drial ultrastructure and reduced cytochrome oxidase activity. MATERIALS AND METHODS

Culture Conditions

Wild type stocks of both mating types from strain 137c of Chlamydomonas reinhardtii were used to derive dk mutants. Cells were grown in 300-ml shake cultures at 25"C in high salt (HS) medium (47) or high salt medium supplemented with 0.12% sodium acetate (anhydrous) (HSA) as follows: Phototrophic, HS medium under 15,000 Ix cool white fluorescent light bubbled with 5 % CO2 in air; Mixotrophic, HSA medium under ~15,000 lx cool white fluorescent light; Heterotrophic, HSA medium in darkness using flasks wrapped with black electrical tape. For growth of cells on Petri plates, these liquid media were solidified with 1.5% or 4.0% agar (Difco Laboratories, Detroit, Mich. or Meer Corp., New York) and incubated either under ~6,000 Ix cool white fluorescent light or in darkness.
Nitrosoguanidine Mutagenesis
Wild type cells, pregrown to around 1 x 108 cells/ml, were collected in a tabletop centrifuge at room temperature and washed twice with citrate buffer (20.5 ml citric acid, 0.1 M, 29.5 ml sodium citrate, 0.1 M, 50 ml water, final pH 5.5), after which the cells were resuspended to around 2.5 x 10~ cells/ml. The cells were then distributed to two 125-ml Erlenmeyer flasks. To one flask, Nmethyl-N'-nitro-N-nitrosoguanidine (MNNG, Aldrich Chemical Co., Inc., Milwaukee, Wis.) in a sufficient amount of citrate buffer was added so that the final concentration of MNNG was either 5/~g/ml or 50 ttg/ml and the cell concentration was 1.0 10a cells/ml. To the other flask, an appropriate volume of additional citrate buffer was added to give 10 ml of ceils at 1.0 x 10~ cells/ ml. Both flasks were gently shaken in a 25~ water bath for 15 min in the dark. The flasks were then brought into the light, the cells transferred to centrifuge tubes, pelleted in a tabletop centrifuge, and washed twice with HS medium. The cells were then diluted to a density of 250 cells/ml for controls, 2,500 cells/ml if 50 ~.g/ml MNNG was used or 1,500 cells/ml if 5 /~g/ml MNNG was used. Aliquots (0.2 ml) of cells were then delivered to 10 plates of HS agar for controls, and to 50-100 plates for the experimentals. Cells were grown phototrophically and allowed to form colonies, after which they were replica-plated to score for the dk phenotype. In several mutagenesis experiments, cells were collected on Millipore filters (Millipore Corp., Bedford, Mass.) rather than by centrifugation, according to the method of Lee and Jones (27).

The dk Phenotype

Presumptive dk mutations were selected by replica-
WISEMAN El" AL. Nuclear Mutations Affecting Mitochondria in Chlamydomonas
plating clones derived from mutagenized cells to HS plates incubated in the light and to HSA plates incubated in the dark. After one week, presumptive dk mutants were scored as clones whose dark replicas showed no detectable growth, but whose light replicas had undergone many doublings and were wild type in appearance. Presumptive mutants were repeatedly retested by streaking suspensions of single cells onto HS plates placed in the light and HSA plates placed in the dark. After one week, these populations of single cells of each presumptive dk mutant were stringently scored under a dissecting microscope for complete lack of cell division on the darkincubated plates. Stable dk mutants were maintained under phototrophic conditions, i.e., in the absence of acetate, to reduce any selective advantages to wild type revertants.

Enzyme Assays

Mitochondrial enzyme activities were assayed spectrophotometrically on whole cell homogenates, using either a Zeiss DMR 21 split beam spectrophotometer (Carl Zeiss, West Germany) or an Aminco DW-2 spectrophotometer in the split beam mode (American Instrument Co., Silver Spring, Md.). Enzyme activities were determined at room temperature in crude cell homogenates since we were unable to isolate functional mitochondria from Chlamydornonas. Cells growing logarithmically at a density of around 1 10~ ceils/ml were collected, concentrated to a density of 1.0 x 107 cells/ml or 5.cells/ml by centrifugation at 12,000 g for 10 rain at 4"C, resuspended in 0.03 M phosphate buffer, pH 7.4, plus 0.1% bovine serum albumin (BSA), and broken twice in a French press (American Instrument Company, Silver Spring, Md.) at 5,000 pounds per square inch. Isocitrate lyase activity (EC 4.1.3.1) was determined by the method of M~iller et al. (32). The millimolar extinction coefficient of glyoxylate phenylhydrazone at 324 nm was determined empirically to be 16.6. The homogenates for assays of cytochrome oxidase (EC 1.9.3.1) were prepared in 0.03 M phosphate buffer, pH 7.4, plus 0.1% BSA. Cytochrome oxidase was assayed according to the method of Cooperstein and Lazarow (9) modified by the addition of 1% (final concentration) Triton X-100 (32). Under conditions where substrate was not limiting, the change in absorbance at 550 nm was not linear with time (45). Thus, all activity measurements were calculated from the change in absorbance during the first minute. Cytochrome oxidase activity measured in this fashion was completely inhibited by 1 mM cyanide. Specific activity was calculated using the miUimolar extinction coefficient of 19.6 of Yonetani (54). Antimycin- and rotenone-sensitive NADH-cytochrome c reductase (EC 1.6.2.1) activities were assayed by the method of Hatefi and Rieske (20). A medium containing 0.05 M Tris, pH 8.0, 0.67 M sucrose, and 1.0 mM histidine (20) was found to be essential to preserve the activity of NADH-cytochrome c reductase. Antimycin-sensitive activity was measured by the addition to the reaction mixture (3 ml final volume) of 0.1 ml antimycin solution to give a final concentration of 600 nM antimycin. The antimycin solution was made from a stock solution (10 mg/ml in ethanol) which was diluted to a concentration of 10 /zg/ml antimycin in an aqueous solution containing 0.1 M PO4, pH 8.0, and 0.1% BSA (20). Rotenone-sensitive activity was determined by the addition to the reaction mixture (3 ml final volume) of 0.1 ml rotenone solution to give a final concentration of 850 nM rotenone. The rotenone solution was made by diluting a stock solution (10 mg/ml in ethanol) 1:100 in water. 0.1 ml of a 0.1% ethanol solution was added to the reaction mixture as a control to determine the effect, if any, of the small amount of ethanol present in the antimycin or rotenone solutions on the enzyme activity. Antimycin and rotenone were pur-

