Biosynthesis of rat liver transhydrogenase in vivo and in vitro.

The biosynthesis of pyridine dinucleotide transhydrogenase, a homodimeric inner mitochondrial membrane redox-linked proton pump, has been studied in isolated rat hepatocytes. Newly synthesized transhydrogenase, having an apparent molecular weight identical to the enzyme of isolated liver mitochondria, was selectively immunoprecipitated from detergent extracts of isolated hepatocytes which were labeled with [35S]methionine. That the enzyme is a nuclear gene product is indicated since 1) synthesis was inhibited by cycloheximide, but not by chloramphenicol and 2) no synthesis could be demonstrated in hepatocyte ghosts which are competent only in mitochondrial translation. In addition to the mature form of the enzyme, a species about 2000 daltons larger was also immunoprecipitated from pulse-labeled cells. The half-life of the larger form during a subsequent chase at 37 degrees C was about 2 min, whereas the mature form was not degraded. The relationship between the two forms of the enzyme was established by in vitro studies. A protein approximately 2000 daltons larger than mature transhydrogenase was immunoisolated from a rabbit reticulocyte lysate system programmed with sucrose gradient fractionated rat liver mRNA. This protein was converted to a species having the same size as mature enzyme after incubation with either intact rat liver mitochondria or a soluble matrix fraction derived from mitoplasts. These studies indicate that transhydrogenase is synthesized in the cytoplasm as a higher molecular weight precursor which is post-translationally processed to the mature protein by a soluble matrix protease during or after membrane insertion.

The biosynthesis of pyridine dinucleotide transhydrogenase, a homodimeric inner mitochondria~ membrane redox-linked proton pump, has been studied in isolated rat hepatocytes. Newly synthesized transhydrogenase, having an apparent molecular weight identical to the enzyme of isolated liver m i t~h o n~i a , was selectively immunoprecipitated from detergent extracts of isolated hepatocytes which were labeled with [35S]methionine. That the enzyme is a nuclear gene product is indicated since 1) synthesis was inhibited by cycloheximide, but not by chloramphenicol and 2) no synthesis could be demonstrated in hepatocyte ghosts which are competent only in mitochondrial translation, In addition to the mature form of the enzyme, a species about 2000 daltons larger was also immunoprecipitated from pulse-labeled cells. The half-life of the larger form during a subsequent chase at 37 "C was about 2 min, whereas the mature form was not degraded. The relationship between the two forms of the enzyme was established by in vitro studies. A protein approximately 2000 daltons larger than mature transhydrogenase was immunoisolated from a rabbit reticulocyte lysate system programmed with sucrose gradient fractionated rat liver mRNA. This protein was converted to a species having the same size as mature enzyme after incubation with either intact rat liver mitochondria or a soluble matrix fraction derived from mitoplasts.
These studies indicate that transhydrogenase is synthesized in the cytoplasm as a higher molecular weight precursor which is post-translationally processed to the mature protein by a soluble matrix protease during or after membrane insertion.
Although the mitochondrial genome is capable of coding for the synthesis of a limited number of proteins, approximately 90% of the protein of the organelle is coded by nuclear genes and translated on cytoplasmic ribosomes. The cytoplasmic gene products destined for import to the inner membrane and matrix are generally, though not always, higher molecular weight precursors that are cleaved post-translationally by a matrix-processing protease (1). Recent studies indicate that processing may occur subsequently to, rather than concomitantly with, energy-dependent membrane insertion or translocation (2,3).
