|Table of Contents|
Table of Contents
List of Tables
List of Figures
Materials and methods
THE BIOSYNTHESIS OF MITOCHONDRIAL
GRETCHEN LYN SCHIEBER
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN
PARTIAL FULFILLMENT OF THE REQUIREMENTS
FOR THE DEGREE OF DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
In my many years in graduate school several people have had an
influence on my understanding and appreciation of both science and
life. I would like to acknowledge their contribution to this
In Dr. O'Brien's laboratory I have had the honor to work with
Bill Cattaneo, Robert Cheatham, Susan Collingwood, Mary Conde, Gary
Daniels, Jane Edwards, Terry Harville, Robert Hessler, Carolann
Johnson, Maria Kucharska, Mark Lijewski, Vince LiCata, Mietek
Piatyszek, Sal Pietromonaco, Joyce Plugge, Bobby Sheehan, and John
Turse. Dr. Nancy Denslow deserves special mention for her never
ending support, optimism, and scientific discussion.
The members of the department have all affected my graduate
training. I would like to note the useful discussion and advice and
the extended support of my supervisory committee, Dr. .Charles Allen,
Dr. Peter McGuire, and Dr. Parker Small. My committee chairperson,
Dr. Thomas O'Brien, has, perhaps, taught me more than even he
realizes. He has made an invaluable contribution to this work.
My brother and sister-in-law, Marc and Marsha, and especially my
parents, Don and Dorothy Schieber, have been a tireless source of
encouragement. They have provided the intellectual and emotional
foundation for this work.
To all of you, my profound thanks.
TABLE OF CONTENTS
ACKNOWLEDGEMENTS ............................................... ii
LIST OF TABLES ................................................ v
LIST OF FIGURES ............................................... vi
ABBREVIATIONS .................................................. vii
ABSTRACT ....................................................... ix
INTRODUCTION ................................................ ... 1
Background ............................................. 1
The Nucleo-Cytoplasmic Protein Synthetic Campartment and
Mitochondrial Import .......................... .. ...... 1
The Mitochondrial Gemcme .................. .......... 4
Mitochondrial Ribosomes ................................. 9
Mitochondrial Ribosome Biosynthesis ..................o 12
Objectives of this Work .......................... ....... 14
MATERIALS AND MET'HODS ............... ................. ......... 16
Materials ................................................... 16
Experiments Using Mitochondria Isolated from Bovire
Liver .............................. ... ... ..... ...... 18
Preparation of Bovine Liver Mitochondria ............... 18
Conditions for Mitochondrial Incubation ............. 19
Experiments Using Cells Grown in Tissue Culture ........... 21
Conditions for Incorporation of Radiolabeled
Methionine ................ ........................ 21
Preparation of Mitochondria from Cultured Cells ........ 25
Isolation of Mitochondrial Ribosomes ................... 26
Assay for Inhibitor Response .......................... 28
Inmmnunologic Methods ..................................... 29
Preparation of Antisera ......................... ........... 29
Preparation of Immune Precipitates ..................... 30
Electrophoretic Methods ................................o 31
Table of Contents continued
RESULTS ................................................ ........ 33
Experiments with Isolated Bovine Mitochondria ............. 33
Synthesis of Protein by Isolated Bovine Mitochondria.... 33
Analysis of Radiolabeled Mitochondrial Products by
T.D-Dimensional PAGE ............................... 35
Experiments with Cells in Culture ......................... 38
Mitoribosomal Proteins from Cells Labeled in the Absence
of Cytoplasmic Protein Synthesis ................... 38
Mitoribosomal Proteins from Cells Labeled in the Absence
of Mitochondrial Protein Synthesis ................. 42
Effect of Inhibitors .................................. 56
Precursors of Mitoribosomal Proteins ...................... 60
DISCUSSION ..................................................... 63
Protein Synthesis in Isolated Mitochondria ................ 64
Protein Synthesis in Cultured Bovine Cells ................ 67
Pools of Unassembled Mitoribosomal Proteins ............... 75
Implications for Mitochondrial Evolution and Biogenesis ... 77
REFERE CES ................. .................................9.... 80
BIOGRAPHICAL SKETCH ............................................ 91
LIST OF TABLES
1. Reading Frames of Mmmalian Mitochondrial DNA ............. 7
2. Solutions Cited .................o ...................... 17
3. In Vitro Mitochondrial Incubation Mixture ................. 20
4. Synthesis of Defined Mitoribosomal Proteins in the Presence
of Chloramphemicol ............................................ .. .. ... 52
LIST OF FIGURES
1. Electropherograms of the Proteins of the Bovine
Mitochondrial Ribosome .................................... 10
2. Schematic diagrams of the Proteins of the Bovine
Mitochondrial Ribosome ......................................... 11
3. Schematic representation of the labeling protocols for MDBK
cells ..................................................... 24
4. Sucrose density gradient profiles of the preparation of
mitochondrial ribosomes and ribosomal subunits ............ 27
5. Incorporation of [ H]leucine into protein by isolated
bovine liver mitochondria ............................................ 34
6. Synthesis of protein by isolated bovine mitochondria ....... 36
7. Fluorograms of mitoribosomal subunits from cells labeled in
the absence of cytoplasmic protein synthesis .............. 40
8. Fluorogram of mitoribosomal subunits from cells labeled in
the absence of cytoplasmic protein synthesis showing high
molecluar weight contaminants ............................... 41
9. Fluorograms of mitoriboscmal subunits from cells labeled in
the absence of mitochondrial protein synthesis ............ 44
10. Fluorograms of 28 S mitoribosomal subunits froman cells
labeled in the absence of mitochondrial protein synthesis,
no chase period .......................................... 48
U. Fluorograms of 39 S mitoribosomal subunits franm cells
labeled in the absence of mitochondrial protein synthesis,
no chase period ........................................... 50
12. Total mitochondrial incorporation in the presence and
absence of inhibitors ..................................... 57
13. Effect of inhibitors on MDBK cells ........................ 59
14. Immune precipitation of mitoribosomal proteins from soluble
cell fractions ............................................ 61
The amount of ribosomes which gives an absorbance
of one unit at 260 nm when dissolved in 1 ml of
solution and measured with a light path of 1 an
1 A^ unit of mitoribosomes = 90 pg (60 pg of
proein, 30 pg of RNA)
1 A260 unit of 55 S monosomes = 32 poles
1 A260 unit of 39 S subunits = 55 pmoles
1 A260 unit of 28 S subunits = 84 pmoles
N,N-bis [ 2-hydroxyethyl ]glycine
bovine serum albumin
Curie, 2.2 x 1012 dpm of radioactivity
scintillation counts per minute
disintegrations per minute
[ethylenedinitrilo ] tetraacetic acid
1000 x a force of one gravity
Madin-Darby bovine kidney cells
mitochondrial matrix fraction of cells
messenger ribonucleic acid
mitochondrial deoxyribonucleic acid
polyacrylamide gel electrophoresis
post-mitochondrial supernatant cell fraction
DNA which codes for ribosomal R1
revolutions per minute
ribosamal ribonucleic acid
sodium dodecyl sulfate
transfer ribonucleic acid
unidentified reading frame
micro, x 106
Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
THE BIOSYNTHESIS OF MITOCHONDRIAL
GRETCHEN LYN SCHIEBER
Chairman: Thomas W. O'Brien, Ph.D.
Major Department: Biochemistry and Molecular Biology
This research has addressed questions on the biosynthesis of the
proteins of the mammalian mitochondrial ribosome (miitoriboscme). It
has focused on the site of synthesis of the mitoribosomal proteins;
that is, are any of the proteins of the mammalian mitoribosome
products of mitochondrial protein synthesis? It is already known
that the mitoribosomes of primitive eukaryotes, yeast and
Neurospora, do contain a protein which is a product of
mitochondrial protein synthesis. However, because these systems are
quite different from the mammalian systems, the biosynthetic site of
the proteins might differ.
Two approaches were used to address this problem. In the first,
mitochondria isolated from bovine liver were allowed to synthesize
proteins in an in vitro system and the products were comigrated in
two-dimensional gel electrophoresis with the mitoriboscoal proteins.
The mitochondrial protein synthetic products, visualized by
fluorography, showed several spots of varying intensity. These were
compared to the mitoribosomal proteins, visualized by Coomassie blue
stain. Only one protein, L44, was comigratory with one of the
radioactive spots, but some spots were positioned near mitoribosomal
proteins (S13, S20, and L20). Therefore, these mitoribosomal
proteins are indicated as possible products of mitochondrial protein
In the second approach, bovine (MDBK) cells grown in culture
were labeled with [3 Sjmethionine in the presence and absence of
inhibitors for mitochondrial and cytoplasmic protein synthesis. The
labeling in the absence of cytoplasmic protein synthesis (with the
cells inhibited by cycloheximide) produced a "blank" fluorogram,
indicating that there is no mitochondrial product. This was further
substantiated by cell labeling studies in the absence of
mitochondrial protein synthesis showing that the mitoriboscmal
proteins, including those proteins indicated by experiments in
isolated mitochondria (see above), incorporated [3 S]nethionine in
the presence of chloramphenicol. Finally, it was demonstrated that
mitoribosomal proteins can be both translated and assembled into
complete mitoribosomes in the absence of mitochondrial protein
synthesis. These results indicate that all of the proteins of the
mammalian mitoribosome are products of cytoplasmic protein synthesis,
and that the function of the mitochondrially produced mitoribosomal
protein in lower eukaryotes must be performed by a cytoplasmically
produced protein in mammals.
It is now firmly established that the mitochondrion is a
"semi-autonomous" organelle, simplistically thought of as some sort
of bacteria-like structure which has evolved to exist within, but
somewhat independent of, the eukaryotic cell. While mitochondria
contain their own DNA and a complete protein synthesis system,
including ribosomes, tRNA, rRNA, and all of the accessory proteins,
they synthesize only a handful of the several hundred proteins
necessary for proper mitochondrial function. The vast majority of
mitochondrial proteins are synthesized by the nucleo-cytoplasmic
protein synthesis system; coded for by genes located on nuclear DNA,
translated on cytoplasmic (80 S) ribosomes, and specifically imported
into mitochondria [1,2].
The Nucleo-Cytoplasmic Protein Synthetic Compartment and
Early studies on the cytoplasmic synthesis of mitochondrial
proteins were hampered by the inability to produce clean cell
fractions and by the practice of studying mitochondrial proteins as a
general class instead of studying a specific protein. In the early
and mid 1970's, Butow's group published a series of papers [3-6]
based on electronmicroscopy which indicated that cytoplasmic
ribosomes were specifically and functionally attached to the outside
of mitochondria, implying that mitochondrial proteins might be
vectorially discharged directly into the the mitochondrial matrix.
About the same time, several groups published papers indicating that
mitochondrial proteins were synthesized initially on ribosomes bound
to the endoplasmic reticulum [7-9] and directed in a manner similar
to secretary proteins. Most of these studies relied on antibodies
raised against either a specific mitochondrial protein or a general
extract of mitochondria to follow the position of newly synthesized
mitochondrial proteins in cell fractions which had been produced
after labeling in intact cells. Several authors reported a
microsomal pool early in the synthetic pathway [10,11] and some
traced the kinetics of protein movement to the mitochondria [11,12].
Occasionally, an author reported synthesis of a specific
mitochondrial protein on free ribosomes [13,14]. The general model
for the synthesis of mitochondrial proteins was expected to parallel
that of secretary proteins.
The model of mitochondrial protein import began to change as
cell-free protein synthetic systems were used to produce precursor
mitochondrial proteins. Unexpectedly, precursors of mitochondrial
proteins were directly taken up by mitochondria added to the mixture
[15-25]. In several cases, this was demonstrated in the absence of
translation on 80 S cytoplasmic riboscmes, showing that import was a
post-translational event. Other workers looked at the distribution
of mRNA which coded for mitochondrial proteins and various authors
reported either that it was isolated almost completely from free
ribosomes [26,27] or that is was on both free and mitochondrially
bound ribosomes [28-30]. Thus, for virtually all proteins studied by
these techniques, recognition for import into mitochondria appears to
be mediated by the precursor protein (without any known covalent,
post-translational modifications) and the intact mitochondrion.
Occasionally, those proteins with an N-temnninal "transit peptide"
(see below) may adventitiously begin to import during translation,
producing mitochondrially-bound cytoplasmic ribosomes, although there
are no reports of obligatory co-translational import.
The structure of the precursor mitochondrial proteins has been
studied for many different proteins in many different systems (for
review see ). In most cases, the precursor protein is slightly
larger than the mature mitochondrial form, although the length of the
extension varies. While some authors refer to this extension as a
"signal peptide" , others are starting to use the term "transit
peptide"  both to denote that it is not endoplasmic reticulum
associated (and probably not recognized by the signal recognition
particle ) and because it may encode some assembly instructions,
not just organellar destination. While, in most cases, this peptide
appears to be a short N-terminal extension on each protein [34-37],
that is not always the case. In one case, it was reported by
Poyton's group that all four of the cytoplasmically synthesized
proteins of the F1-ATPase were synthesized as a single polypeptide,
which was imported in one step [38,39]. However, reports from
several other groups [30,40-42] have shown that these four proteins
are translated in independent events. There are also reports of
uncleaved transit peptides. For example, the newly-synthesized form
of the y-subunit of F1-ATPase in yeast has the same SDS-PAGE
mobility as the form isolated frame the mature complex . Finally,
there is at least one report of what appears to be an internal
transit sequence, in cytochrome c. Newly synthesized cytochrcme c
differs from the mature polypeptide only in the replacement of the
N-terminal methionine with an acetyl group. The uptake of newly
synthesized cytochrome c is competitively inhibited by apo-cytochrcnme
c [43-45] and by an internal polypeptide (trypsin fragment 66-104)
. The uptake is not affected by the N-terminal fragment (either
fragment 1-65 or 1-80) or by holo-cytochrome c.
The Mitochondrial Gename
Early characterization of mitochondrial DNA showed that it was
circular and varied in size franom about 16,000 bases in mammals to
over 100,000 bases in higher plants . Other reports noted that
it was replicated in an asynchronous manner  and was transcribed
completely , with the various RNAs then processed out of the
longer transcirpt. Various researchers tried to map the products of
mitochondrial DNA transcription by hybridization, eventually
positioning the ribosomal RNAs and several tRNs . Early efforts
to identify the products encoded by mitochondrial DNN relied on
analysis by one-dimensional SDS-PAGE which revealed 10-12 proteins
[51-54]. When two-dimensional PAGE was used, the patterns became
more complex; usually around 25 proteins were visualized [55-58].
All of these studies were eclipsed in 1981 by the publication of
the complete sequence of a human placental mitochondrial DNA .
The structure was completely analyzed by comparing it to the known
information about the sequences of mitochondrially produced proteins
and mitochondrial rRNA and tRNA [60-64]. The mitochondrial DNA
contains thirteen open reading frames and the coding sequences for
mitochondrial rRNA and for 21 tRNAs. The other major region of the
DNA is the "D-loop" which contains the origin for replication. This
organization has been preserved in the mitochondrial DNA of all of
the mammalian species sequenced, including mouse , cow , and
rat . Indeed, a partial sequence in Xenopus  indicates that
this general organization may be conserved among all of the
One of the most striking features of the mammalian mitochondrial
genome is its extraordinarily high density of coding information.
There are virtually no "unused" bases. The tRNAs serve to punctuate
the sequences for mRNA and rRNA [6-9,70], often without a single
intervening nucleotide between the 3' terminus of the tRNA and the 5'
initiation codon of the message. The mRNA sequences apparently lack
any untranslated regions, either leader sequences or introns .
