly shown by the experiment in Fig. 2, which demonstrates that the absorbance of suspensions of rat liver mit,ochondria is a linear function of the osmola,rity of ornithine chloride in the range of 50 to 250 mosq in a rela- tionship identical with that given in media of KCI, which has

A,, 0.6

0.4

0.2

NH40Ac _ o-o-o-o-o

50 150 250 mOsM

FLG. 2. Osmometric behavior of respirat,ion-inhibited mito- chondria suspended in ornithine chloride (Orn+Cl-). The mit.o- chondria (l.dmg of prot,ein) were suspended in 2.0 ml of t,he indi- cakd salt, solution cont.ainine 1.0 mM Tris-chloride. nH 7.4. 0.5 mM lj:(:TA, and 0.5 PM rot,enc&. Controls contained eyuivaleut concentrations of KC1 and ammonium :tcet)al.e. The absorbance was measured 5 min after mixing (25”).

been ea.rlier established to be impermcitnt, (23). Since the ab- sorbance is inversely related to mitochondrial volume under such conditions (23), it is clear that the rat liver mitochondria behave as nea,rly perfect osmomet,ers when suspended in or- nithine chloride solutions.

Addition of the ionophores gramicidin or valinomycin to res- piration-inhibited mitochondria suspended in isosmotic ornithine media failed to evoke swelling (Table I), whether the anion was impermeant (chloride) or permeant (nitrate). Moreover, no swelling was observed in such systems adjusted to pH 8.3, at which the membrane is permeable to chloride and other anions (24-26). In appropriate control experiments addition of the ionophores to mitochondria suspended in KC1 at pH 8.3 or in KNOs (27) evoked immediate swelling. It was concluded t,hat neither gramicidin nor valinomycin promote the ent,ry of the ornithine cation.

When the protonconducting uncoupling agent FCCP was added to respiration-inhibited rnitochondria suspended in isosmo- Iar ornithine salts, no swelling took placr (Table 1). In control experiment,s, addition of FCCP did cause swelling of mito- chondria suspended in NaNO 3, in confirmation of earlier coli- elusions (27, 28) that rat liver mitochoudrin contain a N&II+ cschange or antiport system, whose activity can be obxervrd in the presence of FCCP to allow exterual protons to enter mito- chondlla to replace internal protons exchanged for external Na+. The failure of FCCP to stimulate swelling in ornithine nitrate thus indicates that rat liver mitochondria do not contain a sys- tem which promotes ornithinc+-H+ a&port,.

Penetration of Ornithinef into Respiring Mitochondria--The

effect of induction of respiration on ornithine entry was then examined. When succinate was added to mitochondria sus- pended in ornithine chloride, no significant swelling ensued. However, when succinate was added to mitochondria suspended in isosmotic ornithine media containing the anions acetate or phosphate, rapid swelling of the mit,ochondria took place (Fig. 3).

Not all permeant anions will support respiration-dependent entry of ornithine+. Although phosphat’e and acetate support

by guest on April 21, 2020

http://ww

w.jbc.org/

Dow

nloaded from

ornithine+ entry, the permeant anions NOs- and SCN- do not (Table I). This observation is of considerable importance (see “Discussion”), since phosphate and acetate are transported across the membrane as the protonated species HzP04- and HOAc, respectively, whereas t.he nitrate and thiocyanate anions pass the membrane as such (23). As will be seen below, bicarbonate also can support entry of ornithine+. However, induction of respiration by succinate failed to cause N-acetylornithine to enter.

The experiments in Fig. 3 also show that the respiratory in- hibitors antimycin A and cyanide inhibit succinate-induced swelling iti ornithine phosphate medium. The uncoupling agent FCCP also completely prevented respiration-induced swelling in ornithine acetate. Oligomycin had no effect, indicating that the coupled formation of ATP was not a requirement for swelling. Respiratory substrates other than succinate also supported or- nithine+ uptake, as will be shown below. Moreover, endogenous respiration of rat liver mitochondria, which is NAD dependent, can also support swelling in ornithine phosphate media when rotenone is omitted.

Direct gravimetric measurements showed that the decrease in light absorption of respiring mitochondria suspended in or- nithine phosphate is act.ually due to uptake of water. Increases in mitochondrial volume up to 2.5.fold were observed before apparent lysis occurred.