Much emphasis has been placed, particularly in lower eukaryotes, on the synthesis, processing, and assembly of the * This work was supported in part by United States Public Health Service Grant GM 22070. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
3 To whom correspondence should be addressed. energy-transducing complexes of oxidative phosphorylation, i.e. cytochrome c oxidase, ubiquinolcytochrome c reductase (b-ci complex), and Fo I F1-ATPase. A general feature of these hetero-oligomeric complexes is that certain of their subunits are cytoplasmically synthesized, whereas others are mitochondrial gene products (1). For example, in rat liver (4, 5 ) as in yeast (6,7) and Neurospora crassa (8) the three largest subunits (I to 111) to cytochrome c oxidase are synthesized in the mitochondria and the smaller subunits (IV to VII) are of cytoplasmic origin. Two of the liver oxidase subunits, IV (9, 10) and V (11), have been reported to be synthesized as precursors being 1500 to 3000 daltons larger than the mature species. Little is known about the biosynthesis of b-cl complex of F'o.F,-ATPase in higher eukaryotes. Although it has been reported that one subunit of the 6-cl complex and two subunits of the Fn segment of Fo.Fl-ATPase are products of mitochondrial translation (5, 12, 13), no information is available concerning the nature or processing of the cytoplasmically synthesized subunits. Analogous studies on yeast (14, 15) and N . crmsa (16) suggest that most of the cytoplasmic subunits of these complexes in higher eukaryotes are likely to be synthesized as extended precursor forms. This communication describes our initial studies on the biosynthesis and mechanism of insertion of pyridine dinucleotide transhydrogenase into the inner membrane of rat liver mitochondria. The enzyme couples the reversible transfer of a hydride ion between matrix NADPH and NAD' to vectorial proton translocation from the matrix to the cytosol (17,18). Unlike the other mitochondrial energy-transducing complexes, bovine heart transhydrogenase of the native membrane possesses a simple subunit structure consisting of a dimer of apparently identical 110-kDa subunits (19). Here, we have employed monospecific antibody prepared against bovine heart transhydrogenase to identify and characterize the enzyme synthesized by isolated rat hepatocytes and by rabbit reticulocyte lysates programmed with partially purified liver mRNA.
Portions of this paper (including "Experimental Procedures" and

6361
Monospecific ant.ihody has heen prepared against purified transhydrogenase (19,22). T o investigate the hiosynthesis of t.ranshydrogenase in rat liver, it was initially necessary to est.ahlish that heart anti-transhydrogenase specifically recognizes t.he corresponding enzyme from rat hepatocytes. T o accomplish this, immunoblot analysis of heart and rat liver suhmit.ochondria1 particle proteins resolved by SDS2-polyacrylamide gel electrophoresis was employed. Fig. 1 illustrates that heart. suhmitochondrial particle transhydrogenase (lane 2) migrates identically with purified heart enzyme (lanes I and -5). Rat, liver suhmitochondrial particles also gave a single immunoreactive species (lane 4 ) that migrated slightly more slowly into t,he gels than heart transhydrogenase. When membranes from both sources were electrophoresed together (lane 3 , a broadened band appeared, indicative of a composite of two proteins having slightly different molecular weights. Under the elect.rophoresis conditions used, resolution of proteins in the 100-kDa range differing by as lit,tle a s 500 daltons would not he ohserved. In other studies (not shown) the antihody was found to effectively inhibit the reduction of the :bacetylpvridine analog of NAD+ hy NADPH (24) catalyzed hy rat liver suhmitochondrial particles, providing further evidence that the antihody specifically complexes with liver transhydrogenase.