Indeed, some of the messages only acquire a termination codon when
the cleaved portion of the transcript is polyadenylated.
Five of the reading frames of mitochondrial DNA have been assigned
to known products of mitochondrial protein synthesis; the other eight
are unassigned reading frames (URFs). The assigned reading frames
include the genes for subunit 6 of the FIATPase, cytochrome b, and
cytochrome oxidase subunits I, II, and III. Despite some early
reports that subunit 9 of the F1ATPase was mitochondrially
synthesized in rats [72,73], this protein is not coded for on
mitochondrial DNA; thus, mammals, like Neurospora, must synthesize
this protein cytoplasmically [74,75] (although it is made
mitochondrially in yeast). The reading frames of the mammalian
(bovine ) mitochondrial gename are summarized in table 1. These
reading frames are considered significant because they are conserved
in all of the mammalian mitochondrial genomes analyzed to date and
because long-lived, polyadenylated cognate RNAs for each of these
reading frames have been isolated from mitochondria. The assigned
reading frames are exactly comparable to the known sequences of their
product proteins. The amino acid sequences of the products of the
unassigned reading frames can be predicted and this information is
currently being exploited to determine if, as expected, these
polypeptides are being translated on mitoribosomes in living systems.
To date, antibodies produced against artificially synthesized
oligopeptides have been used to inmunoprecipitate the products of the
ATPase 6 gene  as well as URFs A6L, 1, and 3 [76,77].
Additionally, the predicted proteolytic fingerprint pattern was used
to demonstrate the in vivo synthesis of URFs 3 and 6 . No
function has yet been assigned to a product of any unidentified
READING FRAMES OF MAMMALIAN MITOCHONDRIAL DNA
The significant reading frames of a manummalian (bovine) mitochondrial genome are determined by homology
with other mammalian species (human and mouse) and by the presence of corresponding polyadenlylated RNAs.
All data are from Anderson et al. .
ATPase subunit 6
Cytochrome oxidase I
Cytochrome oxidase II
Cytochromre oxidase III
Unidentified Reading Frames
4L 297 98t 10.8 13
5 1821 606t 68.2t 39t
6 528 175t 19.0 lit
A6L 201 66t 7.9t 4t
* in kilodaltons
Despite their similar function, the cytoplasmic and
mitochondrial protein synthesis systems are completely distinct.
When compared with cytoplasmic ribosomes, mitochondrial ribosomes
generally are smaller and appear in some ways (e.g. antibiotic
susceptibility) to be more like prokaryotic ribosomes [79,80]. While
there is variability in size and physical-chemical properties among
mitoribosomes of the other eukaryotes, the mitochondrial ribosomes of
mammals all appear to be of the 55 S type [81,82]. Mammalian
mitoribosomes are slightly larger than bacterial ribosomes  and
their composition is much more protein rich (about 85 proteins ),
with their rRNA content correspondingly reduced. This change in
composition is reflected in the lower sedimentation coefficient of
mammalian mitoriboscmes, 55 S versus 70 S for bacterial ribosomes and
80 S for eukaryotic cytoplasmic ribosomes [81,82].
The proteins of the bovine mitoribosome have been characterized
by two-dimensional PAGE. Typical electropherograms are shown in
figure 1. The small (28 S, 1.1 x 106 daltons) and large (39 S,
1.65 x 106 daltons) subunits of the mitoriboscme are analyzed
separately. In figure 2 are shown the schematic diagrams of these
separations, as analyzed by Matthews et al. . Proteins which are
numbered appear in most of the electropherograms. However same of
them are variable and sometimes appear weakly or not at all. The
small subunit schematic also shows fourteen lettered proteins which
appear infrequently in these electropherograms.
Figure 1. Electropherograms of the Proteins of the Bovine Mitochondrial Ribosome.
Proteins of the small, 28 S (A) and large, 39 S (B) subunits of bovine liver mitochondrial ribosomes
are shown. The subunits were derived froman 55 S ribosomss as described in "Materials and Methods". The
first electrophoretic dimension, from left to right in the figures, is electrophoresis in 9 M urea at pH
4.3. The second dimension, from top to bottom, is electrophoresis in 0.5% SDS and 5 M urea at pH 7.2.
, 2 4-50 50
ftf -44.5 l w4 44.5
8 % -6 6W
4W IA7 7o
I1t- d P- ,1 30 12 13 4 14 30
"3sC dv 14 Z-000'3
i 15w* 15 17019
1f, 17 20 921
S200 19 q.i" 230 h 2
j,21.22,, 31 0 W2- 2824,
23N 24 25 26 14.4 347233 ;2 03
10 0 4b -144. 3700_ 036 qb40-14
27 29 32 1* 440
28 3 3.1 O* -38 390q*44 4
m -12.5 460* 494 -12.5
1033 4W* o47
48 4 50
Figure 2. Schematic diagrams of the Proteins of the Bovine Mitochondrial Ribosome.
Schematic diagrams showing the positions of bovine mitoribosomal proteins when analyzed by
two-dimensional PAGE as described in "Materials and Methods". Proteins of the small, 28 S ribosomal
subunit are shown in A and of the large, 39 S ribosomal subunit are, shown in B. Numbered proteins
appeared in the majority of electrophoretic separations analyzed. lettered proteins appeared in less
than half of the electropherograms analyzed. The schematic diagrams are taken from Matthews et al., 1982
Mitochondrial Riboscme Biosynthesis
How mitochondria synthesize and assemble their ribosomes has
been studied primarily in two primitive eukaryotic systems, yeast and
In yeast, it was discovered that there is one product of
mitochondrial protein synthesis that exhibits a strain specific
molecular weight polymorphism [85,86]. This protein, named "var 1,"
varies in molecular weight from 40 to 44 kilodaltons. An early
report by Groot [87,88] demonstrated that var 1 is associated with
the small mitoribosomal subunit, but concluded that the protein could
not be a "ribosomnal protein" since yeast could be grown for several
generations in the mitochondrial specific inhibitor, chloramphenicol,
without affecting mitochondrial protein synthesis. Because
chloramphenicol does not completely inhibit mitochondrial protein
synthesis in yeast, a second study was performed by Butow's group,
using the more effective antibiotic erythromycin. In the presence of
erythromycin, it was demonstrated that mitochondrial protein
synthesis was affected and that the synthesis of the small
mitoribosomal subunit was inhibited [89, 90]. In fact, the synthesis
of var 1 is highly resistant to inhibition by chloramphenicol and
largely accounts for the leakage of yeast mitochondrial protein
synthesis in the presence of this inhibitor. Inhibiting the
synthesis of var 1 with antibiotics gives rise to small mitoribosomal
subunits which are missing some proteins including var 1 . It is
now known that different gene conversions give rise to the variation
in the molecular weight of this protein [92-95] and that a sequence
homologous to var 1, as well as other mitochondrial DM sequences,
can be detected in the yeast nuclear gencme [96-98].
In Neurospora, there is also one and only one mitochondrially
produced mitoribosomal protein. This protein, originally identified
as S-4a , is now designated S-5 . Like var 1, it is one of
the larger proteins (52,000 daltons) of the small mitoribosomal
subunit. In an early study , Lambowitz, Chua, and Luck
demonstrated that mitoribosomes are synthesized via a series of
precursor intermediates which contain some of the ribosomal proteins
and incompletely processed rRNA. They also showed that the synthesis
of only S-5 is inhibited by chloramphenicol and permitted in the
presence of anisomycin (a specific inhibitor of cytoplasmic protein
synthesis). In other studies, Lambowitz's group has shown that
mitoribosomal small subunits from Neurospora grown in the presence
of chloramphenicol are deficient in S-5 plus other proteins which
apparently assemble onto the particle after S-5 , that the
removal of an intron in the transcript for the 25S large subunit rRNA
is preceded by specific binding of some of the cytoplasmically
synthesized mitoribosomal proteins [102,103], and that mitochondrial
mutations which directly affect the production of energy also affect
the synthesis of small mitoriboscmal subunits .
Objectives of this Work
When this work was begun, the presence of mitochondrially
produced proteins in the mitochondrial ribosomes of yeast and
Neurospora had only recently been established. The study of
mitoribosomal proteins in those systems was aided by the ease with
which comparatively large amounts of mitoribosames could be isolated
and by the adaptability of these unicellular organisms for labeling
studies. In contrast, the proteins of mammalian (bovine)
mitoribosonmes had been characterized (unpublished at the time, )
but there had been no studies on the biosynthesis of these ribosomes.
The main focus of this work has been on determining which, if
any, of the mammalian (bovine) mitoribosomal proteins are
mitochondrially produced. Because of the presence of var 1 and S-5,
it seemed logical to assume that the manmalian mitoriboscme would
contain at least one protein synthesized intramitochondrially.
Indeed, since the bovine mitoribosome contained so many proteins, it
seemed feasible that there could be more than one mitochondrially
produced protein. Identifying which of the mitochondrial proteins
were mitochondrial products would allow questions concerning the need
for intramitochondrial synthesis to be addressed. Would these
proteins be involved in anchoring the mitoribosome to the membrane,
indicate i ng that they might be too hydrophobic to be imported? Would
they be closely associated with the rRNA, indicating that they might
be required for early assembly? When the mammalian mitochondrial
genome was published in 1981 , it added a new aspect to these
studies. Since the predicted sequence of all possible mitochondrial
products became available, if a mitoribosomally produced protein was
identified, it could be assigned to an unidentified reading frame by
sequencing only a short stretch of the primary amino acids. Finally,
the extent to which mammalian mitoribosomes were synthesized
similarly to, or differently from, the ribosomes of lower eukaryotes,
could lead to insights as to the selective constraints involved in
This work has examined the site of synthesis of mitoribosromal
proteins by two techniques. The protein synthetic products of
isolated bovine mitochondria incubated in an in vitro system were
analyzed to determine if they were canomigratory by two-dimensional
PAGE analysis with the enumerated proteins of bovine mitoriboscmes.
In a more powerful technique, bovine cells in culture were treated
with specific inhibitors of cytoplasmic and mitochondrial protein
synthesis to study the contribution of each system's products to the
This work also deals briefly with another aspect of
mitoribosomal biosynthesis, the mechanism of mitoribosome assembly.
This problem includes several'separate questions of how the
cytoplasmically synthesized proteins are recognized and imported by
mitochondria, how mitoribosomal proteins are processed, and how the
proteins and rRNA are assembled into a functional particle. In a
brief study, an immune precipitation experiment was used to determine
if soluble pools of unassembled mitoribosomal proteins can be
detected in the cell cytoplasm or mitochondrial matrix fractions of
MATERIALS AND METHODS
Solutions. Solutions cited are defined in table 2.
Reagents. [3 S]methionine, >600 Ci/mmiol, was purchased from
Ammersham. [3,4,5- H]leucine, >11UO Ci/nmmol, was purchased form New
England Nuclear. The inhibitors, chloramphenicol, erythromycin, and
cycloheximide, were obtained from Sigma, as were digitonin and ultra
pure sucrose. All reagents for the growth of cells in culture were
from Gibco. Adjuvants were obtained franom Difco. Formalin-fixed S.
aureus cells were obtained from Bethesda Research Laboratories. All
other reagents were obtained from standard laboratory suppliers.
Tissue culture cells. Madin and Darby bovine kidney (MDBK)
cells  were used for all experiments in tissue culture. This is
a "non-transformed," continuous cell line. The initial innoculum of
cells was kindly provided by Dr. Paul Klein. The cells were grown in
Dulbecco's modified Eagle medium with high glucose containing 10%
fetal calf serum, and exhibited an average doubling time of
approximately 20 h.
A 0.25 M sucrose, 1 mM EDTA, 5 mM Tris, pH 7.5
B 0.34 M sucrose, 5 mM Tris, pH 7.5
C 100 mM KC1, 20 mM MgCl2, 10 mM triethanolamine, 5 mM B-mercaptoethanol, pH 7.5
D 200 mM KC1, 2 mM MgC12, 10 mM Tris, 5 nMM 8-mnercaptoethanol, pH 7.5
E 300 nmM KC1, 5 mM MgCl2, 10 mM Tris, 5 mM B-mercaptoethanol, pH 7.5
F 25 mM KC1, 10 mM MgCl2, 10 mM Tris, 5 mM B-mercaptoethanol, pH 7.5
G 8 M urea, 12.5 nM KC1, 5 mM MgCl2, 5 mM Tris, 2.5 nmM 6-mercaptoethanol, pH 7.5
H 0.15 M NaCI, 10 mM Tris, pH 7.5
I 9 M urea, 3 M LidC, 0.001% butylated hydroxy-toluene, pH 3
J 9 M urea, 3 M LiCI, 0.001% butylated hydroxy-toluene, 100 mM B-mercaptoethanol, pH 7
K 9 M urea, 60 mM KAc, 0.01% aminoethanethiol, pH 6.7
L 5 M urea, 2% SDS, 10 MrM sodium phosphate, pH 7.2
M 5 M urea, 2% SDS, 10 mM sodium phosphate, 100 mM 8-mercaptoethanol, pH 7.2
Experiments Using Mitochondria Isolated from Bovine Liver
Preparation of Bovine Liver Mitochondria
Mitochondria were prepared frman livers of freshly killed cows by
two procedures. In the first procedure, 2 kg of ground liver tissue
was mixed with 4 L of buffer A and the cells were lysed using a
Tekmar Super Dispax polytron homogenizer. Cell debris was removed by
centrifuging in a Beckman JA-10 rotor at 3000 rpn (1000 x g) for 10
min and then mitochondria were collected by centrifugation at 8000
rpm (7000 x g) for 10 min. The mitochondria were washed three times
by resuspension and centrifugation (7000 x g for 10 min) in buffer A;
during the second wash, the buffer was supplemented with 0.1 g/L
digitonin and the slurry (20 mg mitochondrial protein/ml) was stirred
for 15 min before centrifuging. (This procedure is standardized for
the laboratory and was performed by the laboratory staff.)
The second procedure was a modification of the first, designed
to reduce bacterial contamination in the mitochondrial preparation.
Sterile solutions and glassware were used throughout this
preparation. A small lobe of fresh liver was bathed in 0C 70%
ethanol before use. Liver tissue (100 g) removed from the center of
this lobe was minced and added to 400 ml of ice cold buffer A. A
Tekmar SDT Tissuemizer polytron homogenizer was used to lyse the
cells (> 90% cell lysis). Cell debris was removed by centrifuging in
a Beckman Type 35 rotor at 2000 rpm (300 x g) for 5 min and then
mitochondria were collected by centrifugation at 8000 rpm (5000 x g)
for 5 min. The mitochondria were washed four times by resuspension
in buffer A and centrifugation (5000 x g for 5 min); as before, the
second wash was supplemented with digitonin and the slurry was
stirred for 15 min before oentrifuging.
The mitochondria from either preparation were diluted before use
to 10 mg mitochondrial protein/ml in buffer A. In order to insure
complete dispersion of the mitochondria, this preparation was
homogenized for 5 complete strokes with a TenBroeck homogenizer. The
mitochondrial solution was oxygenated for 15-30 sec and stored sealed
under oxygen at 0C until use (within 1 h).
Conditions for Mitochondrial Incubation
Mitochondria were incubated at a concentration of 1 mg
mitochondrial protein/ml at 37C. The total incubation volume
varied from 3 to 7 ml per condition, depending on the experiment.
