A Cytoplasmic Male Sterility–Associated Mitochondrial Peptide in Common Bean Is Post-Translationally Regulated

Correspondence to: Sally A. Mackenzie, smackenz{at}purdue.edu (E-mail), 765-494-6508 (fax).

	ABSTRACT

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION METHODS REFERENCES

Cytoplasmic male sterility in the common bean plant is associated with a dominant mitochondrial mutation designated pvs-or f 239 (for Phaseolus vulgaris sterility sequence open reading frame 239). The sequence is transcribed in both vegetative and reproductive tissues, but the translation product, ORF239, is present only in reproductive tissues. We present evidence to support a model of post-translational regulation of ORF239 expression based on the following observations. In organello translation experiments using purified mitochondria from young seedlings demonstrated accumulation of ORF239 only when a protease inhibitor was included. Proteolytic activity against ORF239 was observed in mitochondrial extracts fractionating with the mitochondrial inner membrane. The DNA sequence encoding a serine-type protease, similar to the lon protease gene of Escherichia coli, was cloned from the Arabidopsis genome. The expression product of this sequence demonstrated proteolytic activity against ORF239 in vitro, with features resembling the activity detected in mitochondrial inner membrane preparations. Antibodies generated against the overexpressed Lon homolog reduced proteolytic activity against ORF239 when added to mitochondrial extracts. Our data suggest that ORF239 was undetected in vegetative tissue due to rapid turnover by at least one mitochondrial protease that acts against ORF239 post-translationally.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION METHODS REFERENCES

Features common to plant mitochondrial genomes, distinguishing them from other eukaryotes, are most likely the result of selective constraints imposed by the interactions of nuclear and cytoplasmic genomes throughout the life cycle of the plant. These features range from genome organization, size variation, and recombinational activity to nearly all aspects of gene expression (Bonen and Brown 1993 ; Wolstenholme and Fauron 1995 ). Gene regulation in plant mitochondria is not yet well understood. Relatively little evidence is available to suggest that transcriptional regulation serves as the predominant means of gene modulation, although a promoter consensus motif has been described (Rapp and Stern 1992 ; Rapp et al. 1993 ). Even though most plant mitochondrial transcripts undergo extensive editing (Hiesel et al. 1994 ) and, in some cases, cis- and/or trans-splicing of intron sequences (Wolstenholme and Fauron 1995 ), the rate and extent that these processes limit gene expression are not clear. Likewise, no compelling evidence has been reported thus far for translational regulation of plant mitochondrial genes, although considerable information regarding translational activation exists for yeast (Fox 1996 ). Here, we present evidence to suggest that one level of mitochondrial gene regulation that may be important in plants is post-translational proteolysis.

Cytoplasmic male sterility (CMS) in higher plants is a maternally inherited phenotype of pollen sterility (Hanson 1991 ). Characterization of CMS in several species has shown that the phenotype arises as a consequence of dominant mutations in the mitochondrial genome. In each CMS case examined, the particular mitochondrial mutation is distinct. The CMS system in common bean differs from other well-defined systems in several regards. Most important to this study is that the CMS-associated mitochondrial mutation in common bean, designated pvs-or f 239 (for Phaseolus vulgaris sterility sequence open reading frame 239), is transcribed in both vegetative (young seedling) and reproductive tissues (Chase 1994 ). The transcript of pvs-or f 239 has been demonstrated to associate with polysomes at both stages (Chase 1994 ); however, its product, ORF239, accumulates only in reproductive tissues (Abad et al. 1995 ). The ORF239 product, which is ~27 kD in size, ultimately is localized within the callose layer of the premeiotic pollen mother cell wall (Abad et al. 1995 ; He et al. 1996 ).

The observation of tissue-specific expression of a dominant mitochondrial mutation led us to investigate further the means of ORF239 regulation. Here, we present evidence to suggest that ORF239 expression in CMS common bean is regulated post-translationally, and we identify one mitochondrial protease in plants that appears to be involved in ORF239 turnover.

	RESULTS

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION METHODS REFERENCES

or f 239 Encodes a Translationally Competent Transcript in Vegetative (Young Seedling) Tissues in CMS-Sprite
Mitochondria of the cytoplasmic male-sterile line of common bean, CMS-Sprite, perform transcription of the sterility-associated mitochondrial sequence pvs-or f 239 in both vegetative and reproductive tissues (Chase 1994 ). The DNA sequence of pvs-or f 239 encodes a polypeptide of ~27 kD (Johns et al. 1992 ). Yet, the translation product of this mutation, ORF239, is detected only during the development of reproductive organs (Abad et al. 1995 ). To test whether the pvs-or f 239 transcript detected in vegetative tissues is translationally competent, in organello translations in the presence of ³⁵S-cysteine and an energy source were performed using purified mitochondria from etiolated 10-day-old seedlings. Figure 1 shows that mitochondria from CMS-Sprite seedlings produced no detectable ORF239 protein upon immu-noprecipitation by using polyclonal anti-ORF239 anti-bodies. However, when mitochondrial incubations of CMS-Sprite included a protease inhibitor, an ~30-kD mitochondrial polypeptide was immunoprecipitated that comigrated with the in vitro–translated ORF239 control (Figure 1B). The specificity of the anti-ORF239 antibodies had been extensively tested previously (Abad et al. 1995 ; He et al. 1996 ). Of the inhibitors used, all known to be membrane permeable, vanadate acted as the most effective inhibitor of ORF239 turnover, whereas o-phenanthroline and N-ethylmaleimide (NEM) were less effective (Figure 1B). The inhibitor hemin interfered with in organello translation (data not shown) so that it was not possible to determine its effect on ORF239 stability.

View larger version (24K): [in this window] [in a new window]   Figure 1. In Organello Translation Products of Mitochondria Prepared from CMS-Sprite and Isonuclear Fertile Revertant Lines. Mitochondria were prepared from etiolated young seedling tissues, and in organello reactions were incubated in the absence and presence (+) of the protease inhibitors vanadate (Van; 4 mM), o-phenanthroline (o-Phen; 5 mM), and N-ethylmaleimide (NEM; 2 mM). (A) Total in organello translation products from fertile revertant WPR-3 (isogenic to CMS-Sprite but lacking pvs-or f 239) produced in the absence and presence of an added protease inhibitor (vanadate, +Van). No Inhib indicates no inhibitor. (B) Translation products from CMS-Sprite immunoprecipitated with anti-ORF239 antibodies and a control lane (from the same gel; No Inhib [no inhibitor]) containing in vitro–transcribed/translated ORF239 (IVT-ORF239) immunoprecipitated with the anti-ORF239 antibody. The expected location of ORF239 is indicated at right, although two additional higher molecular mass proteins also precipitated. These larger proteins, which are present in both in vitro–transcribed/translated and in organello translation products, appeared to be newly synthesized, to cross-react with anti-ORF239 antibodies, and to accumulate only in the presence of protease inhibitors. We believe these larger products represent multimeric or aggregated forms of the ORF239 product, and they demonstrate a similar susceptibility to proteolysis. (C) The first two lanes contain products recovered with the preimmune (+Preim) and anti-ORF239 antibodies after in vitro transcription and translation of the ORF239 clone. The third lane shows the recovery of ORF239 after immunoprecipitation of in vitro–translated mitochondrial RNA by using anti-ORF239 antibodies. In (A) to (C), numbers at left indicate molecular mass standards in kilodaltons.