Site of Transhydrogenuse Synthesis in Rat Hepatocytes-T h e low rate of protein synthesis in isolated higher eukaryotic cells has hampered investigations on the c-ytoplasmic hiosynthesis, import, and processing of low abundance inner mitochondrial membrane proteins in mammals, although a few have heen reported (9-11, 31). The emphasis on that aspect of higher eukaryote mitochondrial hiogenesis directed by the nuclear genome has heen focused on the uptake of relatively high abundance soluble matrix and intramemhrane space enzymes (31-39). Transhydrogenase represents less than 1% of the protein of the inner mitochondrial membrane (20) and, therefore, may he classified as a low abundance protein of the organelle. Although sequence analysis of mammalian mitochondrial DNA provides no evidence for reading frames that could code for the synthesis of a protein having a molecular weight in excess of ahout fi7 kDa (40), Rhat et al. (41) have reported that isolated rat liver mitochondria are capable of synthesizing 19-24 polypeptide species ranging in size to approximately 110 kDa. I t is noteworthy that the highest molecular weight putative mitochondrial translation product reported corresponds to that of transhydrogenase. To determine the site of transhydrogenase synthesis, isolated hepatocytes were continuously laheled for 2 h with [:"S]methionine under conditions where either cytoplasmic or mitochondrial protein synthesis was blocked. During the incubation at 37 "C total protein synthesis was inhibited by 80-90% in the presence of 0.7 mM cycloheximide and by 515% in the presence of 0.1 mM chloramphenicol. The enzyme was immunoprecipitated from detergent extracts of laheled cells by addition of heart. anti-transhydrogenase followed by adsorption to Protein A-bearing Staphylococcus aureus cells. Newly synthesized transhydrogenase was visualized by fluorography following SDS-polyacrylamide gel electrophoresis of the immune complexes. As shown in Fig. 2, a single "'S-labeled protein was immunoprecipitated from hepatocytes incuhated in the absence of protein synthesis inhihitors (lane 2) that co-migrated with purified heart ['251]iodotranshydrogenase marker added to and immunoprecipitated from an unlabeled hepatocyte extract (lnnr 1 ) . Hepatocytes labeled in the presence of cycloheximide did not express transhydrogenase (lane 3 ) . However, The ahl~reviation used is: SDS, sodium dodecyl sulfate. chloramphenicol did not suppress synthesis of the enzyme (lane 4 ) . These results strongly suggest that transhydrogenase is a nuclear gene product. To eliminate any ambiguity relating to the intracellular site of transhydrogenase synthesis, it was of interest to demonstrate that the enzyme could not he immunoprecipitated from hepatocyte ghosts that are capable only of mitochondrial protein synthesis ( 2 6 ) . To this end, hepatocyte ghosts were prepared by digitonin treatment of intact cells and laheled with ['"Slmethionine according to the procedure of Kuzela et al. ( 2 6 ) . As shown in Fig. 3 protein synthesis by the ghosts (lnnr 2 ) was inhibited totally hy chloramphenicol (lanc. 3 ) but not inhihited hy cycloheximide (lane 4 ) . These data demonstrate that the ghost preparation is completely incapable of cytoplasmic protein synthesis. That transhydrogenase was not immunoprecipitated from laheled ghosts (lane .5) confirms that the enzyme is of cytoplnsmic origin.
The Presence of TranshydroRenase PrQcursor in Hepatocytes-Although relatively long pulses of hepatoc-ytes with [,"'SS]methionine resulted in the immunoprecipitation of a single laheled protein having an apparent molecular weight corresponding to the mature form (Fig.  2 ) , shorter pulses revealed an additional radioactive hand of higher apparent molecular weight ( A M , = 2000). This is illustrated in Fig. 4, where the two bands are clearly visihle after a IO-min pulse (lane 1 ) . As expected for a precursor. the larger species disappeared during a suhsequent chase in the presence of unlaheled methionine and cycloheximide. giving an apparent halflife of ahout 2 min (lunrs [2][3][4][5][6]. Hv contrast, the amount of F I~; .

Evidence for the occurrence of transhydrogenase percursor in hepatocytes.
Hepatocytes were pulsed with [."'SI methionine for 1 0 min and then chased in the presence o f cold methionine and cycloheximide as described under "Experimental I'roredures." At various times during the chase, aliquots were taken for detergent extraction and immunoprecipitation. Transhydrogenase was visualized as described in Fig. 3. IAnr I contained the immunoprecipitate after the pulse. Lanes 2-6 contain immunoprecipitates ohtained after 2. 4, 6. 8, and 10 min of chase, respectively. mature form of the enzyme remained stable during the chase after a perceptible increase during the first 2 min. This apparent half-life is similar to those reported for the processing of mammalian matrix precursor proteins (cf. Ref. 42). The addition of purified heart transhydrogenase to the labeled cell extract totally prevented the immunoprecipitation of both forms of the enzyme.