The complete composition of the incubation mixture is shown in table
3. Basically, it is a bicine buffered salt solution supplemented
with sucrose for osmotic stability, an ATP generating system, and all
of the common amino acids. (The amino acid used as the radiolabel
was omitted from the unlabeled amino acid pool.) Inhibitors, when
used, were added at concentrations of 200 pg/ml for chloramphenicol
and 300 pg/ml for erythromycin and cycloheximide. Mitochondria were
warmed and preincubated in the incubation mixture for 5 min before
incorporation was started by the addition of the radiolabel
([3H]leucine at 100 PCi/ml or [35S]methionine at 50 VCi/ml).
The incubations were monitored using the TCA disk assay .
Aliquots of 10 pl were absorbed into Whatman 3MM paper disks and
IN VITRO MITOCHONDRIAL INCUBATION MIXTURE
Bicine, pH 7.6
KH2PO4, pH 7.6
Amino Acid Mixturet
tAmino Acid Mixture. As defined in Roodyn et al., 1961 ,
with amino acid to be used for the radiolabel emitted.
precipitated in ice cold 20% TCA. The disks were washed once at 0C
in 10% TCA, once for 20-30 min at 80C in 10% TCA, and again at 0C
in 10% TCA before being dried and counted. The hot TCA wash is used
to hydrolyze amino acids from aminoacylated tRNAs so that the data
reflect only incorporation into protein.
At the end of the incubation period (60-90 min), 1 ml aliquots
for analysis by 2D-PAGE were drawn. The mitochondria were removed
from the incubation solution by centrifugation in a Beckman Microfuge
B (9000 x g) for 5 min. The supernatant was discarded and the
pellets were stored at -70C until used.
Experiments Using Cells Grown in Tissue Culture
Conditions for Incorporation of Radiolabeled Methionine
MDBK cells to be used for labeling were grown on plastic roller
bottles (850 cm2/bottle). Generally, two bottles of cells were
used for each labeling condition (approximately 1-2 x 108 cells per
condition). At the beginning of the labeling procedures, the cells
were between 60% and 80% confluent. The medium was removed from the
cells and replaced with Dulbecco's minimal essential medium with
reduced methionine (10 pM) containing 10% dialyzed calf serum.
After the addition of antibiotics, cells were preincubated to allow
the antibiotics to inhibit protein synthesis, before the addition of
a sterile solution of [35S]methionine in media (10 uM methionine)
to a concentration of 4-8 pCi/ml.
Cycloheximide (300 pg/ml) was used to inhibit cytoplasmic
protein synthesis. The protocol is summarized graphically in figure
3, A. The antibiotic was added as a 1.5 mg/ml, freshly made solution
in the low miethionine medium, with a 15 min preincubation before the
addition of [35S]methionine. Because of the tendency of
mitochondrial protein synthesis to decrease rapidly under these
conditions [52,108], the cells were labeled for only 1-2 h. At the
end of the labeling period, the cells were washed for 15 min in
medium containing cycloheximide and 1 mM unlabeled methionine (to
remove the [35S]methionine) and then with normal medium (200 pM
methionine, no antibiotics). The cells were then grown for 2-4 h to
allow newly synthesized proteins to be incorporated into
mitochondrial ribosomes before harvesting.
Chloramphenicol (100 or 200 pg/ml) was used to inhibit
mitochondrial protein synthesis. The antibiotic was added as a 50
mg/ml solution in 95% ethanol. Since cytoplasmic protein synthesis
is not affected by chloramphenicol, these cells could be grown for
longer periods in the presence of inhibitor and [ 35S]mrethionine.
Two protocols were used (figure 3, B and C). In the first, after a
preincubation of at least 15 min with chloranmphenicol, the cells were
labeled for 8-12 h in the presence of inhibitor and
[35S]methionine. After the incubation the cells were washed in the
presence of inhibitor and of 1 mM unlabeled methionine and grown for
a 2-4 h chase period in normal medium (as above). Alternatively, the
cells were preincubated with chloramphenicol for 2-3 h (to allow for
depletion of any existing mitochondrial products). Then, the cells
Figure 3. Schematic representation of the labeling protocols for MDBK cells.
Shown as one-dimensional graphs with the axis representing time are the protocols for labeling in
the presence of cycloheximide, A, in the presence of chloramnphenicol with a chase period, B, and in the
presence of chloramphenicol without a chase period, C. Note the change in scale of the axis between
figure A and figures B and C.
I ~CYCLOHEXIMIDE !
ILOW METHIONINE MEDIA
LOW METHIONINE MEDIA
2 4 6
LOW METHIONINE MEDIA
were incubated for 8-12 h in the presence of inhibitor and
[35S]methionine, before they were washed in the presence of
chloramphenicol and harvested (no chase period).
Control cells were grown and labeled using conditions which
paralleled each inhibitor protocol, including mock additions of
Preparation of Mitochondria from Cultured Cells
Cells were removed franom their bottles with a solution containing
0.02% EDTA and 0.05% trypsin. The cells were washed two times in
buffer A: the first wash at roanom temperature and the second at 4C.
All remaining steps in the ribosane preparation were performed at
The cell pellet was resuspended into 16 ml of diluted buffer B
(buffer B:water, 2:3) containing 120-180 pg/ml digitonin. The cells
were homogenized -using a polytron homogenizer (Teckmar SDT
Tissuemizer) until they were 80-99% lysed. An additional 10 ml of
the digitonin solution was used to rinse the homogenizer and this was
added to the cell lysate. Cell debris was removed by centrifugation
at 1250 x g for 5 min. The supernatant was removed and the debris
pellet was re-extracted with 15 ml of diluted buffer B (buffer
B:water, 1:4). After another centrifugation, the supernatants were
combined and mitochondria were collected by spinning at 9500 x g for
10 min. The post-mitochondrial supernatant (PMS) was stored at
-70C if it was to be used in immune precipitations (see below).
The mitochondria (pellet) were washed in 20 ml of buffer A and then
resuspended into 20 ml of buffer B. A 1.0 ml aliquot of this
suspension was centrifuged in a Beckman Microfuge B (approximately
9000 x g); the supernatant was discarded and the pellet was stored
frozen at -70C until it was prepared for electrophoresis as a
sample of intact mitochondria. The remaining 19 ml was centrifuged
at 9500 x g for 10 min and the pellet was used to prepare
mitochondrial ribosomes (see below).
Isolation of Mitochondrial Ribosomes
Mitochondrial ribosomes were isolated by two techniques. For
experiments where the MDBK cells were labeled for 8-16 h (in the
presence of chloramphenicol and the parallel control) the
mitochondria were resuspended into 1.4-1.7 ml of buffer C. The
mitochondria were lysed by the addition of 200 pl of 16% Triton
X-100. In order to improve recovery of radiolabeled mitoribosomes,
15 A260 units of unlabeled bovine 55 S mitoribosomes (prepared from
bovine liver, see above) were added in enough additional buffer C to
bring the total volume to 2.0 ml. The mitochondrial lysate was spun
for 5 min in a Beckman Microfuge B (approximately 9000 x g) in order
to remove debris. The supernatant was loaded onto 10-30% sucrose
density gradients in buffer C and centrifuged in a Beckman SW28 rotor
to isolate 55 S mitoribosomes (figure 4, A). The fractions
containing ribosomes were pooled and the ribosomes were collected by
centrifugation. (If the soluble proteins at the top of the gradient
were to be used in immune precipitations, they were pooled and stored
at -70C.) The pellets were resuspended into 1-2 ml of either
'6 2 JOK/
15 25/ 35 5 5253
Figure 4. Sucrose density gradient profiles of the preparation of
mitochondrial ribosomes and ribosomal subunits.
Mitochondrial ribosomes were isolated from MDBK cells in the
presence of carrier bovine liver mitoribosomes by sedimentation in a
non-dissociating 10 to 30 % sucrose density gradient, A. The fractions
containing 55 S monosomes (cross-hatched) were pooled and the ribosames
were collected by centrifugation. Subunits were prepared by
resuspending the monoscmes into a dissociating buffer (buffer D or E),
and then subunits were isolated in a 10 to 30 % sucrose density
gradient of the corresponding buffer, B. Fractions containing the 28 S
and 39 S subunits (cross-hatched) were separately pooled and the
subunits were collected by centrifugation.
buffer D or E, which dissociate 55 S monosomes into 28 S and 39 S
subunits, and loaded onto 10-30% sucrose density gradient in the
corresponding buffer (figure 4, B). As before, this was centrifuged
in a Beckman SW28 rotor. The fractions containing 28 S and 39 S
mitoribosomal subunits were pooled separately and collected by
The MDBK cells labeled in the presence of cycloheximide (figure
3, A) and their parallel control had a comparatively short incubation
period. The ribosomes frcm these short incubations were expected to
contain a barely detectable amount of radiolabel. Therefore, the
ribosome isolation protocol was modified in a manner that improves
recovery but produces mitoribosomal subunits containing more
contamination with non-ribosomal components. The mitochondrial
pellet was resuspended and lysed as already described excepting that
buffer E was used instead of buffer C. Thus 28 S and 39 S
mitoribosomal subunits were recovered with only one sucrose density
gradient (in buffer E, equivalent to figure 4, B). The fractions
containing 28 S and 39 S mitoriboscmal subunits were again pooled
separately and centrifuged.
The pellets of 28 S and 39 S mitoribosomal subunits were
resuspended into 60 pl of buffer D and prepared for two-dimensional
Assay for Inhibitor Response
MDBK cells were grown to 60-80% confluence in T150 flasks. The
media was removed from the cells and replaced with Dulbecco's minimal
essential media with reduced methionine (10 pM) containing 10%
dialyzed calf serum. After the addition of antibiotics (see figure
13), cells were preincubated for 15 min to allow the antibiotics to
inhibit protein synthesis, before the addition of a sterile solution
of [35Slmethionine in media (10 pM methionine) to a concentration
of 4-8 VCi/ml. After 1.5-2 h incubation at 37C, the cells were
washed with 1 mM methionine (in the presence of the appropriate
antibiotics) and harvested. The harvested cells were washed with
normal media and then with buffer B. The cells were collected by
centrifugation (9000 x g for 5 min) and the supernatant was
discarded. The cell pellets were sometimes stored at -70C until
they were prepared for one-dimensional electrophoresis.
Preparation of Antisera
Antisera were prepared in rabbits by injection of ribosomal
subunits in either a standard buffer (buffer F) or a denaturing
buffer (buffer G) designed to disperse the ribosomal protein from the
rRNA. After collecting preinmmune serum from each animal, they were
initially injected in their footpads with 30 A260 units of
ribosomal subunits (in 40-120 il buffer) mixed with an equal volume
of Freund's complete adjuvant. Collection of antisera began 4 weeks
after the injection. Animals were bled no more frequently than 20 ml
every two weeks and they were booster injected as needed with a
mixture identical to the initial preparation except that incomplete
adjuvant was used. The first booster injection was given in the
footpads and, thereafter, injections were given subcutaneously in the
back. Titer was tested in Ouchterlony diffusion tests against
extracted mitoribosamal proteins.
Preparation of Imnune Precipitates
Antigen preparations used in these experiments were a by-product
of the labeling procedures with MDBK cells. One preparation used was
the post-mitochondr'ial supernatant (PMS). This solution, the
supernatant of the lysed cells after the removal of membraneous
organelles, should include the soluble proteins of the cytoplasm.
The other preparation was the top 3 ml of the SW 28 gradient in
buffer D mixed with 3 ml of buffer H. This solution should contain
the soluble proteins of the mitochondrial matrix (MM) and probably
some extracted membrane proteins as well. In each case, the
preparations used were those prepared fram NDBK cells labeled in the
presence of chloramphenicol until cell harvest, because these
preparations were expected to contain the highest specific activity
of unassembled, cytoplasmically synthesized mitoribosomal proteins.
In order to clear any pre-existing precipitates from these solutions,
they were held at 4C for > 12 h, and then centrifuged at 20,000 x g
for 30 min before being used.
For the precipitation step, 1 ml of prepared antigen solution
was mixed with 20 pl of serum (or buffer H, see figure 14) and 1 mg
BSA (in 0.1 ml buffer H). These solutions were incubated with mixing
overnight at 4C. Then 25 mg of pre-washed, formalin-fixed
Staphlococcus aureus cells were added. After incubation with
mixing at 4C for >12 h, the cells were washed four times, twice in
0.5 ml buffer H, once in 0.75 ml buffer H + 0.5% Triton X-100, and
finally in 0.75 ml buffer H. The antibodies and bound proteins were
then eluted with 100 pl of buffer M (see below) and analyzed by
Samples were prepared for two-dimensional polyacrylamide gel
electrophoresis (two-dimensional PAGE) by two extraction steps in
buffer I or buffer J. Routinely, 3.5 A260 units of ribosomal
subunits (200 pg of ribosomal protein) ware extracted in 120 P1 of
buffer. This mixture was stirred for 12-16 h at 4C and then the
solubilized protein was collected in the supernatant fraction after
centrifugation at 20,000 x g for 1 h. The pellet was then
re-extracted in an additional 80 pl of buffer for 4-6 h and
centrifuged; the supernatants were combined and dialyzed versus
buffer K before electrophoresis. This procedure was frequently
modified. Ribosomal subunits prepared from tissue culture cells,
with added carrier bovine liver mitoriboscmes, were extracted in a
standard volume (120 pl, first extraction; 80 pl, second
extraction), regardless of the actual A260 measurement (1.5-8
A 260) Whole mitochondria samples were supplemented with 3.5
A260 units of isolated mitoribosomal subunits and extracted in the
standard volume or in a doubled volum. Attempts to improve the
standard volume or in a doubled volume. Attempts to improve the
extraction from the mitochondria included the addition of 0.5% Triton
X-100 or 2 M tetramethylurea to the buffers.
Samples for analysis in one-dimensional SDS-PAGE were
solubilized in buffer L or M (25-200 pl, depending on the sample
volume). The samples were incubated at 60-70C for 1 h before
loading onto the acrylamide gel slab.
The conditions for electrophoresis have been described
previously . Briefly, the first dimension was electrophoresis in
potassium acetate buffer with 9 M urea at pH 4.3, 4.6% acrylamide.
The second dimension was electrophoresis with 0.5% SDS and 5 M urea
in phosphate buffer at pH 7.2, 10% acrylamide. The conditions for
one-dimensional electrophoresis were identical to the second
dimension of the two-dimensional PAGE system. After electrophoresis,
the gels were stained with Cocmassie blue R (0.25% in 50% ethanol).
The pattern of labeled proteins was visualized by fluorography
Experiments with Isolated Bovine Mitochondria
Synthesis of Protein by Isolated Bovine Mitochondria
Mitochondria, isolated from bovine liver, were allowed to
synthesize proteins in an osmotically stabilized, buffered salt
solution containing all of the cannn amino acids, one of which was
radiolabeled, and an ATP-generating system. Synthesis of protein was
followed by assaying incorporation of radiolabeled amino acid into
TCA-precipitable material. In order to exclude incorporation into
aminoacyl tRNA, the TCA precipitates, on paper disks, were washed
once in hot (80 90C) TCA solution, in addition to washes in cold
TCA. This procedure hydrolyzes the aminoacyl bond, leaving the tRNA
precipitated, but no longer charged . Figure 5 shows
incorporation of [3H]leucine by mitochondria isolated by the
standard preparation method. The mitochondria actively incorporate
the amino acid for the duration of the incubation period, 1 h.
Cycloheximide does not affect this incorporation, showing that this
incorporation of amino acids is not fram contaminating cytoplasnic
ribosomes. The slight increase in the presence of this inhibitor
probably represents sampling error. Chloramphenicol eliminates
virtually all of the incorporation, as expected.
0 30 60
Figure 5. Incorporation of [ H]leucine into protein by isolated
bovine liver mitochondria.
Bovine mitochondria, freshly isolated from bovine liver, were
incubated at a concentration of 1 mg mitochondrial protein/ml with
[ H]leucine in the incubation mixture described in table 3.
