Full text of DOI:10.1078/0176-1617-00292

J. Plant Physiol. 158. 787–794 (2001)
Urban & Fischer Verlag
http://www.urbanfischer.de/journals/jpp
Proteomic approach for investigation of cytoplasmic male sterility
(CMS) in Brassica
Christina Mihr, Maja Baumgärtner, Jens-Holger Dieterich, Udo Klaus Schmitz, Hans-Peter Braun*
Abteilung für Angewandte Genetik, Universität Hannover, Herrenhäuserstr. 2, D-30419 Hannover, Germany
Received November 14, 2000 · Accepted February 5, 2001
Summary
The alloplasmic ‹Tournefortii› CMS system from B. napus was chosen for a comprehensive investi-
gation of the protein content of the mitochondrial compartment in male sterile and fertile near iso-
genic lines. As proteins involved in the male sterile phenotype may be expressed in a tissue-specific
manner, mitochondrial preparations were made from different tissues and checked for the purity of
the organelles with a new and highly sensitive method (BN-PAGE). In Brassica, only mitochondrial
preparations from 4–5-day-old seedlings and from flower buds were found to be sufficiently pure.
High resolution, two-dimensional gel electrophoresis of total protein from anthers and from mito-
chondrial protein of male sterile and fertile seedlings of near isogenic lines revealed distinct differ-
ences in the protein patterns. Several explanations for the occurrence of these differences, which
may be the starting point for a detailed molecular analysis and a better understanding of the CMS
syndrome, are discussed.
Key words: cytoplasmic male sterility – 2-D-gel electrophoresis – proteomics mitochondria – rape-
seed (Brassica napus L.)
restore pollen fertility (restorer genes, Rf genes) are neces-
Introduction
sary if seeds are harvested from the F1 plants. CMS plants
maintain normal vegetative growth and have normal female
fertility, but fail to shed functional pollen. Pollen development
may be interrupted at different stages, either prior to or after
meiosis. In Brassica napus for example, the ‹Tournefortii›
and ‹Polima› cytoplasm cause inhibition of microsporogene-
sis, resulting in complete failure to produce pollen. In con-
trast, the ‹Ogura› and ‹Tokumasu› cytoplasms, which have
been introduced from the genus Raphanus (Pelletier et al.
1983, Paulmann and Röbbelen 1988), exhibit postmeiotic
interruption of pollen development, leading to the production
of sterile microspores.
Cytoplasmic male sterility (CMS) is the maternally inherited
inability of higher plants to produce functional pollen. CMS is
a widespread phenomenon in the plant kingdom, caused by
an incompatibility between the nucleus and the cytoplasm.
Thus, it is often induced by intra- or interspecific crosses in
which nuclear genes are combined with a new set of cyto-
plasmic genes. CMS has been widely used for the produc-
tion of hybrid seed from crop plants. Nuclear genes which
* E-mail corresponding author: braun@genetik.uni-hannover.de
0176-1617/01/158/06-787 $ 15.00/0

788
Christina Mihr et al.
Alterations of mitochondrial genome organization are asso-
ciated with nearly every type of CMS examined so far (re-
viewed in: Conley and Hanson 1995). They arise from recom-
bination events through direct or inverted repeats, and cause
the expression of chimeric mitochondrial genes which have
deleterious effects on respiratory function. Such novel, open
reading frames, which are thought to be responsible for the
CMS phenotype, have been described in maize (Dewey et al.
1986), petunia (Young and Hanson 1987), sunflower (Horn et
al. 1991), and other genera.
Materials and Methods
Plant material
Male sterile and fertile lines of Brassica originated from experiments
carried out at the University of Göttingen in the early 90th under su-
pervision of G. Röbbelen. The pollen sterile Brassica line used was
produced by a protoplast fusion between B. tournefortii and B. napus
(Stiewe 1993), and contained due to repeated backcrossing with B.
napus (Duplo), a B. napus nuclear background, and a recombined
cytoplasm of B. tournefortii and B. napus. It was denoted ‹cmstour
25–143› and crossed back with fertile B. napus (Duplo) four times in
Göttingen and once in our Institute (BC 5 line). The cytoplasm of B.
tournefortii causes male sterility in a B. napus or B. juncea nuclear
background (Mathias 1985). Etiolated seedlings were produced in
growth chambers at 22˚C/18˚C (day/night) in the dark.