Genetic Analysis

Standard techniques for crossing, tetrad analysis, and diploid formation were used (11, 18, 28). Gametes were differentiated overnight on N- agar medium in the light (HS without NH4CI). Zygotes were matured in the light on N- plates containing 4% agar and germination was induced by returning them to HS plates containing NH4CI. x This method gave a large increase in zygote viability for crosses involving dk mutants, compared to the usual dark maturation technique (28). AUelism was scored in pairwise crosses between dk mutants by determining the frequency of zygote clones which contained wild type recombinants. For several pairs of dk mutants, spontaneous diploids were selected by plating mating mixtures under restrictive conditions (HSA in the dark) in a fashion analogous to the methods of Harris et al. (19) and Wang et al. (49). Under these conditions, complementing diploids form colonies in 1-2 wk because they are able to grow in the dark on acetate, while dk gametes are unable to grow, and zygotes do not germinate.
[~4C]Acetate Assimilation
[~4C]Acetate assimilation was measured essentially by the method of Alexander et al. (1) except that triplicate Whatman no. 3 filters were spotted, soaked in 10% ice cold TCA for 3 min, then in three changes of 5% icecold TCA for 10 min. Filters were air dried and counted in a toluene scintillation fluid containing 4 g/l 2,5-diphenyloxazole (Packard Instrument Co., Inc., Downers Grove, I11.) and 0.1 g/1 1,4-bis[2-(5-phenyloxazolyl)]benzene (Packard Instrument Co., Inc.).

Whole Cell Respiration

The total and cyanide-sensitive respiration rates of whole cells were measured according to Alexander et al. (1). I K. Van Winkle-Swift, personal communication.
THE JOURNAL OF CELL BIOLOGY" VOLUME 73, 1977
chased from Sigma Chemical Co. (St. Louis, Mo.). For protein determinations, homogenates were routinely treated with 80% acetone in water for 20 min to remove chlorophylls, centrifuged, and the pellet was assayed for total protein by the method of Lowry et al. (29).

Electron Microscopy

Samples were prepared for electron microscopy according to Harris et al. (19), except that the fixation buffer contained HS instead of HSA growth medium. Diaminobenzidine (DAB) was used as a histochemical stain for cytochrome oxidase (24, 42). After samples of each mutant were fixed with glutaraldehyde and washed with phosphate buffer according to the methods of Harris et al. (19), they were resuspended in 4.5 ml of phosphate buffer, 0.1 M, pH 7.4, and 4.5 ml of culture medium, and treated with the followingreagents at room temperature for 15 min: DAB, 5 mg (Sigma Chemical Co.), catalase, 1 ml of 20/zg/ml (Worthington Biochemicals Corp., Freehold, N. J.), and H202, 0.03 ml of a 30% solution. The samples were then processed as previously described (8, 19) except that staining with uranyl acetate and lead citrate was omitted. The DAB staining reaction was completely inhibited in all samples which showed activity by a 10-rain pretreatment of the cells with 10 mM cyanide. The staining of mitochondrial cristae by DAB, expressed as the distance of stained cristae per mitochondrial profile or per mitochondrial area, was quantitated on a Hewlett-Packard Model 20 (Hewlett-Packard Co., Palo Alto, Calif.) equipped with a Hewlett-Packard digitizer. Between 29 and 119 mitochondrial profiles per genotype representing between 2.4 and 7.8 p.m2 mitochondrial area were measured on electron micrographs at a magnification of 60,000. RESULTS

Genetic Analysis o f Mendelian Mutants
Nine mendelian dk mutants were selected for
Wild type cells in HS liquid phototrophic culture Mutagenesis with nitrosoguanidine Plate 300 cells per plate on HS in light Repli!a plate
Medium and culture conditions HS (light) HSA (dark)

Growth response + +

Obligate photoautotrophs ( dk mutants) select stable mutants
Medium and culture conditions HS (light) HS + fluoroacetate (light) Test for sensitivity to fluoroacetate to eliminate permeability mutants: Growth response
Mutants able to Permeability and take up acetate glyoxylate cycle mutants ~.p attem of inheritance t in crosses to wild type: Genetic analysis
2dk + : 2dk mendelian mutants 1. Additional genetic studies: Test for allelism and complementation. 2. Additional biochemical studies: Measure citric acid cycle and electron transport enzyme activities; cyanide-, antimycin-, and rotenone-sensitive respiration, and uptake of [14C]acetate. 3. Electron microscope studies: Investigate whole cell and mitochondrial ultrastructure and test for an active cytochrome oxidase by using diaminobenzidine histochemical stain.
FIGURE 1 Protocol for the selection and analysis of obligate photoautotrophic (dark dier: dk) mutants. Wild type (dk +) cells were pregrown phototrophically in liquid to a density of around 1 10~ cells/ml.
Others nonmendelian mutants?
1. Additional genetic studies: Do backcrosses and investigate somatic segregation during vegetative growth. 2. Additional biochemical and electron microscope studies: None were done as the mutants were found to be unstable.
further study because they were extremely stable and, in preliminary crosses to wild type, had shown clear-cut mendelian inheritance. Additional crosses to wild type, allele testing, and complementation analysis were performed with each of these nine mendelian d k mutants. When these mutants are crossed to wild type, the d k phenotype in all or a large majority of the tetrads segre-
gates in a mendelian fashion, as does a second mendelian marker, mating type (Table II). The occasional d k progeny recovered in crosses of these mutants as well as in crosses of the wild type controls result in part because there are a few dk cells in the wild type stock and in part because of the variable expression of the d k phenotype in occasional d k progeny (53).

