‘Po .
dc"
”

1:“ 3 Ti
«:34, {.‘:£~‘\:x_ ' 9» _ I‘VE."
" 74:15:; v.{-"~'."‘ '5. . H -‘
‘ : " ‘ ‘ ' {7 °
‘ ‘ c

313‘ '0}? ‘ ‘.‘ .il. , .
«'1" ‘2' K h . ‘ ' ‘ vi V “'V'h
9"‘5'933‘ «m . ‘7 v.» 1%. «m
“‘3;
I

‘ 89;:-
“ . ' _ i . ‘ u~~‘_::..*‘\:“-I
”-4 *‘ g“ {4* , =3 3‘ %~",:~w“?:'z»t~?”
‘Y W 1‘ “ ‘ J. U. w
“:13.

_ ‘0‘.

1‘
v‘

r.
..

 

 

\4-\..

 

 

‘ v-uLPij-f' us
‘ V
«

 

’iL‘
yin"
.‘P .

.3:
J'H‘EW‘A’“
"$- ‘ "1‘
‘x‘r’n'f‘. '. ‘
\ '~.:;' ?‘-~"

1'.) A: .

n , ‘
“9.8'

A1.

I

f}, . -
V $1,333

’1'.“ '
....
f ‘ 3""

' .o'
1"! '
,.u‘, "u I,
.nu', (4"

CL;-
’t

R. 4'."

..~v-"I .- ‘fivlv .v "V

"- u n . ~-. ‘1..-

.n'v
A

~‘p
‘0'

‘-
-

A:

 

I);

 

 

 

 

 

 

 

 

 

 

 

 

___

 

H m m I"wt:nunufilm/IMMUNE
LIERAEZY w

J‘li ' ‘ \ 'l V/ ' i
II! n:
3 1293 00548 0334
mchigan State
University

 

 

 

This is to certify that the

dissertation entitled

Physiological and MoleCular Characterizations of
Plastids from a Plastome Mutator-Induced Mutant of
Oenothera hookeri str Iohansen

 

presented by
Ellen Margaret Johnson

has been accepted towards fulfillment
of the requirements for

Doctor of Philosorahy degree in Botany and Plant Pathology
7 and the
Program in Genetics

124mm 05-3mm.

Major professor

 

MS U is an Affirmative Action/Equal Opportunity Institution 0-12771

, var .'. .1.” _-

 

 

 

MSU RETURNING MATERIALS:
Place in book drop to
LIBRARJES remove this checkout from

lug—nan. your record. FINES will
be charged if book is

 

 

 

returned after the date
stamped below.

 

 

 

PHYSIOLOGICAL AND MOLECULAR CHARACTERIZATIONS OF
PLASTIDS FROM A PLAS’I‘OME MUTATOR-INDUCED MUTANT OF

W HQQKEBI STR JOHANSEN

By
Ellen Margaret Johnson

A DISSERTATION

Submitted to p
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY
Department of Botany and Plant Pathology
and

the Program in Genetics

1988

/' "

ABSTRACT

PHYSIOLOGICAL AND MOLECULAR CHARACTERIZATIONS OF
PLASTIDS FROM A PLASTOME MUTATOR-INDUCED MUTANT OF
QENSIEHEBA HQQKEBI STR J CHANSEN

By
Ellen Margaret Johnson

m1 is one of a number of chloroplast mutants derived from the
plastome mutator (nm) line of W, strain J ohansen.
This mutant appears chlorotic, and displays little or no internal
membrane development in the plastid. Biochemical analysis of m
has shown that the mutant has reduced but detectable amounts of
chlorophyll, and levels of carotenoids. fatty acids and lipids which are
within the range of those found in the wild-type. Immunoblotting
analysis has shown that the mutant membrane fraction contains a
number of gene products which are larger in size than the
corresponding mature proteins. Three proteins appear to be
improperly processed in M: cytochrome f and the 23-kD and 16-kD
subunits of the oxygen-evolving sub-complex of photosystem 11.
Additionally, the D1 subunit of photosystem II and a 28-kD
chlorophyll-binding protein from photosystem II were found to be
completely absent in the mutant plastids. These proteins may be

separated into catagories representing both chloroplast and nuclear

DNA gene products. The M mutation is inhertited in a non-
Mendelian fashion. thus restriction patterns of mutant and wild-type
chNAs. as well as the DNA sequence of the wild-type and p_n_i_7_ PETA
regions which encode cytochrome f. were examined and no
differences were found. Since the mutation can undergo reversion in
the homozygous pm nuclear background, these results suggest that a
plastid DNA-encoded function is required for appropriate processing

and subsequent chloroplast development.