Incorporation into protein was followed by hot TCA precipitation of
10 pl aliquots. Incorporation was assayed in the absence of
inhibitors (* ) and in the presence of the inhibitors cycloheximide
( i ) and chloramphenicol ( A ).
Analysis of Radiolabeled Mitochondrial Products by Two-Dimensional
An aliquot of the uninhibited mitochondrial incubation mixture
containing 1 mg of mitochondrial protein was prepared for 2D-PAGE.
The sample was divided in half and, to one half, 28 S mitoriboscmal
subunits were added. To the other half, 39 S mitoribosomal subunits
were added. These samples were then prepared for electrophoresis by
the standard method. After electrophoresis, the stained gels showed
the pattern of genuine mitoribosomal proteins in addition to several
prominent mitochondrial proteins. The pattern of an electropherogram
with 28 S mitoribosomal carrier is shown in figure 6, A.
Flourography was used to study the pattern of mitochondrially
synthesized proteins. As expected, both electropherograms produced
the same pattern of radiolabeled spots, shown in figure 6, B.
The pattern of radiolabeled spots from the fluorogram can be
compared to the stained pattern of genuine mitoriboscmal proteins in
order to determine if any of the radioactive
(mitochondrially-produced) proteins are comigratory with, and
possibly therefore identical to, the mitoriboscmal proteins. Figure
6, C and D show the superimposed positions of the radiolabeled
proteins with both the 28 S (figure 6, C) and the 39 S (figure 6, D)
mitoribosomal protein patterns. There is a radiolabeled spot that is
comigratory with protein L44 (protein 44 of the large, 39 S,
ribosomal subunit). In addition, there are radiolabeled spots in the
vicinity of proteins S13, S20, and L20. The pale and indistinct
nature of these proteins  makes it impossible to determine if
these spots are actually ccmaigratory. Thus, these four mitoribosomal
O .. .
C : 28S D : 39S
-- zo """ r,> o
0 ,? *%44
a, a a0 F
0 0 '44
Figure 6. Synthesis of protein by isolated bovine mitochondria.
Proteins synthesized by isolated bovine mitochondria in an in
vitro incubation system were analyzed in the presence of carrier
proteins from bovine mitoriboscmal subunits by two-dimensional PAGE
and fluorography. The Coomassie blue stained pattern, A, shows the
pattern of the carrier 28 S proteins and several high molecular
weight mitochondrial proteins. In the corresponding fluorogram, B,
the pattern shows several scattered spots of varying intensity. In
the corresponding schematic diagrams, the positions of the
mitochondrial protein synthetic products (cross-hatched spots) are
shown in respect to the positions of the enumerated proteins of the
28 S, C, and 39 S, D, mitoribosncmal subunits (filled circles).
proteins are identified as possible products of mitochondrial protein
synthesis. However, this result must be considered cautiously in
view of its irreproducibility (next paragraph) and of the results
obtained with cells grown in culture (see below).
Two attempts were made to repeat this result. In one, the
mitochondria labeled in the presence of cycloheximide or in the
presence of chloramphenicol (see figure 5) were prepared and
electrophoresed in a manner identical to that used for the
uninhibited sample. The fluorograms which were produced franom either
sample contained one major spot in the high molecular weight region
of the gel, which was comigratory with a non-ribosomal protein. The
fact that this protein labeled in the presence of either
chloramphenicol and cycloheximide indicates that this result may be
artefactual. In the other experiment, mitochondria prepared by the
semi-sterile method were labeled in an identical manner excepting
that [ S]methionine was used and erythromycin was added to same of
the samples. As before, the incorporation was followed by a
TCA-precipitation assay and showed that the mitochondria incorporated
the radiolabel throughout the incubation period (1.5 h) and that
erythromycin did not affect this incorporation (data not shown). The
presence of erythromycin demonstrated both that this mixture was not
contaminated with bacteria and also that the mitochondria prepared by
this method were intact. (Mitochondrial ribosomes are sensitive to
erythromycin but the mammalian mitochondrial membrane is not
permeable to this antibiotic ). Aliquots of these labeled
mitochondria ware prepared for electrophoresis by a diluted standard
extraction as well as extractions supplemented with 0.5% Triton X-100
or with 2 M tetramethylurea. In all cases, fluorograms of the
two-dimensional separations were blank. One-dimensional PAGE
analysis of the residual pellets from the extractions and of the top
of the first dimension stacker gel indicated that the radiolabeled
protein primarily failed to extract, and, to a lesser extent,
aggregated during preparation for electrophoresis.
Experiments with Cells in Culture
Mitoribosomal Proteins from Cells Labeled in the Absence of
Cytoplasmic Protein Synthesis
A continuous line of bovine cells growing in culture, MDBK
cells, was used to study the biosynthesis of mitoriboscmes in vivo.
In order to inhibit labeling of the products of cytoplasmic protein
synthesis, MDBK cells were labeled in the presence of the
cycloheximide. The protocol is shown schematically in figure 3, A.
In the presence of cycloheximide, only proteins which are products of
mitochondrial protein synthesis should incorporate radioactivity.
During the chase period, in the absence of both radioactivity and the
inhibitor, any components required for assembly of ribosomes can be
synthesized and the particles should assemble, but components made by
the nucleo-cytoplasmic system will not be radiolabeled. The
ribosomal large and small subunits were prepared from the
mitochondria of these cells directly using dissociating conditions
and a single sucrose density gradient. They were analyzed by
two-dimensional electrophoresis and fluorography.
The results are shown in figure 7 together with the results from
a parallel control (cells labeled in the absence of inhibitor). No
spots are observed on fluorograms from either large or small
ribosomal subunits labeled in the presence of cycloheximide (figure
7, A and B). Th i s is in contrast to the subunit preparations from
parallel control cells (figure 7, C and D) which show a great deal of
incorporated radioactivity. The most prominent spots in these
autoradiograms are, in fact, not ribosomal proteins, but are
contaminating mitochondrial proteins of high specific radioactivity
which are included in this preparation because the subunits were
prepared by the single gradient method. The known proteins of the
mitoriboscmre are visible as the background of paler spots.
This experiment was performed twice. In one case a few high
molecular weight proteins are visible in the fluorograms of ribosomal
subunits froman cells labeled in the presence of cycloheximide (figure
8). These spots appear to be mitochondrial contaminants which have
"leaked" through the cycloheximide inhibition. They do not appear to
be mitoribosanal proteins for two reasons. They are quite similar in
the fluorograms from both subunits, but it is known that the proteins
of the two subunits are distinct . Additionally, they correspond
in position to some of the intensely labeled spots of the control
Thus, conditions of the labeling period were adequate for the
incorporation of [35S]methionine into mitochondrial ribosamal
proteins, but that incorporation is completely blocked by
cycloheximide, an inhibitor of cytoplasnic protein synthesis.
A. o .
Figure 7. Fluorograms of mitoribosamal subunits from cells labeled
in the absence of cytoplasmic protein synthesis.
Mitochondrial ribosomal subunits froman cells labeled in the
presence of cycloheximide were analyzed by two-dimensional PAGE and
fluorography. Fluorograms of the small, A, and large, B, subunits
are shown. Ribosomal subunits from cells labeled in parallel in the
absence of inhibitors are shown in C (28 S) and D (39 S).
Figure 8. Fluorogram of mitoribosomal subunits from cells labeled in
the absence of cytoplasmic protein synthesis showing high molecluar
Mitoribosomal Proteins from Cells Labeled in the Absence of
Mitochondrial Protein Synthesis
Chloramphenicol is an inhibitor of mitochondrial (and bacterial)
protein synthesis. The cytoplasmic ribosames of eukaryotic cells are
not affected by this inhibitor. In a converse experiment to the one
described above, MDBK cells were labeled with [35S]methionine in
the presence of chloramphenicol, allowing label to be incorporated
only into those proteins produced by the nucleo-cytoplasnic system
(figure 3, B). As before, proteins made during the chase period (by
the mitochondrial protein synthesis system) should not have any
incorporated radioactivity. Since this inhibitor has a low toxicity
to these cells, the labeling period could be much longer (8-16 h),
producing ribosomes of higher specific activity. This allowed the
ribosomes to be prepared using two sucrose density gradients; the
first to isolate 55 S monosomes, which are then collected and
dissociated into 28 S and 39 S ribosonal subunits. The subunits are
then purified in a second sucrose density gradient. As before, the
ribosomal subunit proteins were separated by two-dimensional PAGE and
analyzed by fluorography.
The results are shown in figure 9. The fluorograms of 28 S and
39 S subunits labeled in the presence of chloramphenicol (figure 9, A
and D) show the expected pattern for mitoriboscmal proteins. Indeed,
the patterns show no significant differences from the fluorograms of
ribosomal subunits from cells labeled in the absence of inhibitor
(figure 9, C and F). In the experiment illustrated, the major
difference is in the small subunit pattern, where the control
fluorogram is weaker than the one produced in the presence of the
Figure 9. Fluorograms of mitoribosomal subunits from cells labeled in the absence of mitochondrial
protein synthesis, with chase period.
Mitochondrial ribosomal subunits from cells labeled in the presence of chloramphenicol and chased in
normal medium were analyzed by two-dimensional PAGE and fluorography. Fluorograms of the 28 S small, A,
and 39 S large, D, subunits are shown, together with corresponding schematic diagrams showing the
positions and numbers of the mitoribosomal proteins in the fluorograms, B and E. Ribosoanl subunits from
cells labeled in parallel in the absence of inhibitors are shown in C (28 S) and F (39 S). Dotted
circles indicate the position of carrier proteins in the corresponding Coomassie blue stained pattern
l ie 1.5
*9 28 3I 430
3 33% 3
7 5 06 4
81 *-11 *146
~31 25 '238 o.26
-:.' 28_32 035
41 42ol 0
-, : .46*047
inhibitor. This is at least partially due to reduced recovery of the
uninhibited sample, the stained pattern of the carrier proteins
serving as a gauge of recovery. Other factors may be involved (see
"Discussion"). The accompanying schematic diagrams (figure 9, B and
E) enumerate the proteins in the fluorogram shown which label in the
presence of chloramphenicol. The schematic diagrams show the
location for the more faintly labeled proteins (for example, the low
molecular weight proteins of the large subunit) which are marginally
observable in the photographic prints. (Some of these faint proteins
are more easily observed in figures 10 and 11.) This experiment was
performed three times without ever detecting any proteins which could
be mitoribosomal products, that is, a protein, the labeling of which
is blocked by chloramphenicol.
The patterns of mitoriboscmal proteins in these fluorograms are
virtually the same as the stained pattern characteristically formed
by the carrier bovine liver mitoribosomal proteins revealed by
Coomassie blue stain, differing only in a few proteins which failed
to label in both the inhibited and control ribosomes. These proteins
include S26, L13, L24, L33, L34, L39 and L40. (An additional group
of proteins, S12, S13, S25, L27, and L45, failed to incorporate
radiolabel in both the inhibited and control ribosames of most
samples, but gave an ambiguous result in at least one experiment.
That is, in one case either both the inhibited and the control
ribosomes, or only the inhibited riboscmes appeared to be labeled.)
Proteins which never showed incorporation of radiolabeled methionine,
probably do not contain that amino acid. They must, therefore, be
products of cytoplasmic protein synthesis, because with the known
genome of bovine mitochondria, it is predicted that all mitoribosamal
products contain methionine (see table 1).
In some cases, for example, proteins Sl and S4 in the experiment
shown, the radiolabeled spot was not ccmigratory with the stained
spot on the electropherogram. In figure 9, B, the area of the
stained spot is shown by the dotted line, while the spot on the
fluorogram is shown as a filled spot. This change in mobility may
indicate that the radiolabeled protein is a contaminant which is
comigratory with the mitoribosomal protein. Another possible
explanation is that the mitoriboscmnal protein of the MDBK cells is
slightly modified from that of the bovine liver carrier protein, such
that it occupies a similar, but not identical, place in the
electrophoretic separation. This labeling pattern was reproducible
for Sl, but S4 showed comigratory label and stain in all other
A second protocol for labeling cells in the presence of
chloramphenicol was designed to test a prediction. That is, if all
of the mitoriboscmal proteins are cytoplasnically synthesized, then
it should be possible to completely synthesize and assemble
mitoribosomes in the absence of mitochondrial protein synthesis.
This protocol is shown schematically in figure 3, C. This protocol
differs froman the one already described in two respects. First, after
the chloramphenicol is added, the cells are preincubated for 2 h to
allow any pre-existing mitochondrial protein synthetic products to be
utilized. Secondly, there is no chase period; chloramphenicol is
kept in the solutions until the end of the harvesting procedure.
The results of this protocol for proteins franom 28 S subunits are
shown in figure 10. Two data sets for this experiment are
illustrated, with 28 S subunits labeled in the presence of
chloramphenicol shown in A and C, and the respective parallel
controls shown in B and D. The two experiments are illustrated to
demonstrate the variability of the results with this protocol. In
fact, this experiment was repeated three times without ever producing
a fluorogram of the quality shown in figure 9, A. However, in all
cases, the experimental pattern was predominantly comparable to that
produced from the control cells and the "flaws" were not
reproducible. For example, in the experiment illustrated in A and B,
the experimental fluorogram is less intense than its control. This
could be due to reduced overall recovery of the subunits during the
centrifugation steps, but other factors may be involved (see
"Discussion"). Additionally, protein S14, seems surprisingly weak in
the control fluorogram (indicated in figure 10, A and B). In the
experiment illustrated in C and D, the fluorograms, both inhibited
and control, are of equal intensity but show several highly labeled
contaminating proteins. These contaminants include one protein which
is apparently a mitochondrial protein synthetic product (indicated by
the large arrowhead) since it is more easily observed in the control
fluorograms. Based on its molecular weight (approximately 19,000
daltons) it may be a product of URF 6. However, several factors are
inconsistent with its being a mitoriboscmal protein: 1) it is in a
- .*.*- ** D
Figure 10. Fluorograms of 28 S mitoriboscmnal subunits from cells
labeled in the absence of mitochondrial protein synthesis, no chase
The results for 28 S mitoribosomal subunits from two experiments
with MDBK cells labeled and harvested in the presence of
chloramphenicol are shown, A and C, together with their corresponding
parallel controls, B and D. Note the variation in the two results,
largely due to contamination in the second experiment (C and D). In
A and B, a change in protein S14 is indicated. In the second
experiment, one of the contaminating proteins (large arrowhead)
appears to be a mitochondrial product since it is more easily
obsevred in the control fluorogram. Another protein (*) curiously
appears only in the presence of chloramphenicol.
region of the gel where no mitoribosomal proteins have ever been
observed, 2) it appears in both subunits (although more prominently
in the contaminated 28 S fluorograms), and 3) it appeared only in
this experiment and not in the other two. The other protein
(indicated by an asterisk), which surprisingly appears only in the
mitoribosomal subunits from cells labeled in the presence of
chloramphenicol, also appears in both subunits (see below) but only
in this experiment.
In figure 11 are shown proteins froman 39 S subunits (froman the
same experiment as the 28 S subunits shown in figure 10, C and D,
above) labeled using this protocol in the presence of chloramphenicol
(A) and their parallel control (B). As predicted, the fluorograms
look very similar whether produced with mitoriboscmes from cells
grown in the presence of chloramphenicol or from the parallel
controls. In fact, the most obvious difference in the experiment
illustrated is the presence of the protein (indicated by an asterisk)
which appeared only in these fluorograms of mitoriboscmal proteins
labeled in the presence of chloramphenicol. More importantly, the
labeling of all of the mitoriboscmal proteins was not inhibited by
chloramphenicol in any of the three trials of this experiment.