TABLE I

Recovery and Preliminary Characterization of dk Mutants after Nitrosoguanidine Mutagenesis, Fluoroacetate Sensitivity, and Preliminary Genetic Analysis of Confirmed dk Mutants

Parameter Exp 1 Exp 2

Concentration of MNNG in mutagenesis, gg/ml Cells plated/plate No. of plates Viability, % Killing due to MNNG, % No. of presumptive dk mutants Frequency of presumptive dk mutants (% of total colonies analyzed) No. of confirmed stable dk mutants Frequency of stable dk mutants (% of total colonies analyzed) Total stable dk mutants isolated in eight exp: 62

Sensitivity to fluoroacetate No. %

52 113

13 0.2

Inheritance pattern

Sensitive Resistant Leaky resistant
Mendelian Nonmendelian Unclear Not analyzed
Wild type (dk -) cells were pregrown phototrophically in liquid to a density of I x 10e cells/mi, and treated with 5 /zg/ml nitrosoguanidine (MNNG) for 15 min at 250C, pH 5.5. Fluoroacetate sensitivity was measured by plating suspension of cells on HS agar medium containing 10 mM fluoroacetate whichis 10 x the concentration that kills wild type. The pattern of segregation ofdk mutants was determined by analyzing6-8 tetrads from each mutant crossed to wild type. To ascertain whether any of the nine mendelian dk mutants were allelic, we determined the frequency of recombinant zygote clones from crosses of dk mutants taken in pairwise combination (Table III). Each zygote clone which contained at least one wild type recombinant among the four meiotic products was scored as a recombinant clone, regardless of whether the parental genotypes were also present (as in recombinant clones arising from tetratype tetrads). Thus, the frequency of recombination could exceed 50% and in some cases approached 70%. Although these frequencies do not represent classical map distances, mutants with a frequency of recombinant clones between 50 and 70% are either unlinked or very distantly linked, whereas mutants with a frequency below 1-2% are tightly linked or allelic. The results presented in Table III indicate that dk32 and dk-34, with fewer than 1% recombinant clones, are tightly linked and may be allelic, while the remaining seven mutants are distantly linked or unlinked. Three of the pairwise combinations, dk-32 x dk-32, dk-32 dk-76, anddk-105 x dk105, yielded virtually no viable progeny even after repeated crosses with various subclones. In five additional combinations, dk-32 x dk-52, dk-32 x dk-80, dk-34 x dk-52, dk-34 x dk-80, and dk-52 x dk.llO, the initial cross yielded few or no viable zygotes, although subsequent crosses did produce more viable progeny. The differences in the frequencies of recombinant zygote clones noted for two of these crosses, dk-32 x dk-52 and dk-52 x dk-llO, probably result from large variations in zygote viability. In order to test whether these mutants at eight different gene loci affect the same or different functions, we constructed a complementation matrix by genetically synthesizing diploids containing pairwise combinations o f d k mutants, using a standard selection technique employing two tightly linked arginine auxotrophs (11). Those diploids containing two complementing dk mutations were detected by their restored wild type function, scored as the ability to grow heterotrophically. In addition, each mutant was tested for dominance by synthesizing diploids containing only one dk mutation and its wild type allele. The results of this analysis indicate that, with the exception of

WISEMAN ET AL. Nuclear Mutations A f f e c t i n g M i t o c h o n d r i a in C h l a m y d o m o n a s

TABLE V

Fluoroacetate Sensitivity, p4C]Acetate Assimilation, and lsocitrate Lyase Activities of Mendelian dk Mutants and Wild Type (dk +)
Sensitivity to 10 mM fluoroacetare

Genotype

Aoetate uptake

cpm/lO s cells per h

Isocitrate lyase activity
pmol glyoxflate producedlrag whole cell protein per rain
dk dk-32 dk-34 dk-52 dk-76 dk-80 dk-97 dk-105 dk-llO dk-148