Little is known about the polypeptides involved in the
expression of the CMS phenotype. Only a few of them have
been characterized in more detail such as the T-urf 13 protein
(Dewey et al. 1988). The way in which these proteins cause
the CMS phenotype still remains an enigma. In the past, the
use of two-dimensional gels for the separation of mitochon-
drial proteins was complicated, due both to a limited resolution
capacity and to problems concerning reproducibility. Thus, in
organello translation in the presence of radiolabeled amino
acids was the main technique for the identification of the
polypeptides encoded by novel open reading frames in
mitochondria from male sterile plants. However, this method
has several shortcomings, including e.g., the fact that it does
not answer the question of whether the novel polypeptide is
really synthesized in male sterile or restored plants. In recent
years, the improvement of two-dimensional gel techniques,
e.g. the introduction of immobilised pH-gradients which are
commercially available, sample applications during rehy-
dration of dried gel strips, and usage of high voltages during
focusing (Görg et al. 2000), has allowed more than a thou-
sand different polypeptides to be resolved on a single, two-
dimensional gel. Proteomics is the analysis of a proteome
which represents all expressed polypeptides of an organism,
a tissue, a cell, or a subcellular compartment under defined
conditions (Kahn 1995). The proteomic approach might add
a new dimension to the analysis of CMS. Especially in cyto-
plasms, where no polypeptides specific to the mitochondria
from male sterile plants could be identified, e.g. in the Bras-
sica cytoplasms ‹Tournefortii› and ‹Tokumasu›, the proteomic
approach may be promising.
Isolation of mitochondria
Mitochondria were prepared following a modified version of the pro-
cedure of Boutry et al. (1984) and Douce (1985). Plant tissues
(leaves, stems, flower buds) were disrupted in a Waring Blendor in an
icecold buffer (pH 7.8) containing 0.4 mol/L mannitol, 5 mmol/L EDTA,
25mmol/L MOPS, 1% PVP 40, 0.1% BSA, 15mmol/L β-mercaptoetha-
nol, and 0.05 mmol/L phenylmethylsulfonyl fluoride (PMSF). Seedlings
were ground with a mortar and pestle in the same type of buffer. The
homogenate was passed through two layers of mull and centrifuged
at 3400×g at 4˚C. The supernatant was centrifuged again for 30 min-
utes at 17000 × g. The pellet was resuspended in a buffer (pH 7.2)
containing 0.4 mol/L mannitol, 1 mmol/L EGTA, 10 mmol/L KH2PO4,
and 0.2mmol/L PMSF, and then carefully dispersed in a Dounce homo-
genizer by applying 2 strokes. Between 2.5–3 mL of the suspension
was centrifuged through a Percoll step gradient (14 %–26 %–45 %
Percoll) for 45 minutes at 70000 ×g. Mitochondria were collected from
the 26 %–45 % interphase, washed twice in resuspension buffer, and
pelleted at 15000×g for 10 minutes at 4˚C.
Extraction of total protein from rapeseed plants
Total protein was extracted as described by Damerval et al. (1986), with
the following modifications: instead of β-mercaptoethanol, 20 mmol/L
dithiothreitol (DTT) was used, and for drying the extracted protein, no
vacuum was applied.
An investigation of CMS in Brassica is highly desirable
since this genus is of major economic importance. The Ogura
CMS system, which is now covered by several patents, proved
really useful in hybrid breeding, and has been well character-
ised at the molecular level (Grelon et al. 1994, Bellaoui et al.
1997). In contrast, the ‹Tokumasu› and ‹Tournefortii› allo-
plasmic male sterility systems are not patented and have
good potential for hybrid breeding. An investigation of CMS
‹Tournefortii› was performed by Landgren et al. (1996), who
found differences in the transcription pattern of the atp6 gene
region between male sterile and fertile lines. An orf down-
stream of the atp6 gene (orf 263) potentially encodes a 29
kDa protein and may be associated with CMS. Here we de-
scribe a proteomic approach for the molecular characteriza-
tion of CMS ‹Tournefortii› in Brassica napus.