An unexpected observation emerged from these assays; although the ORF239 monomeric form immunoprecipitated in small amounts upon in organello translation, two larger forms of the protein were evident in greater amounts. We identified these higher molecular mass bands as ORF239. These bands most likely represent multimeric forms based on the following observations. The larger forms immunoprecipitated with anti-ORF239 antibodies in both in organello translations and in vitro translations (Figure 1C). Protein gel blot analysis of in organello and in vitro translation products indicated strong cross-reactivity of the larger forms with anti-ORF239 antibodies (data not shown). The accumulation of the higher molecular mass forms demonstrated a direct correlation with accumulation of the ORF239 monomeric form, and proteolytic turnover of ORF239 was proportional to turnover of the larger forms (Figure 1C and data not shown). A strong propensity to aggregate is routinely observed in the purification of ORF239 (data not shown).

The results of the in organello translation experiments demonstrate that the pvs-or f 239 transcripts detected in mitochondria from vegetative tissues are competent for translation and suggest that some degree of regulation of expression in vivo might be post-translational. Furthermore, the observation of several labeled proteins accumulating in mitochondrial incubations without the addition of protease inhibitors (Figure 1A) suggests that rapid proteolysis under these conditions has some degree of specificity to the ORF239 protein.

Plant Mitochondrial Extracts Contain Proteolytic Activity against the ORF239 Substrate
To characterize further the proteolysis of ORF239 in CMS-Sprite, we identified the location of the protease activity in the organelle and determined some of the properties of the protease. For this purpose, purified mitochondria from CMS-Sprite etiolated seedlings were fractionated to matrix, inner membrane, and outer membrane components. The proteolytic activity of these fractions was tested using as substrates in vitro–transcribed/translated ³⁵S-ORF239 and resorufin-labeled casein, a well-defined protease substrate. Proteolytic activity against ³⁵S-ORF239 was quantitated by the amount of acid-precipitable counts per minute accumulated after translation of the in vitro–transcribed/translated ORF239 (Table 1) and by SDS-PAGE (data not shown). To determine the proteolytic activity against resorufin-labeled casein, the amount of resorufin remaining was quantitated spectrophotometrically after trichloroacetic acid (TCA) precipitation. Figure 2 and Table 1 show that the proteolytic activity was associated primarily with the inner mitochondrial membrane. Fractionations involving sonication or more extensive detergent treatment (>15-min incubation with lubrol) of the inner membrane consistently resulted in the detection of activity in the matrix fraction (data not shown). These results suggest that the association of the detected protease activity with inner mitochondrial membrane is easily disrupted and likely does not involve membrane integration.

View larger version (16K): [in this window] [in a new window]   Figure 2. Protease Activity Detected in CMS-Sprite Mitochondrial Inner Membrane Fractions. Resorufin-labeled casein (90 µg per reaction) was incubated as substrate with purified and fractionated mitochondrial extracts (10 µg per reaction) at 37°C for 120 min in the presence of various protease inhibitors or under conditions of ATP depletion. Substrate turnover was assayed by measuring absorbance of TCA-nonprecipitable peptide. The data used are the average of two experiments, and the bars indicate standard errors obtained at each time point.

The protease activity detected in inner membrane fractions tested against a casein substrate demonstrated ATP dependence and sensitivity to NEM, phenylmethylsulfonyl fluoride (PMSF), vanadate, and EDTA (Figure 2); these properties can be indicative of a serine-type protease. Furthermore, although some sensitivity of the protease to o-phenanthroline was detected, this sensitivity was not influenced by subsequent addition of Mn²⁺ (data not shown), suggesting that detected activity was not likely due to the presence of a metalloprotease. Both serine-type and metalloproteases have been identified in mitochondria of animal and yeast systems (Rep and Grivell 1996 ), and we cannot exclude the possibility that metalloprotease activity may also be present in inner membrane fractions.

Variable results and sometimes undetectable levels of turnover were obtained when mitochondrial fractions were incubated with ³⁵S-labeled in vitro–transcribed/translated ORF239 post-translationally (Table 1, IVT-PT ORF239), requiring a more sensitive assay of protease action. When the in vitro–transcribed/translated ORF239 substrate was heated to 65°C before incubation, substrate turnover rate increased. This change in turnover rate suggested either that the ORF239 protein precipitates rapidly and is unavailable to the protease or that the rate of proteolysis is dependent on structure of the substrate, as has been observed by others (Wagner et al. 1994 ; Van Melderen et al. 1996 ). The first possibility was disregarded because ORF239 is completely undetectable in vivo, even when sensitive ELISA procedures are used (Abad et al. 1995 ). We investigated the second possibility and tested whether the proteolytic activity detected in purified mitochondria would act cotranslationally. The addition of prepared inner-membrane mitochondrial extracts to in vitro–transcribed/translated ORF239 concurrent with its translation (see Methods) resulted in the increase of ORF239 turnover (Table 1, Inner membrane fraction).

This result is consistent with our current observations of highly rapid ORF239 turnover in organello. Although such in vitro observations cannot be assumed a priori to reflect in vivo events without further investigation, our results suggest that the efficient turnover of the ORF239 substrate in vivo may be the consequence of cotranslational proteolysis or particular chaperonin activity that could not be duplicated in our in vitro assay. Thus, it is important to note that the results of proteolysis assays presented in Table 1 involving cotranslational turnover of ORF239 present results in terms of accumulation of ORF239 translation product rather than its degradation. Such an assay required the inclusion of additional controls that are detailed in Methods.

A Mitochondrial Protease Is Encoded in the Plant Nuclear Genome
The ATP-dependent protease activity associated with the inner mitochondrial membrane in CMS-Sprite has properties in common with the Escherichia coli–like Lon protease detected in human and yeast mitochondria (Gottesman 1989 ; Goldberg 1992 ; Wang et al. 1993 , Wang et al. 1994 ; Suzuki et al. 1994 ; Van Dyck et al. 1994 ). Based on this observation, we tested whether the plant protease might be a lon homolog. Using degenerate primers designed from the most conserved regions of the eukaryotic lon homologs, we were able to use polymerase chain reaction (PCR) to amplify a discrete product from the genome of Arabidopsis. The rationale for the amplification of the Arabidopsis sequence, rather than the common bean sequence, was based on the amenability of Arabidopsis to subsequent transformation and mutagenesis experiments. Furthermore, antibodies raised against the Arabidopsis gene product are likely to cross-react with the common bean product because Lon homologs are highly conserved.

The 1.8-kb degenerate PCR amplification product from Arabidopsis (data not shown) was cloned and used to probe an Arabidopsis cDNA library. A selected cDNA clone of 3.1 kb was identified and sequenced to confirm homology to the LON genes previously identified in other species. A rapid amplification of cDNA ends (RACE) and PCR procedure was used to obtain the 5' end of the gene. Multiple primers were used in the RACE-PCR amplifications to ensure that the complete 5' sequence had been obtained (data not shown). The lon-homologous sequence from Arabidopsis has GenBank accession number ATU88087.