Participation of Phosphate Carrier in Respiration-dependent

Penetration of Ornithine Phosphate-Phosphate enters the matrix compartment of rat liver mitochondria in a specific carrier-me- diated process, probably through H,P04--OH- antiport (23). ‘This carrier is inhibited by mersalyl (29-31), p-chloromercuri- brnzoate (32), formaldehyde (33), and NE;11 (32). Table I sl~ov~s data on the effect of NEM on the succinatc-dependent

Orn+CI-

Succinate

4

Orn*NO; 4 I

OrcSCN- +

A?00

\, \ \ \ 0.15 1 ~ \ Oligomycin

\

~ \ \

lmin \ \

FIG. 3. Requirement of energy-coupled respiration and phos- phat.e or acetate for ornithine+ entry. Mitochondria (1.0 mg of prot.ein) were suspended in 2.0 ml of isosmot,ic salt media as shown, supplemented with Tris buffer, pH 7.4, EGTA, and rotenone. Other additions were potassium succinate (5 mM), antimycin A (1.25~~1), FCCP (0.5~~). oligomycin (~.O,UM), and cyanide (1 mM).

613

swelling of rat liver mitochondria in ornithine phosphate and ornithine acetate media. NEM blocked swelling in ornithine phosphate but had no inhibitory effect on energized swelling in the ornithine acetate medium. These results therefore show that entry of ornithine+ into respiring mitochondria requires entry of the phosphate anion via the NEM-sensitive phosphate carrier. However, the entry of acetate, which is believed to enter by simple nonmediated diffusion of free acetic acid, is in- sensitive to NEM.

Isotopic Measuremenis of Entry of Orn&hine+--Data in Table II show that W-labeled ornithine+ enters rat liver mitoehondria when t,hey are supplemented with succinate and phosphate. Omission of either succinate or phosphate reduced the uptake of isotope to very low levels. Uptake of isotope was also com- pletely inhibited by FCCP, antimycin A, and cyanide, but not by oligomycin. The results obtained with radioactive ornithine thus fully agree with those observed with the osmotic swelling method.

Data in Table II also show that valinomycin in the presence of K+ completely inhibits ornithine uptake, in agreement with

Tnnrz II

Rcspirution-coupled uplalce of [YY]ornithine

The test system for Experiments I to 3 consist,ed of 210 rnM mannitol, 70 MM sucrose, 5 mM potassium phosphate, pH 7.4, 10 mM succinate, 1 mn Tris-chloride, pH 7.4, rat liver mitochondria (1.0 mg), and 5 rnM [14C]ornithine chloride, incubated at 37”. The uptake data were corrected for ornithine in the nonmatrix water of the pellets. The time of incubation was 15 min in Experi- ments 1 and 4 and 20 min in Experiments 2,3, and 5.

System

Experiment 1 Complete. ..................................

Phosphate omitted ........................ Succinate omitted ......................... Roth omitted .............................

Complete + FCCP (1 ELM). .................. Complete + antimycin A (1.5 ,uM). ..........

Complete + KCN (1.0 rnM). ................ Experiment 2

Complete (phosphate, 5 rniw) ................ Complete (acetate, 30 mM). ................. Complete (nitrate, 5 mM) ................... Complete (thiocyanate, 5 mM). .............. Complete (bicarbonate, 2 mM). ..............

Experiment 3 Complete ................................... Complete + oligomycin (1 pg per ml). ...... Complete + valinomycin (0.3 pg) + KC1 (10

rnM) ......................................

Complete + CaClz (1 mM). ................. Experiment 4

Complete (succinate) ....................... Complete (g1utamat.e). ...................... Complete (a-ketoglutarate). ................. Complete (p-hydroxybutyrate). ............. Complete (pyruvate + malate). ............. Complete (pyruvate + malate + FCCP). ....

Experiment 5 Complete (succinate). ....................... Comp1et.e + 0.25 mM ADP. ................. Comp1et.e + 1.0 mM ADP. .................. Complete + 3.0 rnnl ADP. ..................