S.vnthesi.s of Transhydrogenase Precursor by Cell-free Translation-To verify that transhydrogenase is synthesized in the cytoplasm as a higher molecular weight precursor, rat liver mRNA was translated in a rabbit reticulocyte lysate system in the presence of ["%]methionine. Initial attempts to selectively immunoprecipitate the enzyme from lysates programmed with total mRNA were unsuccessful due to contamination with many protein species of both higher and lower molecular weights. Accordingly, it was found to be necessary to further fractionate the poly(A) RNA on a sucrose gradient to both enrich the transhydrogenase message and to obtain contaminant-free immunoprecipitates. Fig. 5A shows a typical elution profile of RNA after separation on a 5-2576 discontinuous sucrose gradient. Residual 18 and 28 S ribosomal RNA species were conveniently present and used as markers for mRNA. Selected fractions of the gradient, designated a through f, were pooled and the mRNA recovered from each pooled fraction after ethanol precipitation. Fig. 5R illustrates the total translation products obtained from each mRNA fraction. A trend of increasing molecular weights of the products encoded from fraction a to fraction f is apparent. The translation products obtained in Fig. 5R were subjected to immunoprecipitation to locate transhydrogenase mRNA (Fig.  5C). No immunoprecipitable labeled protein was detected in fractions a through c (only fraction c is shown, lane 2 ) . Fraction d (lane 3 ) gave a highly contaminated precipitate, which may arise from the nonspecific binding of many products to the Protein A adsorbent, with no indication for the presence of a species with a mobility similar to transhydrogenase or its putative precursor. Fraction e (lane .5) gave a single faint hand, which was more intense in fraction f (lane 6). The immunoprecipitated protein of fractions e and f had a slightly lower mobility in SDS-polyacrylamide gels than heart [1Z51]iodotranshydrogenase marker (lanes I , 4, and 7). T o better define the difference in apparent molecular weight between mature hepatocyte transhvdrngenase and its putative in vitro synthesized precursor, both immunoprecipitated ""Slabeled products were electrophoresed under conditions developed to optimize separation of proteins in the 100-kI)a range. As shown in Fig. 6 , bovine heart [""ljiodotranshvdrogenase (lanes I and 3 ) and mature hepatocyte transhydrogenase (lane 2 ) migrated identicallv, whereas the in Litrn svnthesized product migrated slower (/una 4 ) . The specificity of the transhydrogenase antibodv is indicated by the complete prevention of in vitro product precipitation in the presence of added purified bovine heart transhvdrogenase (InnrJ 5 ) . The difference in apparent molecular weights hetween the in L'iro and in vitro products, as indicated by molecular weight markers including [j-galactosidase ( 1 16,000) and phosphorylase h (97,400), was calculated to he approximatelv 2000.
Processing of Transhydrogennsa Pwcursor hy Intact Rat Liver Mitochondria and a Matrix Subfraction-In order to establish the precursor relationship of the in uitro svnthesized product to mature transhydrogenase the cell-free translation products were exposed to intact rat liver mitochondria. After incubation, the mitochondria were separated from the suspending medium by centrifugation, and transhydrogenase was then immunoprecipitated from both fractions. Fig. 7.4 shows that the hand present in the supernatant fraction migrates identically with that of the translation mixture, whereas all of the radiolabeled product found in the mitochondrial fraction migrates faster and corresponds to the mature form of the enzyme. In other studies (data not shown), transhydrogenase taken up by mitochondria, hut not the protein of the supernatant fraction, has been shown to be resistant to degradation by trypsin. This indicates, hut does not prove, that the protein has heen inserted into the inner memhrane. €+ecursor processing protease activity has heen t.ypically found to reside in the matrix compartment of both higher and lower eukaryotes (1). Fig. 7 R illustrates t.hat this is true also for transhydrogenase. After incubation of translation products with a matrix fraction derived from detergent lysis of mitoplasts. the larger transhydrogenase form, under conditions employed, is partially converted into a smaller form that corresponds in apparent, mass to mature enzyme.
These studies strongly indicate that the in vitro synthesized transhydrogenase represents a precursor form of the enzyme, identical to that found in short-term pulsed intact cells. However, it remains unknown if the in vitro processing of the precursor occurs at t.he N terminus of the protein, if processing faithfully produces mature enzyme, if processing results in the proper insertion of the protein in the inner membrane, and if processing is required to generate substrate-binding sites characteristic of an active enzyme. These aspects are currently being invest,igated.