The results of all of the labeling of mitoriboscmes in the
presence of chloramphenicol are summarized in table 4. Most of the
enumerated proteins of the bovine mitoriboscme were consistently
labeled in all conditions (the 3/3/4 result). An additional group of
proteins (e.g. S2, S3, or S9) do not always appear on the
electropherograms, as judged by the Coomassie blue stain pattern of
Figure 11. Fluorograms of 39 S mitoribosonal subunits from cells
labeled in the absence of mitochondrial protein synthesis, no chase
Figure ii. Fluorograms of 39 S mnitoribosomal subunits from cells
labeled in the absence of mitochondrial protein synthesis, no chase
The results for 39 S mitoriboscnal subunits frman MDBK cells
labeled and harvested in the presence of chloramphenicol are shown,
A, together with the corresponding parallel control, B. Note the
presence of a protein (*) which only labeled in the presence of the
the carrier, but, when present, were always labeled. As mentioned
above, some proteins were never labeled (the 0/0/0 result produced by
S26, LI3, L24, L33, L34, L39 and L40) indicating that they probably
contain no methionine. Another group of proteins (S12, S13, S25,
L27, and L45), was not labeled in most experiments, but seemed
comigratory or nearly ccmigratory with a radioactive spot in at least
one experiment. These proteins might not contain methionine, but the
results make this conclusion suspect. Importantly, when they were
labeled, these proteins labeled in the pattern expected for products
of cytoplasmic protein synthesis. Some of the proteins are difficult
to resolve in the electropherograms; for those experiments where the
identification is questionable, this is indicated in the table by the
second (decimal) position. A good example of this type of ambiguity
is protein S4. As already discussed, in the experiment shown in
figure 9, stained, carrier S4 was easily observed and there is a
radiolabeled spot which migrates at the rim of the stained spot in
both the inhibited and control conditions. However, in other
experiments, radiolabeled S4 was comigratory with its stained
carrier. Thus, it is listed as "2.1," clearly observed in two
experiments and ambiguously identified in one. Also, the cracked gel
shown in figure 10, B, made S4 impossible to identify in that control
experiment, so only three out of the four possible control conditions
are reported. Likewise, it was too pale to identify in one of the
+CAP/-chase conditions so that column is a "2" instead of a "3." In
contrast, the radiolabeled spot nearest carrier protein Sl was always
below and to the left of its expected position, as shown in
SYNTHESIS OF DEFINED MITORIBOSOMAL PROTEINS
IN THE PRESENCE OF CHLORAMPHENICOL
Results are reported for each defined mitoribosomal protein (as in
ref. ) as the nmber of separate experiments in which the
incorporation of [ S]methionine was detected. Proteins were
identified by exact comigration of the spots revealed by fluorography
( S labeled MDBK cells) and by Coomassie blue stain (bovine liver
carrier mitoribosomal proteins). Where two numbers are given (A.B)
the first number represents detection of spots which were exactly
comigratory and the second number represents spots that were in
approximately the correct area of the gel but were not clearly
comigratory with the stained spot. "N.D." (no data), indicates the
protein failed to appear on all fluorograms and in all stained
electropherograms of that condition. The prefixes "S" and "L" refer
to the small and large mitoribosomal subunits.
Protein +CAP +CAP No
Number +chase -chase Inhibitor
Sl 0.3 0.3 0.3
52 3 2 2
53 3 1 2
S4 2.1 2 2.1
55 3 2 2.1
S6 3 3 4
S7 2.1 2.1 3.1
58 3 3 4
59 1 1 2
510 3 3 3.1
Sll 1.1 0.1 2.1
S12 1.1 0 0.1
S13 0.1 0.1 0.1
Table 4 continued
Protein +CAP +CAP No
Number +chase -chase Inhibitor
S15 3 3 4
S16 3 3 4
S17 0.2 0.2 0.2
S18 2 1 2
S19 0.3 0.2 0.1
S20 1.1 0.1 2
S21 3 2.1 4
S22 1.2 3 3
S23 3 3 4
S24 N.D. N.D. N.D.
S25 1.1 0 0
S26 0 0 0
S27 2 1.1 2
S28 3 3 4
S29 3 3 4
S30 3 2.1 3
S31 3 3 4
S32 2 N.D. 2
S33 3 3 4
L1 3 3 4
L2 3 3 4
L3 3 3 4
L4 3 3 4
Table 4 continued
Protein +CAP +CAP No
Number +chase -chase Inhibitor
L5 3 3 4
L6 3 3 4
L7 3 3 4
L8 3 3 4
L9 2 3 3
L10 3 3 4
LII 3 3 4
L12 3 3 4
L13 0 0 0
L14 3 3 4
L15 3 3 4
L16 3 3 4
L17 3 3 4
L18 3 3 4
L19 3 3 4
L20 3 3 4
L21 3 3 4
L22 3 3 4
L23 3 3 4
L24 0 0 0
L25 2.1 3 3.1
L26 3 3 4
L27 0 0 0.1
L28 3 3 4
Table 4 continued
Protein +CAP +CAP No
Number +chase -chase Inhibitor
L29 2 3 4
L30 2 3 4
L31 3 3 4
L32 3 3 4
L33 0 0 0
L34 0 0 0
L35 3 3 4
L36 N.D. N.D. N.D.
L37 3 3 4
L38 3 3 4
L39 0 0 0
L40 0 0 0
L41 3 3 4
L42 3 3 4
L43 3 3 4
L44 3 3 4
L45 0 0.1 0.1
L46 3 3 4
AL47 3 3 4
L48 2 2 4
L49 N.D. 3 4
L50 2 3 3.1
L51 N.D. 3 4
L52 N.D. 3 4
figure 9, B. However, it was observed in all of the experiments
except the cracked gel (figure 10, B) and, therefore, was scored as
0.3/0.3/0.3. This type of analysis gave results consistent with a
cytoplasmic site of synthesis for each of the enumerated proteins of
the mammalian mitochondrial ribosome.
Effect of Inhibitors
The interpretation of the ribosomal pattern is determined by the
effect of the inhibitor under each condition. In order to
demonstrate that the inhibitors produce the expected results, it is
necessary to show both that mitochondrial protein synthesis continued
in the presence of cycloheximide and that mitochondrial protein
synthesis ceased completely in the presence of chloramphenicol.
Aliquots of intact mitochondria franom the labeled cells (see
above) were analyzed by one-dimensional PAGE to show mitochondrial
protein synthetic products in cells labeled in the presence of
cycloheximide. In figure 12, lane A shows mitochondria from cells
labeled in the presence of cycloheximide. The expected limited band
pattern of mitochondrial products is produced from cells labeled in
the presence of cycloheximide. Because this system (modified from
Leister and Dawid, 1974 , see "Methods") has not been charac-
terized, specific assignment of known mitochondrial products to the
bands is not possible. This demonstrates that cytoplasmic protein
synthesis was eliminated by the cycloheximide but that mitochondrial
incorporation continued. In a comparatively short exposure, the
mitochondria from cells labeled in the presence of chloramphenicol
Figure 12. Total mitochondrial incorporation in the presence and
absence of inhibitors.
A sample of mitochondria from each labeling condition was
analyzed by one-dimensional SDS-PAGE and fluorography. Mitochondria
from cells labeled in the presence of cycloheximide, A, exposed for
70 days. Mitochondria from cells labeled in the presence of
chloramphenicol, B, and their parallel control, C, exposed for 18 h.
display a band pattern identical to their parallel control (figure
12, B and C). The pattern of these control cells shows a saturating
pattern of total cellular protein synthesis. Because of the
overwhelming pattern produced by cytoplasmic protein synthesis, the
absence of mitochondrial products is not observable in lane B.
The mitochondrial protein synthetic products are only observable
with one-dimensional PAGE in the absence of cytoplasmic protein
synthesis (in cells also labeled with cycloheximide). Therefore, the
effect of chloramphenicol could not be observed in the cells prepared
for the isolation of mitochondrial ribosomes. The effect of
chloramphenicol was studied using cells grown in T150 flasks and
labeled under a variety of inhibitor conditions. After a 2 h
labeling period with [35S]nmethionine, the cells were harvested and
analyzed by one-dimensional SDS-PAGE. The results are presented in
figure 13. (Because each lane is the total SDS-extractable material
from the cells of one T150 flask, the lanes are overloaded and the
band pattern is somewhat distorted.) Cells labeled in the absence of
inhibitors or in the presence of chloramphenicol show a complex,
heavily labeled pattern (lanes A and B, respectively). In the
presence of only cycloheximide, this pattern is eliminated and a
limited band pattern, the mitochondrial protein synthetic products,
is observed (lane C). This pattern is completely eliminated by the
additional presence of chloramphenicol at either 100 (lane E) or 200
(lane D) pg/ml. At very long exposures, the one-dimensional PA3E
separations from cells labeled in the presence of both inhibitors
show a complex band pattern (lanes F and G). This pattern is
A BC D E
Figure 13. Effect of inhibitors on MDBK cells.
MDBK cells were labeled with [ Simethionine in the presence
or absence of the inhibitors cycloheximide and chloramphenicol. The
total cellular material was analyzed by one-dimensional SDS-PAGE and
fluorography. A, no inhibitor; B, 200 pg/ml chloramphenicol; C,
300 pg/ml cycloheximide; D and F, 300 pg/ml cycloheximide and
200 pg/ml chloramphenicol; E and G, 300 pg/ml cycloheximide and
100 pg/ml chloramphenicol. Fluorograms were exposed for 17 h (A
and B), 4 days (C-E), and 14 days (F and G).
similar to a very short exposure of the products of uninhibited
cells, demonstrating that the cycloheximide inhibition is not
complete. However, there is no indication of the limited band
pattern of mitochondrial protein synthetic products at either
concentration of chloramphenicol, even at this long exposure.
Precursors of Mitoriboscmal Proteins
Antibodies raised against the proteins of the large (39 S)
mitoribosomal subunit were used in an attempt to detect unassembled
ribosomal proteins in the soluble fractions of the MDBK cells. TWo
cell fractions were tested, the post-mitochondrial supernatant (PMS)
which should contain the soluble proteins of the cytoplasm, and the
soluble fraction froman the mitochondrial lysate, which should contain
the soluble proteins of the mitochondrial matrix (MM) as well as the
Triton X-100 soluble proteins of the mitochondrial membranes. Each
fraction was immunoprecipitated with two different preparations of
antiserum, as well as with the corresponding preimmune sera and with
buffered saline solution (buffer H). The imnuine precipitates were
analyzed by one-dimensional SDS-PAGE and fluorography.
The results are presented in figure 14. Because of non-specific
precipitation and trapping, there are radiolabeled proteins
observable in all of the fractions. Of interest however, are those
proteins (see arrowheads in figure 14) which appear only in the
immune sera precipitations, suggesting that they are specifically
mitoribosomal proteins. Some of these proteins were
EF G H
Figure 14. Immune precipitation of mitoribosomal proteins from
soluble cell fractions.
Cell fractions isolated from cells labeled with
[35S]methionine in the presence of chloramphenicol were
precipitated with antisera against the proteins of 39 S mitoriboscmal
subunits and analyzed by one-dimensional SDS-PAGE and fluorography.
A-D, post-mitochondrial supernatant fraction; E-H, mitochondrial
matrix fraction. In A and E, the antigen was analyzed directly; in
the other lanes the antigen was first treated with buffer H (B and
F), pre-imnmune serum (C and G), and inine serum (D and H), and then
with fixed S. aureus cells. The inmune precipitates were analyzed
by one-dimensional SDS-PAGE and fluorography. The arrowheads
indicate the positions of bands specifically precipitated by the
inunmmunoprecipitated from each of the cell fractions, indicating that
there is an unassembled pool of at least same mitoribosomal proteins
in the cytoplasm and in the mitochondria. Because digitonin was used
during the cell lysis, the internal contents of some of the
membraneous organelles (Golgi, endoplasmic reticulum, etc.) may have
been released at that step, so it cannot be positively concluded that
the cytoplasmic proteins are free of proteins from other
Theoretically, this type of analysis should also be able to
indicate the possibility of transit peptide cleavage, that is,
proteins which are immmunoprecipitated from both the
post-mitochondrial supernatant and the mitochondrial matrix
fractions, but which appear to be of lower molecular weight when
isolated from the latter. This experiment did not indicate that type
of change. Two factors may be involved; 1) the antisera used were
raised against total 39 S ribosomal subunits and their broad
recognition spectrum did not allow a specific protein to be
distinguished so that its molecular weight in the two compartments
could be directly compared, and 2) the patterns of non-specific
precipitation are so complex that they could obscure that type of
band position shift.
The mitochondrion participates in its own biosynthesis. It
contains within it a complete protein synthetic system, including
ribosomes, mRNA, tRNA, and the necessary accessory proteins. The
intraorganellar location of the mitochondrial ribosome raises several
questions about how its components are synthesized and assembled.
Based upon what is known about the synthesis of other mitochondrial
proteins and of mitoribosomes in yeast and Neurospora, it can be
envisioned that most or all of the mitoribosomal proteins are
synthesized individually in the cytoplasn and imported into the
mitochondrion where they begin a specific, ordered assembly onto an
incompletely cleaved transcript of the complete mitochondrial rDN.
region. Specific proteins assembling onto the transcript might cause
it to assume secondary or tertiary structures which signal for
cleavage of the rRNA or for assembly of still more of the
mitoriboscmal proteins. This process of structural rearrangement and
cleavage of the rRNA and of addition of mitoribosomal proteins
continues until the riboscme is completed.
Some of the 85 proteins of the mammalian mitoribosome
undoubtedly serve in a functional capacity during protein synthesis,
but others might serve a purely structural function, either creating
a structural assembly site for rRNA or protein during mitoribosomal
synthesis, or perhaps adding necessary size to the organelle in place
of some of the rRNA seen in other classes of ribosomes. The current
data on the mitochondrially produced proteins of the yeast and
Neurospora mitoribosones indicate that they may serve as a rRNA or
protein assembly site during the synthesis of the small subunit. If
so, this leads to questions as to whether this function is obligatory
and if intramitochondrial synthesis is absolutely required for a
protein serving that function.
This work has addressed questions of the synthesis of
mitochondrial ribosomes in nammalian systems, using the bovine system
as a model. The major concentration has been on determining whether
or not any of the mitoribosomal proteins are products of
mitochondrial protein synthesis. Additionally, one experiment has
looked at the presence of soluble pools of mitoriboscmal proteins
both intra- and extramitochondrially.
Protein Synthesis in Isolated Mitochondria
One approach used was to examine the protein synthetic products
of isolated bovine mitochondria. The radiolabeled proteins
synthesized by intact mitochondria isolated from bovine liver were
analyzed by two-dimensional PAGE and compared to a pattern of
authentic bovine mitoribosomal proteins added as visible carrier.
Mitoribosomal proteins which are comigratory with radiolabeled
proteins are identified as possible mitochondrial protein synthetic
products. One radioactive pattern was produced by this technique,
and its spots included one which was ccmigratory with mitoribosomal
protein AL44 and spots which migrated in the vicinity of proteins S13,
S20 and L20.
The advantage of this technique is that it studies the
mitochondrial protein synthesis system independently of the rest of
the cell. If any cytoplasmic (80 S) ribosomes are present (as trace
contamination) they cannot incorporate the radioactive amino acids
because they lack an energy source and the necessary soluble
components (e.g. charged tRNA's). If growing bacteria are present in
the mixture, however, they can incorporate the radiolabel so care
must be taken to avoid introducing bacteria during the preparation of
the mitochondria, or inhibitors can be used which differentiate
bacterial protein synthesis.