Sequence analysis identified an ORF of 2825 bp encoding a polypeptide of 110 kD. Figure 3 shows that the polypeptide has a significant amino acid sequence similarity to the E. coli Lon protease (61.1%) and its homolog in yeast (64.3%) and to a recently identified sequence in maize (75%). The first ATG after the start of transcription presents a conserved Kozak-like sequence (Kozak 1986 ) at this site and also marks the beginning of a putative mitochondrial transit sequence of 55 amino acids. A well-conserved ATP binding domain (positions 464 to 471) and a catalytic domain (positions 839 to 847) with the requisite serine residue (position 842) also demonstrate high sequence conservation (Figure 3). No obvious membrane-spanning domains were evident in the sequence. Computer analysis with the PSORT program (Nakai and Kanehisa 1992 ) predicted most probable localization of the encoded gene product to the mitochondrial matrix with a discrimination value of 0.614; a discrimination value of 0.317 was obtained for localization to the mitochondrial inner membrane, intermembrane space, and outer membrane. These results also support our hypothesis that a mitochondrial lon homolog is encoded within the plant nuclear genome.

View larger version (66K): [in this window] [in a new window]   Figure 3. Amino Acid Sequence Comparison of the Arabidopsis Lon Homolog (AtLON) with Products of Other Previously Cloned lon Genes. Sequence of a full-length cDNA clone (3331 bp; GenBank accession number ATU88087) reveals an ORF that encodes 941 amino acid residues. The deduced amino acid sequence was aligned with Lon homologs from maize (Zmays2), yeast, and E. coli. The AtLON peptide contains highly conserved ATP binding and catalytic domains (underlined). The serine residue (asterisk), which is typical of this serine protease, is also present in the catalytic domain. Black boxes indicate amino acid identity in relation with AtLON. Spaces (no amino acid) optimize alignment.The maize sequence, with GenBank accession number U85495, has been reported by Barakat et al. 1998 .

The selected cDNA clone representing a putative lon homolog from Arabidopsis (AtLON) was used to probe the common bean genome. DNA gel blot analysis with moderately stringent conditions (described in Methods) allowed us to detect hybridization to what appeared to be a single-copy locus in the common bean genome with several restriction enzymes (data not shown). To confirm that the hybridizing band(s) in common bean represented a single locus, we mapped the locus based on two distinct restriction fragment length polymorphisms (RFLPs) detected between parental bean lines Calima (C) and Jamapa (J) with the enzymes HaeIII (fragments of 6.6 kb [C] and 8.0 kb [J]) and HpaII (fragments of 10.7 kb [C] and 9.1 kb [J]). Both RFLPs mapped to linkage group A, ~1.4 centimorgans (LOD 17.7) away from marker Bng23 on the common bean map developed by Vallejos et al. 1992 . Mapping was conducted using a recombinant inbred population of 76 lines. These data confirm that a common bean lon-homologous gene is present as a single-copy locus. Moreover, Figure 4 shows the results of RNA gel blot hybridization experiments using the homologous cDNA clone as a probe. The transcripts detected were of the expected lengths of 3.2 kb in Arabidopsis and 3.1 kb in common bean leaf tissues. This observation further confirmed that we had identified the full-length cDNA sequence from Arabidopsis.

View larger version (50K): [in this window] [in a new window]   Figure 4. Expression of a lon Homolog in Common Bean and Arabidopsis. RNA gel blot analysis of total mRNA prepared from Arabidopsis (5 µg) and CMS-Sprite (10 µg) leaf tissues demonstrates hybridization with 3.2- and 3.1-kb transcripts, respectively, when probed with the cloned AtLON sequence. Numbers at left indicate molecular lengths in kilobases.

Evidence That Some Portion of Mitochondrial Proteolytic Activity against ORF239 Is Encoded by the AtLON Gene
In vitro transcription and translation reactions were conducted with the cloned AtLON sequence to test for proteolytic activity. The in vitro–transcribed/translated AtLON product was incubated with both resorufin-labeled casein and in vitro–transcribed/translated ³⁵S-ORF239 substrates to characterize activity relative to that observed using bean mitochondrial extracts. Figure 5 and Table 1 (see IVT AtLON) demonstrate that similarities were observed between the IVT AtLON product and mitochondrial fractions in ATP dependence, response to different inhibitors, casein degradation, and cotranslational turnover of the ORF239 substrate.

View larger version (14K): [in this window] [in a new window]   Figure 5. Proteolytic Activity Is Detected in in Vitro–Transcription/Translation Products from the Cloned Plant LON Homolog. Incubation of resorufin-labeled casein substrate (90 µg per reaction) with in vitro–transcribed/translated LON prepared using the full-length AtLON clone (10 µL of product per reaction) demonstrated proteolytic activity. Substrate turnover was monitored spectrophotometrically as absorbance of TCA-nonprecipitable peptide. The addition of protease inhibitors resulted in a pattern of response similar to that observed using mitochondrial extracts (see Figure 2). Note that proteolytic activity observed by using in vitro–transcribed/translated LON in the labeled-casein assay was significantly lower than was the activity detected using mitochondrial extracts; this is most likely accounted for, in part, by difficulties in accurately quantifying the protease present in the in vitro–transcribed/translated AtLON preparations versus mitochondrial extracts. The values used correspond to the average of two experiments, and the bars in each time point correspond to the standard error.

It was noted that activity of the LON in vitro–transcribed/translated product was much lower than was the activity detected in mitochondrial extracts. In vitro proteolysis assays using in vitro–transcribed/translated protease product have not been reported previously using a eukaryotic Lon homolog, and in vitro assays have been problematic using the E. coli lon gene (Zehnbauer and Markovitz 1980 ; Goff and Goldberg 1987 ; Sonezaki et al. 1994 ); thus, we have limited basis for comparison with other systems. It is reasonable to assume that important mitochondrial components are absent from our in vitro reaction. Furthermore, it was not possible to quantify accurately the amounts of protease present in mitochondrial extracts relative to that produced in vitro. These factors most likely contribute to the lowered activity observed. Nevertheless, our results confirm that the gene cloned encodes a protease, and its similarity in behavior to mitochondrial preparations supports our hypothesis that the cloned gene contributes at least a portion of the protease activity detected in mitochondrial extracts.

Rabbit polyclonal antibodies were raised against an E. coli overexpression product derived from a partial clone of the AtLON sequence (representing bases 304 to 1330 of AtLON sequence in the database). Protein gel blot analysis confirmed cross-reactivity of the prepared antibodies (anti-AtLON) against the cloned overexpression product used as antigen (data not shown) and against an ~100-kD polypeptide in common bean mitochondrial protein preparations, as seen in Figure 6. Crude cell fractionations prepared from bean tissues by using differential centrifugation to nuclear, plastid, and mitochondrial fractions resulted in detection of >90% of the Lon homolog within the mitochondrial fraction (data not shown). Fractionation of highly purified mitochondrial preparations from common bean to matrix, inner membrane, and outer membrane components demonstrated that the cross-reacting protein was detected mainly in inner membrane fractions. Coincubation of mitochondrial inner membrane fractions with anti-AtLON antibodies during the assays for protease activity against in vitro–transcribed/translated ORF239 substrate resulted in an increase in accumulation of the ORF239 translation product, rising from 47 to 81% (Table 1). Incubation of the in vitro–transcribed/translated AtLON product with antibodies during incubation with in vitro–transcribed/translated ORF239 resulted in a similar increase in substrate accumulation, rising from 64 to 87% (Table 1). These results imply that antibodies against the homolog AtLON interfere with the protease activity detected in common bean mitochondria.

