Current Genetics
Current Genetics 1, 75-83 (1979)
© by Springer-Verlag 1979
Fine Structure of OXI1, the Mitochondrial Gene Coding for Subunit II of Yeast Cytochrome c Oxidase* B. Weiss-Brummer, R. Guba, A. Haid, and R. J. Schweyen Genetisches Institut der Universit~itMtinchen, Maria-Ward-Strasse la, D-8000 Miinchen, Federal Republic of Germany
Summary. Genetic and biochemical studies have been performed with 110 mutants which are defective in cytochrome a-a 3 and map in the regions on mit DNA previously designated 0 X l l and 0)(12. With 88 mutations allocated to 0 X l l fine structure mapping was achieved by the analysis of r h o - deletions. The order of six groups of mutational sites (A1, A2, B1, B2, CI, C2) thus determined was confirmed by o x i i X oxij recombination analysis. Analysis of mitochondrially translated polypeptides of o x i l mutants by SDS-polyacrylamide electrophoresis reveals three classes of mutant patterns: i) similar to wild-tpye (19 mutants); ii) lacking SU II of cytochrome c oxidase (53 mutants); iii) lacking this subunit and exhibiting a single new polypeptide of lower Mr (16 mutants). Mutations of each of these classes are scattered over the OXI1 region without any detectable clustering; this is consistent with the assumption that all oxil mutations studied are within the same gene. New polypeptides observed in oxil mutants of class iii) vary in Mr in the range from 10,500 to 33,000. Those of Mr 17,000 to 33,000 are shown to be antigenically related to subunit II of cytochrome c oxidase. Colinearity is established between the series of new polypeptides of Mr values increasing from 10,500 to 31,500 and the order of the respective mutational sites on the map, e.g. mutations mapping in A 1 generate the smallest and mutations mapping in C2 the largest mutant fragments. From these data we conclude that i) all mutations allocated to the OXI1 region are in the same gene;ii) this gene codes for subunit II of cytochrome c oxidase; iii) the direction of translation is from CAP to OXI2. Out of 19 mutants allocated to 0 X I 2 three exhibit a new polypepOffprint requests to." B. Weiss-Brummer * Abbreviations. mitDNA mitochondrial DNA; SU I, SU II, SU III
subunits of cytochrome c oxidase; SDS sodium dodecyl sulfate; Mr Molecular weight; d Dalton.
tide; these and all the other oxi2 mutants lack subunit III of cytochrome oxidase. This result provides preliminary evidence that the OXI2 region harbours the structural gene for this subunit III.
Key words: Genetic mapping - Mutant polypeptides Direction of translation.
Introduction The synthesis of functional cytochrome c oxidase is affected specifically by mutations mapping in three distinct regions on mitDNA of Saccharomyces cerevisiae; these regions are designated OXI1, OXI2 and OXI3 (Slonimski and Tzagoloff, 1976). In addition, some of the mutations in the COB-BOX region on mitDNA also disturb the synthesis of cytochrome c oxidase. Their effect, however, is pleiotropic, leading to a lack of apocytochrome b and of subunit I of cytochrome c oxidase simultaneously (Claisse et al., 1978; Haid et al., 1979). Evidence has accumulated during the past few years that the three O X I regions correspond with the three larger, mitochondrially made subunits of cytochrome c oxidase (SU I, SU II, SU III) in that each region harbours one of the structural genes. This correlation is based on the observation that mitDNA or transcripts of the O X I regions stimulate the in vitro synthesis of peptides which are immunologically related to subunits of cytochrome c oxidase (Grivell and Moorman, 1977; Moorman et al., 1978) and on the study of phenotypic expression of oxil , oxi2 and oxi3 mutations (Slonimski and Tzagoloff, 1976; Mahler et al., 1977; Cabral et al., 1977;Eccleshall et al., 1978). The results obtained in these studies allow a tentative attribution of SU I, SU II and SU III to OXI3, OXI1 and OXI2 , respectively. O172-8083/79/0001/0075/$02.00
76
B. Weiss-Brummeret al.: Fine Structure of OXI 1 were obtained from T. D. Fox (Biozentrum Basel). Other rhoclones were either spontaneously arisen subclones of these two rho- clones or isolated after MnCl2 treatment of rho + strains.
Evidence is most striking for the location of the structural gene coding for SU II ofcytochrome c oxidase within the region 0 X l l . Mutants mapping in this region have been shown to lack SU II and - in a few cases - to replace it by polypeptides of lower Mr which were found to be biochemically related to wild-type SU II (Cabral et al., 1978). A restriction map of this region was established recently including the localization of some mutational sites which most likely are part of the structural gene (Fox, 1979). Large numbers of mutations in O X I 1 and O X I 2 and improved genetics of fine structure mapping now permit extension of the studies on the genetic information encoded by this segment of mitDNA. The aim of the present study is to relate alterations observed in mitochondrially translated subunits of cytochrome c oxidase to mutations in O X I 1 and O X I 2 . The results obtained strictly correlate mutations in O X I 1 with the lack ofSU II and mutations in O X I 2 with the lack of SU III. Some mutations in OXI1 and in O X I 2 are found to generate shorter polypeptides replacing the wild-type subunit. This prorides evidence for the location of the respective mutations in one or the other structural gene. Colinearity was observed between a series of such "new" polypeptides and mutational sites on a fine structure map of the OXI1 region. This result allows the allocation of SU II to the map Of this region and to determine the direction of translation.
Analysis o f Mitochondrial Translation Products. Growth and
labelling of cells with 35SO~- (carrier free) in the presence of cycloheximide and analysis of mitochondrial translation products by electrophoresis on SDS-polyacrylamide gradient gels was carried out according to Douglas and Butow (1976) with the following modifications: i) Diploid cells from crosses 777-3.4 rho + m i t - x R M 590-9D rho 0 were grown in 2% galactose complete medium and labelled in a low sulphate medium containing 0.3% galactose after preincubation with chloramphenicol (Haid et al., 1979). ii) In order to obtain strong labelling of mutant mitochondrial translation products of Mr lower than 18,000, diploid ceils were grown and labelled in media containing 0.5% and 0.3% glucose, respectively (Claisse et al., 1977), without preincubation with chloramphenicol. Immunological Identification o f Mutant Polypeptides. All reac-
tions were carried out with antiserum raised against subunit II of cytochrome c oxidase, kindly provided by Prof. G. Schatz, Biozentrum Basel. Immune-overlay experiments Were performed essentially according to the procedure of Showe et al. (1976) with modifications as described by Cabral et al. (1978). Itnmunoprecipitation by use of glutaraldehyde fixed cells of Staphylococcus aureus was carried out as described by Maccechini et al. (1979). Other Methods. Cytochrome spectra were obtained by the method described by Haid et al. (1979). Crosses for the r h o - x m i t - deletion mapping and for the rnit- x m i t - recombination
analysis were performed as described by Schweyen et al. (1978) and Haid et al. (1979). "Cytoduction" was performed by mating mutant strains 777-3A c~ m i t - with strain 1CS/AA1 a leu- karl rho 0 (Lancashire and Mattoon, 1979) on glucose minimal medium. The mating mixture was grown overnight, resuspended and plated in appropriate dilutions on minimal glucose medium supplemented with leucine. On this medium diploid cells form large colonies, a leu- rho ° cells form small colonies and a leu- m i t - form colonies of intermediate size.
Material and Methods Strains. A list of strains (Saccharomyces cerevisiaej used is given in Table 1. The isolation of m i t - mutants was described by Sehweyen et al. (1978). The r h o - clones TFC45 and TF8 6
Table 1. List of strains Strain
_Ge_n_o_ty lo_e. . . . . . . . . . . . . . . . nuclear mitochondrial
Remarks
Reference
777-3A
o~ a d e l , o p l
rho + mit +
"wild-type"
K o t y l a k and S l o n i r n s k i ( 1 9 7 7 )
M1273 e t c .
c~ a d e l , o p l
rho + lr,it-
d e r i v e d f r o m 777 3A
S c h w e y e n et al.
(1978)
M8-171
o~
rho +. mit-
derived
T r e m b a t h et al,
(1977)
a_ trpl
rho °
i)
S e h w e y e n et al.
(1978)
a leul, karl
rho °
RM
etc.
590-9D
IC8/AAI
used
from
for karl
D273-10B
mediated
Lancashire
and IVIattoon (1979)
"cytodustiorl' + TF2
a_ his4
+
rho, rnit
to generate C45and 8-6
F o x (1979)
+
KL
14-4A
a hisl,trpl
+ rho, mit , cap r, olilr,par r
used rho-
used to generate shown in Fig la
rho-
1) used in crosses with 777-3A mit- strains to generate diploid mit- strains
H a i d et al.