17 119

11 - 22 - 11 - 5* 11 8* 2* - 9 --+ 5* - 19"
-+ 6 - 4 - 7* - 9* --+ - 3 - 14
Fluoroacetate sensitivity (+) was measured on cell suspensions plated on minimal media containing 10 mM fluoroacetate. The incorporation of [14C]acetate into material insoluble in cold TCA was measured as the increase in counts from h 1 to h 2 after addition of acetate at h 0 to phototrophic cultures. The data are corrected for background and are expressed as counts per minute per 103 cells. Isocitrate lyase was assayed on crude homogenates of cells grown to a density of around 1 x 100 cells/ml. Differences noted in [14C]acetate incorporation or isocitrate lyase activity between the mutants and wild type were analyzed by t-tests and the results denote the probabilities that the differences are due to chance. * P < 0.01. The [~4C]acetate and isocitrate lyase data are presented as a mean and a standard deviation. as the cells increase in n u m b e r from early log phase (105 eells/ml) through stationary phase (107 cells/ml) (Fig. 2). Furthermore, the relative proportion of the total rate that is cyanide-sensitive shifts from about 50% during log phase to about 20% at stationary phase. Because the observed rates of respiraton depend on the growth stage, comparisons between the mutants and wild type could best be made by measuring respiration at various times in a growing culture (Fig. 2). Four of the mutants, dk-80, dk-97, dk-llO, and dk-148, have very low cyanide-sensitive respiration rates over most of the growth curve. Concurrent with this, the total respiration rate in each of these mutants is also low. The mutants dk-32, dk-34, dk-76, and dk-105 have intermediate levels of cyanide-sensitive respi-
ration, while that of dk-52 is similar to wild type. The total respiratory rate of wild type and five dk mutants is inhibited - % by 1 mM SHAM, while the respiratory rate of the remaining four mutants, dk-32, dk-34, dk-97, and dk-llO, is inhibited ~ %. The total respiratory rate of wild type and each mutant is inhibited - % by 1 mM S H A M plus 1 m M cyanide. Although the concentration of S H A M used does not completely inhibit the remaining cyanide-insensitive respiration, the rates of cyanide-insensitive respiration in the mutants appear to be similar to that of wild type. Thus, each mutant is thought to possess a functional cyanide-insensitive pathway. Except for dk-148, the reduction in total respiration rate in each mutant is largely due to the reduction in cyanide-sensitive respiration. For these mutants, the level of respiratory substrates is probably not greatly reduced, but rather one or more components of the cyanide-sensitive branch of the respiratory chain are not functioning normally. The reduction in both cyanide-sensitive and total respiration rates observed in dk-148 could be due to either reduced levels of respiratory substrates or a defect which affects both branches of the respiratory chain. Although none of the mutants will grow in darkness, Fig. 2 demonstrates that each mutant grows at near wild type rates in the light under strictly photosynthetic conditions. In agreement with the respiration data (Fig. 2), cyanide-sensitive cytochrome oxidase activity is reduced or absent in each dk mutant when assayed in whole cell homogenates (Table VI). Concurrent with the assays of cytochrome oxidase activity in vitro, the histochemical stain D A B was employed to assess the activity of this enzyme in situ. D A B reacts with cytochrome oxidase (42) in glutaraldehyde-fixed cells in the presence of catalase, and a polymerized oxidation product of D A B is precipitated along the surfaces of the mitochondrial cristae as electron-dense deposits. In addition, D A B reacts with other intermediates involved in cellular oxidation-reduction reactions, resulting in electron-dense deposits on the outer mitochondrial membrane and on the chloroplast lamellae in Chlamydomonas. However, only the reaction with cytochrome oxidase is specifically inhibited by cyanide. Mitochondria from phototrophically grown wild type cells treated with D A B but not contrasted with uranyl acetate and lead citrate are seen in Fig. 3 a and b. The mitochondrial cristae "stain" intensely with DAB, whereas cells pretreated with cyanide before the D A B reaction

THE JOURNAL OF CELL BIOLOGY" VOLUME ,

wildlype 3O"

d_k- 7 s~ ,.o---o ,I0

9 s IO.i0 s

I50 dk- 34

6 ,b 2'o ~o 4'o ~o

-,0" Sl.,,~=~.o.o 107

"lOS h

o ,;G~o4b~'o

.,o, o J CI0, ,20 0

.iOs 50

u~ c" 0

d k - 80

3" ,,,0

.10' iOs

o. " -0"

,,,o"

0 ,d ~

l ' d ~

z'o Xo 4'o ~o

dk- 97

.IOs I50

dk-105 -107

0 m G) Cr

-Io"

2' I"

0 I0 50

"10

O"
d~,- I10 ~t -I0;' dk- 148

.o9o-o.d

o,-"

c." "

abgogogo
FIouaE 2 Total and cyanide-sensitive respiration of nine mendelian dk mutants and wild type. The total and cyanide-sensitive (&) respiration rates of nine mendelian dk mutants and wild type, expressed as ~.consumed per hour per 106 cells, were measured with the oxygen electrode. The growth curve ( 9 is expressed as the cell density in cells per milliliter at various times. These curves demonstrate that both the total and cyanide-sensitive rates of respiration can change during the growth of a culture and that the cyanide-sensitive respiration of many of the mutants is much lower than wild type during log growth.
WISEMAN ET AL. Nuclear Mutations Affecting Mitochondria in Chlamydomonas
TABLE VI Comparison of Cytochrome Oxidase and Antimycin- or Rotenone-Sensitive NADH-Cytochrome c Reductase Activities of Mendelian dk Mutants and Wild Type (dk +)
NADH-cytochromec reductase Rotenone-sensitive activity

Cytochrome oxidase

pJ'nol cytochrorae c oxidized~rain per rag whole cell protein

Total activity

Antimycin-sensitive activity
i.u,nol cytochrorae c reduced/rain per rag whole cell protein x 10 -2
0.35 - 0.03 0.01 --- 0.01" 0.01 0.01" 0.10 - 0.02* 0.09 -+ 0.03* 0.01 --- 0,01" 0.01 0.01" 0.21 0.06* 0.002 - 0.002* 0.10 0.04*