Two-dimensional gel electrophoresis
Blue native gel electrophoresis was carried out as described by
Jänsch et al. (1996). For isoelectric focusing, total protein extracts or
purified mitochondria (100 µg protein in a volume of 10 µL) were in-
cubated for 1 hour at room temperature in a buffer containing 8 mol/L
urea, 4 % Triton-X-100, 40 mmol/L Trisbase, 1mmol/L PMSF, and sub-
sequently supplemented with 340 µL of rehydration solution (8 mol/L
urea, 2 % Triton-X-100, a trace of Bromophenol blue, 20 mmol/L DTT,
0.5 % IPG buffer, pH 3–10/NL [supplied by Amersham/Pharmacia]).
Total protein extracts were centrifuged for 20 minutes at 15000 × g
prior to addition of the rehydration solution to get rid of unsoluble par-
ticles or aggregates. After rehydration for 12 hours at 30 V, isoelectric
focussing was performed in 18 cm strip holders, with IPG dry strips

Proteomic approach for CMS
789
pH 3–10/NL in an IPGphor apparatus (Amersham/Pharmacia) at
20 ˚C, by increasing the voltage from 500 V to 8000 V in four steps:
500 V (1 hour), 500–1000 V (1 hour), 1000–8000 V (4 hours), and
8000 V (6 hours). Gelstrips were equilibrated for 15 minutes in a solu-
tion containing 50 mmol/L Tris-HCl pH 8.8, 6 mol/L urea, 30 % glyc-
erol, 2 % SDS, 70 mmol/L DTT, and a trace Bromphenol blue. After
washing the strips briefly in Tricine-electrophoresis-buffer, they were
transferred horizontally onto Tricine-SDS polyacrylamide gels as de-
scribed by Berkelman and Stenstedt (1998). A protocol for Tricine-
SDS-PAGE is given in Schägger et al. (1994). Two-dimensional gels
were stained with silver nitrate as described by Heuckeshoven and
Dernick (1986). Individual 2-D separations were repeated at least
three times in order to identify random variations of protein content.
Results
Analysis of total protein from seedlings
Our first approach in this investigation of cytoplasmic male
sterility in Brassica was a direct comparison of total protein
from male sterile and fertile plants on two-dimensional gels.
The high resolution 2-D gel electrophoresis used is based on
ready-made polyacrylamide gels for the first dimension, con-
taining an immobilised pH gradient. Different pH ranges were
tested and the optimal for our purposes was found to be a
non-linear range between pH 3 and pH 10. In the second di-
mension, Tricine-SDS-PAGE gave the best resolution. Total
protein was prepared from 5-day-old etiolated seedlings and
subjected to isoelectric focusing in the first dimension and
Tricine-SDS-PAGE in the second dimension. Figure 1 shows
two silver stained gels of near isogenic male sterile (A) and
fertile (B) Brassica lines. No differences in strongly stained
protein spots are visible between the two gels, and a com-
puter aided comparison did not detect substantial differ-
ences between the protein pattern of the male sterile and the
fertile seedlings (not shown).
Figure 1. Two-dimensional gel electrophoresis of total protein
(250µg) from 5-day-old male sterile (A) and fertile (B) etiolated B. na-
pus seedlings. The silver stained polypeptides were separated in the
first dimension according to their isoelectric points and in the second
dimension under denaturing conditions (Tricine-SDS-PAGE).
Analysis of total protein from anthers
As CMS exclusively affects the development of the male
flower organs, total protein from anthers of pollen sterile and
fertile plants was analysed in the next step. Again, the protein
patterns are rather similar, but clear-cut differences between
the polypeptides from male sterile and fertile anthers are visi-
ble (Fig. 2). The arrows indicate proteins which specifically
occur in the male sterile and the fertile anthers.
chondrial polypeptides. Therefore, mitochondria from fertile
and pollen sterile near isogenic lines were purified and sub-
sequently analysed.