(1979)
B. Weiss-Brummer et al.: Fine Structure of OXI l
77 The majority o f mutations fall in a group containing several oxi] mutations previously mapped. All o f these fulfill the criteria of OXI1 mutations as will be shown below (cf. Table 2). A second group comprises 19 m i t mutations which map together with a reference markers of the OXI2 locus. A single mutation (M5401) maps between these two groups; two others (M5391 and M6681) are located outside of the O X I I - O X I 2 span, between CAP and TSM-8. None o f these three mutations appears to affect the synthesis o f cytochrome oxidase (data not shown). The large set comprising the o x i l mutations has been further subdivided by use o f two r h o - clones (TFC45 and T F 8 - 6 ) which have been shown previously to reconstitute differentgroups o f o x i l mutations only ( F o x , 1979) and by use of four subclones which arose spontaneously in cultures of these two r h o - clones. As shown in Fig. l b , the OXI1 region is subdivided into six groups of mutational sites, A1, A2, B1, B2, C1 and C2.
Results The Genetic Map o f the OXI1 Region as Determined by r h o - deletion Analysis
A set of 110 mutations was allocated by preliminary experiments as described earlier (Schweyen et al., 1978) to the segment of the mitochondrial genome which is known to harbour the regions OXI1 and OXI2 (Slonimski and Tzagoloff, 1976). These m i t - mutants then were crossed with several r h o - clones which had been selected to retain one, two or more o f the genetic markers previously m a p p e d in this region (cap r, tsm-8, o x i l , oxi2; Slonimski and Tzagoloff, 1976; Monnerot et al., 1977; Trembath et al., 1977). The results o f these crosses are represented schematically by Fig. la. cap
~,5~9~
M5587
rsu-8
OXI I
Ms~ol 0)(12
Recombinant Frequencies in m i t - x m i t - Crosses
92 OX/ ! mutont5
All o x i l mutants were crossed to a collection o f mutants from this and from other regions. This collection contained several mutants from each of the subgroups A1 to C2 and the neighbouring marker tsm-8 as well as an oxi2 marker. By this test two mutations, M6771 and M8-171,
19 O X I 2 mulont5
0Xl
1
'',
C&5 8-6 $1 p1 R4 R5 M2"71 M2481 N4971 N5411 M5441 MSL81 M5501 M5611 M5531 M 5641
N56fA M5671 M5701 M5721 M57L1 N5751 M5761 M7662
1,44234 M,'-526 N4246 Nl't533 M4364 M45&7 M4408 N4551 M4449 M45T3 M4462 M5642 M4/-,81 M5711 N4510
N 8 - 171
M1237 M2511 M5301 M5322 M5371 M5451 N 5561 N 5821
M&611 M5351 M5541 M5681 M9-69 M9-94 M13-249
/
M224.~ M5303 M5392 M5553 M5591 MS59& N5651 M 56-Crl
M5781 M5791 MSB&I M5851 M5861 M5881 M6/,01
M1524N5421N5731 M2012MS431M5782 M2121MS461M5801 M23&1M5511M5811 M4360M5531MS831 M/.636H5551H5842 N4983M5571M 5671 M5311 M5581M 12-246 M5341 N 5 5 ~ M5381 M 5621
M6771
Fig. 1 a and b. Mapping of m i t - mutations in the CAP-OXII-OXI2 segment of the mitochondrial genome, a A series of rho- clones was selected which retain various combinations of the mutational sites capr, oxi1-M9-94, oxil-M8-171, oxi2-M9-3 (Trembath et al., 1977) and tsm-8 (Monnerot et al., 1977). The series of newly isolated mit- mutants then was crossed with these rho- clones in order to determine retention of the respective sites in the rho- genomes. The new mutational sites were allocated to the known regions OXI1 and OXI2 or located outside (M5401; M6681, M5391) depending on co-retention with or separation from the known sites, b Fine structure mapping of mutations allocated to OXI1 was obtained by analyzing retention/deletion of mutational sites in rho- clones TFC45, TF8-6 (Fox, 1979) and four rho- subclones derived from them. Retention of mutational sites in rho- clones is represented by horizontal bars. The open box in the lower part of Fig. lb indicates deletion of all oxil mutational sites by mutations M8-171 and M6771
78
B. Weiss-Brummer et al.: Fine Structure of OXI 1
were found to be large deletions: recombinants were detected only in pairwise crosses with mutants mapping outside of the 0 X l l region. The conclusion that the deletions cover all oxil mutational sites is in agreement with findings of Trembath et al. (1977) concerning M 8 - 1 7 1 , and is consistent with the r h o - deletion mapping. M 8 - 1 7 1 recently was shown to lack a restriction fragment of 2,400 base pairs (Fox, 1979). In Fig. lb the deletions are represented by long open boxes. In order to obtain a quantitative determination of recombinant frequencies, five m i t - mutants were crossed with a series of oxil mutants mapping in the six groups
A1 to C2. Frequencies of mit + recombinants observed range between 0% and more than 10%. These latter values signify absence of genetic linkage; they are dlose to the maximal values expected for each of the two possible recombinant classes (mit +, which only can be scored, and mit~- mitT) in the absence of polarity o f recombination (Dujon et al., 1974). Recombinant frequencies observed in five series of crosses are given in Fig. 2. The results confirm the order of mutational sites as determined by r h o - deletion mapping in that recombinant frequencies increase in the order of groups o f mutational sites A1 to C2. Recombinant frequencies between the most distant oxil muta-
Table 2. Synopsis of mutant phenotypes Mutant
Map pos~tSon
Cytochrome a'a 3 ( 6 0 3 nm)
Growth on ethano~
oxidase II
subunit III
M r of "new" polypeptide
OXII M5761
AI
leaky
(+)
+
10 5 0 0
M5661
AI
leaky
(+)
+
10 5 0 0
M5701
AI
leaky
(+)
+
10 5 0 0
M5631
AI
leaky
(+)
+
12 5 0 0
M5411
AI
-
+
12 5 0 0
M5721
AI
leaky
(+)
+
12 5 0 0
M4364
A2
-
-
+
12 5 0 0
M4575
A2
-
+
12 5 0 0
M2511
BI
+
15 0 0 0
M5821
BI
+
17 0 0 0
M5301
BI
+
33 000
M4611
B2
+
17 0 0 0
M5681
B2
+
18 0 0 0
M5351
B2
+
32 0 0 0
M5861
CI
+
27 0 0 0
M5341
C2
+
31 5 0 0
M5671
AI
599.5
n m 1)
-
+
M5661
AI
600
n m I)
-
+
M5451
BI
601
n m I)
-
+
M5431
C2
601.5
n m I)
-
+
o t h e r s 2)
AI t o C 2
-
-
+
others
AI t o C 2
~
+
+
M5491
-
+
-
M5321
-
+
-
15 5 0 0
M7582
-
+
-
22 7 0 0
3)
OXI2
others
4)
-
_
13 5 0 0
+
1) Absorption spectra were recorded from whole yeast cells in liquid nitrogen. 2-4) A total number of 53, 19, and 16 mutants, respectively. Mr values of SU II and SU III of cytochrome c oxidase are 33,500 and 23,000, respectively. (+): reduced amounts of SU II of Mr 33,500 present.
79
B. Weiss-Brummer et al.: Fine Structure of OXI1 OXI 1 CA P
-_
M 5391 M 5681
/ ~ /
TSM-8
~
M5561
I
1
A,
ML354
M5631
I
I
M 1273 M2511
I
A2
I
I
I
B1
i
M4611
I
I
M5681
=2
I
M5861
I
c,
M4638
I
I
I
M4350`
c2
I
OXI 2
1
1ZO 4 6
5,4
4,5
5,0
63
5,5
8,5
7,5
9,0 14,0
5O 0,1
- - 0 , 2
3,7
~2
2,9
18 3,3
3,4
20,0 3,5
16,9
2,0
0,2 -
-
- -
3,I
3,4
43
87
14,0
11,8 10,6
9,0
8,3
7,9'
11,7
5,5 3
lZ,,6
9,4
Z1
6,0
10,6
3,4
10,2
7,7"
6,0
.
3
"
4,3-
-
-
0,9 -
-
-
-
0,1
.15,4
.20.0
Fig. 2. Mapping by mit F x mitT recombination analysis. Frequencies of mit+ recombinants were determined in palrwise crosses of mit- mutants of opposite mating type as described previously (Hald et al., 1979). Strains of mating type a and carrying rnit- mutations were obtained by "cytoduction" in crosses 777-3,4 ~ mit- x 1C8 a karl rho °. A series of horizontal bars and figures at their ends refer to crosses of a given mit- mutant with a series of others and to the percentage of mit + recombinants observed. Groups A1 to C2 are defined by Figure lb; the order of mutational sites within A1, B1, B2 and C2 is ambiguous
tions involved are close to the maximal values expected although the length of this region is only about 1.000 base pairs (Fox submitted for publ.). Similarly high frequencies have been observed with mutations in the COBB O X region of mitDNA (Kotylak and Slonimski, 1977; Haid et al., 1979; Slonimski, Kotylak and Schweyen, in prep.). In a previous study of Trembath et al. (1977), however, recombinant frequencies in crosses between oxil mutants were found to be low (maximum of 1.6%) and not useful to order mutations in this region. Differences in the genetic background of strains used in different laboratories may account for these discrepancies in frequencies of recombinants.