To all the Johnsons and the Caspars.
in recognition of their encouragement
and support.

iv

ACKNOWLEDGMENTS

This project would not have been possible without the generous
contribution of antibodies from a large number of researchers: Terry
Bricker, Alice Barkan, Christer Jansson, Yossi Hirshberg. Charles
Yocum, Cathy Chia. Richard Malkin, Sylvia Darr, Dave Husic, Micha
Volakita. Nancy Artus, Jack Preiss. John Boynton, Ken Keegstra. Erin
Bell, Carol Schumann, Rick Vierstra and Tom Fox. Barbara Sears.
Hans Kende, Shauna Somerville, Tom Friedman and Charles Arntzen
have each served as valuable members of my guidance committee. I
would especially like to thank Barbara and Charlie and the other past
and present members of the Sears and Arntzen laboratories who
provided enjoyable and constuctive environments in which to work. I
would also like to acknowledge Karen Klomperans, Nancy Artus,
Ljerka Kunst. Mike Stine and Dan Keathley for helpful advise and
instruction in various experimental techniques and analysis. Finally.
Tim Caspar deserves my great respect and gratitude for his undying

support and encouragement for this endeavor.

TABLE OF CONTENTS

List of Tables ................................................ ix
List of Figures ............................................... x
Chapter
1 Introduction .................................... 1-5
Use of Chloroplast Mutants ..................... l
Plastome Mutator System of W .......... 2
Introduction to Thesis ......................... 5

2 Examination of Heteroplasmy among Wild-type and

Mutant chNAs from W Plastome Mutator
Lines .......................................... 6-37
Introduction ................................. 6
Materials and Methods ........................ 9
Plant Material. ......................... 9
Isolation of Chloroplasts and chNA. ...... 13

Restriction and Hybridization Analysis. . . . 14

Results ..................................... 19
Preliminary Analysis of Restriction Patterns
ofm pr_n1 chNA. ........................ 19
Analysis of Variable Regions in Plastome
Mutator Lines ......................... 24

vi

Discussion .................................. 32

Immunochemical and Molecular Analysis of the
Phenotypes of nail and a Putative Revertant of

m ........................................ 38- 105
Introduction ............................... 38
Materials and Methods ...................... 46

Plant Material ........................ 46
M13 Cloning and Dideoxy Sequencing

Strategy ............................. 46
Chloroplast Preparation and Fractionation of
Membrane / Soluble Proteins ............. 50

SDS-Polyacyrylamide Gel Electrophoresis 5 1
Preparation and Sources of Antibodies Specific

for Chloroplast Proteins ................ 5 1
Immunoblotting Analysis .............. 54
Results .................................... 55

Cloning of Fragments from the Bam 3a
Region of 321112 and Wild-type chNAs. . . . 55
Analysis of Mutant and Wild-type DNA

Sequences in the 9113 Region .......... 55
Analysis of Putative cyt f Precursor Specific
Antibodies ........................... 58
Analysis of Heme Staining Activity ....... 62
Immunoanalysis of Wild-type and Mutant
Chloroplast Proteins ................... 64
Immunoanalysis of a Putative Revertant of
21112 ................................ 73
Discussion ................................. 77