T H E JOURNAL OF CELL BIOLOGY" VOLUME ,
FIGURE 4 Ultrastructure of phototrophicaUy grown wild type and dk-llO. (a) Median section through a phototrophicaUy grown wild type rot- cell showing the central nucleus (N) and nucleolus (Nu), mitochondrion (M), Golgi apparatus (G), and the peripheral chloroplast (C) containing a pyrenoid (P). x 18,000. (b) Higher magnification showing a wild type mitochondrion with well-defined cristae and a uniformly staining matrix, x 60,000. (c) Median section through a phototrophically grown dk-llO mt+ cell. Bar, 1 p.m. x 18,000. (d) Higher magnification showing mitochondria with normal morphology. Bar, 0.1 p.m. 60,000.
FIGURE 5 Ultrastructure of phototrophically grown dk-32 and dk-34. (a and c) Median sections through phototrophically grown dk-32 mt and dk-34 mt +. Bar, 1 /zm. x 18,000. (b and d) Higher magnifications of mitochondria showing the grossly disorganized cristae in each mutant and the ultrastructural similarities between the two mutants. Bar, 0.1 /xm. 60,000. 70
FIGURE 6 Ultrastructure of phototrophically grown dk-80 and dk-97. (a) Median section through a phototrophicaily grown dk-80 rot- cell showing a large peripheral vacuole (V) or swollen endoplasmic reticulum, x 18,000. (b) A higher magnification showing distorted saclike cristae, x 60,000. (c) Median section through a phototrophically grown dk-97 mr- cell. Bar, 1 v.m. x 18,000. (d and e) Higher magnifications showing small but normal mitochondria. Bar, 0.1 ttm. x 60,000.
Because each mutant grows at near wild type rates in the light, a reduction in the activity of the Krebs cycle, the major source of respiratory substrates, seems unlikely. Because of the central role of the Krebs cycle in intermediary metabolism, any significant reduction in its activity should affect not only cyanide-sensitive respiration and acetate assimilation, but phototrophic growth rates as well. Each mutant appears to have a normal cyanideinsensitive pathway, based on the sensitivity of its respiration to SHAM alone or in combination with cyanide. The cyanide-insensitive pathway in dk mutants with significantly reduced cyanide-sensitive respiration may provide the essential function of oxidizing respiratory substrates, especially N A D H produced via the Krebs cycle. In the absence of the cyanide-insensitive pathway, mutants with defective respiratory chains and significantly reduced respiration rates might be lethal because oxidized respiratory substrates which could accept electrons from Krebs cycle intermediates would rapidly become depleted. To place the nine mendelian dk mutants in perspective with wild type and with one another, we have summarized in Table VIII all of the basic biochemical and uitrastructural findings of the investigation reported here. This summary demonstrates that dk-32 and dk-34 are biochemically and ultrastructurally similar and supports the conclusion, based on genetic evidence, that these two

mutants are alleles. Differences in both biochemical and ultrastructural parameters for each of the remaining seven dk mutants demonstrate that each possesses a unique phenotype and support the genetic observations that these mutants are nonallelic and affect separate functions. This investigation demonstrates that eight of nine mendelian dk mutants have diminished cytochrome oxidase activity, and three of these also have reduced antimycin- or rotenone-sensitive NADH-cytochrome c reductase activity. Seven of the eight dk mutants with low levels of cytochrome oxidase activity both in vitro and in situ also have greatly reduced cyanide-sensitive respiration rates. These findings suggest that in many of the mutants the reduced respiratory activity results from defective respiratory chains incapable of oxidizing respiratory substrates at normal rates. In addition, four mutants possess defects in mitochondrial ultrastructure including grossly disorganized mitochondrial cristae and/or the precipitation of matrix material along the surfaces of the cristae. These observations suggest that the primary genetic lesion in many of the dk mutants is not in a gene coding for specific cytochrome oxidase peptides but rather in a gene that indirectly reduces cytochrome oxidase activity either by limiting its capacity to integrate functionally into the mitochondrial membrane or by altering the machinery

TABLE VIII

Summary o f Biochemical and Ultrastructural Properties of Nine Mendelian dk Mutants and Wild Type (dk +)
Antimycin- or rotenone sensitive Fluoroacetate sensitivity

yes yes

dk + dk-32

Acetate incorporation

lsocitrate lyase activity
Cyanidesensitive rcspiration
Cytochrome oxidase activity
NADHcytochrome c reductase
Mitochondria Whole cell ultrastructure (except mitocbondria)

normal normal

UItrastrocture

normal grossly altered

DAB staining
dk-34 dk-52 dk*76 dk-80 dk-97 dk-105 dk-llO dk-148
yes no yes yes yes yes yes yes

+ -~ + -+ ++ ++

+ + + + + + + +

+ -+ -+ -

_+ -+_+ -+

+ -+ + + + + +

normal reduced electron density of membranes normal swollen endoplasmic reticulum normal normal normal normal
grossly altered reduced electron density of membranes normal swollen cristae normal normal precipitated matrix material precipitated matrix material

+ -+ +

The biochemical and ultrastructural properties of nir~ mendelian dk mutants are summarized from Tables V to VII and Figs. 2 to 5, as well as unpublished observations. The symbol + , indicates >75 % of wild type levels of activity, the symbol -+, - - % of wild type levels of activity, and the symbol - , < % of wild type levels of activity.