Mutant Pheno types Cytochrome Spectra. Four mutants exhibit a shift in the absorption band of cytochrome a-a 3 (cf. Table 2 and Fig. 3). All the other mutants allocated to the regions OXI1 and OXI2 lack the absorption band of cytochrome a-a 3 ; other cytochromes appear not to be affected. Mitochrondrial Translation Products in oxil Mutants.
Consistent with previous studies of Cabral et al. (1978) three classes of mutant patterns of mitochondrial translation products can be recognized. (cf. Fig. 4, Table 2). i) SU II ofcytochrome oxidase (M r 33,500) is lacking; no further alterations in pattern of translation products are detected. This class comprises 53 out of 88 mutants, ii)
SU II is replaced by a single new polypeptide of Mr lower than 33,500. This phenotype is detected in 16 of the mutants studied, iii) Mutants exhibit a pattern similar to wild-type; 19 mutants show this phenotype. New polypeptides exhibited by mutants of class ii) vary in the range from 10.500 to 33.000 d. A correlation between these Mr values and map positions of respective mutations reveals a pattern of strict colinearity in the following order: 10.5 and 12.5 kd: A1; 12.5 kd: A2; 15 and 17 kd: B 1 ; 1 7 a n d 1 8 k d : B 2 ; 2 7 k d : C l ; 3 1 . 5 kd: C2. An obvious conclusion is that the series of polypeptides of 10.5 to 31.5 kd replacing SU II are generated by premature chain termination due to nonsense mutation or frame shift mutations followed by a nonsense codon at a short distance. According to this interpretation mutations in A1 to C2 in the OXI1 region are within the coding sequence of the same polypeptide which most likely is SU II of cytochrome oxidase. Only two mutations generating new polypeptides of M r closest to wild-type subunit II (M5301, 33 kd; M 5 3 5 1 , 3 2 kd) do not fit into this pattern of colinearity. Alterations in charge and/or conformation of polypeptides generated by mis-sense mutation may account for the slightly lower M r . (This interpretation also can explain modified electrophoretic mobility of revertant SU II as will be discussed below.) Immunological Identification o f N e w Polypeptides Replacing S U I1. More direct information on the relation-
ship between mutant and wild-type mitochondrial translation products can be obtained by an immunological ap-
80
B. Weiss-Brummer et aL: Fine Structure of OXI1
r
i 0
520
ii
I ,' rh
5C0
56O
i
I,:
I
580
600
820
Wavelength (rim)
Fig. 3. Low temperature absorption spectra of wild-type and mutant M 5671. Absolute spectra were recorded as described by-Haid et al. (1979). Absorption maxima of cytochrome a.a 3 (wild-type: 603 nm; mutant M 5671:599.5 nm), cytochrome b (558.5 rim), cytochrome c 1, (553 nm) and cytochrome c (547 rim) are indicated Fig. 4. Mitochondrial translation products of wild-type and oxil mutants. Cells were labelled with 3 S s o 2 - in the presence of cycloheximide (cf. Material and Methods). Separation of translation products on 10 to 15% SDS polyacrylamide gradient gels was followed by radioantography. Arrows identify SU I, SU II and SU III of cytochrome c oxidase, apocytochrome b and "new" polypeptides detected in mutants. In case of mutants M5861 and M2511 "new" polypeptides are weakly labelled but clearly detectable on the original radioautographs. *Indicates labelling of cells in 0.3% glucose instead of galactose; this causes weak labelling of apocytochrome b
proach as applied by Cabral et al. (1978) in case o f two
oxil mutants, by Kreike et al. and Solioz and Schatz (both in press) in case of cob-box mutants. Therefore, labelled mitochondrial protein was immunoprecipitated by use o f antisera raised against SU II of cytochrome c oxidase. Electrophoresis followed by radioautography revealed that wild-type subunit II and m u t a n t polypeptides o f M r 17,000 to 33,000 were precipitated in significant amounts (cf. Fig. 5). In contrast, with m u t a n t polypeptides of M r 10.500 to 15.000 no positive response was detectable (not shown). Ai_i:i.... ~ of the immuneoverlay technique gave the same resuhs 0~ot shown). This clearly relates the series of new polypeptides with M r values from 17,000 to 33,000 to subunit II of cytochrome c oxidase. It supports the conclusion that these polypeptides are fragments o f this subunit, generated by premature chain termination. Fragments smaller than 17,000 d apparently lack antigenic sites accessible to the antibodies applied. However, the fact that they form together with the larger, antigenic fragments a continuous series, colinear with map positions of mutations, strongly supports the idea that these also are fragments o f SU II.
Modified SU H in Revertants. Several respiratory competent (oxt~) revertants o f each one of the 16 mutants exhibiting "new" polypeptides were isolated. Revertants derived from 13 of these mutants showed patterns o f mitochondrial translation products indistinguishable from wild-type. This finding is consistent with the assumption that the short "new" polypeptides typical for these mutants are generated by chain terminating codons and that reversion restores the wild-type codon or another legible codon leading to a SU II o f wild-type electrophoretic mobility. With the other three mutants, M2511, M5681, M5351, however, revertant SU II different in electrophoretic mobility from that o f wild-type was detected. M r values of these revertant polypeptides are variable even amongst revertants of the same oxil mutant; some are slightly higher, the majority lower in M r than wild-type SU II. Results obtained with revertants of m u t a n t M2511 are shown in Fig. 6. Similar observations were made by Cabral et al. (1978) with revertants of two oxil mutants (M 1 3 - 249, M 9 - 6 9 , for map positions cf. Fig. lb). Sequencing of mitDNA o f mutant M 1 3 - 2 4 9 and of some o f its revertants reveals that M 1 3 - 2 4 9 is a frame shift mutation
B. Weiss-Brummeret al.: Fine Structure of OXI1
8l and that reversion restores the proper reading frame. A series of mis-sense codons generated thereby causes alterations in charge of SU II which are correlated with modified Mr values in revertants (Fox, submitted for puN.). The assumption that sequence dependent secondary structures persist in SU II even in SDS and change its electrophoretic mobility could explain modified Mr values in two of the mutants (cf. Table 2) and in three series of revertants studied here. Mitochondrial Translation Products in oxi2 Mutants. All of the 19 mutants allocated to the OXI2 region were found to lack SU III ofcytochrome c oxidase. Other mitochondrial translation products appear to be present although amounts ofSU I of cytochrome c oxidase seem to be low in some mutants. There oxi2 mutants exhibit a new polypeptide of Mr lower than SU III (M5491, M5321, M7582, cf. Fig. 7 and Table 2). In order to demonstrate that new polypeptides are related to SU III, their immunoprecipitation was attempted by use of the antiserum raised against SU II which co-precipitates SU II1 in small amounts (cf. Fig. 5). With the two mutants tested (M5491 and M5321) significant amounts o f p o l y -
Fig. 5. Immunological identification of "new" polypeptides in oxil mutants. Cells were labelled in the presence of cycloheximide with 35SO2 - , mitochondrial particles were isolated and treated with antiserum against SU II of cytochrome c oxidase and with fixed cells of Staphylococcus aureus (cf. Material and Methods). Immunoprecipitates were analyzed on 12% SDS-polyacrylamide gels followed by radioautography. The figure combines radioautographs derived from two different electrophoretic experiments. The serum cross-reacts weakly with SU III of cytochrome oxidase
Fig. 6a and b. Mitochondrial translation products in revertants of oxil mutant M 2511. a Translation products are visualized as described in Fig. 4. b Polypeptides reacting with antiserum against SU II were identified on the gel slabs with the immune-overlay technique (cf. Material and Methdds) followed by radioautography. R1, R2, R3, R4 designate revertants of M 2511 (cf. Fig. 4), WT: wildtype
82 peptides of Mr 13.500 and 15,500, respectively, were found to be precipitated (not shown). Although conclusive evidence for the gene - product relationship in the 0)212 region will be provided only by more detailed studies, the results obtained show that mutations in this region specifically alter SU III o f cytochrome c oxidase. Most likely these mutations are in the structural gene.