vii

4 Physiological Analysis: Structure. Compostion and
Activity of Mutant and Wild-type Plastids ....... 106- 135
Introduction .............................. 106
Materials and Methods ..................... 108
Plant Material ....................... 108
Photosynthetic Activity Measurements. . 108
Methods for Ultrastructural Analysis. . . . 109
Determination of Chlorophyll Content. . 1 10
Determination of Carotenoid Content. . 1 10
Determination of Fatty Acid Content. . . . l 1 1
Measurement of Lipid Synthesis ........ l 1 1
Results .................................. 1 13
Photosynthetic Electron Transport
Activity ........................... 113
Ultrastuctural Analysis ............... 1 13
Analysis of Chlorophyll and Carotenoid
Accumulation ........................ 1 16
Analysis of Fatty Acid Accumulation and
Lipid Synthesis ...................... 122
Discussion ................................ 129
5 Summary and Conclusions .................... 136- 141
Bibliography ..................................... 142-159

viii

Table

LIST OF TABLES

Sources of antibodies used in immunoanalysis of pgil

and wild-type chloroplast proteins ................... 53
Summary of immunological analysis of m plastid

proteins ......................................... 74
Photosynthetic electron transport activity in W
wild-type and 91111 mutant chloroplasts ............. 1 14

Results of statistical analysis using split plot analysis of
variance tests ............................... 1 19- 120

ix

LIST of FIGURES

Figure

10

11
12

13

14

15

Pedigree of plant lines .................................. 10

Restriction map of the variable Bam 12 region of the
Johansen strain of W W from lines D.
C1 and C2 ............................................ 16

Restriction map of the variable Bam 3b region of the
Johansen strain of 9mm 119915911 from lines D.
C1 and 02 ............................................ 17

Southern hybridization of 91111 and C1 wild-type chNAs
probed with total wild-type chNA. ...................... 20

Southern hybridization of p191 and C1 wild-type chNAs
probed with a spinach fragment containing the gem gene. . .21

Southern hybridization of nail and C1 wild-type chNAs

probed with a Bam 3a. ................................ 22
Southern hybridization of 121111 and C1 wild-type chNAs
probed with a Barn 3b .................................. 23
Physical map of mm mm plastome I ........... 24
Restriction digest of chNA from (green) wild-type

variants C1 and Cg .................................... 26

Pedigree of plant lines, indicating C1 and Oz classes of
plastome mutator line ................................. 28

Restriction digest and Southern hybridization with Bam 3b. 29

Southern hybridization of Hian-digested PILL].- 21.1.11- and
C2 wild-type chNA with Bam 3b and Bam12 probes ...... 31

Coomassie blue staining and immunoanalysis with cyt f
antibodies of mutant and wild-type chloroplast proteins. . . 39

Map of the Barn 3a region fi‘om the Johansen strain of
Qmotheza hogkeu chNA. ............................ 48

Strategr employed to sequence the 5' end and region
upstream of DEA from pm? and line C2 wild-type chNA. . 49

X

16

17

18

19

20

21

22

23

24

25

26

27

28

Partial sequence of m and upstream region of chNA
from W mam and the Johansen and Standard
strains of m 11532122121 ........................ 56-57

Immunoblot of wild-type and 11:91 total chloroplast
proteins probed with antibodies against synthetic cyt f
precursor-conjugate and cyt f ........................... 60

Heme-stained SDS-polyacrylamide gel (IO-15% acrylamide
yadient, 4 M urea) of mutant and wild-type chloroplast

proteins ............................................. 63
Coomassie-stained gel of membrane and soluble fractions of
mutant and wild-type chloroplast proteins ............... 65

Immunoblot of pm7 and wild-type membrane fractions
probed with antibodies against components of the cyt b6 / f
complex. ........................................... 66

Immunoblot of nail and wild-type membrane fractions
probed with antibodies against components of the DEC
subcomplex of PS 11 .................................. 68

Immunoblot of 121111 and wild-type membrane fractions
probed with antibodies against components of PS II and
PS I ................................................ 7O

Immunoblot of 9.1111 and wild-type membrane (A) or soluble
(B) fractions probed with antibodies against a ribosomal
protein and two stromal proteins ....................... 72

Immunoblot of 111111 and wild-type chloroplast proteins in
stable and unstable plastome mutator nuclear backgrounds
probed with antibodies against PS 11-22, cyt f and DEC-23. .75

Amino acid sequences of the N-terminal extensions of the
precursors of cyt c1, cyt f. and the components of the DEC
subcomplex of PS 11 ................................ 95-96