Purity and composition of mitochondrial protein from
different tissues
As CMS is correlated with the mitochondrial compartment,
an estimation of the content of mitochondrial protein in total
protein from seedlings and anthers was made. Information
from sequencing of individual protein spots from 2-D gels of
Brassica and Arabidopsis (Kruft et al. 2001) revealed that the
mitochondrial protein represents a minor part of total protein.
Thus, a proteomic approach using total protein for the analy-
sis of CMS has severe limitations due to big quantitative dif-
ferences between abundant cellular proteins and rare mito-
First, the purity of isolated Brassica mitochondria was investi-
gated since mitochondrial preparations from plants tend to
be contaminated by proplastids and chloroplasts. Monitoring
organellar contaminations by enzymes, which are character-
istic for both mitochondria and chloroplasts, is a widely used,
although not very predicative method, which draws con-
clusions from looking at a minor segment of organellar pro-
teins. However, contamination of mitochondria with different

790
Christina Mihr et al.
peptide spots (Jänsch et al. 1996, Kügler et al. 1997). Com-
parison of the mitochondrial and chloroplast proteins, in ei-
ther the first or in the second dimension, allows a good esti-
mation of the degree of cross-contamination of the organellar
fractions. The mitochondria isolated from etiolated seedlings
of rapeseed exhibit only very small contaminations by plastid
proteins. Minor amounts of the ribulosebiphosphate carboxyl-
ase/oxygenase (Rubisco) are visible, while the preparation is
more or less free of the major thylakoid protein complexes
such as photosystems I and II or the light harvesting complex
(LHC II). In contrast, mitochondrial fractions isolated from
green seedlings always exhibit contaminations from chloro-
plasts, which increase dramatically with the age of the seed-
lings (Fig. 4). While in mitochondrial fractions of 5-day-old
green Brassica napus seedlings only Rubisco caused some
contamination, mitochondria prepared from older green seed-
lings contained substantial amounts of thylakoid protein com-
plexes (Fig. 4, arrows). Different methods described in the liter-
ature, which are supposed to yield pure mitochondrial frac-
tions from green tissues, were tested (Day et al. 1985, Hamasur
et al. 1990), but none of them resulted in a drastic reduction
of chloroplast contaminations in Brassica. Self-generating
Percoll gradients containing linear gradients of polyvinylpyro-
lidone 25 (0–10 % w/v) led to a decrease of contaminating
thylakoid proteins in comparison to Percoll step gradients;
however, contaminations of soluble proteins like Rubisco in-
creased strongly (data not shown). This indicates that chloro-
plasts tend to disintegrate on discontinous gradients used for
the preparation of plant mitochondria, but thylakoids or vesi-
cles containing them accumulate at the 26 %/45 % Percoll in-
terphase. Also, mitochondria isolated from different green tis-
sues of rapeseed plants are contaminated by chloroplast
proteins (Fig. 5). Mitochondrial preparations from flower buds
possibly contain equal amounts of chloroplast proteins and
those from stems or leaves of adult rapeseed plants contain
more chloroplast protein than mitochondrial protein.
Figure 2. Two-dimensional gel electrophoresis of total protein
(250 µg) from anthers of male sterile (A) and fertile (B) B. napus
plants. The silver stained polypeptides were separated in the first di-
mension according to their isoelectric points and in the second di-
mension under denaturing conditions (Tricine-SDS-PAGE). Arrows in-
dicate proteins which are specifically present in the sterile or fertile
line.