Discussion Alterations in Mr of SU II ofcytochrome c oxidase observed with two oxil mutants were taken as first evidence for the localization of the sequence coding for SU II
Fig. 7. Mitochondrial translation products of oxi2 mutants. Labelled polypeptides were visualized as in Fig. 4. Arrows identify SU I, SU II and SU III of cytochrome c oxidase, apocytochrome b and "new" polypeptides detected in mutants
B. Weiss-Brummer et al.: Fine Structure of OXI1 within the OXI1 region by Cabral et al. (1978). In the present paper 16 further mutants are described which exhibit a new mitochondrial translation product replacing SU II. M r values of these mutant polypeptides vary in the range from 10,500 to 33,000. Those of M r 17,000 and higher are found to be immunologically related to wild-type SU II. This is consistent with the previous observation and strongly supports the conclusion of Cabral et al. (1978). New products of M r in the range from 10,500 to 15,000 are found not to react with antibodies directed against SU II. This lack of immunological response may reflect a loss of antigenic determinants due to shortening of SU II polypeptides to less than 50%. The alternative explanation that these short polypeptides are not related to SU II is most unlikely for the following reason: New polypeptides of Mr 10,500 to 15.000 and 17,000 to 31.500 form a unique series which correlates perfectly with the series of mutations on the map in the way that mutations mapping in A1, i.e. proximal to CAP (cf. Fig. 1), generate polypeptides of the smallest Mr observed and mutations mapping more distant to CAP, i.e. in A2 to C2, give rise to polypeptides of higher M r values up to 3 1 5 0 0 . This observation of colinearity strongly suggests that premature chain termination due to nonsense mutation causes the appearance of fragments of SU II. As alternative explanations for the variety of mutant polypeptides one might envisage proteolysis or post-transcriptional modification or alteration of electrophoretic mobility due to mis-sense mutation in the gene for SU 1I. None of these, however, readily accounts for the series of mutant polypeptides of at least seven groups of M r values between 1 0 5 0 0 and 31,500 or for their colinearity with mfitational sites on the genetic map. Minor modification of the electrophoretic mobility due to mis-sense mutation is a likely explanation only for the occurence of polypeptides with Mr values close to that of wild-type SU II in case of two mutants which do not fit in the general scheme of colinearity (M5301, M5351, cf. Table 2) and in case of three series of revertants (cf. Fig. 6). Indications for modified electrophoretic mobility of SU II in the presence of SDS due to alterations in charge of revertant polypeptides come from recent experiments (Fox, submitted for publ.). Based on the pattern of colinearity observed it is concluded that i) the structural gene for SU II of cytochrome c oxidase is in the OXI1 region and ii) the direction of translation is from the CAP proximal end (A 1) to the CAP distal end (C2). Whether all mutations mapping in A1 are located within the structural gene or in sequences regulating its expression cannot be decided. Consistent with the conclusion that 0 X l l contains only one structural gene are two further observations: i) Mutations generating one or the other of four different phenotypic classes observed (cf. Table 2) are scattered
B. Weiss-Brummer et al.: Fine Structure of OXI1
over all subgroups mapped in the OXI1 region, ii) Linkage in terms of recombinant frequencies is observed between and within all subgroups of oxil mutations and between mutations in A1 and the neighbouring marker tsm-8 (cf. Fig. 2). A comparison between the percentage of recombinants observed with two mutations and the length difference of SU II fragments generated by early chain termination gives an estimate of 1% recombinants for a distance of roughly 50 base pairs. Similar estimates were obtained with the more elaborate genetics of the COB-BOX region (Haid et al., 1979; Kotylak, Slonimski, Schweyen, in prep.). Application of this estimate to recombinant frequencies shown in Fig. 2 reveals that the OXI1 region is relatively short and that the coding sequence is not interrupted by long intervening sequences, i.e. exceeding several hundred base pairs, as found in the COB-Box region. It is remarkable that out of 88 oxil mutants 53 lack wild type SU II without exhibiting a detectable new polypeptide. This contrasts with a very low frequency of such mutants in the COB-BOX region (Haid et al., 1979) where, with few exceptions, mutations mapping in the coding regions generate a fragment of apocytochrome b. A possible explanation is that many SU II polypeptides which are terminated by nonsense mutation or modified by mis-sense mutation are not as firmly integrated in the mitochondrial membrane and/or are rapidly degraded. Therefore, they may not have been detectable in preparations mainly composed of submitochondrial particles as used in the present study. A similar explanation may apply to the puzzling observation that mitochondrial particles of four mutants lack detectable SU II whereas whole cells of these mutants exhibit the spectral absorption bands of cytochrome a.a 3, although shifted to some extent (cf. Table 2 and Fig. 3). Mahler et al. (1977) described one mutant mapping in the OXI2 region and having replaced SU III of cytochrome c oxidase by a slightly shorter polypeptide. In the study presented here 19 oxi2 mutants are described out of which three replace SU III (23,000 d) by a polypeptide of lower Mr. The other oxi2 mutants also lack SU III but show no detectable amounts ofnewlyarising mutant polypeptides. These data together with the observation of Mahler et al. (1977) are highly suggestive that SU Ill is encoded in the 0X12 region and that at least some of the oxi2 mutations, namely those generating a mutant polypeptide shorter than 23,000 d, are in the structural gene for SU III. Acknowledgement. We thank Dr. T. D. Fox and P. Schnittchen for giving valuable yeast strains and for communicating results prior to publication and Dr. A. Lewin for helpful discussion. The excellent technical assistance of Mrs. S. Pollinger is highly appreciated. This work was supported by the Deutsche Forschungsgemeinschaft.
83
References Cabral, F., Solioz, M., Deters, D., Rudin, Y., Schatz, G., Clavilier, L., Groudinski, O., Slonimski, P. P.: In: Mitochondria 1977; Genetics and Biogenesis of Mitochondria. Bandlow, W., Schweyen, R. J., Wolf, K., Kaudewitz, F. (eds.), pp. 4 0 1 414. Berlin: de Gruyter 1977 Cabral, F., Solioz, M., Rudin, Y., Schatz, S., Clavilier, L., Slonimski, P. P.: J. Biol. Chem. 253,297-304 (1978) Claisse, M., Spyridakis, A., Slonimski, P. P.: In: Mitochondria 1977; Genetics and Biogenesis of Mitochondria. Bandlow, W., Schweyen, R. J., Wolf, K., Kaudewitz, F. (eds.), pp. 337-344, Berlin: de Gruyter 1977 Claisse, M., Spyridakis, A., Wambier-Kluppel, M. L., Pajot, P., Slonirnski, P. P.: In: Biochemistry and Genetics of Yeast. Bacila, M., Horecker, B. L., Stoppani, A. K. M. (eds.), pp. 369-390. New York: Academic Press 1978 Douglas, M., Butow, R. A.: Proc. Nat. Acad. Sci. USA. 73, 1083-1086 (1976) Dujon, B., Slonimski, P. P., Weill, L.: Genetics 78, 415-437 (1974) Eccleshall, T. R., Needleman, R. B:, Storm, E. M., Buchferer, B., Marmur, J.: Nature 273, 67-70 (1978) Fox, T. D.: J. Mol. Biol. 130, 63-82 (1979) Grivell, L. A., Moorman, A. F. M.: In: Mitochondria 1977, Genetics and Biogenesis of Mitochondria. Bandlow, W., Schweyen, R. J., Wolf, K., Kaudewitz, F. (eds.), pp. 371-384. Berlin: de Gruyter 1977 Haid, A., Schweyen, R. J., Bechmann, H., Kaudewitz, F., Solioz, M., Schatz, G.: Eur. J. Biochem. 94,451-464 (1979) Kreike, J., Bechman, H., Van Hemert, F. J., Boer, P. H., Schweyen, R. J., Kaudewitz, F., Groot, G. S. P.: Eur. J. Biochem. (in press) Kotylak, Z., Slonimski, P. P.: In: Mitochondria 1977, Genetics and Biogenesis of Mitochondria. Bandlow, W., Schweyen, R. J., Wolf, K., Kaudewitz, F. (eds.), pp. 161-172. Berlin: de Gruyter 1977 Lancashire, W. E., Mattoon, J~R.: Mol. Gen. Genet. 170, 333344 (1979) Maccechini, M. L., Rudin, Y., Blobel, G., Schatz, G.: Proc. Nat. Acad. Sci. USA 76, 343-347 (1979) Mahler, H. R., Hanson, D., Miller, D., Bihnski, T., Ellis, D. M., Alexander, N. J., Perlman, P. S.: In: Mitochondfia 1977, Genetics and Biogenesis of Mitochondria. Bandlow, W., Schweyen, R. J., Wolf, K., Kaudewitz, F. (eds.), pp. 345-370. Berlin: de Gruyter 1977 Monnerot, M., Schweyen, R. J., Fukuhara, H.: Mol. Gen. Genet. 152, 307-309 (1977) Moorman, A. F. M., Van Ommen, G. J. B., Grivell, L. A.: Mol. Gen. Genet. 160, 13-24 (1978) Schweyen, R. J., Weiss-Brummer, B., Backhaus, B., Kaudewitz, F.: Mol. Gen. Genet. 159, 151-160 (1978) Showe, M., Isobe, E., Onorato, L.: J. Mol. Biol. 107, 55-69 (1976) Slonimski, P. P., Tzagoloff, A.: Eur. J. Biochem. 61, 27-41 (1976) Solioz, M., Schatz, G.: J. Biol. Chem. (in press) Trembath, K. M., Macino, G., Tzagoloff, A.: Mol. Gen. Genet. 158, 35 45 (1977)
Communicated by F. Kaudewitz Received August 20, 1979
Current Genetics 1, 75-83 (1979)
© by Springer-Verlag 1979
Fine Structure of OXI1, the Mitochondrial Gene Coding for Subunit II of Yeast Cytochrome c Oxidase* B. Weiss-Brummer, R. Guba, A. Haid, and R. J. Schweyen Genetisches Institut der Universit~itMtinchen, Maria-Ward-Strasse la, D-8000 Miinchen, Federal Republic of Germany
Summary. Genetic and biochemical studies have been performed with 110 mutants which are defective in cytochrome a-a 3 and map in the regions on mit DNA previously designated 0 X l l and 0)(12. With 88 mutations allocated to 0 X l l fine structure mapping was achieved by the analysis of r h o - deletions. The order of six groups of mutational sites (A1, A2, B1, B2, CI, C2) thus determined was confirmed by o x i i X oxij recombination analysis. Analysis of mitochondrially translated polypeptides of o x i l mutants by SDS-polyacrylamide electrophoresis reveals three classes of mutant patterns: i) similar to wild-tpye (19 mutants); ii) lacking SU II of cytochrome c oxidase (53 mutants); iii) lacking this subunit and exhibiting a single new polypeptide of lower Mr (16 mutants). Mutations of each of these classes are scattered over the OXI1 region without any detectable clustering; this is consistent with the assumption that all oxil mutations studied are within the same gene. New polypeptides observed in oxil mutants of class iii) vary in Mr in the range from 10,500 to 33,000. Those of Mr 17,000 to 33,000 are shown to be antigenically related to subunit II of cytochrome c oxidase. Colinearity is established between the series of new polypeptides of Mr values increasing from 10,500 to 31,500 and the order of the respective mutational sites on the map, e.g. mutations mapping in A 1 generate the smallest and mutations mapping in C2 the largest mutant fragments. From these data we conclude that i) all mutations allocated to the OXI1 region are in the same gene;ii) this gene codes for subunit II of cytochrome c oxidase; iii) the direction of translation is from CAP to OXI2. Out of 19 mutants allocated to 0 X I 2 three exhibit a new polypepOffprint requests to." B. Weiss-Brummer * Abbreviations. mitDNA mitochondrial DNA; SU I, SU II, SU III
subunits of cytochrome c oxidase; SDS sodium dodecyl sulfate; Mr Molecular weight; d Dalton.