Ornithine uptaLe

nmoles/ng protein

73.4 12.3 12.1

6.0 4.8

11.8 5.7

47.9 47.2 12.7 lo.s 51 .o

47.9 55.0

3.6 5.6

11.6 18.4

9.0 12.2 12.0

2.0

68.0 25.1

7.0 4.9

by guest on April 21, 2020

http://ww

w.jbc.org/

Dow

nloaded from

614

many observations that this ionophore imposes a drain on res- The initial rate of ornithine entry at low initial concentrations piratory energy, making it unavailable for oxidative phosphoryla- of ornithine is difficult to measure precisely because of the rela- tion and other respiration-dependent activities of mitochondria. tively long times required to filter or centrifuge the mitochondria

Succinate, /3-hydroxybutyrate, and pyruvate plus malate were in relation to the initial rate of uptake. However, data in Fig.

about equally effective as substrates in supporting ornithine 5 suggest that the initial rate of entry of ornithine+ in the first uptake; cY-ketoglutarate was somewhat less effective (Table II). minute approaches a saturation level as t,he ornithme concentra- Glutamate was significantly more active than other substrates. tion is increased. The concentration of ornithine+ yielding This observation may be accounted for by the fact that con- approximately half-maximal initial rates of isotope uptake ap-

version of ornithine to citrulline in the mitochondrial matrix pears to be less than 500 pM, which may be compared with the required, in addition to ATP, sources of ammonia and bicar- concentration of ornithine in the cell water of rat liver, estimated bonate, bot.h of which can be supplied by oxidative degradation to be in the range of 133 to 484 pM (35). On increasing the ex- of glutamate. However, addition of ammonium chloride to the ternal ornithine concentration to much higher levels (5 to 10 mM) system did not stimulate succinate-supported ornithine uptake. and the incubation time to 15 to 20 min, the saturation effect

It is also seen (Table II) that the permeant anions bicarbonate was lost. Under the latter conditions, which lead to extensive and acetate can replace phosphate in supporting ornithine up- swelling of the mitochondria, some of the ornithine was evidently take; however, the permeant anions nitrate and thiocyanate entering by unmediated concentration-dependent diffusion; pre- are relatively inactive, in agreement with the osmotic swelling sumably stretching of the membrane leads to a generalized in- data shown above and with earlier experiments on support of crease in its permeability. respiration-dependent Ca2+ uptake by proton-yielding anions Efect of ADP and Ca2+ on Uptake of Omithine-Data in Table

(34). II show that respiration-dependent [V]ornithine upta,ke sup-

Rate and Extent of Ornithine Uptake-Fig. 4 shows that the ported by oxidation of succinate is severely inhibited by ADP. energized uptake of [W]ornithine+ is linear with time up to Approximately half-maximal inhibition was given by as little 20 min at 37”. The observed rate of uptake, starting from 5 as 0.2 mM ADP. Ca2+ (1.0 mu) iuhibited ornithine uptake mu ornithine+, was 2.9 nmoles per min per mg of protein. In nearly completely. These data indicate that phosphorylation other experiments rates exceeding 5 nmoles per min per mg of of ADP and energy-requiring accumulation of Ca2f t,ake prec- protein were observed. Although this is relatively low com- edence over ornithine transport for energy generated by electron pared to the rate of energy-linked Ca2+ uptake, which may ex- transport. However, other factors may be involved in the effect teed 350 nmoles of Ca2+ per mg per min under similar conditions, of ADP on ornithine+ uptake (see “Discussion”). it is of the same order of magnitude as the rate of urea synthesis Effect of Omithine+ on Oxygen Uptake and Hi Ejection-Addi- in rat liver (see “Discussion”). tion of ornithine+ in high concentrations (1 to 10 mM) yielded

In the experiment of Fig. 4 about 60 nmoles of ornithine per no detectable stimulation of the state 4 rate of oxygen uptake mg of protein were accumulated, but in other experiments up- of rat liver mitochondria respiring on succinate, in comparison takes exceeding 160 nmoles of ornithine per mg of protein were with the large stimulation yielded by ADP or Ca2+. This find- observed. If it is assumed that the volume of the matrix water ing is consistent with the relatively low rate of ornithine uptake after maximum ornithine uptake is 2.0 ~1 per mg of protein, as compared to the rate of oxidative phosphorylation or Ca2f the internal concentration of ornithine achieved in the matrix accumulation. Addition of 10 rnM ornithine to state 4 mito- water in such experiments may exceed 80 mM, some Id-fold higher than the external concentration of 5 mM. In similar experiments it was found that when the external ornithine concentration was initially at 5 PM, internal to external ornithine ratios greatly exceeded 200: 1. These calculations assume that all the radio- activity recovered in the mitochondria is in the form of ornithine.