View larger version (33K): [in this window] [in a new window]   Figure 6. Protein Gel Blot Analysis of Fractionated Common Bean Mitochondrial Proteins. Mitochondria from etiolated CMS-Sprite seedlings were purified over Percoll gradients and fractionated by gentle detergent treatment, and the proteins were separated by SDS-PAGE. Protein gel blot analysis was performed with anti-AtLON polyclonal antibodies prepared against the overexpression product of the cloned AtLON homolog. A protein of ~100 kD in the inner membrane (IM) fraction was detected. Note that a small amount of the protein is detectable in matrix (MAT) and outer membrane/inner membrane space (OM) fractions as well. This is assumed to be the consequence of slight cross-contamination of the fractions during preparation.

To test this interaction more directly, competition experiments were conducted by preincubating anti-AtLON antibodies with the in vitro–transcribed/translated AtLON antigen before their addition to the protease reaction. The addition of the preincubated anti-AtLON antibodies to the protease reaction had measurably less influence on proteolysis (Table 1), suggesting that the antibody interacts specifically with the mitochondrial protease. However, preincubation of the antibodies with the in vitro–transcribed/translated AtLON antigen did not eliminate completely the effect of the antibodies; this is likely due to the fact that antibodies were raised against a truncated form of the antigen, minus the active site, so that antibody-to-antigen cross-reactivity was less efficient. This assumption is based on the similar results obtained with both mitochondrial extracts and in–vitro transcribed/translated AtLON proteolysis assays. These experiments provide further evidence that the AtLON homolog that was cloned is responsible, at least in part, for the proteolytic activity detected in mitochondrial extracts.

	DISCUSSION

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION METHODS REFERENCES

Tissue-specific expression of plant mitochondrial genes has not been reported widely, although some recent evidence does suggest possible transcriptional modulation in different tissues (Li et al. 1996 ). To our knowledge, the common bean pvs-or f 239 sequence is the only demonstrated example of a sequence regulated post-translationally or cotranslationally in plant mitochondria. Notwithstanding, protein turnover likely represents a significant mode of mitochondrial gene regulation in higher plants, based on the following observations.

Over the past several years, it has become increasingly evident that the majority of plant mitochondrial genes produce transcripts that require editing before proper translation (Hiesel et al. 1994 ). Only recently, however, has it become clear that at least some of these transcripts are translated before editing (Lu et al. 1996 ). Because these aberrant translation products do not appear to accumulate in the mitochondrion, it is likely that they are targeted for rapid degradation. Likewise, the plant mitochondrial genome is unique in its propensity to undergo intergenic and intragenic recombination as well as DNA insertions and rearrangements (Wolstenholme and Fauron 1995 ). Several examples exist whereby such genomic rearrangements generate unusual ORFs that are translated (Dewey et al. 1987 ; Nivison and Hanson 1989 ; Krishnasamy and Makaroff 1994 ; Abad et al. 1995 ). The pvs-or f 239 appears to be such a sequence, and proteolysis is an important means of regulating its expression.

Another example of post-translational regulation in mitochondria is its influence on the coordination of nuclear-mitochondrial gene expression. It is well known that mitochondrially encoded subunits of the electron transport apparatus, when expressed in the absence of their nuclear-encoded counterparts, demonstrate unusually rapid turnover (reviewed in Rep and Grivell 1996 ). The question we have not yet addressed is whether the protease that we have characterized here is involved in all of these necessary protein turnover functions.

The Lon protease in E. coli is an ATP-dependent, serine-type protease that is responsible for much of the protein turnover that occurs due to mistranslation or misassembly of proteins (Chung and Goldberg 1981 ). Homologous genes in eukaryotes, bearing significant similarity to the E. coli lon sequence, have been cloned from the human and yeast genomes and demonstrated to encode a mitochondrial protease. The function of this protease in eukaryotes and its natural substrates have not yet been fully determined. In yeast, the mitochondrial Lon homolog has been shown to be essential to respiratory function, and its loss results in a petite phenotype (Suzuki et al. 1994 ; Van Dyck et al. 1994 ). In the cases of human and yeast, LON protease activity is present in matrix fractions (Suzuki et al. 1994 ; Van Dyck et al. 1994 ; Wang et al. 1994 ). Consequently, the association of the plant Lon homolog with the inner mitochondrial membrane was unexpected. Upon further analysis, however, this observation is, perhaps, fully in keeping with its apparent capability to target substrates for cotranslational degradation. Our data in support of cotranslational activity are based strictly on in vitro analyses, and it will be technically challenging to determine definitively whether the plant LON protease acts cotranslationally in vivo. If our observations do mimic in vivo activity, the cotranslational turnover of ORF239 would be unique to the plant mitochondrial LON protease. Because ribosomes localize to the inner mitochondrial membrane (Spithill et al. 1978 ; Marzuki and Hibbs 1986 ; McMullin and Fox 1993 ), the observed membrane association most likely would facilitate its activity in association with active ribosomes.

An alternative interpretation of our results, however, would implicate a necessary chaperonin function missing from our assay. It is clear that LON-mediated proteolysis is dependent on secondary structure of the substrate (reviewed in Gottesman et al. 1997 ). The possibility exists that our in vitro reaction conditions do not supply adequately the necessary chaperonin activity. Thus, the observed cotranslational activity in vitro would simply signal the necessity of proper folding to the reaction.

We have identified only one component of the plant mitochondrial proteolysis system, and we do not assume this to be the only protease present. However, others investigating plant mitochondrial proteases that are involved in protein turnover have detected little or no activity in mitochondrial matrix fractions; nearly all activity reported to date has been localized to the inner membrane fractions (Knorpp et al. 1995 ). Furthermore, it is clear that activity in plant mitochondrial preparations is difficult to detect using in vitro assays. It is for this reason that we have presented data from proteolysis assays by using both resorufin-labeled casein, a highly sensitive assay for substrate turnover, and in vitro–transcribed/translated ORF239. The low activity detected, relative to that reported in human or yeast systems tested, may be the consequence of less than optimal reaction conditions coupled with the unique features of the plant mitochondrial LON protease.