tide; these and all the other oxi2 mutants lack subunit III of cytochrome oxidase. This result provides preliminary evidence that the OXI2 region harbours the structural gene for this subunit III.
Key words: Genetic mapping - Mutant polypeptides Direction of translation.
Introduction The synthesis of functional cytochrome c oxidase is affected specifically by mutations mapping in three distinct regions on mitDNA of Saccharomyces cerevisiae; these regions are designated OXI1, OXI2 and OXI3 (Slonimski and Tzagoloff, 1976). In addition, some of the mutations in the COB-BOX region on mitDNA also disturb the synthesis of cytochrome c oxidase. Their effect, however, is pleiotropic, leading to a lack of apocytochrome b and of subunit I of cytochrome c oxidase simultaneously (Claisse et al., 1978; Haid et al., 1979). Evidence has accumulated during the past few years that the three O X I regions correspond with the three larger, mitochondrially made subunits of cytochrome c oxidase (SU I, SU II, SU III) in that each region harbours one of the structural genes. This correlation is based on the observation that mitDNA or transcripts of the O X I regions stimulate the in vitro synthesis of peptides which are immunologically related to subunits of cytochrome c oxidase (Grivell and Moorman, 1977; Moorman et al., 1978) and on the study of phenotypic expression of oxil , oxi2 and oxi3 mutations (Slonimski and Tzagoloff, 1976; Mahler et al., 1977; Cabral et al., 1977;Eccleshall et al., 1978). The results obtained in these studies allow a tentative attribution of SU I, SU II and SU III to OXI3, OXI1 and OXI2 , respectively. O172-8083/79/0001/0075/$02.00
76
B. Weiss-Brummeret al.: Fine Structure of OXI 1 were obtained from T. D. Fox (Biozentrum Basel). Other rhoclones were either spontaneously arisen subclones of these two rho- clones or isolated after MnCl2 treatment of rho + strains.
Evidence is most striking for the location of the structural gene coding for SU II ofcytochrome c oxidase within the region 0 X l l . Mutants mapping in this region have been shown to lack SU II and - in a few cases - to replace it by polypeptides of lower Mr which were found to be biochemically related to wild-type SU II (Cabral et al., 1978). A restriction map of this region was established recently including the localization of some mutational sites which most likely are part of the structural gene (Fox, 1979). Large numbers of mutations in O X I 1 and O X I 2 and improved genetics of fine structure mapping now permit extension of the studies on the genetic information encoded by this segment of mitDNA. The aim of the present study is to relate alterations observed in mitochondrially translated subunits of cytochrome c oxidase to mutations in O X I 1 and O X I 2 . The results obtained strictly correlate mutations in O X I 1 with the lack ofSU II and mutations in O X I 2 with the lack of SU III. Some mutations in OXI1 and in O X I 2 are found to generate shorter polypeptides replacing the wild-type subunit. This prorides evidence for the location of the respective mutations in one or the other structural gene. Colinearity was observed between a series of such "new" polypeptides and mutational sites on a fine structure map of the OXI1 region. This result allows the allocation of SU II to the map Of this region and to determine the direction of translation.
Analysis o f Mitochondrial Translation Products. Growth and
labelling of cells with 35SO~- (carrier free) in the presence of cycloheximide and analysis of mitochondrial translation products by electrophoresis on SDS-polyacrylamide gradient gels was carried out according to Douglas and Butow (1976) with the following modifications: i) Diploid cells from crosses 777-3.4 rho + m i t - x R M 590-9D rho 0 were grown in 2% galactose complete medium and labelled in a low sulphate medium containing 0.3% galactose after preincubation with chloramphenicol (Haid et al., 1979). ii) In order to obtain strong labelling of mutant mitochondrial translation products of Mr lower than 18,000, diploid ceils were grown and labelled in media containing 0.5% and 0.3% glucose, respectively (Claisse et al., 1977), without preincubation with chloramphenicol. Immunological Identification o f Mutant Polypeptides. All reac-
tions were carried out with antiserum raised against subunit II of cytochrome c oxidase, kindly provided by Prof. G. Schatz, Biozentrum Basel. Immune-overlay experiments Were performed essentially according to the procedure of Showe et al. (1976) with modifications as described by Cabral et al. (1978). Itnmunoprecipitation by use of glutaraldehyde fixed cells of Staphylococcus aureus was carried out as described by Maccechini et al. (1979). Other Methods. Cytochrome spectra were obtained by the method described by Haid et al. (1979). Crosses for the r h o - x m i t - deletion mapping and for the rnit- x m i t - recombination
analysis were performed as described by Schweyen et al. (1978) and Haid et al. (1979). "Cytoduction" was performed by mating mutant strains 777-3A c~ m i t - with strain 1CS/AA1 a leu- karl rho 0 (Lancashire and Mattoon, 1979) on glucose minimal medium. The mating mixture was grown overnight, resuspended and plated in appropriate dilutions on minimal glucose medium supplemented with leucine. On this medium diploid cells form large colonies, a leu- rho ° cells form small colonies and a leu- m i t - form colonies of intermediate size.
Material and Methods Strains. A list of strains (Saccharomyces cerevisiaej used is given in Table 1. The isolation of m i t - mutants was described by Sehweyen et al. (1978). The r h o - clones TFC45 and TF8 6
Table 1. List of strains Strain
_Ge_n_o_ty lo_e. . . . . . . . . . . . . . . . nuclear mitochondrial
Remarks
Reference
777-3A
o~ a d e l , o p l
rho + mit +
"wild-type"
K o t y l a k and S l o n i r n s k i ( 1 9 7 7 )
M1273 e t c .
c~ a d e l , o p l
rho + lr,it-
d e r i v e d f r o m 777 3A
S c h w e y e n et al.
(1978)
M8-171
o~
rho +. mit-
derived
T r e m b a t h et al,
(1977)
a_ trpl
rho °
i)
S e h w e y e n et al.
(1978)
a leul, karl
rho °
RM
etc.
590-9D
IC8/AAI
used
from
for karl
D273-10B
mediated
Lancashire
and IVIattoon (1979)
"cytodustiorl' + TF2
a_ his4
+
rho, rnit
to generate C45and 8-6
F o x (1979)
+
KL
14-4A
a hisl,trpl
+ rho, mit , cap r, olilr,par r
used rho-
used to generate shown in Fig la
rho-
1) used in crosses with 777-3A mit- strains to generate diploid mit- strains
H a i d et al.