I I I I I 5 IO I5 20

MINUTES

FIG. 4. Rate of uptake of 1’4CJornithine. Mitochondrja (1.0 mg of protein) were suspended in 1.0 ml of a medium of 210 rnM mannitol, 70 mM sucrose, 10 mM: potassium succinate, 5 rnM KP;, DH 7.4. and 500 UM [Wlornithine. Thev were incubated at 37”.

I 0.5 Ornlthlne mM

At the’indicated times -the reaction was stopped by rapid cen- FIG. 5. Effect of ornithine concentration on initial rate of trifugation and the pellets counted after drying. The uptake ornithine uptake. Mitochondria (7.5 mg of protein) were incu- data were corrected for ornithine present in the nonmatrix space bated 1.0 min in 1.0 ml of a medium of 0.25 M sucrose, 5 mM suc- of the pellets by subtraction of controls incubated in the presence cinate, 5 mM phosphate, pH 7.4, and labeled ornit.hine in the of 1.0 PM FCCP (see Table II). concentrations shown. Temperature, 2F.

by guest on April 21, 2020

http://ww

w.jbc.org/

Dow

nloaded from

615

TABLE III Osmotic swelling of mitochondria in isosmolar citrulline

and other amino acids

The test media contained 250 mosM amino acids, 1.25 PM rote- none to suppress endogenous respiration, 0.5 mM EGTA, mito- chondria (1 mg of protein per ml), and other additions as shown. The temperature was 25”.

Source of mitochondria Amino acid

Experiment 1. Rat liver

Experiment 2. Rat liver Rat heart

Ornithine chloride 0.000

Citrulline 0.080

Citrulline + succinate 0.080 Citrulline + FCCP 0.085 Alanine 0.350 Valine 0.380 Proline 0.410 Glycine 0.500

Citrulline Citrulline

0.120 0.000

II Rate of swelling

-AA,oo/min

M lNu?ES LO

FIG. 6. Relative rates of efflux of isotopic citrulline and orni- thine from preloaded liver mitochondria. Rat liver mitochondria (10 mg of protein) were preloaded with ornithine by incubation for 15 min at 25” in 6.0 ml of a medium containing 10 rnM potassium succinate, 10 mM phosphate, pH 7.4, and 500 pM [14C]ornithine. Preloading with citrulline was carried out in a simple medium of 250 mosM [l%]citrulline. The loaded mitochondria were re- covered by centrifugation and resuspended in a fresh medium of 250 mosM KC1 containing 0.5 mM EGTA, 1.25 pM rotenone, and 1.0 mM Tris-chloride. DH 7.4. At the indicated times samnles were filtered on Millipore dishes which were washed, dried, and counted.

chondria also yielded little or no ejection of H+ into the medium, in contrast to the large and rapid ejection of Hf given by Ca*+.

Xpecijkity of Ornithine Uptake-Table I shows the results of experiments on the rate of succinate-induced swelling of rat liver mitochondria when suspended in isosmotic solutions of the phosphate salts of some cationic metabolites similar in struc- ture to ornithine’. Addition of succinate produced no signifi- cant mitochondrial swelling in isosmolar arginine’ medium and only a slight swelling in lysine+ and choline+ media, compared to the rate given in an ornithine+ medium. N-Acetylornithine also failed to enter. The failure of arginine+ to enter the en- ergized mitochondria not only shows the specificity of ornithine entry but also is of biological interest since this intermediate of the urea cycle is normally formed and hydrolyzed in the cytosol.

The L stereoisomer of ornithine, the biologically occurring form, enters rat liver mitochondria far more rapidly than the unnatural D isomer (Table I), strongly suggesting that the mito- chondria contain a stereospecific membrane transport system for L-ornithine+. Data in Table I also demonstrate that respir- ing mitochondria from rat heart and beef heart do not undergo swelling in L-ornithine phosphate media, consistent with the fact that heart tissue cannot bring about the synthesis of urea.