Hydrophilicity plots for the presequences of cyt f and the
components of the DEC subcomplex of PS 11 .......... 97-98

Model for processing of cyt f. DEC-23, and DEC-16 in the
chloroplast ......................................... 101

Hydrophilicity plots for the presequences of cyt f and
cyt c1 .............................................. 103

xi

29

3O

31

32

33

34

Electron micrographs of plastids from leaf-tip cultures of
mutant (A. B) and wild-type (wt. C) ..................... 1 15

Chlorophyll content of tissue containing wild- -type and
11191 plastome type I plastids in A/A angd A/C nuclear
backgrounds ....................................... 1 18

Total carotenoid content of tissue containing wild-type and
m plastome type I plastids in A/A and A/ C nuclear

backgrounds ........................................ 121

Major carotenoids in tissue containing wild-type and
121111 plastome type I plastids in A/A and A/C nuclear
backgrounds .................................... 123-124

Fatty acid content of tissue containing wild-type and
111111 plastome type I plastids in A/A and A/C nuclear
backgrounds .................................... 125-126

Lipid synthesis in tissue containing wild-type and

121111 plastome type I plastids in A/A and A/C nuclear
backgrounds ........................................ 128

xii

CHAPTER 1

Intoduction

W

The genome of higher plant chloroplasts encodes approximately 50
known genes for various soluble and membrane-bound chloroplast
proteins involved in photosynthesis, various components of the plastid
translational system, a complete set of tRNA genes and a number of yet
uncharacterized open reading frames (ORFs) (Parthier, 1982;
Herrmann et al., 1985; Rochaix. 1985; Shinozaki et al., 1986; Ohyama
et al., 1986). As a consequence of this. there are ample "targets" for.
mutagenesis of the plastid genome (or "plastome") . Mutation analysis
can be a valuable complement to standard biochemical and gene
cloning techniques in the analysis of structure and function of gene
products and chloroplast development. Thus. the ability to increase
mutation levels beyond the naturally low frequency of spontaneous
mutation could be advantageous because it would allow the isolation of
valuable plastome mutants.

A number of plastome mutants have been very useful in elucidating the

identities and functions of chloroplast gene products (Borner and

Sears. 1986). These mutants were recovered by isolating rare

spontaneous mutants or by screening chemically mutagenized

populations. Another alternative to these approaches may be found in
1

several higher plant isolates which have elevated rates of plastome
mutation, including mm W (Redei and Plurad, 1973).
maize (Shumway and Weier. 1967: Walbot and Coe, 1979: Thompson
et al., 1983), Petunia (Potrykus, 1973) and the evening primrose.

Qmszihm (Epp. 1973).

WW

Confirming the initial findings of Epp (1973). the plastome mutations
analyzed in our laboratory are inherited in a non-Mendelian fashion
and represent a variety of phenotypes. Difi‘erent mutants show varying
degrees of chlorosis (Epp, 1973), similar to the chloroplast mutator-
induced mutant of W W and in contrast to the mutants
induced by the 191311 type mutators. They also differ in the extent of
plastid development (Epp and Parthasarthy. 1987) and in the

accumulation of chloroplast proteins (Sears, personal communication).

The maize plastome mutants induced by the 1913,11 mutator system
show only a single phenotype. These mutants and similar mutants
induced by the barley W115 system (Bdmer et al., 1976) all
display a complete loss of chloroplast ribosomes. It has been proposed
that the nuclear gene product encoded by 121311 is essential for
assembly or function of chloroplast ribosomes, based upon the lack of
chloroplast polyribosomes, 168 and 23S chloroplast rRNAs and plastid
protein synthetic capacity (Walbot and Coe, 1979). In the absence of
chloroplast ribosomes, none of the chNA-encoded products can be

synthesized, including a number of chloroplast ribosomal proteins.

Since these proteins cannot be made, even if the plastids are removed
from the 19139, nuclear background. they will never be able to
regenerate a functional ribosome nor a functional chloroplast. Hence,
these mutator systems result in an irreversible and heritable alteration
of the chloroplast, presumably without afi‘ecting chNA.