The results of this study demonstrate that plant mitochondria contain a LON-like protease that is present in mitochondria from vegetative tissues and is able to degrade the mitochondrial sterility-associated mutant protein ORF239. If our observations account, at least in part, for the absence of ORF239 in seedling tissues, what remains to be deduced are the events occurring during microsporogenesis. During pollen mother cell development, the ORF239 protein is apparently stable, accumulating within the callose layer and the primary cell wall. We have not yet determined whether the LON protease is expressed or active during pollen mother cell development. Although protein gel blot analysis of proteins prepared from young, premeiotic and postmeiotic buds indicates that LON is present (data not shown), it is not feasible to achieve sufficient resolution by using immunocytological techniques to evaluate expression of the protease in the pollen mother cell itself. We currently are implementing transgenic analyses to examine in more detail the pattern of lon promoter activity during floral development. However, our observation of tissue-specific differences in AtLON transcript levels does support the notion of developmental influence on expression.

Other nuclear suppressors of pvs-or f 239 expression, known as fertility restorer genes, have been identified in common bean. One, designated Fr2 (Mackenzie 1991 ), most likely acts post-transcriptionally, as does the AtLON homolog; transcription of the pvs-or f 239 region appears to be unaffected in an Fr2-restored line (Chase 1994 ), but the ORF239 protein is no longer present in developing anthers (Abad et al. 1995 ). Therefore, one question raised during the course of this study was whether the lon homolog and Fr2 might, in fact, be one and the same. Genetic mapping of the lon homolog indicated that the locus present on linkage group A of the bean genome is not allelic to Fr2, which is located on linkage group K (Jia et al. 1997 ). However, we have not excluded the possibility that Fr2 also might regulate pvs-or f 239 expression post-translationally or perhaps interact with the Lon homolog to activate proteolysis specifically during anther development. These investigations continue.

	METHODS

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION METHODS REFERENCES

Mitochondrial Preparations and Fractionations
Mitochondria were isolated from 10-day-old, dark-grown cytoplasmic male sterility (CMS)–Sprite (Phaseolus vulgaris) seedlings, as reported by Forde et al. 1978 , using a 26% Percoll gradient for purification. Mitochondria were resuspended in one-tenth of the final pellet volume (approximate protein concentration of 40 µg/µL). Subfractionation was conducted using detergent treatment, as described by Ragan et al. 1987 . More extensive detergent treatment of mitoplast was performed by extending lubrol incubation time to 15 min or by increasing lubrol concentration to 0.2 mg/mL.

DNA Cloning, Library Screenings, and Polymerase Chain Reactions
Degenerate primers 5'-ATHT TRGAYGARGAYCAYTAYGG-3' and 5'-GCTYGGYCCGTCCT TYGGRGT-3' were used to amplify a conserved region of the lon protease gene. (H, R, and Y designate variable nucleotide sites.) The 1.8-kb fragment obtained was cloned in the PCRII vector (Invitrogen, Carlsbad, CA). BLAST analysis was performed to determine identity of the fragment. This fragment was used subsequently to screen a PRLZ, Ziplox derivative (Bethesda Research Laboratories, Gaithersburg, MD; see Focus, volume 14, pages 76 to 79) Arabidopsis thaliana (ecotype Columbia, nonglaborous) cDNA library that was provided by the Arabidopsis Research Center (ABRC, Columbus, Ohio). The library contains SalI-NotI cDNA inserts and was generated by mixing equal amounts of mRNA from several plant tissues and stages.

The 3.1-kb Ara12 clone was isolated and excised as described by D'Alessio et al. (Bethesda Research Laboratories; Focus, volume 14, pages 76 to 79) and subcloned for double-strand DNA sequence analysis. The insert from Ara12 was cloned into the pBS+ vector (Stratagene, La Jolla, CA) as an SalI-XbaI fragment (Aralon12). The missing 5' end portion of the gene was obtained by using the 5' and 3' rapid amplification of cDNA ends (RACE) kit from Boehringer Mannheim. Primers 5'-TGAAATCTGAGCAAGTGTGCCAAC-3' (SP1), 5'-TCAGCTGATGGGTCATCCT TCAGA-3' (SP2), and 5'-CTATGTGGCACCGGCAATGCTAGA-3' (SP3), located 365, 276, and 136 bp, respectively, from the 5' end of Aralon12, were used. A 450-bp fragment was obtained, cloned in the PCRII vector (clone 210F), and sequenced.

To determine the accuracy of the sequence obtained at the new 5' end, a second RACE reaction was performed using SP3, and two additional upstream primers designed from the 210F sequence were obtained. They are 5'-TCTCCAT TGGTAGGCTCAGA-3' (SP4) and 5'-AGGCAAGAAACCAT TGGAG-3' (SP5). Primers SP4 and SP5 were located 300 and 50 bp from the 5' end of clone 210F, respectively; SP4 spans the 3' end of clone 210F and the 5' end of clone Ara12. The ATG start codon for the prepeptide was bordered by a highly conserved Kozak sequence (Kozak 1986 ). Prediction of a putative mitochondrial targeting sequence was made using the PSORT program (World Wide Web) by employing the discriminant analysis from values of partial amino acid composition (Nakai and Kanehisa 1992 ).

The clone designated lon, in which most of the putative transit sequence was not included, was made by amplifying the sequence between the primers 5'-GGGCATGCCT TCTCCAGTGGAGTCTCC-AT T-3' and SP3 from clone 210F. An SphI site was introduced at the 5' end of the upstream primer. The polymerase chain reaction (PCR) product was then SphI-BlnI digested and cloned into the Aralon12 construct. Primer 5'-GAAT TCGAGCCTACCAATGGAGAGGCGGCG-3' (ExplonF) containing an EcoRI site and primer 5'-GTCGACTCA-CGT TAAACTCGCT TGA-3' (ExplonR) containing an SalI site were used to amplify from 304 to 1330 bp of the gene sequence. The PCR product was EcoRI-SalI digested and cloned into the pMALp2 expression vector (New England Biolabs, Beverly, MA). The clone generated was named Explon. Pvs-or f 239 was cloned as a KpnI-SacI fragment into pBluescript SK- (Stratagene).

DNA sequencing was done using the ALFexpress DNA sequencer system from Pharmacia Biotechnology. Sequence analysis was conducted using Expassy tools (World Wide Web). All PCR reactions were performed in an Amplitron I thermocycler (Thermoline, Dubuque, IA). Pwo DNA polymerase used for the PCR reactions was purchased from Boehringer Mannheim.

RNA Gel Blot Analysis
Plant total RNA extractions were performed as described by Pawlowski et al. 1994 . Samples (5 to 10 µg per lane) were separated in a formaldehyde-containing 1% agarose gel, as described by Sambrook et al. 1989 . Hybridization buffer was 0.25 M NaHPO₄, pH 7.2, 7% SDS, and 1 mM EDTA. Hybridization conditions were 65°C for 16 hr, followed by two washes for 20 min each with 20 mM NaHPO₄, pH 7.2, 1% SDS, and 1 mM EDTA at 65°C.