(1979)
B. Weiss-Brummer et al.: Fine Structure of OXI l
77 The majority o f mutations fall in a group containing several oxi] mutations previously mapped. All o f these fulfill the criteria of OXI1 mutations as will be shown below (cf. Table 2). A second group comprises 19 m i t mutations which map together with a reference markers of the OXI2 locus. A single mutation (M5401) maps between these two groups; two others (M5391 and M6681) are located outside of the O X I I - O X I 2 span, between CAP and TSM-8. None o f these three mutations appears to affect the synthesis o f cytochrome oxidase (data not shown). The large set comprising the o x i l mutations has been further subdivided by use o f two r h o - clones (TFC45 and T F 8 - 6 ) which have been shown previously to reconstitute differentgroups o f o x i l mutations only ( F o x , 1979) and by use of four subclones which arose spontaneously in cultures of these two r h o - clones. As shown in Fig. l b , the OXI1 region is subdivided into six groups of mutational sites, A1, A2, B1, B2, C1 and C2.
Results The Genetic Map o f the OXI1 Region as Determined by r h o - deletion Analysis
A set of 110 mutations was allocated by preliminary experiments as described earlier (Schweyen et al., 1978) to the segment of the mitochondrial genome which is known to harbour the regions OXI1 and OXI2 (Slonimski and Tzagoloff, 1976). These m i t - mutants then were crossed with several r h o - clones which had been selected to retain one, two or more o f the genetic markers previously m a p p e d in this region (cap r, tsm-8, o x i l , oxi2; Slonimski and Tzagoloff, 1976; Monnerot et al., 1977; Trembath et al., 1977). The results o f these crosses are represented schematically by Fig. la. cap
~,5~9~
M5587
rsu-8
OXI I
Ms~ol 0)(12
Recombinant Frequencies in m i t - x m i t - Crosses
92 OX/ ! mutont5
All o x i l mutants were crossed to a collection o f mutants from this and from other regions. This collection contained several mutants from each of the subgroups A1 to C2 and the neighbouring marker tsm-8 as well as an oxi2 marker. By this test two mutations, M6771 and M8-171,
19 O X I 2 mulont5
0Xl
1
'',
C&5 8-6 $1 p1 R4 R5 M2"71 M2481 N4971 N5411 M5441 MSL81 M5501 M5611 M5531 M 5641
N56fA M5671 M5701 M5721 M57L1 N5751 M5761 M7662
1,44234 M,'-526 N4246 Nl't533 M4364 M45&7 M4408 N4551 M4449 M45T3 M4462 M5642 M4/-,81 M5711 N4510
N 8 - 171
M1237 M2511 M5301 M5322 M5371 M5451 N 5561 N 5821
M&611 M5351 M5541 M5681 M9-69 M9-94 M13-249
/
M224.~ M5303 M5392 M5553 M5591 MS59& N5651 M 56-Crl
M5781 M5791 MSB&I M5851 M5861 M5881 M6/,01
M1524N5421N5731 M2012MS431M5782 M2121MS461M5801 M23&1M5511M5811 M4360M5531MS831 M/.636H5551H5842 N4983M5571M 5671 M5311 M5581M 12-246 M5341 N 5 5 ~ M5381 M 5621
M6771
Fig. 1 a and b. Mapping of m i t - mutations in the CAP-OXII-OXI2 segment of the mitochondrial genome, a A series of rho- clones was selected which retain various combinations of the mutational sites capr, oxi1-M9-94, oxil-M8-171, oxi2-M9-3 (Trembath et al., 1977) and tsm-8 (Monnerot et al., 1977). The series of newly isolated mit- mutants then was crossed with these rho- clones in order to determine retention of the respective sites in the rho- genomes. The new mutational sites were allocated to the known regions OXI1 and OXI2 or located outside (M5401; M6681, M5391) depending on co-retention with or separation from the known sites, b Fine structure mapping of mutations allocated to OXI1 was obtained by analyzing retention/deletion of mutational sites in rho- clones TFC45, TF8-6 (Fox, 1979) and four rho- subclones derived from them. Retention of mutational sites in rho- clones is represented by horizontal bars. The open box in the lower part of Fig. lb indicates deletion of all oxil mutational sites by mutations M8-171 and M6771
78
B. Weiss-Brummer et al.: Fine Structure of OXI 1
were found to be large deletions: recombinants were detected only in pairwise crosses with mutants mapping outside of the 0 X l l region. The conclusion that the deletions cover all oxil mutational sites is in agreement with findings of Trembath et al. (1977) concerning M 8 - 1 7 1 , and is consistent with the r h o - deletion mapping. M 8 - 1 7 1 recently was shown to lack a restriction fragment of 2,400 base pairs (Fox, 1979). In Fig. lb the deletions are represented by long open boxes. In order to obtain a quantitative determination of recombinant frequencies, five m i t - mutants were crossed with a series of oxil mutants mapping in the six groups
A1 to C2. Frequencies of mit + recombinants observed range between 0% and more than 10%. These latter values signify absence of genetic linkage; they are dlose to the maximal values expected for each of the two possible recombinant classes (mit +, which only can be scored, and mit~- mitT) in the absence of polarity o f recombination (Dujon et al., 1974). Recombinant frequencies observed in five series of crosses are given in Fig. 2. The results confirm the order of mutational sites as determined by r h o - deletion mapping in that recombinant frequencies increase in the order of groups o f mutational sites A1 to C2. Recombinant frequencies between the most distant oxil muta-
Table 2. Synopsis of mutant phenotypes Mutant
Map pos~tSon
Cytochrome a'a 3 ( 6 0 3 nm)
Growth on ethano~
oxidase II
subunit III
M r of "new" polypeptide
OXII M5761
AI
leaky
(+)
+
10 5 0 0
M5661
AI
leaky
(+)
+
10 5 0 0
M5701
AI
leaky
(+)
+
10 5 0 0
M5631
AI
leaky
(+)
+
12 5 0 0
M5411
AI
-
+
12 5 0 0
M5721
AI
leaky
(+)
+
12 5 0 0
M4364
A2
-
-
+
12 5 0 0
M4575
A2
-
+
12 5 0 0
M2511
BI
+
15 0 0 0
M5821
BI
+
17 0 0 0
M5301
BI
+
33 000
M4611
B2
+
17 0 0 0
M5681
B2
+
18 0 0 0
M5351
B2
+
32 0 0 0
M5861
CI
+
27 0 0 0
M5341
C2
+
31 5 0 0
M5671
AI
599.5
n m 1)
-
+
M5661
AI
600
n m I)
-
+
M5451
BI
601
n m I)
-
+
M5431
C2
601.5
n m I)
-
+
o t h e r s 2)
AI t o C 2
-
-
+
others
AI t o C 2
~
+
+
M5491
-
+
-
M5321
-
+
-
15 5 0 0
M7582
-
+
-
22 7 0 0
3)
OXI2
others
4)
-
_
13 5 0 0
+
1) Absorption spectra were recorded from whole yeast cells in liquid nitrogen. 2-4) A total number of 53, 19, and 16 mutants, respectively. Mr values of SU II and SU III of cytochrome c oxidase are 33,500 and 23,000, respectively. (+): reduced amounts of SU II of Mr 33,500 present.
79
B. Weiss-Brummer et al.: Fine Structure of OXI1 OXI 1 CA P
-_
M 5391 M 5681
/ ~ /
TSM-8
~
M5561
I
1
A,
ML354
M5631
I
I
M 1273 M2511
I
A2
I
I
I
B1
i
M4611
I
I
M5681
=2
I
M5861
I
c,
M4638
I
I
I
M4350`
c2
I
OXI 2
1
1ZO 4 6
5,4
4,5
5,0
63
5,5
8,5
7,5
9,0 14,0
5O 0,1
- - 0 , 2
3,7
~2
2,9
18 3,3
3,4
20,0 3,5
16,9
2,0
0,2 -
-
- -
3,I
3,4
43
87
14,0
11,8 10,6
9,0
8,3
7,9'
11,7
5,5 3
lZ,,6
9,4
Z1
6,0
10,6
3,4
10,2
7,7"
6,0
.
3
"
4,3-
-
-
0,9 -
-
-
-
0,1
.15,4
.20.0
Fig. 2. Mapping by mit F x mitT recombination analysis. Frequencies of mit+ recombinants were determined in palrwise crosses of mit- mutants of opposite mating type as described previously (Hald et al., 1979). Strains of mating type a and carrying rnit- mutations were obtained by "cytoduction" in crosses 777-3,4 ~ mit- x 1C8 a karl rho °. A series of horizontal bars and figures at their ends refer to crosses of a given mit- mutant with a series of others and to the percentage of mit + recombinants observed. Groups A1 to C2 are defined by Figure lb; the order of mutational sites within A1, B1, B2 and C2 is ambiguous
tions involved are close to the maximal values expected although the length of this region is only about 1.000 base pairs (Fox submitted for publ.). Similarly high frequencies have been observed with mutations in the COBB O X region of mitDNA (Kotylak and Slonimski, 1977; Haid et al., 1979; Slonimski, Kotylak and Schweyen, in prep.). In a previous study of Trembath et al. (1977), however, recombinant frequencies in crosses between oxil mutants were found to be low (maximum of 1.6%) and not useful to order mutations in this region. Differences in the genetic background of strains used in different laboratories may account for these discrepancies in frequencies of recombinants.