J O U R N A L OF C E L L

BIOLOGY' VOLUME 73,
necessary for the transcription or translation of the enzyme or membrane peptides. On the basis of their pleiotropic phenotypes as summarized in Table VIII, each dk mutant with reduced cytochrome oxidase activity can be provisionally assigned to one of three possible phenotypic classes. Mutants in Class I may affect genes which code for components involved directly or indirectly in mitochondrial protein synthesis, including those genes which regulate the biosynthesis of enzymes such as cytochrome oxidase. Mutants in Class II may affect genes which code for nonenzymatic proteins of the inner mitochondrial membrane that are essential for the proper organization of the membrane as a whole or the proper integration into the membrane of other enzymatic proteins such as cytochrome oxidase or the ATP synthetase. Mutants in Class III may alter genes which code for peptides of the enzyme cytochrome oxidase. Mutants with altered mitochondrial protein synthesis, or mutants affected in membrane biogenesis, particularly the biogenesis of the inner mitochondrial membrane, would likely possess the broadest syndrome of defects in mitochondrial structure and function (Class I). dk-32, dk-34, and possibly dk-76 with grossly abnormal mitochondrial structure and/or function and reduced cytochrome oxidase activity most likely exemplify such mutants. The allelic mutants dk-32 and dk-34 possess grossly altered inner mitochondrial membranes which appear to be completely disorganized with the membrane components forming random sheets, threads, and amorphous clusters. That these two mutants have virtually no cytochrome oxidase activity and reduced antimycinand rotenone-sensitive NADH-cytochrome c reductase activities is not surprising. The mutant dk-76 has moderately reduced antimycin- and rotenone-sensitive NADH-cytochrome c reductase and cytochrome oxidase activities and cyanide-sensitive respiration but normal mitochondrial uitrastructure. This mutant is likely to be a leaky mutant of Class I that makes functional respiratory chains, but in reduced amounts, and assembles them into morphologically normal mitochondria. Mutants such as dk-52, dk-80, and dk-148 with structural and/or functional abnormalities of the mitochondrial inner membrane and reduced cytochrome oxidase activity may contain altered nonenzymatic membrane proteins which affect the assembly or function of cytochrome oxidase pep-
tides (Class II). The mitochondrial cristae of dk-80 are distorted and swollen, the mitochondrial matrix material is unevenly distributed along these cristae surfaces, and cytochrome oxidase activity is only 2-4% of the wild type level. The mutant dk-148, which also contains precipitated matrix material along the surfaces of the cristae, has about 30% of normal cytochrome oxidase activity, but very low total and cyanide-sensitive respiration. While the low respiration rate may be due to a lack of respiratory substrates, this is unlikely because the mutant has very high levels of acetate assimilation. More likely, dk-148 has a defect in its respiratory chain at a site that has not been studied, possibly cytochrome c or succinate dehydrogenase, and this may result in the observed precipitation of matrix material along the mitochondrial cristae. The mitochondrial membranes of dk-52 are stained very poorly by both uranyl acetate and lead citrate, which suggests the generalized loss of one or more membrane components. In addition, while the mutant's cytochrome oxidase activity in vitro and in situ is at most 30% of normal, its total and cyanide-sensitive respiration rates are as high as wild type's. The reason for this apparent anomaly is not clear but may be due to an uncoupling of respiration from oxidative phosphorylation which is known to increase the respiratory rate (30). The third class of mutants, those in which only cytochrome oxidase is affected, include dk- 97, dk105, and dk-llO. All three mutants have reduced cytochrome oxidase activity, which may account for their reduced rates of cyanide-sensitive respiration, but normal levels of antimycin- and rotenone-sensitive NADH-cytochrome c reductase. Although dk-llO has some precipitated matrix material along its cristae, the morphology of the mitochondrial cristae of these three mutants appears normal, especially when compared to the more extreme defects in mitochondrial ultrastructure in dk-32, dk-34, or dk-80. The simplest explanation of why the dk mutants die in the dark is that they produce insufficient ATP via mitochondrial oxidative phosphorylation for growth and survival, although we have not as yet demonstrated this directly. The nine mutants with reduced or very low cytochrome oxidase activities most likely produce less ATP, at least at the third coupling site. The three dk mutants which also have reduced antimycin- or rotenonesensitive NADH-cytochrome c reductase activities may also synthesize reduced amounts of ATP at