~@ZUX and Efluz of C&&line-Experiments in Table III show that mitochondria swell in the presence of citrulline, which is predominantly in the electrically neutral zwitterionic species at pH 7.0, without a requirement for either respiratory energy or phosphate. The rate of swelling in isotonic citrulline media was unaffected by either permeant (acetate, thiocyanate, or phosphate) or impermeant (chloride) anions. Moreover, the entry of citrulline was not influenced by respiratory inhibitors such as cyanide or rotenone or by uncoupling agents such as FCCP. Other neutral (zwitterionic) amino acids (proline, valine, alanine, and glycine) also enter mitochondria in a similar passive manner under these conditions (Table III). The rate of citrulline entry is relatively modest compared to that of gly- tine, the most active. Rat liver mitochondria preloaded with [‘4C]citrulline and then added to an isotonic sucrose medium were found to undergo a rapid loss of isotope to the medium (Fig. 6), showing that citrulline can readily pass through the mitochondrial membrane in either direction. On the other

hand, rat liver mitochondria preloaded with [‘“C]ornithine did

Data in Table III show that rat heart mitochondria, which not lose isotope under these conditions.

cannot form citrulline from ornithine, do not swell when sus- pended in an isosmotic citrulline medium. The tissue specificity suggests the possibility that citrulline t’ransport is also mediated by a carrier in rat liver mitochondria.

DISCUSSIOK

This exploratory investigation of the transport of ornithine and citrulline across the mitochondrial membrane provides in- formation regarding (a) the driving force for transport (b) the mechanism of transport, and (c) the metabolic organization of the urea cycle in the liver.

The observations reported here show t,hat, the membrane of respiration-inhibited rat liver mitochondria is impermeant to ornithine, whether as the cation, the electrically neutral zwit- terion, or as N-acetylornithine. However, the ornithine+ cation readily enters the mitochondrial matrix when electron transport is taking place, in the presence of certain permeant anions. The permeant anions promoting entry of ornithinef, which include phosphate, acetate, and bicarbonate, have as common denomi- nator the ability to yield protons when they enter into the matrix. Acetate is believed to cross the membrane as the undissociated free acid (23), phosphate is transported as either HP0a2- or H2P04-, most likely the latter (23, 27), and bicarbonate crosses the membrane as dissolved CO,; the dissolved CO2 presumably is rehydrated in the matrix to yield the proton-bearing bicarbon- ate ion (34). On the other hand, ornithine+ entry is not sup- ported by the permeant anions SCN- and NO,, which cross the membrane as such and do not occur to any significant extent as protonated species at pH 7.4.

These observations on the entry of the ornithine+ cation closely resemble the observations of Brierley and his colleagues (36, 37) on the respiration-dependent entry of K+ and other cations into the matrix of beef heart mitochondria, and the observations of Hansford and Lehninger (38) on entry of K+ into the matrix of respiring blowfly muscle mitochondria. In both cases, phos- phate or the anions of weak acids were required for respiration- dependent entry of K+. Moreover, the anion requirements for

by guest on April 21, 2020

http://ww

w.jbc.org/

Dow

nloaded from

616

e-

H+* I *OH- HOH

OH-4 n

i’ ,H,PO;

e-

1

I H+*

,““I+ HOH +OAc-

OAc--HOAc * HOAc

MEMBRANE

e- FIG. 8. Electrophoretic uniport of ornithinef cation in response to negative-inside potential generated by respiration-dependent inward movement of phosphate.

Ii+ OH-

HCO;

ii

H&O, CO, t- C% H,C%

HOH+HCO; The hypothesis of an ornithine+-specific carrier or transport,

f I”

system functioning in response to an electrochemical gradient

H20 H20

generated by electron transport is not,, however, sufficient to explain all our findings. WC have found that nonrespiring rat

Fro. 7. Generation of negative-inside potential by respiration- liver rnitochondria do not swell when placed in isosmolar or-

dependent entry of proton-carrying anions. nithine nitrate or thiocyanate, conditions in which both the ornithine+ cation and the pcrmeant anion would be expected