In theory, the major limitation of this system is that it has no
specific selection criterion for mitochondrial ribosomal proteins.
This in vitro system does not process or assemble proteins into
ribosomal particles, so the analysis includes all radiolabled
products extracted from intact mitochondria. The fluorogram reveals
a pattern of several scattered spots and there is no method for
determining coincidental comigration. Furthermore, since not all of
the authentic mitoribosomal proteins are always clearly resolved
, some of the radiolabeled spots are in an area where comigration
cannot be clearly delineated. Additionally, the presence of so many
spots on the electropherogram indicates that not all of the spots are
simple translation products of the mtDNA reading frames. One
possibility is that some of the spots could represent modified forms
of these proteins. It is not known whether or not isolated
mitochondria process proteins exactly like their in vivo
counterparts. While there have been some reports of modifications to
mitochondrial protein synthetic products [114-116], the identified
proteins produced by mammalian mitochondria correspond to their
reading frames [59-67,117,118].
In practice, the most disconcerting aspect of this approach was
its irreproducibility. Two subsequent attempts to reproduce the
first result failed, apparently because the radiolabeled proteins
were not extracted from the mitochondria. Interestingly, the 9 M
urea extraction removes virtually all mitoribosanal proteins from the
isolated ribosomal particles. If a mitoriboscmal protein is one of
the mitochondrial products, its extraction is probably blocked by the
presence of mitochondrial membranes. In general, systems for the
electrophoretic analysis of the mitochondrial protein synthetic
products rely on detergents for solubilization. However, the
electrophoresis system used in these experiments which characterizes
the mitoribosomal proteins is not compatible with detergents, so
their use was avoided. In one case, where Triton X-100 was added to
the extraction mixture, the extraction of these proteins appears to
have been improved, but when the detergent was removed during the
first electrophoretic dimension, the proteins aggregated. These
results suggest that the original "successful" experiment, might not
be a separation of ribosamal product proteins. One possibility is
that this radioactivity was incorporated into easily extracted
proteins of contaminating bacteria. Thus, this technique identifies
mitoribosomal proteins which might be products of mitochondrial
protein synthesis, but it cannot be conclusive.
Protein Synthesis in Cultured Bovine Cells
By using specific inhibitors, mitochondrial and cytoplasmic
protein synthesis were "isolated" and studied in a complete living
system, bovine tissue culture cells (MDBK cells). By studies in the
MDBK cells it was shown that the proteins of mammalian mitoribosomes
label in a manner expected for products of cytoplasmic protein
synthesis. That is, they fail to label in the presence of
cycloheximide and do incorporate label in the presence of
chloramphenicol. Furthermore, correctly sedimenting ribosomal
subunits were synthesized completely in the presence of
chloramphenicol, and the analysis of their proteins indicated normal
assembly of mitoribosomes even in the absence of mitochondrial
This experimental system offers several advantages. Firstly,
the tissue culture cells provide a complete system for studying
mitochondrial biosynthesis. Both cytoplasmic and mitochondrial
protein synthesis can be examined. Although not included in this
study, it is of course possible to manipulate other aspects of
mitoribosame biosynthesis (e.g. rRNA synthesis or protein import), as
Secondly, there is a built in selection criterion for
mitoribosomal proteins. In other systems, where one or a few
specific proteins have been studied, the analysis was performed by
solubilizing the mitochondria with detergents and then precipitating
the protein of interest with a specific antibody . For the
mitoribosomal proteins, however, a complete spectrum of
characterized, specific antibodies does not exist. Therefore
selection for mitoriboscmal proteins was done by allowing the cells
sufficient time, either in the absence or presence of inhibitors, to
assemble proteins into mitoriboscmal particles. Then, the ribosomal
subunits could be prepared in sucrose gradients and their protein
composition analyzed. The analysis is aided by the use of bovine
cells, as opposed to some other species, because the electrophoretic
pattern of the enumerated mitoribosomal proteins is well
Finally, bovine liver mitoribosomes are easily obtained for
protein carrier. When working with the small amounts of
mitoribosomal protein from tissue culture cells, these carrier
mitoriboscmes serve several uses. Added at the time of mitochondrial
lysis, after the riboscmes are labeled and assembled, they add needed
bulk to the material being isolated. This not only makes the
manipulations easier, but is particularly important in reducing
adsorptive and handling losses. During the purification of
ribosomes, it is the A260 measurement of the carrier which is
followed and used to judge fractions to be pooled and to judge total
recovery. Thus, by definition, mitoribosomal proteins are those
which are associated with the correct gradient positions for intact,
fully assembled mitoribosomal subunits. Precursor particles, which
if present would run at a higher position in the gradient, would not
be detected by this method. During the electrophoretic analysis, the
carrier pattern shows the correct position for the ribosomal
proteins, and exact comigration can be used to identify mitoribosomal
proteins and distinguish contamination.
In analyzing these results, several factors must be considered.
The blank fluorograms which resulted with the inhibitor
cycloheximide would also have been produced if the conditions for the
labeling period (media, length of incubation, specific activity of
radiotracer, etc.) or for recovery and analysis of ribosomnal subunits
(mitochondrial isolation, length of fluorography, etc.) were
inadequate to visualize synthesis into mitoribosomal proteins. A
parallel control experiment, however, where all such factors were the
same except that cycloheximide %s emitted, demonstrated that the
synthesis of mitoribosomal proteins was observable under these
conditions. Additionally, it is known that mitochondrial protein
synthesis is sensitive to the inhibition of cytoplasmic protein
synthesis, that during prolonged incubations in the absence of
cytoplasmic protein synthetic products, mitochondrial protein
synthesis becomes inhibited and eventually mitochondrial products are
degraded. In order to insure that the cycloheximide labeling
condition was not excessive, a sample of intact mitochondria from
these cells was examined. It showed the expected limited number of
mitochondrially synthesized protein bands in one-dimensional
SDS-PAGE, demonstrating that the products of mitochondrial protein
synthesis were translated during the incubation period and that they
were not degraded during the labeling period, but, in fact, they were
intact and could be assembled during the chase period. Additionally
the electrophoretic band pattern of proteins produced by mitochondria
from cells labeled in the presence of cycloheximide (and then chased
and used for the preparation of mitoriboscmes, figure 12, A) is
roughly similar to that of total protein synthesized by cells labeled
under the same conditions to test the inhibitor effectiveness (no
chase period, figure 13, C). However, the possibility exists that a
mitoribosomal protein might be a minor product of mitochondrial
protein synthesis which is not discernable by one-dimensional
SDS-PAGE analysis. Additionally, sequestration inside the
mitochondrial membrane might give added protection from proteolysis
to electron transport proteins, which would not necessarily be true
for a riboscmal protein. Since the effect of these factors cannot be
evaluated, the blank fluorograms are not conclusive data.
Another important factor is the effectiveness of chloramphenicol
inhibition of mitochondrial protein synthesis in these cells. In
these experiments, the analysis examines only those proteins which
assemble into properly sedimenting mitoriboscmes. Because of the
small amount of material available and the addition of carrier liver
mitoribosomes, the results would not distinguish between unaffected
incorporation and partial incorporation caused by incomplete
chloramphenicol inhibition. Figure 13 demonstrates that
chloramphenicol is very effective in MDBK cells, and reduces
mitochondrial protein synthesis below detectable levels even at 100
ig/ml. One problem with this control experiment is that
chloramphenicol inhibition could not be tested directly with the
cells used to produced the riboscmes shown in figures 9, 10, and 11,
but had to be tested in a separate set of cells grown specifically
for that purpose. However, all indications are that the
chloramphenicol inhibition in the MDBK cells was complete.
All of these experiments were performed using [3 S]methionine
as the radioactive amino acid. This analysis is limited to detection
by 35S (either methionine or cysteine) because only this isotope
can be used to synthesize amino acids of adequate specific activity
for detection by these methods. Fortunately, since the amino acid
sequence of all possible mitochondrial products can be predicted from
the known DNA sequence (see table 1), it appears that all
mitochondrial protein synthetic products contain a high proportion of
methionine residues. Additionally, the methionine residues of these
predicted polypeptides are distributed throughout the sequences,
making it unlikely that all of them would be removed by any
hypothetical processing of the transcript or of the protein. Indeed,
those proteins (e.g. L13) missing on fluorograms of mitoribosomes
from cells labeled in both the presence and absence of
chloramphenicol probably simply lack methionine residues. These
unlabeled proteins, therefore, cannot be mitochondrial products.
Finally, this analysis is based upon the ability to detect all
of the mitoribosomal proteins by a specific system of two-dimensional
PAGE. Yet a few of the mitoribosomal proteins are variable in their
appearance in this system of analysis. For example, protein L36 was
never detected on any of the electropherograms or fluorograms from
these experiments. Additionally, the possibility must be considered
that there are ribosomal proteins that are never detected in the
two-dimensional PAGE analysis. A protein would be consistently
overlooked in this analysis if it was comigratory with another
protein (one "spot" which is actually two proteins) or if it was not
solubilized by the extraction conditions. By analogy to lower
eukaryotes, these data also address the possibility that a
mitochondrially produced mitoribosonal protein might be missed in
this analysis because it was not displayed in these two-dimensional
PAGE separations. It is known that in yeast and Neurospora, the
small ribosomal subunits formed in the absence of mitochondrial
protein synthesis are defective. They migrate incorrectly in
sedimentation analysis, and they specifically lack a few proteins
which assemble after the mitochondrial product. In our bovine
system, the subunits examined were those which migrated to the
position of assembled subunits in sucrose density gradients. In the
protocol shown in figure 3, B (cells labeled in the presence of
chloramphenicol and then chased) if chloramphenicol blocks
mitoribosome synthesis at a precursor stage, then the particles made
during the labeling in the presence of chloramphenicol would have
been completed during the chase period. Proteins added during the
chase period should not be labeled, though. So the particles
recovered from the gradients would be deficiently labeled in those
proteins which were added after any mitochondrial product.
Therefore, it is important to note that the subunits produced by this
experiment (figure 9) produce the same pattern of spot intensities as
their parallel control.
However, that experiment does not address the possibility that
proteins made cytoplasmically during the labeling period, might still
be available for assembly during the chase period. This situation
could occur either for proteins which required several minutes for
import into the mitochondria, or if unassembled proteins inside the
organelle are not quickly degraded. If such proteins are retained
into the chase period, then, when the inhibition is released, a
mitochondrially synthesized protein could be produced and assembly
could continue. These scenarios are addressed by the second protocol
for labeling in the presence of chloramphenicol (no chase period,
figure 3, C). As seen in figures 10 and 11, mitoribosomes from cells
labeled and assembled in the presence of chloramphenicol show the
same patterns as their uninhibited counterparts. This indicates that
the complete absence of detectable mitochondrial protein synthesis
did not block synthesis or assembly of both subunits of the
mitoribosome. If mammalian mitoribosomes contained a protein
analogous to either S-5 in Neurospora or var 1 in yeast, this
experiment should have produced a distorted or deficient pattern.
These data do not completely rule out the possibility of
"hidden" mitoribosomal proteins, proteins which were not detected
either because they are not present in the electropherogram or
because they are not distinguished from another similarly migrating
protein. However, if such a mitoriboscmal protein does exist, it
does not affect the ability of any other protein to assemble onto the
mitoribosome, and, therefore, is functioning in some manner which
differs from the behavior of the protein of the mitoribosomes in
Continuously disconcerting aspects of these experiments were the
problems in producing good quality fluorograms, especially of the
small ribosomal subunit. Problems were anticipated in incorporating
sufficient radioactivity and in obtaining adequate recovery of the
subunits (see above). However, some experiments produced excellent
fluorograms of the subunits (e.g. figure 9). Several of the small
subunit fluorograms were also damaged by differing degrees of
contamination. This latter factor probably indicates that the
preparation method for obtaining mitoriboscmes franom tissue culture
cells is not optimized. However, this contamination is independent
of mitoribosome synthesis, since, when it occurred, it was present
equally in both the experimental and control fluorograms.
In general, the small ribosomal subunits produced weak
fluorograms. The two exceptions were fluorograms of subunits froman
cells labeled in the presence of chloramphenicol and then given a
chase period in normal media. These fluorograms were significantly
better than those of small subunits either frman cells labeled
continuously in the presence of chloramphenicol or from control
cells. Two factors may be involved. Firstly, it has frequently been
reported that there is a burst of mitochondrial synthetic activity in
cells being released from chloramphenicol inhibition. This increased
level of synthesis may maximize assembly of mitoriboscmes in cells
labeled by this protocol. Secondly, Lambowitz's group has noted that
especially the synthesis of the small mitoribosomal subunit in
Neurospora, more so than the other mitochondrially produced
proteins, is sensitive to a mutation which causes a defect in the
F ATPase . That is, weak fluorograms produced from cells
labeled continuously in the presence of chloramphenicol may reflect a
repression of synthesis which is not directly caused by inhibition of
the synthesis of a mitoribosomal protein.
In summary, the experiments in cultured bovine cells provide the
most complete analysis of the site of synthesis of the proteins of
the mammalian mitochondrial ribosome. All of the experiments in this
system indicate that in mammals, as opposed to the lower eukaryotes,
all of the proteins of the mitoribosome are cytoplasmically
Pools of Unassembled Mitoribosomal Proteins
Knowing the site of synthesis of the mitoriboscomal proteins,
cytoplasm or mitochondrial matrix, answers only one of several
questions raised by mitoribosome biosynthesis. Other questions
raised include 1) Are the proteins synthesized on bound or free
ribosomes and, if bound, to what membrane? 2) In what cellular
compartment do unassembled mitoribosomal proteins exist? 3) How much
time elapses between translation in the cytoplasm and import into the
mitochondria?, 4) What is the mechanism of import into the
mitochondria?, and 5) What is the assembly pattern of mitoribosomes?
Virtually all of these questions can theoretically be addressed by
immune precipitations of mitoribosomal proteins and their precursors
in carefully prepared cellular fractions.
This type of study was not a major part of this work, but one
experiment was performed which utilized cell fractions prepared
during the labeling studies in MDBK cells. This imminune precipitation
with a broad spectrum antiserum raised against the proteins of the
large mitoriboscmnal subunit detected some unassembled mitoribosomal
proteins in both the post-mitochondrial supernatant and the
mitochondrial matrix (figure 14). Furthermore, since mitoribosomal
proteins could be detected among the soluble proteins of the
post-mitochondrial supernatant, this suggests that these proteins are
free in the cytoplasm and possibly synthesized on free polysomes.
Two problems limit the amount of information which can be
obtained from this experiment. First, the conditions used for the
precipitation were not optimized and the one-dimensional SDS-PAGE
profiles showed several contaminating proteins. In some cases, what
appear to be specific precipitation bands might just be increased
aggregation of contaminating proteins in the lattice of antibodies
and protein A. Second, the cells were lysed using a hypotonic
digitonin solution. Digitonin is a detergent which solubilizes all
cellular membranes except the inner mitochondrial memnbrane .
Although the amount of digitonin used was below the concentration
necessary to disrupt membranes thoroughly, it may have caused some
leakage of, for example, the contents of the endoplasmic reticulum or
the Golgi into the cytoplasm. Therefore, the "soluble proteins of
the cytoplasm" nay include proteins which were actually sequestered
during translation on bound riboscmes.