tion in b o t h C h l a m y d o m o n a s and yeast should clarify the role of b o t h nuclear a n d mitochondrial genetic systems in the biogenesis of the mitochondrion, particularly with respect to key c o m p o n e n t s of the respiratory chain such as cytochrome oxidase a n d the A T P a s e. We thank Ms. Sue Fox and Dr. Elizabeth H. Harris for their considerable help in preparing the manuscript for publication. This work was supported by a traineeship to A. Wiseman from Public Health Service Training Grant GM02007, by National Institutes of Health (NIH) Grant GM19427 to J. E. Boynton and N. W. Gillham, and by NIH Research Career Development Awards GM70453 to J. E. Boynton and GM70437 to N. W. Gillham. Received for publication 16 April 1976, and in revised form 22 November 1976. REFERENCES 1. ALEXANDER, N. J., N. W. GILLHAM, and J. E. BOYNTON. 1974. The mitochondrial genome of Chlamydomonas. Induction of minute colony mutations by acriflavin and their inheritance. Mol. Gen. Genet. 130:275-290. 2. ASHWELL, M., and T. W. WORK. 1970. The biogenesis of mitochondria, Annu. Rev. Biochem. 39:251-290. 3. BASTIA, D., K.-S. CHIANG, H. SWIFT, and P. SIERSMA. 1971. Heterogeneity, complexity and repetition of the chloroplast DNA of Chlamydomonas reinhardtii. Proc. Natl. Acad. Sci. U. S. A. 68:1157-1161. 4. BECK, J. C., J. H. PARKER, W. X. BALCAVAGE,and J. R. MArrOON. 1971. Mendelian genes affecting development and function of yeast mitochondria, In Autonomy and Biogenesis of Mitochondria and Chloroplasts. N. K. Boardman, A. W. Linnane, and R. M. Smillie, editors. North-Holland Publishing Co., Amsterdam. 194-204. 5. BORST, P. 1972. Mitochondrial nucleic acids. Annu. Rev. Biochem. 41:333-376. 6. BOYNTON, J. E., N. W. GILLHAM, and J. F. CHABOT. 1972. Chloroplast ribosome deficient mutants in the green alga Chlamydomonas reinhardi and the question of chloroplast ribosome function. J. Cell Sci. 10:267-305. 7. BRADY, R. O. 1955. Fluoroacetyl coenzyme A. J. Biol. Chem. 217:213-224. 8. CONDE, M. F., J. E. BOYNTON, N. W. GILLHAM, E. H. HARRIS, C. L. TINGLE, and W. L. WANG. 1975. Chloroplast genes in Chlamydomonas affecting organelle ribosomes. Genetic and biochemical analysis of antibiotic-resistant mutants at several gene loci. Mol. Gen. Genet. 140:183-220.
9. COOP~RSrEIN, S. J., and A. LAZAROW. 1951. A microspectrophotometric method for the determination of cytochrome oxidase. J. Biol. Chem. 189:665-670. 10. DROOP, M. R. 1974. Heterotrophy of carbon. In Algal Physiology and Biochemistry. W. D. P. Stewart, editor. Blackwell Scientific Publications, Ltd., Oxford. 530-559. 11. EBERSOLD,W. T. 1967. Chlamydomonas reinhardi: heterozygous diploid strains. Science (Wash. D. C. ). 157:447-449, 12. EaNER, E., L. MENNUCCI, and G. SCrtATZ. 1973. Mitochondrial assembly in respiration-deficient mutants of Saccharomyces cerevisiae. I. Effect of nuclear mutations on mitochondrial protein synthesis. J. Biol. Chem. 248:5360-5368. 13. EaNER, E., T. L. MASON, and G. SCHATZ. 1973. Mitochondrial assembly in respiration-deficient mutants of Saccharomyces cerevisiae. If. Effect of nuclear and extrachromosomal mutations on the formation of cytochrome c oxidase. J. Biol. Chem. 248:5369-5378. 14. EnNER, E., and G. SCHATZ. 1973. Mitochondrial assembly in respiration-deficient mutants of Saccharomyces cerevisiae. III. A nuclear mutant lacking mitochondrial adenosine triphosphatase. J. Biol, Chem. 241t:5379-5384. 15. EDWARDS, D. L., and F. KWIEONSlO. 1973. Altered mitochondrial respiratory chain in a chromosomal mutant of Neurospora crassa. J. Bacteriol. 116:610-618. 16. EDWARDS, D. L., and E. ROSENBERG. 1976. Regulation of cyanide-insensitive respiration in Neurospora. Eur. J. Biochem. 62:217-221. 17. ERNSTER, L., and B. KUYLENS~ERNA. 1970. Outer membrane of mitochondria. In Membranes of Mitochondria and Chloroplasts. E. Racker, editor. Van Nostrand Reinhold Co., New York. 172-212. 18. GILLHAM, N. W. 1963. Transmission and segregation of a non-chromosomal factor controlling streptomycin resistance in diploid Chlamydomonas. Nature (Lond.). 200:294. 19. HARRIS, E. H., J. E. BOVNTON, and N. W. GILLHAM. 1974. Chloroplast ribosome biogenesis in Chlamydomonas. Selection and characterization of mutants blocked in ribosome formation. J. Cell Biol. 63:160-179. 20. HATE~, Y., and J. S. RJESKE. 1967. The preparation and properties of DPNH-cytochrome c reductase (complex I-III of the respiratory chain), Methods Enzymol. 10:225-231. 21. HEBER, U., and G. H. KRAUSE. 1971. Transfer of carbon, phosphate energy and reducing equivalents across the chloroplast envelope. In Photosynthesis and Photorespiration. M. D. Hatch, C. B. Osmond, and R. O. Slatyer, editors. John Wiley & Sons, Inc., New York. 218-225. 22. HENRY, M.-F., and E.-J. NYNS. 1975. Cyanide-

23. 24.