Linkage Analysis
A recombinant inbred family (Burr et al. 1988 ) between the common bean Mesoamerican cultivar Jamapa and the Andean breeding line Calima was used as the mapping population. This population consisted of 76 lines. Restriction fragment length polymorphism (RFLP) analysis (DNA extraction, DNA gel blotting, probe labeling, hybridizations, and autoradiography) was conducted essentially as described earlier (Vallejos et al. 1992 ). Hybridizations with AtLON were conducted at 50°C for 16 hr; after hybridization, the blots were washed successively for 15 min at 50°C in 2 x, 1 x, and 0.1 x SSPE (1 x SSPE is 0.15 M NaCl, 10 mM sodium phosphate, and 1 mM EDTA, pH 7.4) containing 0.1% SDS. MAPMAKER 3.0 (Lander et al. 1987 ; Lincoln et al. 1992 ) was used to analyze the linkage relationships of the AtLON homolog with 36 previously mapped DNA marker loci distributed throughout the common bean genome (Vallejos et al. 1992 ; Boutin et al. 1995 ).

In Vitro Transcription and Translation
Clones AtLon and orf239 were translated using the coupled transcription/translation (TNT) reticulocyte lysate system (Promega, Madison, WI). DNA template (1 µg) orf239 was transcribed using T7 RNA polymerase, and AtLon was transcribed using T3 RNA polymerase. Incubation time was 90 min. Translation products were analyzed by PAGE, immunoblotting, or trichloroacetic acid (TCA)–precipitable counts (Mans and Novelli 1961 ). Mitochondrial RNA was translated using the reticulocyte lysate system (Promega).

Protein Gel Electrophoresis, Protein Gel Blotting, and Immunodetection
Samples were run in an SDS-PAGE discontinuous system, as described by Laemmli 1970 . Protein blotting was done using polyvinylidene difluoride membranes (Bio-Rad) in a tank system, as described by Gallagher et al. 1995 . Transfer buffer was 25 mM Trizma base, 19 mM glycine, and 15% [v/v] methanol.

Immunodetection was conducted using the Vectastain ABC alkaline phosphatase kit from Vector Laboratories (Burlingame, CA). Visualization was achieved by using 5-bromo-4-chloro-3-indolyl phosphate/nitro blue tetrazolium substrates from Vector Laboratories.

Protein Overexpression and Antibody Production
Overexpression isolation and purification of Explon as a maltose binding protein fusion were done following the manufacturer's recommendations (New England Biolabs). After factor Xa cleavage, 100 µg of the purified AtLON peptide in complete Freund's adjuvant was injected into a rabbit, followed by a second dose 3 weeks later. After three inoculation boosters, the antiserum was used to determine the titer and specificity of the antibodies against the in vitro–translated AtLON protease in protein gel blot assays. Preimmune serum was taken before the first AtLON inoculation and used as a control in titrations of the anti-AtLON antibodies. The IgG fraction was purified from anti-AtLON and anti-ORF239 antisera and preimmune sera by using the Affi-Gel protein A MAPS II system from Bio-Rad. The ORF239 antibody production was described previously (Abad et al. 1995 ).

In Organello Translations
Mitochondrial translations were done as described by Forde et al. 1978 without modifications. Total product of translation was used for immunoprecipitation of ORF239 products. In organello translations were performed with and without vanadate (4 mM), o-phenanthroline (5 mM), and N-ethylmaleimide (NEM) (2 mM).

Immunoprecipitations
After in organello translations, mitochondria were treated with digitonin and lubrol, as described by Ragan et al. 1987 , without fractionation. Antiserum (6 µL) was added before incubation for 1 hr at room temperature. Subsequently, 15 µL of Affi-Gel protein A–agarose resin (Bio-Rad) was added to the mix and incubated for 1 hr at 4°C. Samples were centrifuged at 1500g for 1 min and washed twice with binding buffer (Bio-Rad). Samples were resuspended in 20 µL of Laemmli sample buffer (Laemmli 1970 ) and boiled at 85°C for 10 min. After a 15-sec centrifugation in an Eppendorf microcentrifuge, supernatants were loaded onto the gel. Amplify (Amersham, Arlington Heights, IL) treatment of the gel was performed before gel drying and exposure.

Proteolysis Assays
Resorufin-labeled casein (Boehringer Mannheim) was used in proteolysis assays. The labeled casein (90 µg) in 100 mM Tris-HCl, pH 7.5, 2 mM ATP, and 5 mM MgSO₄ was used in each reaction for proteolysis. Sources of the LON protease were either 10 µL of in vitro translation product or 10 µg of mitochondrial protein from inner membrane or matrix fractions. Inhibitors used were NEM (2 mM), vanadate (4 mM), o-phenanthroline (5 mM), and phenylmethylsulfonyl fluoride (PMSF) (4 mM) as well as apyrase (2 units). Incubations were performed at 37°C for 2 hr. Samples were taken every 30 min to monitor degradation reactions. Determination of proteolytic activity was done as recommended by Boehringer Mannheim. Absorbance of the TCA-nonprecipitable peptides was measured in a Beckman Instruments (Chicago, IL) DU640 spectrophotometer.

Proteolytic activity of the in vitro–translated lon clone and the inner membrane and matrix fractions against the ORF239 substrate was determined by adding the prepared extracts to an ORF239 in vitro translation reaction. The in vitro–translated lon clone (10 µL) or 10 µg of protein from a mitochondrial fraction was incubated with 25 µL of TNT reticulocyte lysate (Promega), 2 µL of TNT buffer, 1 µL of an amino acid mixture (1 mM) minus cysteine, 4 µL of ³⁵S-cysteine at 10 mCi/mL (Amersham), 40 units of RNasin, 1 µL of T7 polymerase, 100 mM Tris-HCl, pH 7.5, 5 mM MgSO₄, 2 mM ATP, and 1 µg of cloned orf239 DNA. Inhibitors were used at concentrations indicated above, and incubation time was 90 min, with samples being taken every 30 min. Post-translational turnover of ORF239 was monitored by incubating 10 µL of in vitro–translated Aralon or 10 µg of a mitochondrial fraction with 5 µL of ³⁵S-cysteine–labeled in vitro–translated ORF239. The incubation was in the presence of 100 mM Tris-HCl, pH 7.5, 5 mM MgSO₄, and 2 mM ATP. Inhibitor concentrations were as indicated above. Reactions were sampled every 30 min for a total incubation time of 90 min.

To conduct assays for cotranslational turnover of the ORF239 substrate, it was necessary to demonstrate that the mitochondrial inner membrane protease activity did not target components of the reticulocyte lysate necessary for the translation of ORF239. A control experiment was run in which the transcription/translation reaction mix was preincubated with the inner membrane mitochondrial fraction for 60 min at 30°C in the absence of the Pvs-or f 239 clone and the protease inhibitor PMSF (4 mM). The treatment minus the orf239 clone and PMSF was subsequently incubated for 90 min at 30°C in the presence of the orf239 clone and PMSF (4 mM) and then analyzed by PAGE and autoradiography to confirm accumulation of the ³⁵S-cysteine–labeled ORF239 peptide and little or no proteolytic activity. A second control reaction was performed as described above, except that the mitochondrial fraction was omitted; this produced an equal accumulation of the ORF239 peptide. In a third treatment, the orf239 clone was added at the start of the incubation, but no PMSF was included during the incubation to confirm that proteolytic activity was present in the mitochondrial fractions used. As expected, the amount of ³⁵S-cysteine–labeled ORF239 accumulated in this treatment was considerably lower and proportional to what is presented in Table 1.