Mutant Pheno types Cytochrome Spectra. Four mutants exhibit a shift in the absorption band of cytochrome a-a 3 (cf. Table 2 and Fig. 3). All the other mutants allocated to the regions OXI1 and OXI2 lack the absorption band of cytochrome a-a 3 ; other cytochromes appear not to be affected. Mitochrondrial Translation Products in oxil Mutants.
Consistent with previous studies of Cabral et al. (1978) three classes of mutant patterns of mitochondrial translation products can be recognized. (cf. Fig. 4, Table 2). i) SU II ofcytochrome oxidase (M r 33,500) is lacking; no further alterations in pattern of translation products are detected. This class comprises 53 out of 88 mutants, ii)
SU II is replaced by a single new polypeptide of Mr lower than 33,500. This phenotype is detected in 16 of the mutants studied, iii) Mutants exhibit a pattern similar to wild-type; 19 mutants show this phenotype. New polypeptides exhibited by mutants of class ii) vary in the range from 10.500 to 33.000 d. A correlation between these Mr values and map positions of respective mutations reveals a pattern of strict colinearity in the following order: 10.5 and 12.5 kd: A1; 12.5 kd: A2; 15 and 17 kd: B 1 ; 1 7 a n d 1 8 k d : B 2 ; 2 7 k d : C l ; 3 1 . 5 kd: C2. An obvious conclusion is that the series of polypeptides of 10.5 to 31.5 kd replacing SU II are generated by premature chain termination due to nonsense mutation or frame shift mutations followed by a nonsense codon at a short distance. According to this interpretation mutations in A1 to C2 in the OXI1 region are within the coding sequence of the same polypeptide which most likely is SU II of cytochrome oxidase. Only two mutations generating new polypeptides of M r closest to wild-type subunit II (M5301, 33 kd; M 5 3 5 1 , 3 2 kd) do not fit into this pattern of colinearity. Alterations in charge and/or conformation of polypeptides generated by mis-sense mutation may account for the slightly lower M r . (This interpretation also can explain modified electrophoretic mobility of revertant SU II as will be discussed below.) Immunological Identification o f N e w Polypeptides Replacing S U I1. More direct information on the relation-
ship between mutant and wild-type mitochondrial translation products can be obtained by an immunological ap-
80
B. Weiss-Brummer et aL: Fine Structure of OXI1
r
i 0
520
ii
I ,' rh
5C0
56O
i
I,:
I
580
600
820
Wavelength (rim)
Fig. 3. Low temperature absorption spectra of wild-type and mutant M 5671. Absolute spectra were recorded as described by-Haid et al. (1979). Absorption maxima of cytochrome a.a 3 (wild-type: 603 nm; mutant M 5671:599.5 nm), cytochrome b (558.5 rim), cytochrome c 1, (553 nm) and cytochrome c (547 rim) are indicated Fig. 4. Mitochondrial translation products of wild-type and oxil mutants. Cells were labelled with 3 S s o 2 - in the presence of cycloheximide (cf. Material and Methods). Separation of translation products on 10 to 15% SDS polyacrylamide gradient gels was followed by radioantography. Arrows identify SU I, SU II and SU III of cytochrome c oxidase, apocytochrome b and "new" polypeptides detected in mutants. In case of mutants M5861 and M2511 "new" polypeptides are weakly labelled but clearly detectable on the original radioautographs. *Indicates labelling of cells in 0.3% glucose instead of galactose; this causes weak labelling of apocytochrome b
proach as applied by Cabral et al. (1978) in case o f two
oxil mutants, by Kreike et al. and Solioz and Schatz (both in press) in case of cob-box mutants. Therefore, labelled mitochondrial protein was immunoprecipitated by use o f antisera raised against SU II of cytochrome c oxidase. Electrophoresis followed by radioautography revealed that wild-type subunit II and m u t a n t polypeptides o f M r 17,000 to 33,000 were precipitated in significant amounts (cf. Fig. 5). In contrast, with m u t a n t polypeptides of M r 10.500 to 15.000 no positive response was detectable (not shown). Ai_i:i.... ~ of the immuneoverlay technique gave the same resuhs 0~ot shown). This clearly relates the series of new polypeptides with M r values from 17,000 to 33,000 to subunit II of cytochrome c oxidase. It supports the conclusion that these polypeptides are fragments o f this subunit, generated by premature chain termination. Fragments smaller than 17,000 d apparently lack antigenic sites accessible to the antibodies applied. However, the fact that they form together with the larger, antigenic fragments a continuous series, colinear with map positions of mutations, strongly supports the idea that these also are fragments o f SU II.
Modified SU H in Revertants. Several respiratory competent (oxt~) revertants o f each one of the 16 mutants exhibiting "new" polypeptides were isolated. Revertants derived from 13 of these mutants showed patterns o f mitochondrial translation products indistinguishable from wild-type. This finding is consistent with the assumption that the short "new" polypeptides typical for these mutants are generated by chain terminating codons and that reversion restores the wild-type codon or another legible codon leading to a SU II o f wild-type electrophoretic mobility. With the other three mutants, M2511, M5681, M5351, however, revertant SU II different in electrophoretic mobility from that o f wild-type was detected. M r values of these revertant polypeptides are variable even amongst revertants of the same oxil mutant; some are slightly higher, the majority lower in M r than wild-type SU II. Results obtained with revertants of m u t a n t M2511 are shown in Fig. 6. Similar observations were made by Cabral et al. (1978) with revertants of two oxil mutants (M 1 3 - 249, M 9 - 6 9 , for map positions cf. Fig. lb). Sequencing of mitDNA o f mutant M 1 3 - 2 4 9 and of some o f its revertants reveals that M 1 3 - 2 4 9 is a frame shift mutation
B. Weiss-Brummeret al.: Fine Structure of OXI1
8l and that reversion restores the proper reading frame. A series of mis-sense codons generated thereby causes alterations in charge of SU II which are correlated with modified Mr values in revertants (Fox, submitted for puN.). The assumption that sequence dependent secondary structures persist in SU II even in SDS and change its electrophoretic mobility could explain modified Mr values in two of the mutants (cf. Table 2) and in three series of revertants studied here. Mitochondrial Translation Products in oxi2 Mutants. All of the 19 mutants allocated to the OXI2 region were found to lack SU III ofcytochrome c oxidase. Other mitochondrial translation products appear to be present although amounts ofSU I of cytochrome c oxidase seem to be low in some mutants. There oxi2 mutants exhibit a new polypeptide of Mr lower than SU III (M5491, M5321, M7582, cf. Fig. 7 and Table 2). In order to demonstrate that new polypeptides are related to SU III, their immunoprecipitation was attempted by use of the antiserum raised against SU II which co-precipitates SU II1 in small amounts (cf. Fig. 5). With the two mutants tested (M5491 and M5321) significant amounts o f p o l y -
Fig. 5. Immunological identification of "new" polypeptides in oxil mutants. Cells were labelled in the presence of cycloheximide with 35SO2 - , mitochondrial particles were isolated and treated with antiserum against SU II of cytochrome c oxidase and with fixed cells of Staphylococcus aureus (cf. Material and Methods). Immunoprecipitates were analyzed on 12% SDS-polyacrylamide gels followed by radioautography. The figure combines radioautographs derived from two different electrophoretic experiments. The serum cross-reacts weakly with SU III of cytochrome oxidase
Fig. 6a and b. Mitochondrial translation products in revertants of oxil mutant M 2511. a Translation products are visualized as described in Fig. 4. b Polypeptides reacting with antiserum against SU II were identified on the gel slabs with the immune-overlay technique (cf. Material and Methdds) followed by radioautography. R1, R2, R3, R4 designate revertants of M 2511 (cf. Fig. 4), WT: wildtype
82 peptides of Mr 13.500 and 15,500, respectively, were found to be precipitated (not shown). Although conclusive evidence for the gene - product relationship in the 0)212 region will be provided only by more detailed studies, the results obtained show that mutations in this region specifically alter SU III o f cytochrome c oxidase. Most likely these mutations are in the structural gene.