Specific features of the mitochondrial proteome from
male sterile plants
The mitochondrial proteome from rapeseed was first charac-
terized in etiolated seedlings. The 2-D analysis was again
based on Tricine-SDS-PAGE in the second dimension since
all characterized mitochondrial polypeptides involved in male
sterility in different plant systems have a molecular weight be-
low 35 kDa. Figure 6 shows a 2-D gel of mitochondrial poly-
peptides from the male sterile line CMS-tour 25–143 and a
fertile near isogenic B. napus (Duplo) maintainer line. In con-
trast to the more or less identical patterns of total protein from
male sterile and fertile lines (Fig. 1), the mitochondrial poly-
peptides of the same lines exhibit clear-cut differences. Sev-
eral proteins which are not present in the fertile line form dis-
tinct spots in the sterile line (see arrows), but vice versa,
subcompartments of plastids varies considerably (see be-
low). Therefore, we used the resolution of organellar protein
complexes by Blue native gel electrophoresis (BN-PAGE) to
monitor the purity of mitochondrial fractions from different de-
velopmental stages and tissues. Figure 3 shows the major
protein complexes from mitochondria (A) and chloroplasts
(B) separated by BN-PAGE. The subunits of the organellar
protein complexes were resolved in the second dimension
under denaturing conditions in the presence of SDS. The
most prominent protein complexes of both organelles have
been identified by N-terminal sequencing of individual poly-

Proteomic approach for CMS
791
Figure 3. Two-dimensional resolution of mitochondrial (A) and chloroplast (B) protein complexes from etiolated (A) and green (B) Brassica na-
pus seedlings. The protein complexes were separated by blue native gel electrophoresis in the first dimension (horizontal top lane) and by Tri-
cine-SDS-PAGE in the second dimension (two-dimensional gel).
Figure 4. Analysis of the purity of mitochondrial
fractions from green rapeseed seedlings by
BN-PAGE. Increasing amounts of prominent
chloroplast protein complexes (arrows) appear
in the mitochondrial fractions of seedlings older
than five days (5d).

792
Christina Mihr et al.
Figure 5. Mitochondrial preparations from different tissues of
adult rapeseed plants: flower buds (f), stems (s), leaves (l). The
protein complexes from mitochondria are indicated on the left
and those from chloroplasts on the right.
there are also some proteins which seem to be specific to the
mitochondria of the fertile line.
Discussion
The data presented in this paper suggests that the proteomic
approach is a suitable method for the analysis of cytoplasmic
pollen sterility in higher plants. The alloplasmic ‹Tournefortii›
male sterility (Stiewe et al. 1995) differs from the CMS ‹Tour-
nefortii› described by Landgren et al. (1996), which origi-
nated from six backcrossings of the male sterile B. juncea
line of Rawat and Anand (1979). Total protein of seedlings
from the male sterile BC 5 line used in this report exhibits no
distinct differences in comparison to fertile lines (Fig. 1). In
contrast, total protein from anthers of the same lines led to
the detection of several variant polypeptides which are spe-
cific to the fertile anthers (Fig. 2). However, it is possible that
these variant proteins only reflect indirect effects of male ste-
rility since there are also morphological differences between
the male organs from pollen sterile and fertile plants.
For the analysis of the mitochondrial proteome from plants,
a thorough check of the purity of these organelles is required.
In our experiments, we tested the mitochondrial preparations
for contaminating chloroplast or proplastid proteins with BN-
PAGE, another powerful 2-D-technique. An investigation of
mitochondrial preparations from different plant tissues and
developmental stages from Brassica napus revealed that the
mitochondrial fractions from green tissues of adult plants in
particular, but also from young green seedlings, are strongly
contaminated with chloroplast proteins. These contamina-
tions would not be detected if only a single chloroplast pro-
tein was analysed, which is a common practice in reports on
leaf mitochondria (e.g. Hamasur et al. 1990). Conventional
methods for the determination of the purity of organellar frac-
tions are based on an analysis of the activity of enzymes
characteristic for chloroplasts or mitochondria. However, the
amount of these enzymes may easily be underestimated due
to activity loss. Furthermore, sensitive detection with BN-
PAGE shows that it is not possible to predict the amount of
Figure 6. Two-dimensional gel electrophoresis of mitochondrial pro-
tein (250µg) from 5-day-old male sterile (A) and fertile (B) etiolated B.
napus seedlings. The silver stained polypeptides were separated in
the first dimension according to their isoelectric points and in the sec-
ond dimension under denaturing conditions (Tricine-SDS-PAGE). Ar-
rows indicate proteins which specifically occur in the male sterile or
fertile lines.