In all proteolysis assays that involved the addition of anti-AtLON antibodies or preimmune serum, 3 µL of purified IgG fraction was used for incubation at room temperature with LON-containing extracts 30 min before the initiation of the proteolysis reactions. In competition experiments, the in vitro–transcribed/translated AtLON product was warmed to 65°C for 5 min before the addition of the antibody. In all assays using ³⁵S-cysteine–labeled ORF239, TCA-precipitable counts were determined using the procedure by Mans and Novelli 1961 . Duplicated samples for every time point were included and read in an LS6500 Beckman scintillation counter.

Proteolysis data with the respective standard errors were plotted using the Sigmaplot program from Jandel Corporation (San Rafael, CA).

	ACKNOWLEDGMENTS

We gratefully acknowledge Dr. Andre Abad for his important contributions at the beginning of this study and Dr. Mark Hermodson for his critical review of the manuscript. We also thank Ms. Joanne Cusumano and Dr. Clinton Chapple for their technical assistance and prepared Arabidopsis materials for RNA gel blot analysis. Antibody production was conducted through the Purdue Cancer Center Antibody Production Facility, and we thank John Wilder for his assistance. This work was supported by a National Science Foundation grant (No. 9630252-MCB) to S.A.M.

Received December 29, 1997; accepted April 16, 1998.

	REFERENCES

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION METHODS REFERENCES

Abad, A.R., Mehrtens, B.J., and Mackenzie, S.A. (1995) Specific expression in reproductive tissues and fate of a mitochondrial sterility-associated protein in cytoplasmic male-sterile bean. Plant Cell 7:271-285[Abstract].

Barakat, S., Pearce, D.A., Sherman, F., and Rapp, W.D. (1998) Maize contains a Lon protease gene that can partially complement a pim1-deletion mutant. Plant Mol. Biol. 37:141-154[CrossRef][Medline].

Bonen, L., and Brown, G.G. (1993) Genetic plasticity and its consequences: Perspectives on gene organization and expression in plant mitochondria. Can. J. Bot. 71:645-660.

Boutin, S., Young, N., Olson, T., Yu, Z.H., Shoemaker, R., and Vallejos, C.E. (1995) Genome conservation among legume genera detected with DNA markers. Genome 38:928-937.

Burr, B., Burr, R.A., Thompson, K.H., Albertsen, M.C., and Tuber, C.W. (1988) Gene mapping with recombinant inbreds in maize. Genetics 118:519-526[Abstract/Free Full Text].

Chase, C.D. (1994) Expression of CMS unique and flanking mitochondrial DNA sequences in Phaseolus vulgaris L. Curr. Genet. 25:245-251[CrossRef][Medline].

Chung, C.H., and Goldberg, A.L. (1981) The product of the lon (capR) gene in Escherichia coli is the ATP-dependent protease, protease La. Proc. Natl. Acad. Sci. USA 78:4931-4935[Abstract/Free Full Text].

Dewey, R.E., Timothy, D.H., and Levings, C.S., III (1987) A mitochondrial protein associated with cytoplasmic male sterility in the T cytoplasm of maize. Proc. Natl. Acad. Sci. USA 84:5374-5378[Abstract/Free Full Text].

Forde, B.G., Olivier, R.J.C., and Leaver, C.J. (1978) Variation in mitochondrial translation products associated with male-sterile cytoplasms in maize. Proc. Natl. Acad. Sci. USA 75:3841-3845[Abstract/Free Full Text].

Fox, T.D. (1996). Genetics of mitochondrial translation. In Translational Control, J.W.B. Hershey, M.B. Mathews, and N. Sonenberg, eds (Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press), pp. 733–758.

Gallagher, S., Winston, S.E., Fuller, S.A., and Hurrell, J.G.R. (1995). Immunoblotting and immunodetection. In Current Protocols in Molecular Biology, F.M. Ausubel, R. Brent, R.E. Kingston, D. Moore, J.G. Seidman, J.A. Smith, and K. Struhl, eds (New York: John Wiley Interscience), pp. 10.8.7–10.8.21.

Goff, S.A., and Goldberg, A.L. (1987) An increased content of protease La, the lon gene product, increases protein degradation and blocks growth in Escherichia coli.. J. Biol. Chem. 262:4508-4515[Abstract/Free Full Text].

Goldberg, A. (1992) The mechanism and functions of ATP-dependent proteases in bacterial and animal cells. Eur. J. Biochem. 203:9-23[ISI][Medline].

Gottesman, S. (1989) Genetics of proteolysis in Escherichia coli.. Annu. Rev. Genet. 23:163-198[CrossRef][ISI][Medline].

Gottesman, S., Wickner, S., and Maurizi, M.R. (1997) Protein quality control: Triage by chaperones and proteases. Genes Dev. 11:815-823[Free Full Text].

Hanson, M.R. (1991) Plant mitochondrial mutations and male sterility. Annu. Rev. Genet. 25:461-486[CrossRef][ISI][Medline].

He, S., Abad, A.R., Gelvin, S.B., and Mackenzie, S.A. (1996) A cytoplasmic male sterility–associated mitochondrial protein causes pollen disruption in transgenic tobacco. Proc. Natl. Acad. Sci. USA 93:11763-11768[Abstract/Free Full Text].

Hiesel, R., Combettes, B., and Brennicke, A. (1994) Evidence of RNA editing in mitochondria of all major groups of land plants except the Bryophyta. Proc. Natl. Acad. Sci. USA 91:629-633[Abstract/Free Full Text].

Jia, M., He, S., Vanhouten, W., and Mackenzie, S. (1997) Nuclear fertility restorer genes map to the same linkage group in cytoplasmic male sterile bean. Theor. Appl. Genet. 95:205-210[CrossRef].

Johns, C., Lu, M., Lyznik, A., and Mackenzie, S. (1992) A mitochondrial DNA sequence is associated with abnormal pollen development in cytoplasmic male sterile bean plants. Plant Cell 4:435-449[Abstract/Free Full Text].

Knorpp, C., Azigyarto, C., and Glaser, E. (1995) Evidence for a novel ATP-dependent membrane-associated protease in spinach leaf mitochondria. Biochem. J. 310:527-531.

Kozak, M. (1986) Bifunctional messenger RNAs in eukaryotes. Cell 47:481-483[CrossRef][Medline].

Krishnasamy, S., and Makaroff, C.A. (1994) Organ-specific reduction in the abundance of a mitochondrial protein accompanies fertility restoration in cytoplasmic male sterile radish. Plant Mol. Biol. 26:935-946[Medline].

Laemmli, U.K. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:680-685[CrossRef][Medline].

Lander, E.S., Green, P., Abrahamson, J., Barlow, A., Daly, M.J., Lincoln, S.E., and Newburg, L. (1987) MAPMAKER: An interactive computer package for constructing primary genetic linkage maps of experimental and natural populations. Genomics 1:174-181[CrossRef][Medline].

Li, X.-Q., Zhang, M., and Brown, G.G. (1996) Cell-specific expression of mitochondrial transcripts in maize seedlings. Plant Cell 8:1961-1975[Abstract].