ornithine+ entry are also identical with those established for to move down rather steep electrochemical gradients into the the respiration-dependent entry of Ca2+ into rat liver mitochon- matrix if the membrane contained an ornithine-specific carrier. dria, which is supported by phosphate, acetate, and bicarbonate, Nor do they swell in media of ornithine phosphate or acetate, but not by nitrate or thiocyanate (34, 39). Thus it appears even in the presence of FCCP to allow protons brought in wit,h that t,he driving force for entry of the ornithine+ cation is the phosphate or acetate to escape from the matrix. Only after same as that involved in the respiration-dependent entry of K+ institution of respiration in the presence of proton-yielding anions and Ca2f. The requirement of a proton-yielding anion is ac- does ornithine enter, whether from concentrated (125 m&l) or counted for by the generation of an alkaline-inside proton gra- dilute (50 pM) media. Again, these findings are similar to those dient by electron transport, for which the model of Mitchell (40) on entry of Ii+, which readily passes through the membrane of provides the simplest but not the only possible rationale. The heart (36, 37) and blowfly (38) mitochondria when they are alkaline-inside proton gradient pulls in the proton-yielding anion respiring in state 4 but do not pass the membrane of respiration- (Fig. 7), which gives up its proton to the excess hydroxyl ions inhibited mitochondria unless an ionophore such as gramicidin in the matrix. The negative-inside potential gradient developed or valinomycin is present. To explain the requirement of res- in this manner is responsible for electrophoretic transport of piration for ornithine+ entry we postulate that in nonenergized, the cation into the matrix, whether it is ornithine+, K+, or Ca2+ respiration-inhibited rat liver mitochondria the ornithine carrier (Fig. 8). Arguments supporting such a role of phosphate and is in an inactive form whereas during state 4 respiration it is in the anions of weak acids have been developed in other commu- an active state and can then transport ornithine+ inward electro- nications from this laboratory (34, 39), and by Brierley and his phoretically in response to the negative-inside potent.ial gen- colleagues (36, 37). erated by electron transfer. This proposal (see also Reference

We may now examine the nature of the process by which the 37) is consistent with much evidence that the mitochondrial ornithine+ cation passes through the membrane of rat-liver mito- membrane undergoes very significant microscopic and molecular chondria in response to the electrophoretic driving force. Our changes in its conformation during transitions between state swelling test results show that the L stereoisomer of ornithinc 4 and state 3, as detected with the electron microscope (41) and enters but the D isomer does not; in addition it was shown that with fluorescent probe methods (42). the &ucturaliy similar amino acids L-lysine and L-arginine do The experiments reported here suggest that citrulline enters not rllter. Moreover, it was found that ornithine enters respir- and leaves rat liver mitochondria passively by a respiration- ing rat liver mitochondria, which participate in urea synthesis, independent process. However, since rat liver mitochondria but not heart mitochondria, which do not. Moreover, ornithine undergo rapid swelling in isosmolar citrulline media, but heart entry occurs by a process that appears to approach saturation mitochondria do not, it appears likely that rat liver mitochondria at relatively low concentrations. All these properties strongly contain a carrier capable of transport’ing cirtulline as well as suggest that the ornithine+ cation enters rat liver mitochondria other neutral amino acids (cf. Reference 43). Tests of the stereo- in response to an electrophoretic driving force via a specific car- specificity of citrulline transport should furnish significant evi- rier or transport system rather than by simple unmediated phys- dence on this point. We have considered the possibility that ical diffusion through the membrane, which would be expected the entry of the ornithine cation and the exit of the citrulline to be nonspecific for the stereoisomers and to be concentration- zwitterion may t,ake place by an obligatory antiport, process, dependent. in keeping with the fact that during urea synthesis one molecule

by guest on April 21, 2020

http://ww

w.jbc.org/

Dow

nloaded from

InnT membrane

Ornithine-

Argika- Succinate

ATP

Aspartate

CYTOPLASM

I Ornithine

MITOCHONDRIAL MATRlX

FIG. 9. Summary of role of electron transport-generated energy in mechanism of urea synthesis. TCA cycle, tricarboxylic acid cycle.

617

may also yield an extra thermodynamic “push” in the direction of urea synthesis by bringing about unidirectional transport of ornithine into the matrix at the expense of respiratory energy (Fig. 9).

Acknowledgments-The authors thank Drs. Richard Hansford and William Jacobus for valuable advice and many useful dis- cussions.

REFERENCES

1. COHEN, I?. P., .~ND HAYANO, M. (1948) J. Biol. Chem. 172,405 2. LEUTHARDT, F., AND MILLER, A. F. (1948) Experientia (Basel)

4, 478 3. GRISOLI.~, S., KORITZ, S. B., AND COHEN, P. P. (1951) J. Biol.