Proteomic approach for CMS
793
Berkelmann T, Stenstedt T (1998) 2-D Electrophoresis using immobi-
lized pH gradients. Principles and Methods. Amersham Pharmacia
Biotech Inc, Sweden
contamination of mitochondrial fractions with stroma proteins
from an analysis of the amount of thylakoid proteins or vice
versa. While in Arabidopsis highly pure mitochondrial frac-
tions may be isolated from green leaves or stems (Kruft et al.
2001), in Brassica the same isolation procedures lead to
strongly contaminated mitochondrial fractions. Only the purity
of mitochondria from etiolated or five-day-old green Brassica
seedlings, or from young flower buds, is satisfactory.
Boutry M, Faber AM, Charbonnier M, Briquet M (1984) Microanalysis
of plant mitochondrial protein synthesis products. Plant Mol Biol 3:
445–452
Conley CA, Hanson MR (1995) How do alterations in plant mitochon-
drial genomes disrupt pollen development. J Bioenerg Biomemb
27: 447– 457
Comparison of the proteome of mitochondria from 5-day-
old etiolated male sterile seedlings with the one from near
isogenic fertile seedlings revealed some distinct differences.
Interestingly, some of the proteins which specifically occur in
male sterile plants are in the size range of the 32 kDa protein
reported by Landgren et al. (1996). However, there are other
explanations for the variation in these proteomes which should
be considered:
Damerval C, de Vienne D, Zivy M, Thiellement H (1986) Technical im-
provements in two-dimensional electrophoresis increase the level
of genetic variation detected in wheat-seedling proteins. Electro-
phoresis 7: 52–54
Day DA, Neuburger M, Douce R (1985) Biochemical characterization
of chlorophyll-free mitochondria from pea leaves. J Plant Physiol
12: 219 – 228
Dewey RE, Levings CS III, Timothy DH (1986) Novel recombinations
in the maize mitochondrial genome produce a unique transcrip-
tional unit in the Texas male-sterile cytoplasm. Cell 44: 439–449
(i) There may still be slight differences in the nuclear back-
ground of the male sterile and fertile lines. These differences
in gene content could be responsible for the expression of
additional proteins in male sterile or fertile plants. Future ex-
periments will focus on further backcrossed material with
even less residual heterozygoty between the lines analysed.
(ii) Some of the differences may occur randomly, as revealed
by the repetitions of individual experiments. Further experi-
ments will focus on the delimitation of proteins which tend to be
variant in consecutive experiments. Finally, only those proteins
should be identified that are worth being sequenced N-termi-
nally and being further characterized at the molecular level.
(iii) As a reaction to the presence of a sterility-inducing pro-
tein in the mitochondria, the cell may regulate several nuclear
encoded mitochondrial proteins up or down. It is well estab-
lished that plants have available to them a variety of reactions
to alterations of internal or external conditions. Here, the great
potential of the proteomic approach for the investigation of
CMS in plants becomes obvious. This approach can not only
identify a mitochondrial protein which causes CMS, but also
all other proteins which are regulated up or down due to the
altered mitochondrial physiology. Thus, proteomics may also
provide a versatile tool for the identification and analysis of
restorer genes.
Dewey RE, Siedow JN, Timothy DH, Levings III CS (1988) A 13-kilo-
dalton maize mitochondrial protein in E. coli confers sensitivity to
Bipolaris maydis toxin. Science 239: 293–295
Douce R (1985) Mitochondria in higher plants: Structure, functions
and biogenesis. Academic Press, Orlando, FL
Görg A, Obermaier C, Boguth G, Harder A, Scheibe B, Wildgruber R,
Weiss W (2000) The current state of two-dimensional electrophor-
esis with immobilized pH gradients. Electrophoresis 21: 1037–1053
Grelon M, Budar F, Bonhomme S, Pelletier G (1994) Ogura cyto-
plasmic male-sterility (CMS)-associated orf138 is translated into a
mitochondrial membrane polypeptide in male-sterile Brassica cy-
brids. Mol Gen Genet 243(5): 540–547
Hamasur B, Birgersson U, Eriksson AC, Glaser E (1990) Large-scale
purification of spinach leaf mitochondria-isolation and immunologi-
cal studies of the F1-ATPase. Physiol Plant 78: 367–373
Hanson MR (1991) Plant mitochondrial mutations and male sterility.