In the plastome mutator system of WW str J ohansen,
plants homozygous for a recessive allele at the 12m locus show mutation
frequencies for plastome genes which are 1000-fold higher than
spontaneous levels (Epp, 1973). A large collection of plastome
mutator-induced chloroplast mutants has been generated by several
laboratories (Epp, 1973: Epp et al., 1987; Sears, 1983). All of the
mutants isolated thus far have been recognized initially by the
chlorotic sectors which appeared in plants homozygous for the 11:11
allele.

Plastome mutator activity appears to be independent of a variety of
growth conditions (Johnson et al., 1988). For example, a comparison
of field-grown and greenhouse-grown plants showed no differences in
the mutation frequencies of plants grown in these two environments.
Furthermore, when plant lines which are inactive for the plastome
mutator activity (pgfimi' or 1311+me are grown in shoot-tip
culture, they are extremely stable. In seven years of growing shoot-tip
cultures of m, we observed only one chlorotic sector in a
heterozygous plant line. In contrast. after about four months of
cultivation as shoot-tip cultures. homozygous m lines produce
mutations at a high frequency (Johnson et al., 1988). Although it is

not possible to accurately compare the frequency of appearance of
sectors in such clones to the frequency of appearance of sectors in
field-grown plants, the activity of plastome mutator is clearly not
dependent upon conditions exclusive to any one of these growth

regimes.

There are a number of possible mechanisms by which the plastome
mutator system could induce such mutations. Just as certain
components of the chloroplast translational apparatus are encoded by
nuclear genes, so are a number of components essential to chNA
replication and repair (Possingham and Lawrence, 1983). Preliminary
experiments designed to address this possibility and identify a
candidate for a pm gene product are currently underway in the Sears
laboratory. A purified product of the 11m gene could ultimately be used
to help clone the pm gene itself. If the pm gene encoded a defective
form of one of these elements, mutations could be introduced into
chNA. creating a plastome mutant. A point mutation in the chNA of
such a mutant could be the result of increased levels of errors in
chNA replication or defective chNA repair processes. Evidence of
these mutations may or may not be obvious at the level of chNA
restriction patterns. depending upon the nature and location of the
mutation event. An alternative mode of action of the plastome mutator
sytem involves the possibilty that the 11m gene product regulates the
activity of a chloroplast transposable element. Such a mechanism
would most likely be detectable as major alterations in chNA

restriction patterns.

W

The primary objective of the work presented in this thesis is to utilize
a number of difl'erent approaches to characterize one particular
plastome mutator-induced mutant of mm; mm. Initial
analyses explored the possibility that major alterations in the mutant
chNA pointed to the site of the primary lesion and were indicative of
the mechanism by which plastome mutator acts. Various aspects of
chloroplast protein accumulation and processing were then addressed
as well as the extent to which these proteins assemble into functional
complexes. The levels of other major components of mature
chloroplasts, such as chlorophyll, carotenoids. fatty acids and lipids
have also been determined as part of this characterization.

CHAPTER 2

Examination of Heteroplasmy among Wild-type and Mutant chNAs
from W Plastome Mutator Lines

Inimducflszn

The plastome mutator-induced mutants in our collection were initially
recognized as chlorotic sectors in plants which were homozygous for
pm, and hence displayed active plastome mutator function. Secondary
screening of these mutants was performed by Dr. Sears by analyzing
two characteristics of plastids isolated from pure mutant cultures. A
number of mutants showed variability in the restriction patterns of
their chNA in comparison to chNA isolated from descendants of the
wild-type, green plastome mutator line. These variable regions
containing the restriction fragment length polymorphisms (RFLPs)
were subsequently localized to the Earn 12 and Earn 3b fragments of
W chNA for two plastome mutator-induced mutants, 121118
and 1111111 (Kaplan, 1987). The 3-kilobasepair (kb) Earn 12 fragment
was shown to be located completely within the inverted repeat region
of the chNA molecule of 9mm 11.9.0311 (Kaplan, 1987; Johnson
et al., 1988).