Lincoln, S.E., Daly, M., and Lander, E.S. (1992). Constructing Genetic Maps with MAPMAKER/EXP 3.0, 3rd ed. (Cambridge, MA: Whitehead Institute).

Lu, B., Wilson, R., Phreaner, C.G., Mulligan, R.M., and Hanson, M.R. (1996) Protein polymorphism generated by differential RNA editing of a plant mitochondrial rps12 gene. Mol. Cell. Biol. 16:1543-1549[Abstract].

Mackenzie, S. (1991) Identification of a sterility-inducing cytoplasm in a fertile line of Phaseolus vulgaris L. Genetics 127:411-416[Abstract].

Mans, R.J., and Novelli, G.D. (1961) Measurement of the incorporation of radioactive amino acids into protein by filter-paper disk method. Arch. Biochem. Biophys. 9:48-53.

Marzuki, S., and Hibbs, A.R. (1986) Are all mitochondrial translation products synthesized on membrane-bound ribosomes? Biochim. Biophys. Acta 866:120-124[Medline].

McMullin, T.W., and Fox, T.D. (1993) COX3 mRNA-specific translational activator proteins are associated with the inner mitochondrial membrane in Saccharomyces cerevisiae.. J. Biol. Chem. 268:11737-11741[Abstract/Free Full Text].

Nakai, K., and Kanehisa, M. (1992) Acknowledge base for predicting protein localization sites in eukaryotic cells. Genomics 14:897-911[CrossRef][ISI][Medline].

Nivison, H.T., and Hanson, M.R. (1989) Identification of a mitochondrial protein associated with cytoplasmic male sterility in petunia. Plant Cell 1:1121-1130[Abstract/Free Full Text].

Pawlowski, K., Kunze, R., De Vries, S., and Bisseling, T. (1994). Isolation of total, poly(A) and polysomal RNA from plant tissues. In Plant Molecular Biology Manual, S.B. Gelvin and R.A. Shilperoort, eds (Dordrecht, The Netherlands: Kluwer Academic Publishers), pp. D5, 1–13.

Ragan, C.J., Wilson, M.T., Darley-Usmar, V.M., and Lowe, P.N. (1987). Subfractionation of mitochondria and isolation of proteins of oxidative phosphorylation. In Mitochondria: A Practical Approach, V.M. Darley-Usmar, D. Rickwood, and M.T. Wilson, eds (Washington, DC: IRL Press), pp. 79–112.

Rapp, W.D., and Stern, D.B. (1992) A conserved 11 nucleotide sequence contains an essential promoter element of the maize mitochondrial atp1 gene. EMBO J. 11:1065-1073[ISI][Medline].

Rapp, W.D., Lupold, D.S., Mack, S., and Stern, D.B. (1993) Architecture of the maize mitochondrial atp1 promoter as determined by linker-scanning and point mutagenesis. Mol. Cell. Biol. 13:7232-7238[Abstract/Free Full Text].

Rep, M., and Grivell, L.A. (1996) The role of protein degradation in mitochondrial function and biogenesis. Curr. Genet. 30:367-380[CrossRef][ISI][Medline].

Sambrook, J., Fritsch, E.F., and Maniatis, T. (1989). Extraction, purification, and analysis of messenger RNA from eukaryotic cells. In Molecular Cloning: A Laboratory Manual (Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press), pp. 7.43–7.45.

Sonezaki, S., Kondo, A., Oba, T., Ishii, Y., Kato, Y., and Nakayama, H. (1994) Overproduction and purification of lon protease from Escherichia coli using a maltose-binding protein fusion system. Appl. Microbiol. Biotechnol. 42:313-318[ISI][Medline].

Spithill, T.W., Trembath, M.K., Lukins, H.B., and Linnane, A.W. (1978) Mutations of the mitochondrial DNA of Saccharomyces cerevisiae which affect the interaction between mitochondrial ribosomes and the inner mitochondrial membrane. Mol. Gen. Genet. 164:155-162[CrossRef][Medline].

Suzuki, C.K., Suda, K., Wang, N., and Schatz, G. (1994) Requirement for the yeast gene LON in intramitochondrial proteolysis and maintenance of respiration. Science 264:273-276[Abstract/Free Full Text].

Vallejos, C.E., Sakiyama, N., and Chase, C. (1992) A molecular marker–based linkage map of Phaseolus vulgaris L. Genetics 131:733-740[Abstract].

Van Dyck, L., Pearce, D.A., and Sherman, F. (1994) PIM1 encodes a mitochondrial ATP-dependent protease that is required for mitochondrial function in the yeast Saccharomyces cerevisiae.. J. Biol. Chem. 269:238-242[Abstract/Free Full Text].

Van Melderen, L., Thi, M.H.D., Lecchi, P., Gottesman, S., Couturier, M., and Maurizi, M.R. (1996) ATP-dependent degradation of CcdA by Lon protease: Effects of secondary structure and heterologous subunit interactions. J. Biol. Chem. 271:27730-27738[Abstract/Free Full Text].

Wagner, I., Arlt, H., Van Dyck, L., Langer, T., and Neupert, W. (1994) Molecular chaperones cooperate with PIM1 protease in the degradation of misfolded proteins in mitochondria. EMBO J. 13:5135-5145[ISI][Medline].

Wang, N., Gottesman, S., Willingham, M.C., Gottesman, M.M., and Maurizi, M.R. (1993) A human mitochondrial ATP-dependent protease that is highly homologous to bacterial Lon protease. Proc. Natl. Acad. Sci. USA 90:11247-11251[Abstract/Free Full Text].

Wang, N., Maurizi, M.R., Emmert-Buck, L., and Gottesman, M.M. (1994) Synthesis, processing and localization of human lon protease. J. Biol. Chem. 269:29308-29313[Abstract/Free Full Text].

Wolstenholme, D.R., and Fauron, C.M.-R. (1995). Mitochondrial genome organization. In The Molecular Biology of Plant Mitochondria, C.S. Levings III and I.K. Vasil, eds (Dordrecht, The Netherlands: Kluwer Academic Publishers), pp. 1–60.

Zehnbauer, B.A., and Markovitz, A. (1980) Cloning of gene lon (capR) of Escherichia coli K-12 and identification of polypeptides specified by the cloned deoxyribonucleic acid fragment. J. Bacteriol. 143:852-863[Abstract/Free Full Text].

This article has been cited by other articles:

				S. Fujii and K. Toriyama Genome Barriers between Nuclei and Mitochondria Exemplified by Cytoplasmic Male Sterility Plant Cell Physiol., October 1, 2008; 49(10): 1484 - 1494. [Abstract] [Full Text] [PDF]

				O. Ostersetzer, Y. Kato, Z. Adam, and W. Sakamoto Multiple Intracellular Locations of Lon Protease in Arabidopsis: Evidence for the Localization of AtLon4 to Chloroplasts Plant Cell Physiol., June 1, 2007; 48(6): 881 - 885. [Abstract] [Full Text] [PDF]

H. V. Tang, J. F. Pedersen, C. D. Chase, and D. R. Pring Fertility Restoration of the Sorghum A3 Male-Ster