Ideally, this type of experiment is probably best performed by
using monoclonal or adsorbed polyclonal antibodies to follow the
progression of a specific protein. Also, the cell fractions should
be prepared differently. For example, inhibitors of mitochondrial
import could be used during a labeling period to prepare a fraction
for the study of the extramitochondrial form of these proteins, or
ethidium bromide could be used to inhibit the synthesis of
mitochondrial ribosomal RNN to study the imported but unassembled
proteins. The major problem with such experiments would probably be
incorporating sufficient radiolabel into the proteins, since the
inhibitors would probably necessitate a relatively short labeling
Implications for Mitochondrial Evolution and Biogenesis
This project was originally designed to determine which of the
proteins of the mammalian mitochondrial ribosome was a mitochondrial
product, in hopes of assigning it to one of the unidentified
mitochondrial genes (URF's). That the manmmalian mitoribosomal
proteins are all products of the nucleo-cytoplasmic system was an
unexpected result. This reflects yet another difference between the
mitochondrial protein synthetic systems of higher and lower
eukaryotes. If indeed, the mitochondrial product protein in lower
eukaryotes functions in ribosome assembly, that function is performed
in mammal s by a protein or proteins imported from the cytoplasm. It
might be that a protein of identical function is imported. This
would argue against theories that the mitochondrially produced
proteins have retained intramitochondrial synthesis because they are
"too hydrophobic", or have some other structural feature which makes
them incapable of import. Because of the differences in the
structure of the two classes of mitoriboscmes, it is also possible
that the ribosomal domain involved is sufficiently changed in mammals
that a different assembly mechanism is used. Considering the highly
proteinaceous nature of the mammalian mitoribosome, perhaps this is
not so surprising. It appears that regions of RNA and protein in
other ribosomal classes have been replaced in the mammalian
mitoribosome with structures containing only protein. It might be
that the ribosomal region of S-5 or var 1 in lower eukaryotes is
functionally the same in mammalian mitoribosaomes but structurally
This result has important implications for mitochondrial
evolution and biogenesis. The only other known example of a change
in mitochondrial products is the "DCCD-binding protein" of the
F ATPase, which is made mitochondrially in yeast and
cytoplasmically in Neurospora and mammals. However, this change
does not seem to affect the general structure of the mitochondrial
gene map in lower eukaryotes. In contrast, the genarme of mammals
seems very efficient. Its reduced size evolved predominantly by
eliminating intervening sequences. Although an exact number of
reading frames of the mitochondrial gencarme of lower eukaryotes is not
yet available, the number of mitochondrially produced structural
proteins seems comparable in lower eukaryotes and mammals. The high
density of information on mammalian mitochondrial DNP. suggests that
mitochondria are under selection for conservation of genetic
material, although the reason for such a selective pressure is
difficult to discern. The close homology of the mitochondrial DNN
reading frames in species as diverse as human and mouse, suggests
that the proteins encoded by these genes are required for proper
mammalian mitochondrial function. Yet, the data presented here
demonstrate that mitochondria are not obligated to make a set of
specific types of proteins, but that the protein synthetic products
are evolving in response to, or in partnership with the proteins
encoded on the nuclear gename.
1. Teintze, M., Hennig, B., Schleyer, M., Schmidt, B., and Neupert,
W. (1982) Biogenesis of mitochondrial membrane proteins. In:
Membranes in Growth and Developmnent (Hoffman, J.F., Giebisch,
G. H., and Bolis, L., eds) pp. 37-47, A. R. Liss Inc., New York.
2. Schatz, G. (1979) How mitochondria import proteins from the
cytoplasm. Fed. Eur. Biochem. Soc. Letters 103, 203-211.
3. Kellems, R. E., and Butow, R. A. (1972) Cytoplasmic-type 80S
ribosomes associated with yeast mitochondria. I. Evidence for
ribosome binding sites on yeast mitochondria. J. Biol. Chem.
4. Kellems, R. E., Allison, V. F., and Butow, R. (1974) Cytoplasmic
type 80S ribosomes associated with yeast mitochondria. II.
Evidence for the association of cytoplasmic ribosomes with the
outer mitochondrial membrane in situ. J. Biol. Chem. 249,
5. Kellems, R. E., and Butow, R. (1974) Cytoplasmic type 80S
ribosomes associated with yeast mitochondria. III. Changes in
the amount of bound ribosomes in response to changes in
metabolic state. J. Biol. Chem. 249, 3304-3310.
6. Kellems, R. E., Allison, V. F., and Butow, R. (1975) Cytoplasmic
type 80S ribosomes associated with yeast mitochondria. IV.
Attachment of ribosomes to the outer membrane of isolated
mitochondria. J. Cell Biol. 65, 1-14.
7. Godinot, C., and Lardy, H. A. (1973) Biosynthesis of glutamate
dehydrogenase in rat liver. Demonstration of its microsamal
location and hypothetical mechanisms of transfer into
mitochondria. Biochemistry 12, 2951-2060.
8. Hallermayer, G., Zimmermann, R., and Neupert, W. (1977) Kinetic
studies on the transport of cytoplasmically synthesized proteins
into the mitochondria in intact cells of Neurospora crassa.
Eur. J. Biochem. 81, 523-532.
9. Harmey, M. A., Hallermayer, G., Korb, H., and Neupert, W. (1977)
Transport of cytoplasmically synthesized proteins into the
mitochondria in a cell free system from Neurospora crassa.
Eur. J. Biochem. 81, 533-544.
10. Kadenbach, B. (1966) Synthesis of mitochondrial proteins:
Demonstration of a transfer of proteins froman microsoames into
mitochondria. Biochim. Biophys. Acta 134, 430-442.
11. Kawajiri, K., Harano, T., and Omura, T. (1977) Biogenesis of the
mitochondrial matrix enzyme, glutamate dehydrogenase, in rat
liver cells. I. Subcellular localization, biosynthesis, and
intracellular translocation of glutamate dehydrogenase. J.
Biochem. 82, 1403-1416.
12. Hallermnayer, G., and Neupert, W. (1976) Studies on the synthesis
of mitochondrial proteins in the cytoplasm and on their
transport into the mitochondrion. In: Genetics and Biogenesis
of Chloroplasts and Mitochondria (Bucher, Th., Neupert, W.,
Sebald, W., and Werner, S., eds) pp. 807-812, North Holland, New
13. Robbi, M., Berthet, J., Trouet, A., and Beaufay, H. (1978) The
biosynthesis of rat liver cytochrome c. I. Subcellular
distribution of cytochrome c. Eur. J. Biochem. 84, 333-340.
14. Robbi, M., Berthet, J., Trouet, A., and Beaufay, H. (1978) The
biosynthesis of rat liver cytochrome c. II. Subcellular
distribution of newly synthesized cytochrome c. Eur. J.
Biochem. 84, 341-346.
15. Ades, I. Z. (1982) Transport of newly synthesized proteins into
mitochondria -- A review. Mol. Cell. Biol. 43, 113-127.
16. Conboy, J. G., and Rosenberg, L. E. (1981) Posttranslational
uptake and processing of in vitro synthesized ornithine
transcarbamylase precursor by isolated rat liver mitochondria.
Proc. Nat. Acad. Sci. USA 78, 3073-3077.
17. Fluckiger, J., Behra, R., Jaussi, R., and Christen, P. (1982)
Intracellular location of the precursor of mitochondrial
aspartate aminotransferase and its in vitro import into
mitochondria. Experientia 38, 740.
18. Marra, E., Passarella, S., Doonan, S., Quagliariello, E., and
Saccone, C. (1980) Protease resistance of aspartate
aminotransferase imported in mitochondria. Fed. Eur. Biochem.
Soc. Letters 122, 33-36.
19. Miralles, V., Felipo, V., Hernendez-Yago, J., and Grisolia, S.
(1983) Transport of the precursor for rat liver glutamate
dehydrogenase into mitochondria "in vitro." Biochem. Biophys.
Res. Coamn. 110, 499-503.
20. Morita, T., Miura, S., Mori, M., and Tatibana, M. (1982)
Transport of the precursor for rat liver ornithine
carbamoyltransferase into mitochondria in vitro. Eur. J.
Biochem. 122, 501-509.
21. Ries, G. Hundt, E., and Kadenbach, B. (1978)
Immunoprecipitation of a cytoplasmic precursor of rat-liver
cytochrcme oxidase. Eur. J. Biochem. 91, 179-191.
22. Sakakibara, R., Kamisaki, Y., and Wada, H. (1981) Import of a
putative precursor of rat liver mitochondrial glutamic
oxaloacetic transaminase into mitochondria. Biochem. Biophys.
Res. Ccmm. 102, 235-242.
23. Shore, G. C., Carignan, P., and Raymond, Y. (1979) In vitro
synthesis of a putative precursor to the mitochondrial enzyme,
carbamyl phosphate synthetase. J. Biol. Chem. 254,
24. Zwizinski, C., Schleyer, M., and Neupert, W. (1983) Transfer of
proteins into mitochondria: Precursor to the ADP/ATP carrier
binds to receptor sites on isolated mitochondria. J. Biol.
Chem. 258, 4071-4074.
25. Maccecchini, M.-L., Rudin, Y., and Schatz, G. (1979) Transport
of proteins across the mitochondrial outer membrane. A
precursor form of the cytoplasmically made intermembrane enzyme
cytochrame c peroxidase. J. Biol. Chem. 254, 7468-7471.
26. Aziz, L. E., Chien, S.-M., Patel, H. V., and Freeman, K. B.
(1981) A putative precursor of rat liver mitochondrial malate
dehydrogenase. Fed. Eur. Biochem. Soc. Letters 133, 127-129.
27. Morita, T., Mori, M., Tatibana, M., and Cohen, P. P. (1981) Site
of synthesis and intracellular transport of the precursor of
mitochondrial ornithine carbamoyltransferase. Biochem. Biophys.
Res. Comm. 99, 623-629.
28. Northemann, W. Schmelzer, E., and Heinrich, P. C. (1981) The
size and distribution of cytochrome c oxidase subunit IV mRNA
between free and membrane-bound polyribosomes. Eur. J.
Biochem. 119, 203-208.
29. Suissa, M., and Schatz, G. (1982) Import of proteins into
mitochondria: Translatable mRNAs for imported mitochondrial
proteins are present in free as well as mitochondria-bound
cytoplasmic polyscmes. J. Biol. Chem. 257, 13048-13055.
30. Heinrich, P. C., Northemann, W., and Schmelzer, E. (1981)
Proteolytic precursor processing in the biosynthesis of
mitochondria. Acta biol. med. germ. 40, 1451-1463.
31. Kaput, J., Golfz, S., and Blobel, G. (1982) Primary structure of
a signal peptide for post-translational translocation into
mitochondria. J. Cell Biol. 95, 275a.
32. Bennett, J. (1982) Sic transit peptide. Trends in Biochem.
Sci. 7, 269.
33. Walter, P., and Blobel, G. (1981) Translocation of proteins
across the endoplasmic reticulum. II. Signal recognition protein
(SRP) mediates the selective binding to microsomal membranes of
in vitro assembled polysomes synthesizing secretary protein.
J. Cell Biol. 91, 551-556.
34. Ades, I. Z., and Harpe, K. G. (1981) Biogenesis of mitochondrial
proteins. Identification of the mature and precursor fonnrms of
the subunit of 6-aminolevulinate synthetase franom embryonic chick
liver. J. Biol. Chem. 256, 9329-9333.
35. Cote, C., Solioz, M., and Schatz, G. (1979) Biogenesis of the
cytochrome bc, complex of yeast mitochondria: A precursor form
of the cytoplasmically made subunit V. J. Biol. Chem. 254,
36. Jaussi, R., Sonderegger, P., and Christen, P. (1981)
Demonstration of a precursor of mitochondrial aspartate
aminotransferase (mAATase) in cultured chicken embryo
fibroblasts (CEF). Experientia 37, 626-627.
37. Mori, M., Morita, T., Ikeda, F., Amaya, Y., Tatibana, M., and
Cohen, P. P. (1981) Synthesis, intracellular transport, and
processing of the precursors for mitochondrial ornithine
transcarbamylase and carbamoylphosphate synthetase I in isolated
hepatocytes. Proc. Nat. Acad. Sci. USA 78, 6056-6060.
38. Poyton, R. 0., and McKenmie, E. (1979) A polyprotein precursor
to all four cytoplasmically translated subunits of cytochrame c
oxidase from Saccharomyces cerevisiae. J. Biol. Chem. 254,
39. Poyton, R. 0., and McKemmie, E. (1979) Post-translational
processing and transport of the polyprotein precursor to
subunits IV and VII of yeast cytochrome c oxidase. J. Biol.
Chem. 254, 6772-6780.
40. Lewin, A. S., Gregor, I., Mason, T. L., Nelson, N., and Schatz,
G. (1980) Cytoplasmically made subunits of yeast mitochondrial
F I-ATPase and cytochrome c oxidase are synthesized as
individual precursors, not as polyproteins. Proc. Nat. Acad.
Sci. USA 77, 3998-4002.
41. Maccecchini, M.-L., Rudin, Y., Blobel, G., and Schatz, G. (1979)
Import of proteins into mitochondria: Precursor form of the
extramitochondrially made F -ATPase subunits in yeast. Proc.
Nat. Acad. Sci. USA 76, 343-347.
42. Mihara, K., and Blobel, G. (1980) The four cytoplasmically made
subunits of yeast mitochondrial cytochrome c oxidase are
synthesized individually and not as a polyprotein. Proc. Nat.
Acad. Sci. USA 77, 4160-4164.
43. Hennig, B., and Neupert, W. (1982) Assembly of cytochrame c.
Apocytochrome c is bound to specific sites on mitochondria
before its conversion to holocytochrcme c. Eur. J. Biochem.
44. Korb, H., and Neupert, W. (1978) Biogenesis of cytochrome C in
Neurospora crassa: Synthesis of apocytochrame C, transfer to
mitochondria and conversion to holocytochrome C. Eur. J.
Biochem. 91, 609-620.
45. Matsuura, S., Arpin, M., Margoliash, E., Sabatini, D., and
Morimoto, T. (1979) Synthesis of cytochrame c in rat liver free
polysomes and in vitro transfer into mitochondria. J. Cell
Biol. 83, 437a.
46. Matsuura, S., Arpin, M., Hannum, C., Margoliash, E., Sabatini,
D., and Morimoto, T. (1981) In vitro synthesis and
post-translational uptake of cytochrome c into isolated
mitochondria: Role of a specific addressing signal in the
apocytochraome. Proc. Nat. Acad. Sci. USA 78, 4368-4372.
47. Borst, P. (1972) Mitochondrial nucleic acids. Ann. Rev.
Biochem. 41, 333-376.
48. Crews, S., Ojala, D., Pasakony, J., Nishiguchi, J., and Attardi,
G. (1979) Nucleotide sequence of human mitochondrial DNA
containing the precisely identified origin of replication.
Nature 277, 192-198.
49. Murphy, W. I., Attardi, B., Tu, C., and Attardi, G. (1975)
Evidence for complete symmetrical transcription in vivo of
mitochondrial DNA in HeLa cells. J. Mol. Biol. 99, 809-814.
50. Amalric, F., Merkel, C., Gelfand, R., and Attardi, G. (1978)
Fractionation of mitochondrial RNA from HeLa cells by high
resolution electrophoresis under strongly denaturing conditions.
J. Mol. Biol. 118, 1-25.
51. Costantino, P., and Attardi, G. (1975) Identification of
discrete electrophoretic components among the products of
mitochondrial protein synthesis in HeLa cells. J. Mol. Biol.
52. Yatscoff, R. W. and Freeman, K. B. (1977) Discrete
electrophoretic products of mitochondrial protein synthesis in
the Chinese hamster ovary cell line. Can J. Biochem. 55,
53. Yatscoff, R. W., Freeman, K. B., and Vail, W. J. (1977) Site of
biosynthesis of mammalian cytochrome c oxidase subunits. Fed.