A second screening procedure involved an immunological analysis of
chNA—encoded proteins present in the mutant plastids. This
examination revealed that one particular plastome mutator-induced

6

mutant accumulated a high molecular weight protein which was
antigenically related to cytochrome f (cyt t). Cytochrome f is a
thylakoid membrane protein encoded on chNA (Doherty and Gray,
1979; Alt et al., 1983; Willey et al., 1983: Phillips and Gray. 1984). It
is also known to be translated as a precursor with an amino-terminal
extension in a number of plants, including Mega (Alt and
Herrmann. 1984: Willey et al., 1984a. 1984b: Tyagi and Herrmann.
1986). Aberrant processing of the mutant cyt f may occur as a result
of a lesion in the gem gene which encodes this protein . Further
analysis of chloroplast proteins in this mutant. denoted 121311, is the
subject of the following chapter. However, this mutant was not
included in the initial RFLP screen and so it was necessary to purify
chNA from and tissue in order to compare the mutant and wild-type
chNArestriction patterns and look for evidence of major

rearrangements in 111111 chNA.

Lindenhahn et a1. (1985) have suggested that chlorotic sectors arising
in an 9339mm Mm line containing an active plastome mutator
background were the result of contaminating cross pollination and
subsequent hybrid plastome-genome incompatibility, where a given
Oenothera plastid type (I, II, III or IV) fails to develop normally in a
particular nuclear background (Stubbe, 1959). In response to this
assertion, Epp et a1. (1987) have recently shown that the chNAs
isolated from several different plastome mutator-induced sectors all
show the plastome I type restriction pattern (Gorden et al., 1982) and
are identical in restriction pattern to that the wild-type. Major
differences in chNA restriction patterns reported by Lindenhahn et

al. were probably, as they proposed, the result of contamination with
pollen with plastome 111 type chNA. In contrast, the results
described here show chNAs from another group of plastome
mutator-induced mutants and their wild-type analogs digested with
more frequently cutting enzymes than those used by Epp et al. This
analysis revealed the occurrence of RFLPs among individual plants all
of which were derived directly from Epp's original 12111112111 line.
However, the subsequent analyses and examination of the pedigree
presented in this chapter suggest that the presence of more than one
type of chNA, or heteroplasmy, occurred independently of plastome

mutator-induced mutagenesis.

Genes encoded on chNA appear to be unusually conserved (Curtis
and Clegg, 1984; Zurawski and Clegg. 1987). However the existence
of heteroplasmy in Oenothera suggests there may be a limitation to the
conserved nature of chNA. Other comparisons of the entire chNA
molecule between closely related plants have indicated that frequent
changes have occurred through insertion/ deletion events (Gorden et
al., 1982; Salts et al., 1984: Schmitz and Kowallik. 1986; Doebley et
al., 1987) and by rearrangements (Palmer, 1985). Similarly,
evolutionary comparisons of mitochondrial DNA (mtDNA) have
revealed many rearrangements (Sederofi', 1984); many mtDNA
polymorphisms within animal species can be explained by sequence

duplications.

Weds

Plant Material

The original 11mm isolate (E-15-7) was obtained after EMS (ethyl
methane sulfonate) mutagenesis of an inbred line of 99.353.111.223 hQchi’l
strain Johansen grown at Cornell University (Epp 1973). This species
has an A/A nuclear background and a type I plastome, according to the
classification scheme of Stubbe (1959). The original 11111412111 line was
inbred several generations to produce the plastome mutator seed
stocks maintained by Prof. W. Stubbe (University of Dusseldorf).

Figure 1 indicates the pedigree for the plastome mutator-induced
mutants (11ml, 121118,, 121119,, 1211111,. 9111.12. p193} and 111335) isolated
from progeny of further self-crosses. In 1982, green siblings of plants
containing p_n_1;32_ and p313; were crossed as the female parent with
pollen from W W strain Johansen carrying plastome 1V
in order to produce a 2111+me line containing wild-type green
plastids of plastome type I for use as a control. The progeny are
heterozygous at the pm locus and carry the A/ A nuclear composition
which is fully compatible with plastome I (Stubbe 1959). As described
further in the results, this wild-type line is classified as the C1 type,

after the Qomell origin of Epp's Mm line of Q. 1352913231, and the
chNA is referred to as type I-C1.