Chem. 191, 181 4. C.\R.~VXA, J., AND GRISOLIB, S. (1960) J. Biol. Chem. 235,684 5. METZENRERG, R. L., HILL, L. M., M.~RsH.\LL, M., AND COHEN,

I’. P. (1957) J. Biol. Chem. 229, 1019 6. T.~TIB.\N,I, M., AND COHEN, P. P. (1965) Proc. Nat. Acad. Sci.

U. S. A. 63, 104 7. ~HAMHAUGH, G. E., III? BALINSlcY, J. B., AND COHEN, P. 1’.

of citrulline must leave the mitochondria for each ornithine molecule entering. However, experiments designed to test this idea have shown no evidence of obligatory antiport.

9. CH.\PPI:LL, J. 13. (1968) firit. Mctl. IZull. 24, 150 10. KLIN<Z~:NRRXG, M. (1970) in Essays ix Biochemistry, vol. 6,

p. 119, Academic Press, New York

Some comments may now be made on the relationship of these findings to the compartmentation, dynamics, and energetics of urea synthesis in the liver. Our experiments show that or- nithine readily enters respirin, v rat liver mitochondria, but ar- ginine, the preceding intermediate of the urea cycle, does not. Whether the remaining urea cycle intermediate, argjninosuc- cinate, is similarly unable to pass the membrane, remains to be est,ablished. The question arises whether the transport of or- nithine int,o the mitochondria by the process studied in this paper proceeds sufficiently rapidly to account for the rate of urea synt’hesis in vivo. The rat synthesizes urea at the rate of 1,000 to 1,600 mg per kg per day (44). Thus a 250-g rat puts out at. least 250 mg per day, equivalent to 3 lmoles per min. rlssuming t,hat the liver weighs 10 g, equivalent to about 350 mg of mitochondrial protein, the rate of urea production is thus about 8.5 nmoles per min per mg of mitochondrial protein. Al- though we have observed initial entry rates exceeding 5 nmoles per min per mg, in most experiments it has been less. It must be emphasized that we have not yet established optimal condi- tions for maximal rates and extent of ornithine entry, which may require the availability in the matrix of ATP, bicarbonate, and ammonia for the synthesis of carbamyl phosphate, with which ornithine must react.. These considerations may also

be responsible for the fluctuations from one preparation of mito- chondria to another in rate and extent of the uptake of isotope from the [14C]ornithine+ media.

II. LI;HNINGI~:K, A. I,. (1971) in Biomewdvanes (MANSON, L. A., ed) vol. 2. D. 147. Plenum Pablishinr Corn.. New York

12. GAM;ILI.:, J.‘&., AND LEHNIIVGI~I~, A. L. (19?lj Fed. Proc. 30, 1238

13. SCIINEID~R, W. C. (1948) J. Biol. Chem. 176, 259 14. PANDIC, S. V., .&ND BL.<NCH.IER, M. C. (1971) J. Biol. Chern. 246,

402 15. W.~DDI~LL, W. J. (1956) J. Lab. Clin. Med. 43, 311 16. MURPHY, J. B., AND KIES, M. W. (1960) Biochim. Biophys.

Acta 45, 382 17. SCHN.UTMIN, C., AND GREENA\%-ALT, J. W. (1968) J. Cell Biol.

38, 158 18. CH.W, T. L., GRGENAX~ILT, J. W., .IND PEDERSEN, P. L. (1970)

J. Cell Biol. 45, 291 19. KERSON, L. 8., AND APPEL, S. H. (1968) J. Biol. Chem. 243,

4279 20. BR.IUF.IY, H., BEND.ILL, D. S., B.~UDHUIN, P., .IND DEDuw:,

C. (1959) Biochem. J. 73, 623 21. SCHNAITMIN, C. A., ERIYIN, V. G., .IND GRISENAW.ZLT, J. W.

(1967) J. Cell Biol. 32, 719 22. SCHIMKI~;, R. T. (1962) J. Biol. Chem. 23’7, 450

23. CH.\PPELL, J. B., .k~n CIZOPTS, A. R. (1966) in Regulatio>h of Metabolic Processes in Afitochondria (T.\GI<:R, J. M., P~\P.‘I, S., (~U.ZGLI.~IZII.:I,LO, E., AND SLATF:R, E. C., eds) p. 293, Ameri- can Elsevier Publishing Company, New York