Ear. Biochem. Soc. Letters 81, 7-9.
54. Yatscoff, R. W., Aujame, L., Freeman, K. B., and Goldstein, S.
(1978) Interspecific variations in proteins synthsized by
mammalian mitochondria. Can. J. Biochem. 56, 939-942.
55. Bhat, N. K., Niranjan, B. G., and Avadhani, N. G. (1981) The
complexity of mitochondrial translation products in mammalian
cells. Biochem. Biophys. Res. Coammn. 103, 621-628.
56. Bhat, N. K., Niranjan, B. G., and Avadhani, N. G. (1982)
Quantitative and comparative nature of mitochondrial translation
products in manummalian cells. Biochemistry 21, 2452-2460.
57. Attardi, G. and Ching, E. (1979) Biogenesis of mitochondrial
proteins in HeLa cells. Meth. Enzymol. 56, 66-79.
58. Ching, E., and Attardi, G. (1982) High-resolution
electrophoretic fractionation and partial characterization of
the mitochondrial translation products from HeLa cells.
Biochemistry 21, 3188-3195.
59. Anderson, S., Bankier, A. T., Barrell, B. G., deBruijn, M. H.
L., Coulson, A. R., Drouin, J., Eperon, I. C., Nierlich, D. P.,
Roe, B. A., Sanger, F., Schreier, P. H., Smith, A. J. H.,
Staden, R., and Young, I. G. (1981) Sequence and organization of
the human mitochondrial genome. Nature 290, 457-465.
60. Attardi, G. (1981) Organization and expression of the mammalian
mitochondrial genome, a lesson in econanomy. Trends in Biochem.
Sci. 6, 86-89.
61. Attardi, G. (1981) Organization and expression of the mammalian
mitochondrial genome, a lesson in economy. (part 2) Trends in
Biochem. Sci. 6, 100-103.
62. Attardi, G., Cantatore, P., Ching, E., Crews, S., Gelfand, R.,
Merkel, C., Montoya, J., and Ojala, D. (1981) Organization and
expression of genetic information in human mitochondrial DNA.
In: International Cell Biology 1980-1981 (Schweiger, H. G.,
ed) pp. 225-238, Springer-Verlag, New York.
63. Borst, P. and Grivell, L. A. (1981) Small is beautiful --
Portrait of a mitochondrial genome. Nature 290, 443-444.
64. Grivell, L. A. (1983) Mitochondrial DNA. Sci. Amer. 248,
65. Bibb, M. J., VanEtten, R. A., Wright, C. T., Walberg, M. W., and
Clayton, D. A. (1981) Sequence and gene organization of mouse
mitochondrial DNA. Cell 26, 167-180.
66. Anderson, S., deBruijn, M. H. L., Coulson, A. R., Eperon, I. C.,
Sanger, F., and Young, I. G. (1982) Complete sequence of bovine
mitochondrial DNA: Conserved features of the mammalian
mitochondrial genome. J. Mol. Biol. 156, 683-717.
67. Pepe, G., Holtrop, M., Gadaleta, G., Kroon, A. M., Cantatore,
P., Gallerani, R., DeBenedetto, C., Quagliariello, C., Sbisa,
E., and Saccone, C. (1983) Non random patterns of nucleotide
substitutions and codon strategy in the manmalian mitochondrial
genes coding for identified and unidentified reading frames.
Biochem. Int. 6, 553-563.
68. Wong, J. F. H., Ma, D.-P., Wilson, R. K., and Roe, B. A. (1983)
Nucleotide sequence analysis of Xenopus laevis mitochondrial
tRNA genes. Fed. Proc. 42, 2240.
69. Battey, J., and Clayton, D. A. (1980) The transcription map of
Human mitochondrial DNA implicates transfer RNA excision as a
major processing event. J. Biol. Chem. 255, 11599-11606.
70. Ojala, D., Montoya, J., and Attardi, G. (1981) tRNA punctuation
model of RNA processing in human mitochondria. Nature 290,
71. Montoya, J., Ojala, D., and Attardi, G. (1981) Distinctive
features of the 5'-terminal sequences of the human mitochondrial
mRNAs. Nature 290, 465-470.
72. Dianoux, A.-C., Bof, M., and Vignais, P. V. (1978) The
dicyclohexylcarbodiimide-binding protein of rat liver
mitochondria is a product of the mitochondrial protein
synthesis. Eur. J. Biochem. 88, 69-77.
73. Kuzela, S., Luciakova, K., and Lakota, J. (1980) Amino acid
incorporation by isolated rat liver mitochondria into two
protein components of mitochondrial ATPase complex. Fed. Eur.
Biochem. Soc. Letters 114, 197-201.
74. Schmidt, B., Hennig, B., Zinrerman, R., and Neupert, W. (1983)
Biosynthetic pathway of mitochondrial ATPase subunit 9 in
Neurospora crassa. J. Cell Biol. 96, 248-255.
75. van den Boogaart, P., Samallo, J., and Asteribbe, E. (1980)
Similar genes for a mitochondrial ATPase subunit in the nuclear
and mitochondrial genome of Neurospora crassa. Nature 298,
76. Chomyn, A., Mariottini, P., Gonzalez-Cadavid, N., Attardi, G.,
Strong, D. D., Trovato, D., Riley, M. and Doolittle, R. F.
(1983) Identification of the polypeptides encoded in the ATPase
6 gene and in the unassigned reading frames 1 and 3 human mtDNA.
Proc. Nat. Acad. Sci. USA 80, 5535-5539.
77. Manottini, P., Chcmyn, A., Attardi, G., Trovato, D., Strong, D.
D., and Doolittle, R. F. (1983) Antibodies against synthetic
peptides reveal that the unidentified reading frame A6L,
overlapping the ATPase 6 gene, is expressed in human
mitochondria. Cell 32, 1269-1277.
78. Oliver, N. A., Greenberg, B. D., and Wallace, D. C. (1983)
Assignment of a polymorphic polypeptide to the human
mitochondrial DNA unidentified reading frame 3 by a new peptide
mapping strategy. J. Biol. Chem. 258, 5834-5839.
79. O'Brien, T. W., and Matthews, D. E. (1976) Mitochondrial
ribosomes. In: Handbook of Genetics (King, R. C., ed) volume
5, pp. 535-580, Plenum Press, New York.
80. Buetow, D. E., and Wood, W. M. (1978) The mitochondrial
translation system. Subcellular Biochem. 5, 1-85.
81. O'Brien, T. W. (1971) The general occurrence of 55S ribosomes in
mammalian liver mitochondria. J. Biol. Chem. 246, 3409-3417.
82. Sacchi, A., Ferrini, U., Londei, P., Camuarano, P., and Miraldi,
N. (1977) Mitochondrial and cytoplasmic riboscmaes fram mammalian
tissues: Further characterization of ribosonal subunits and
validity of buoyant-density methods for determination of the
chemical composition and partial specific volume of
ribonucleoprotein particles. Biochem J. 168, 245-259.
83. Hamilton, M. G., and O'Brien, T. W. (1974) Ultracentrifugal
characterization of the mitochondrial ribosome and subribosomal
particles of bovine liver: Molecular size and composition.
Biochemistry 13, 5400-5403.
84. Matthews, D. E., Hessler, R. A., Denslow, N. D., Edwards, J. S.,
and O'Brien, T. W. (1982) Protein composition of the bovine
mitochondrial ribosome. J. Biol. Chem. 257, 8788-8794.
85. Douglas, M. G., and Butow, R. A. (1976) Variant forms of
mitochondrial translation products in yeast: Evidence for
location of determinants on mitochondrial DNA. Proc. Nat. Acad.
Sci. USA 73, 1083-1086.
86. Perlman, p. S., Douglas, M. G., Strausberg, R. L., and Butow, R.
A. (1977) Localization of genes for variant forms of
mitochondrial proteins on mitochondrial DNA of Saccharcmyces
cerevisiae. J. Mol. Biol. 115, 675-694.
87. Groot, G. S. P. (1974) The biosynthesis of mitochondrial
ribosomes in Saccharomyces cerevisiae. In: The Biogenesis of
Mitochondria (Kroon, A. M., and Saccone, C., eds) pp. 443-452,
Academic Press, New York.
88. Groot, G. S. P., Grivell, L. A., VanHarten-Loosbroek, N.,
Kreike, J., Moorman, A. F. M., and VanOCnmen, G. J. B. (1977) The
role of the mitochondrial genetic system in the biogenesis of
the mitochondrial inner membrane. In: Structure and Function of
Energy Transducing Membranes (vanDam, K., and vanGelder, eds)
pp. 177-186, Elsevier/North Holland, New York.
89. Terpstra, P., Zanders, E., and Butow, R. A. (1979) The
association of var 1 with the 38S mitochondrial ribosncmal
subunit in yeast. J. Biol. Chem. 254, 12653-12661.
90. Terpstra, P., and Butow, R. A. (1979) The role of var 1 in the
assembly of yeast mitochondrial ribosomes. J. Biol. Chem.
91. Maheshwari, K. K., Marzuki, S., and Linnane, A. W. (1982) The
formation of defective small ribosomal subunits in yeast
mitochondria on the absence of mitochondrial protein synthesis.
Biochem. Int. 4, 109-115.
92. Lopez, I. C., Farrelly, F., and Butow, R. A. (1981) Trans action
of the var 1 determinant region on yeast mitochondrial DA;
specific labelings of mitochondrial translation products in
zygotes. J. Biol. Chem. 256, 6496-6501.
93. Hudspeth, M. E. S., Ainley, W. M., Shumard, D. S., Butow, R. A.,
and Grossman, L. I. (1982) Location and structure of the var 1
gene on yeast mitochondrial DNA: Nucleotide sequence of the 40.0
allele. Cell 30, 617-626.
94. Strausberg, R. L., and Butow, R. A. (1981) Gene conversion at
the var 1 locus on yeast mitochondrial DNA. Proc. Nat. Acad.
Sci. USA 78, 494-498.
95. Stausberg, R. L., Vincent, R. D., Perlman, P. S., and Butow, R.
A. (1981) Assymetric gene conversion at inserted segments on
yeast mitochondrial DNA. Nature 276, 577-583.
96. Farrelly, F., and Butow, R. A. (1983) Rearranged mitochondrial
genes in the yeast nuclear gencrme. Nature 301, 296-301.
97. Fox, T. D. (1983) Mitochondrial genes in the nucleus. Nature
98. Reid, R. A. (1983) Can migratory mitochondrial DNA activate
oncogenes? Trends in Biochem. Sci. 8, 190-191.
99. Lambowitz, Am. M., Chua, N.-H., and Luck, D. J. (1976)
Mitochondrial ribosome assembly in Neurospora. Preparation of
mitochondrial ribosomal precursor particles, site of synthesis
of mitochondrial ribosomal proteins and studies on the poky
mutant. J. Mol. Biol. 107, 223-253.
100. Lambowitz, A. M., LaPolla, R. J., and Collins, R. A. (1979)
Mitochondrial ribosome assembly in Neurospora. Two-dimensional
gel electrophoretic analysis of mitochondrial ribosomal
proteins. J. Cell Biol. 82, 17-31.
101. LaPolla, R. J., and Lambowitz, A. M. (1977) Mitochondrial
ribosome assembly in Neurospora crassa. Chloramphenicol
inhibits the maturation of small ribosome subunits. J. Mol.
Biol. 116, 189-205.
102. LaPolla, R. J., and Lambowitz, A. M. (1979) Binding of
mitochondrial ribosomal proteins to a mitochondrial ribosomal
precursor RNA containing a 2.3-kilobase intron. J. Biol. Chem.
103. LaPolla, R. J., and Lambowitz, A. M. (1982) Mitochondrial
ribosome assembly in Neurospora. Structural analysis of mature
and partially assembled ribosomal subunits by equilibrium
centrifugation in CsCl gradients. J. Cell Biol. 95, 26-277.
104. Collins, R. A., Bertrand, H., LaPolla, R. J., and Lambowitz, A.
M. (1981) A novel extracellualr mutant of Neurospora with a
temperature-sensitive defect in mitochondrial protein synthesis
and mitochondrial ATPase. Molec. Gen. Genet. 181, 13-19.
105. Madin, S. H., and Darby, Jr., N. B. (1958) Established kidney
cell lines of normal adult bovine and ovine origin. Proc. Soc.
Exp. Biol. Med. 98, 574-576.
106. Mans, R. J., and Novelli, G. D. (1961) Measurement of the
incorporation of radioactive amino acids into protein by a
filter-paper disk assay. Arch. Biochem. Biophys. 94, 48-53.
107. Roodyn, D. B., Reis, P. J., and Work, T. S. (1961) Protein
synthesis in mitochondria. Requirements for the incorporation of
radioactive amino acids into mitochondrial protein. Biochem.
J. 80, 9-21.
108. Costantino, P., and Attardi, G. (1977) Metabolic properties of
the products of mitochondrial protein synthesis in HeLa cells.
J. Biol. Chem. 252, 1702-1711.
109. Laskey, R. A., and MillI,4A. D. (1975) Quantitative film
detection of H and C in polyacrylamide gels by
fluorography. Eur. J. Biochem. 56, 335-341.
110. Laskey, R. A., ad Mills.A D. (1977) Enhanced autoradiographic
detection of P and -I using intensifying screens and
hypersensitive film. Fed. Eur. Biochem. Soc. Letters 82,
111. Chamberlain, J. P. (1979) Fluorographic detection of
radioactivity in polyacrylamide gels with the water soluble
fluor, sodium salicylate. Analyt. Biochem. 98, 132-135.
112. DeVries, H., Arendzen, A. J., and Kroon, A. M. (1973) The
interference of the macrolide antibiotics with mitochondrial
protein synthesis. Biochim. Biophys. Acta 331, 264-275.
113. Leister, D., and Dawid, I. (1974) Pysical properites and protein
constituents of cytoplasmic and mitochondrial riboscmes of
Xenopus laevis. J. Biol. Chem. 249, 5108-5118.
114. Rascati, R. J., and Parsons, P. (1979) Biosynthesis of
cytochrome c oxidase by isolated rat liver mitochondria. J.
Biol. Chem. 254, 1594-1599.
115. Sevarino, K. A., and Poyton, R. 0. (1980) Mitochondrial membrane
biogenesis: Identification of a precursor to yeast cytochrome c
oxidase subunit II, an integral polypeptide. Proc. Nat. Acad.
Sci. USA 77, 142-146.
116. van't Sant, P., Mak, J. F. C., and Kroon, A. M. (1981) Larger
precursors of mitochondrial translation products in Neurospora
crassa: Indications for a precursor of subunit 1 of cytochrome
c oxidase. Eur. J. Biochem. 121, 21-26.
117. Darley-Usmar, V. M., and Fuller, S. D. (1981) M-values of
mature subunits I and III of beef heart cytochrcme c oxidase in
relationship to nucleotide sequences of their genes. Fed. Eur.
Biochem. Soc. Letters 135, 164-166.
118. Feldman, F., and Mahler, H. R. (1974) Mitochondrial biogenesis.
Retention of terminal formylmethionine in membrane proteins and
regulation of their synthesis. J. Biol. Chem. 249,
119. Hare, J. F., and Hodges, R. (1982) No unique mitochondrial
translation products in respiratory chain-linked NADH
dehydrogenase. Biochem. Biophys. Res. Ccmmn. 105, 1250-1256.
120. Schnaitman, C., Erwin, V. G., and Greenawalt, J. W. (1967) The
submitochondrial localization of monoamine oxidase: An enzymatic
marker for the outer membrane of rat liver mitochondria. J.
Cell Biol. 32, 719-735.