Another wild-type line (C2), carrying a second chloroplast type (I-C2)

was recovered from the 1980 field season. when plants containing the

10

Cornell line
Oenothera hookeri str Johansen
Epp's E15-7 plastome mutator

i

 

1976 1 978
field field

i i

i’ i

1980 1 980
field field

I l l I l
pm7 pm8 pm9 pm11 pm12 (29
+ i

CZ wt 1982
field

l | |

Clwt pm32 pm35

 

 

 

Figure 1. Pedigree of plant lines. Pedigree of plastome mutant and
wild-type lines descended from Epp's Cornell Wm m
m isolate E-15-7. The circled x's denote self-crosses. Line C2
wild-type (wt) was recovered by the selection and purification from
variegated 11mg seedlings of light green tissue containing wild-type
plastids, as described in the Materials and Methods.

11

newly arisen plastome mutant mg were crossed as the female parent
to Wm: (A/ C) plants with plastome IV. As described by
Cleland (1972), the $112325 genome is an alpha-complex (egg-
transmitted) while the was, genome is a beta-complex (pollen-
transmitted); therefore, the seeds resulting from this cross have the
genotype maxim/mm (A/C). In this particular cross, both the
newly arisen mutant plastids and green plastids were transmitted
through the egg, and plastome IV plastids were transmitted by the
pollen, producing progeny carrying different combinations of three
plastid types, as described previously by Stubbe (1957). As shown by
Stubbe (1959), plastome I fails to develop normally in this nuclear
background, resulting in a pale green leaf color. In the seedlings
produced by this cross. three types of tissue were observed: white,
light green and dark green. Respectively. these tissues contained the
m2 mutant plastome, wild-type plastome I transmitted from the
female parent, and wild-type plastome N from the male parent. The
light green line was selected and maintained in shoot-tip culture. as a
wild-type control for the appropriate plastome mutants from line C2.

Wild-type C 1 plants were grown in the greenhouse in sandy soil.
However, as described above, the C2 chNA was available only in an
A/C nuclear background. Since plants of plastome type I are not as
vigorous in combination with this nuclear background as they are in
their natural A/A nuclear background (Kutzelnigg and Stubbe 1974),
they were grown as shoot-tip cultures on a modified NT medium
(Nakata and Takabe 1971, Stubbe and Herrmann 1982) which
contained a ratio of 2:1 (w/w) 6-benzyl-amino purine (13 uM) to

12

naphthalene acetic acid (8 M). Cultures of pure mutant tissue were
maintained by transferring them to fresh media every four weeks.

13

Isolation of Chloroplasts and chNA

W leaf tissue (50-500 g) was homogenized in ice-cold buffer
containing 330 mM sorbitol, 6 mM EDTA, 1 mM ascorbic acid, 3 mM
cysteine, 0.15% (w/v) PVP (polyvinyl pyrrolidone. Mr=40,000), 0.1%
(w/v) BSA (bovine serum albumin) and 50 mM Tris, pH 7.5, (Sigma
Chemical Co.) as given by Herrmann (1982), with modifications
adapted from Somerville et a1. (1981), Loomis (1974) and Galliard

( 1974). All manipulations were performed at 4°C. Following filtration
through 100 um nylon mesh and two layers of Miracloth (Calbiochem)
and differential centrifugation at 6000 x g for 5 minutes, the crude
chloroplast preparation was washed in the homogenization buffer
lacking PVP. This preparation was further purified by layering on a 10
- 80% sucrose step gradient. buffered with 50 mM Tris-HCl, 6 mM
EDTA and 330 mM sorbitol, pH 7.5, which was then centrifuged at
22,500 x g for 10 minutes in a Sorvall $890 vertical rotor. Chloroplast
bands were removed from the gradient, diluted with 20 mM EDTA, 50
mM Tris, pH 7.5 and pelleted by centrifugation at 16,500 x g for 10
minutes. The chloroplast pellets were resuspended in equal volumes
of NET (5 mM NaCl. 100 mM EDTA. 50 mM Tris, pH8.5) for
subsequent lysis in 1% (w/v) sarkosyl and 1 unit/ ml pronase
(Calbiochem) with gentle agitation at 4°C for 4 hours. When only
limited quantities of plant material were available (line C2 wild-type
and some mutant lines), the sucrose gradient was omitted. In both
cases, CsCl was added to give a refractive index of approximately 1 .400
and the mixture was centrifuged for 45 minutes at 10,000 x g in a
Sorvall I-IB4 rotor. The chlorophyll/ protein pellicle which bands at