insensitive respiration. An alternative mitochondrial pathway. Sub-Cell. Biochem. 4:1-65. IKUMA, H. 1972. Electron transport in plant respiration. Annu. Rev. Plant Physiol. 23:419-436. KEVHANI,E. 1972. Localisation cytochimique de la cytochrome oxidase chez le levure Candida utilis. J. Microsc. (Paris). 15:343-352. KOLAROV,J., J. SURIK, and L. KOVAC. 1972. Oxidative phosphorylation in yeast. IX. Modification of the mitochondrial adenine nucleotide translocation system in the oxidative phosphorylation deficient mutant opl. Biochim. Biophys. Acta. 267:465-478. KROON, A. M., and C. SACCONE, editors. 1974. In The Biogenesis of Mitochondria, Transcriptional, Translational, and Genetic Aspects. Academic Press, Inc., New York. 551 pp. LEE, R. W., and R. F. JONES. 1973. Induction of mendelian and nonmendelian streptomycin resistant mutants during the synchronous cell cycle of Chlamydomonas reinhardtii. Mol. Gen. Genet. 121:99108. LEVINE, R. P., and W. T. EBERSOLD. 1960. The genetics and cytology of Chlamydomonas. Annu. Rev. Microbiol. 14:197-216. LOWRy, O. H., N. J. ROSEBROUGH, A. L. FARR, and R. J. RANDALL. 1951. Protein measurement with the folin phenol reagent. J. Biol. Chem. 193:265-275. MAHLER, H. R., and E. H. CORDES. 1971. Biological Chemistry, Harper and Row Publishers, New York. 1009 pp. MORRISON, J. F., and R. A. PETERS, 1954. Biochemistry of fluoroacetate poisoning: The effect of fiuorocitrate on purified aconitase. Biochem. J. 58:473-479. MOLLER, M., J. F. HOGG, and C. DEDUVE. 1968. Distribution of tricarboxylic acid cycle enzymes and glyoxylate cycle enzymes between mitochondria and peroxisomes in Tetrahymena pyriformis. J. Biol. Chem, 243:5385-5395. MUNKRES, K. D., and F. M. RICHARDS. 1965. Genetic alteration of Neurospora malate dehydrogenase. Arch. Biochem. Biophys. 109:457-465. NEILSON, A. H., O. HOLM-HANSEN, and R. A. LEWlN. 1972. An obligately autotrophic mutant of Chlamydomonas dysosmos: a biochemical elucidation. J. Gen. Microbiol. 71:141-148. OGug, M., L. CORER, and S. OGUR. 1964. Glutamate auxotrophs in Saccharomyces. I. The biochemical lesion in the gltl mutants. Biochem. Biophys. Res. Commun. 14:193-197. RAVEN, J. A. 1974. Photosynthetic electron flow and photophosphorylation. In Physiology and Binchemistry. W. D. P. Stewart, editor. Blackwell Scientific Publications, Ltd., Oxford. 391-423. RICHTER, D. 1971. Production of mitochondrial peptide-chain elongation factors in yeast deficient in
mitochondrial deoxyribonucleic acid. Biochemistry. 10:4422-4425. RYAN, R. S., K.-S. CmANG, and H. Swift. 1974. Circular DNA from mitochondria of Chlamydomonas reinhardtii. J. Cell Biol. 63(2, Pt. 2): 293a. (Abstr.). SACCONE, C., and A. M. KROON, editors. 1976. In The Genetic Function of Mitochondrial DNA. North Holland Publishing Co., Amsterdam. 354 PP. SArZrARIOS,K. A., and U. HEBER. 1965. Changes in the intracellular levels of ATP, ADP, AMP and Pl and regulatory function of the adenylate system in leaf cells during photosynthesis. Biochim. Biophys. Acta. 102:39-54. SANTARIUS,K. A., U. HEBER, W. ULLRICH, and W. URBACH. 1964. Intracellular translocation of ATP, ADP and inorganic phosphate in leaf cells of Elodea densa in relation to photosynthesis. BIOchem. Biophys. Res. Commun. 15:139-146. SELICMAN, A. M., M. J. KARNOVSKY, H. L. WASSERKRUG, and J. S. HANKER. 1968. Nondroplet ultrastructural demonstration of cytochrome oxidase activity with a polymerizing osmiophilic reagent, diaminobenzidine (DAB). J. Cell Biol. 38:114, SHERMAN,F., J. W. STEWART,M. JACKSON,R. A. GILMORE, and J. H. PARKER. 1974. Mutants of yeast defective in iso-l-cytochrome c. Genetics 77:255-284. SHERMAN,F., J. W. STEWART,J. H. PARKER, E. INHABER, N. A. SHIPMAN, G. J. PUTI'ERMAN,R. L. GARDISKY, and E. MARGOLL~Sn. 1968. The mutational alteration of the primary structure of yeast iso-l-cytochrome c. J. Biol, Chem. 243:5446-5456. SMrrH, L. 1955. Spectrophotometric assay of cytochrome c oxidase. Methods Biochem. Anal. 2:427434. SOMLO, M. 1970. Respiration-linked ATP formation by an "oxidative phosphorylation mutant" of yeast. Arch. Biochem. Biophys. 136:122-133. SUEOKA,N. 1960. Mitotic replication of deoxyribonucleic acid in Chlamydomonas reinhardi. Proc. Natl. Acad. Sci. U. S. A. 46:83-91. TZAGOLOFF, A., A. ANAl, and R, B. NEEDLEMAN. 1975. Characterization of nuclear mutants of Saccharomyces cerevisiae with defects in mitochondrial ATPase and respiratory enzymes. J. Biol. Chem. 250:8228-8235. WANG, W.-Y., W. L. WANG,J. E. BOYNTON,and N. W. GILLHAM. 1974. Genetic control of chlorophyll biosynthesis in Chlamydomonas. Analysis of mutants at two loci mediating the conversion of protoporphyrin-IX to magnesium protoporphyrin. J. Cell Biol. 63:806-823. WELLS, R., and R. SAGER. 1971. Denaturation and the renaturation kinetics of chloroplast DNA from

THE JOURNAL OF CELL BIOLOGY. VOLUME 73, 1977
Chlamydomonas reinhardi. J. Mol. Biol. $8:611622. 51. WINTERSBF_.,RGER, 1965. Protein Synthese in isoE. lierten Hefe-mitochondrien. Biochem. Z. 341"409419. 52. WISTERSBEn~ER, E. 1970. DNA-dependent RNA polymerase from mitochondria of a cytoplasmic "petite" mutant of yeast. Biochem. Biophys. Res.
Commun. 40:1179-1184. 53. WISEMAN, A., N. W. GILLHAM, and J. E. BOYNTOW. 1977. The mitochonddal genome of Chlamydomonas. II. Genetic analysis of two nonMendelian obligate photoautotrophic mutants. Mol. Gen. Genet. 150:109-118. 54. YONETANI, T. 1967. Cytochrome oxidase: beef heart. Methods Enzymol. 10:332-335.
WlSEMAN ET AL. Nuclear Mutations Affecting Mitochondria in Chlamydomonas