Annu Rev Genet 25: 461–486
Heukeshoven J, Dernick R (1986) In: BJ Radola (ed) Elektrophorese
Forum ’86. Technische Universität München, Germany, pp 22–27
Horn R, Köhler RH, Zetsche K (1991) A mitochondrial 16 kDa protein
is associated with cytoplasmic male sterility in sunflower. Plant Mol
Biol 17: 29–36
Jänsch L, Kruft V, Schmitz UK, Braun HP (1996) New insights into the
composition, molecular mass and stoichiometry of the protein
complexes of plant mitochondria. Plant J 9(3): 357–368
Acknowledgements. We wish to thank Drs. H. Becker, G. Stiewe, W.
Ecke, and C. Möllers, Göttingen, for providing plant material of B. na-
pus. Thanks are also due to D. Lewejohann and G. Kühne for expert
technical assistance. All the experiments described in this paper com-
ply with the current laws of Germany. This work was supported by a
grant to UKS, from the Deutsche Forschungsgemeinschaft (Schm 698/
8-1), and a grant to HPB from the Fonds der Chemischen Industrie.
Kahn P (1995) From genome to proteome: Looking at a cell’s proteins.
Science 270: 369–371
Kruft V, Eubel H, Jänsch L, Werhahn W, Braun HP (2001) Proteomic
approach to identify novel mitochondrial functions in Arabidopsis
thaliana. Plant Cell (submitted)
Kügler M, Jänsch L, Kruft V, Schmitz UK, Braun HP (1997) Analysis of
the chloroplast protein complexes by blue-native polyacrylamide
gel electrophoresis (BN-PAGE) Photosynth Res 53: 35–44
Landgren M, Zetterstrand M, Sundberg E, Glimelius K (1996) Allo-
plasmic male-sterile Brassica lines containing B. tournefortii mito-
chondria express an ORF 3‘ of the atp6 gene and a 32 kDa pro-
tein. Plant Mol Biol 32: 879–890
References
Bellaoui M, Pelletier G, Budar F (1997) The steady-state level of
mRNA from the Ogura cytoplasmic male sterility locus in Brassica
cybrids is determined post-transcriptionally by its 3′ region. EMBO
J 16: 5057–5068
Lottspeich F (1999) Proteom analysis: A pathway of the functional
analysis of proteins. Angew Chem Int Ed 38: 2476–2492

794
Christina Mihr et al.
Mathias R (1985) Transfer of cytoplasmic male sterility from brown
mustard (Brassica juncea (L.) Coss.) into rapeseed (Brassica na-
pus L.). Z Pflanzenzucht 95: 371–374
Schägger H, Cramer WA, von Jagow G (1994) Analysis of molecular
masses and oligomeric states of protein complexes by Blue native
electrophoresisand isolationofmembraneproteincomplexesbytwo-
dimensional native electrophoresis. Anal Biochem 217: 220–230
Stiewe G (1993) Erzeugung von männlicher Sterilität in Brassica na-
pus L. Proc. 8th EUCARPIA Congress, Angers, France, pp 125–126
Stiewe G, Witt U, Hansen S, Theis R, Abel WO, Röbbelen G (1995)
Natural and Experimental Evolution of CMS for Rapeseed Breed-
ing. Adv Pl Breed 18: 59–76
Paulmann W, Röbbelen G (1988) Effective transfer of cytoplasmic
male sterility from radish (Raphanus sativus L.) to rape (Brassica
napus L.). Plant Breed 100: 299–309
Pelletier GC, Primard F, Vedel P, Chetrit R, Remy R, Renard M (1983)
Intergeneric cytoplasmic hybridization in Cruciferae by protoplast
fusion. Mol Gen Genet 191: 244–250
Young EG, Hanson MR, (1987) A fused mitochondrial gene asso-
ciated with cytoplasmic male sterility is developmentally regulated.
Cell 50: 41–49
Rawat DS, Anand IJ, (1979) Male sterility in Indian mustard. Indian J
Genet Breeding 39: 412–414