24. BLONDIN, CT, A., VAIL, W. J., AND GREI:N, D. E. (1969) Arch.

,%&hem. Biophyr. 129, 158 25. BRIICI<I,KY, (:. P. (1970) Biochemists!/ 9, 697 26. Azzr, A., .WD AZZONIG, C:. F. (1967) Biochim. Bioph,ys. Acla

131, 468 27. MITCHELL, P., AND MOYLIC, J. (1969) Eur. J. Biochem. 9, 149 28. BRII.;RLICY, CT. P., JURICO~.ITZ, M., SCOTT, K. M., AND M>;RoL.I,

In the light of the experiments reported here t,he mitochondrion plays three roles in furnishing the energy for urea synthesis. First,, it provides the respiratory energy for synthesis of the two molecules of internal Al’P required in the matrix for the synthesis of one molecule of carbamyl phosphate from bicarbo- nate generated by the tricarboxylic acid cycle and ammonia formed by oxidative deamination of amino acids. Secondly, the mitochondrion also generates one molecule of external ATP required for the formation of argininosuccinate from aspartate and citrulline in the cytoplasm. Thirdly, the mitochondrion

A. J. (1970) J. Biol. Chem. 246, 5404 29. FONYO, A. (1968) Biochem. Biophys. Res. Commun. 32,624 30. TYLXR, D. D. (1969) Biochem. J. 111, 665 31. CHIPPELL, J. B., AND HAARHOFF, K. N. (1967) in Biochemistry

of the LXitochondria @LATER. E. C., KANIUG~. Z.. .~ND W~JTCZAK, L., eds) p. 75, Acabemic Press, New York’

32. GuI<:RIN, B., GUERIN, M., AND KLINGENBERG, M. (1970) Fed. Eur. Biochem. Sot. Lett. 10,265

33. TYLER, D. D. (1968) Biochem. J. 107, 121 34. ELDISR, J. A. (1972) Fed. Proc. 31, 856 35. BRONK, J. R., ?IND FISHIGR, R. B. (1956) B&hem. J. 64,106 36. BRIERLEY, G. P., SRTTLEMIRIZ, C. I., ASD KNIGHT, V. A. (1968)

Arch. Biochem. Biophys. 126, 276

(1969) J. Biol. Chem. 244, 5295 8. RATNXR, S. (1962) in The Enzymes (BOYI:R, P. D., LUU)Y, H.,

IND MYRD~~K, K., eds) 2nd Ed, vol. 6, p. 499, Academic Press, New York

by guest on April 21, 2020

http://ww

w.jbc.org/

Dow

nloaded from

618

37. BRIERLEY,G.P.,JURKOWITZ, M., SCOTT, K. M., ANDMEROLA, 40. MITCHELL, P. (1968) Chemiosmotic Coupling and Energy A, J. (1971) Arch. Biochem. Biophys. 147, 545 Transduction, Glynn Research Ltd., Bodmin, England

38. HANSFORD, R. G., .IND LEHNINGER, A. L. (1972) Biochem. J. 41. HACKENBROCK, C. H. (1966) .I. Cell Biol. 30, 269

126, 689 42. Azzr, A., CHANCE, B., RADDA, G. K., AND LEB, C. I?. (1969)

39. LEHNINGER, .4. L. (1972) in Molecular Basis of Electron Trans- Proc. Nat. Acad. Sci. U. S. A. 62, 612

port (SCHULTZ, J., AND CAMERON, B. F., eds) p. 133, Aca- 43. BUCHANAN, J., POPOVITCH, J. R., AND TAPLEY, D. F. (1969)

Biochim. Biophys. Acta 173, 532 demic Press, sew York 44. PEESIKE, E. C. (1948) Endocrinology 42, 356

by guest on April 21, 2020

http://ww

w.jbc.org/

Dow

nloaded from

James G. Gamble and Albert L. LehningerTransport of Ornithine and Citrulline across the Mitochondrial Membrane

1973, 248:610-618.J. Biol. Chem. 

  http://www.jbc.org/content/248/2/610Access the most updated version of this article at

 Alerts:

  When a correction for this article is posted• 

When this article is cited• 

to choose from all of JBC's e-mail alertsClick here

  http://www.jbc.org/content/248/2/610.full.html#ref-list-1

This article cites 0 references, 0 of which can be accessed free at

by guest on April 21, 2020

http://ww

w.jbc.org/

Dow

nloaded from