Discussion Alterations in Mr of SU II ofcytochrome c oxidase observed with two oxil mutants were taken as first evidence for the localization of the sequence coding for SU II
Fig. 7. Mitochondrial translation products of oxi2 mutants. Labelled polypeptides were visualized as in Fig. 4. Arrows identify SU I, SU II and SU III of cytochrome c oxidase, apocytochrome b and "new" polypeptides detected in mutants
B. Weiss-Brummer et al.: Fine Structure of OXI1 within the OXI1 region by Cabral et al. (1978). In the present paper 16 further mutants are described which exhibit a new mitochondrial translation product replacing SU II. M r values of these mutant polypeptides vary in the range from 10,500 to 33,000. Those of M r 17,000 and higher are found to be immunologically related to wild-type SU II. This is consistent with the previous observation and strongly supports the conclusion of Cabral et al. (1978). New products of M r in the range from 10,500 to 15,000 are found not to react with antibodies directed against SU II. This lack of immunological response may reflect a loss of antigenic determinants due to shortening of SU II polypeptides to less than 50%. The alternative explanation that these short polypeptides are not related to SU II is most unlikely for the following reason: New polypeptides of Mr 10,500 to 15.000 and 17,000 to 31.500 form a unique series which correlates perfectly with the series of mutations on the map in the way that mutations mapping in A1, i.e. proximal to CAP (cf. Fig. 1), generate polypeptides of the smallest Mr observed and mutations mapping more distant to CAP, i.e. in A2 to C2, give rise to polypeptides of higher M r values up to 3 1 5 0 0 . This observation of colinearity strongly suggests that premature chain termination due to nonsense mutation causes the appearance of fragments of SU II. As alternative explanations for the variety of mutant polypeptides one might envisage proteolysis or post-transcriptional modification or alteration of electrophoretic mobility due to mis-sense mutation in the gene for SU 1I. None of these, however, readily accounts for the series of mutant polypeptides of at least seven groups of M r values between 1 0 5 0 0 and 31,500 or for their colinearity with mfitational sites on the genetic map. Minor modification of the electrophoretic mobility due to mis-sense mutation is a likely explanation only for the occurence of polypeptides with Mr values close to that of wild-type SU II in case of two mutants which do not fit in the general scheme of colinearity (M5301, M5351, cf. Table 2) and in case of three series of revertants (cf. Fig. 6). Indications for modified electrophoretic mobility of SU II in the presence of SDS due to alterations in charge of revertant polypeptides come from recent experiments (Fox, submitted for publ.). Based on the pattern of colinearity observed it is concluded that i) the structural gene for SU II of cytochrome c oxidase is in the OXI1 region and ii) the direction of translation is from the CAP proximal end (A 1) to the CAP distal end (C2). Whether all mutations mapping in A1 are located within the structural gene or in sequences regulating its expression cannot be decided. Consistent with the conclusion that 0 X l l contains only one structural gene are two further observations: i) Mutations generating one or the other of four different phenotypic classes observed (cf. Table 2) are scattered
B. Weiss-Brummer et al.: Fine Structure of OXI1
over all subgroups mapped in the OXI1 region, ii) Linkage in terms of recombinant frequencies is observed between and within all subgroups of oxil mutations and between mutations in A1 and the neighbouring marker tsm-8 (cf. Fig. 2). A comparison between the percentage of recombinants observed with two mutations and the length difference of SU II fragments generated by early chain termination gives an estimate of 1% recombinants for a distance of roughly 50 base pairs. Similar estimates were obtained with the more elaborate genetics of the COB-BOX region (Haid et al., 1979; Kotylak, Slonimski, Schweyen, in prep.). Application of this estimate to recombinant frequencies shown in Fig. 2 reveals that the OXI1 region is relatively short and that the coding sequence is not interrupted by long intervening sequences, i.e. exceeding several hundred base pairs, as found in the COB-Box region. It is remarkable that out of 88 oxil mutants 53 lack wild type SU II without exhibiting a detectable new polypeptide. This contrasts with a very low frequency of such mutants in the COB-BOX region (Haid et al., 1979) where, with few exceptions, mutations mapping in the coding regions generate a fragment of apocytochrome b. A possible explanation is that many SU II polypeptides which are terminated by nonsense mutation or modified by mis-sense mutation are not as firmly integrated in the mitochondrial membrane and/or are rapidly degraded. Therefore, they may not have been detectable in preparations mainly composed of submitochondrial particles as used in the present study. A similar explanation may apply to the puzzling observation that mitochondrial particles of four mutants lack detectable SU II whereas whole cells of these mutants exhibit the spectral absorption bands of cytochrome a.a 3, although shifted to some extent (cf. Table 2 and Fig. 3). Mahler et al. (1977) described one mutant mapping in the OXI2 region and having replaced SU III of cytochrome c oxidase by a slightly shorter polypeptide. In the study presented here 19 oxi2 mutants are described out of which three replace SU III (23,000 d) by a polypeptide of lower Mr. The other oxi2 mutants also lack SU III but show no detectable amounts ofnewlyarising mutant polypeptides. These data together with the observation of Mahler et al. (1977) are highly suggestive that SU Ill is encoded in the 0X12 region and that at least some of the oxi2 mutations, namely those generating a mutant polypeptide shorter than 23,000 d, are in the structural gene for SU III. Acknowledgement. We thank Dr. T. D. Fox and P. Schnittchen for giving valuable yeast strains and for communicating results prior to publication and Dr. A. Lewin for helpful discussion. The excellent technical assistance of Mrs. S. Pollinger is highly appreciated. This work was supported by the Deutsche Forschungsgemeinschaft.
83
References Cabral, F., Solioz, M., Deters, D., Rudin, Y., Schatz, G., Clavilier, L., Groudinski, O., Slonimski, P. P.: In: Mitochondria 1977; Genetics and Biogenesis of Mitochondria. Bandlow, W., Schweyen, R. J., Wolf, K., Kaudewitz, F. (eds.), pp. 4 0 1 414. Berlin: de Gruyter 1977 Cabral, F., Solioz, M., Rudin, Y., Schatz, S., Clavilier, L., Slonimski, P. P.: J. Biol. Chem. 253,297-304 (1978) Claisse, M., Spyridakis, A., Slonimski, P. P.: In: Mitochondria 1977; Genetics and Biogenesis of Mitochondria. Bandlow, W., Schweyen, R. J., Wolf, K., Kaudewitz, F. (eds.), pp. 337-344, Berlin: de Gruyter 1977 Claisse, M., Spyridakis, A., Wambier-Kluppel, M. L., Pajot, P., Slonirnski, P. P.: In: Biochemistry and Genetics of Yeast. Bacila, M., Horecker, B. L., Stoppani, A. K. M. (eds.), pp. 369-390. New York: Academic Press 1978 Douglas, M., Butow, R. A.: Proc. Nat. Acad. Sci. USA. 73, 1083-1086 (1976) Dujon, B., Slonimski, P. P., Weill, L.: Genetics 78, 415-437 (1974) Eccleshall, T. R., Needleman, R. B:, Storm, E. M., Buchferer, B., Marmur, J.: Nature 273, 67-70 (1978) Fox, T. D.: J. Mol. Biol. 130, 63-82 (1979) Grivell, L. A., Moorman, A. F. M.: In: Mitochondria 1977, Genetics and Biogenesis of Mitochondria. Bandlow, W., Schweyen, R. J., Wolf, K., Kaudewitz, F. (eds.), pp. 371-384. Berlin: de Gruyter 1977 Haid, A., Schweyen, R. J., Bechmann, H., Kaudewitz, F., Solioz, M., Schatz, G.: Eur. J. Biochem. 94,451-464 (1979) Kreike, J., Bechman, H., Van Hemert, F. J., Boer, P. H., Schweyen, R. J., Kaudewitz, F., Groot, G. S. P.: Eur. J. Biochem. (in press) Kotylak, Z., Slonimski, P. P.: In: Mitochondria 1977, Genetics and Biogenesis of Mitochondria. Bandlow, W., Schweyen, R. J., Wolf, K., Kaudewitz, F. (eds.), pp. 161-172. Berlin: de Gruyter 1977 Lancashire, W. E., Mattoon, J~R.: Mol. Gen. Genet. 170, 333344 (1979) Maccechini, M. L., Rudin, Y., Blobel, G., Schatz, G.: Proc. Nat. Acad. Sci. USA 76, 343-347 (1979) Mahler, H. R., Hanson, D., Miller, D., Bihnski, T., Ellis, D. M., Alexander, N. J., Perlman, P. S.: In: Mitochondfia 1977, Genetics and Biogenesis of Mitochondria. Bandlow, W., Schweyen, R. J., Wolf, K., Kaudewitz, F. (eds.), pp. 345-370. Berlin: de Gruyter 1977 Monnerot, M., Schweyen, R. J., Fukuhara, H.: Mol. Gen. Genet. 152, 307-309 (1977) Moorman, A. F. M., Van Ommen, G. J. B., Grivell, L. A.: Mol. Gen. Genet. 160, 13-24 (1978) Schweyen, R. J., Weiss-Brummer, B., Backhaus, B., Kaudewitz, F.: Mol. Gen. Genet. 159, 151-160 (1978) Showe, M., Isobe, E., Onorato, L.: J. Mol. Biol. 107, 55-69 (1976) Slonimski, P. P., Tzagoloff, A.: Eur. J. Biochem. 61, 27-41 (1976) Solioz, M., Schatz, G.: J. Biol. Chem. (in press) Trembath, K. M., Macino, G., Tzagoloff, A.: Mol. Gen. Genet. 158, 35 45 (1977)
Communicated by F. Kaudewitz Received August 20, 1979