14

the top of the gradient was discarded while the clear supernatant was
processed further. Bisbenzimide (Hoeschst 33258) was added to a
final concentration of 0.01 mg/ml (after Hudspeth et a1. 1980) and the
refractive index was adjusted to l .3960. After buoyant density
equilibrium centrifugation, achieved in 10 -12 hours with a Sorvall
W865 vertical rotor at 42,000 rpm at 19° C., the upper (chNA) band
in the gradient was collected. If the lower band of nuclear DNA was so
broad that it contaminated the upper chNA band during its removal
from the gradient. the chNA band was further purified on a second
gradient under the same conditions. Bisbenzimide was removed by
three extractions with NaCl-saturated isopropanol, the remaining
chNA-CsCl mixture was diluted 10-fold with water, precipitated by
the addition of sodium acetate to 300 mM and two volumes of ethanol
or one volume of isopropanol, The resulting chNA pellet was washed
with 70% ethanol and dissolved in 10 mM Tris, 0.1 mM EDTA, pH
8.0.

Restriction and Hybridization Analysis

Restriction endonuclease digests were performed according to the
manufacturers' specifications with enzymes purchased from Bethesda
Research Laboratories, Boehringer Mannheim or New England Biolabs.
Fragments were separated by agarose gel electrophoresis and
visualized with ethidium bromide staining according to Maniatis et al.
(1982), using HindIII-digested lambda phage DNA fragments as size
markers. The Barn 3a fragment was located on the Mega, chNA
map using a heterologous probe from spinach containing the 1231A

15

gene (a gift from Dr. Warwick Bottomley, CSIRO, Canberra, Australia).
Fragments 3a and 3b were isolated from BamHI digests of mutant or
wild-type 95119113113 chNA by cutting the fragment out of a 0.6%
agarose gel, freezing in a dry ice-ethanol bath for 10 minutes, then
manually crushing using a sterile spatula. An equal volume of
redistilled phenol, pH 8.0 (Maniatis et al., 1982) was added to the
resulting slurry which was frozen for 10 minutes in dry ice-ethanol.
The aqueous phase was recovered from the mixture by centrifugation
(13,000 xg, 5 min.) and the phenol phase was re-extracted with an
equal volume of 10 mM Tris, 0.1 mM EDTA. pH 8.0 buffer. The two
aqueous phases were pooled and extracted three times with phenol,
once with phenol:chloroform:isoamyl alcohol (25:24:1) and three
times with chloroform:isoamyl alcohol (24:1). This was followed by
ethanol precipitation (Maniatis et al., 1982).

These fragments were cloned by conventional procedures (Maniatis et
a1. 1982) into the plasmid vectors pBR322 or pUC 18/ 19 (Norrander
et al., 1983) and nick-translated with a-32P-dATP or a-32P-dCTP
(800 Ci/mMol, New England Nuclear) (Rigby et a1. 1977) and used as
probes in Southern hybridization experiments. Figures 2 and 3 show
restriction maps which illustrate the additional W Barn 12
(Kaplan, 1987: Blasko et al., 1988) and Barn 3b (Pyung-Ok Lim,
personal communication) sub-fi'agments also used as probes against
Hinfl-digested chNA. Transfer of chNA from agarose gels to
nitrocellulose (Millipore) was performed as described by Maniatis et
al. (1982). The nitrocellulose membranes were hybridized, washed
and autoradiographed following procedures described by Southern

16

Bam 12 i

 

 

.-

 

 

 

 

A
B E 089 E (Hc) CB
l I ll — I
500bp
RHf HfAMT
“were . .Li B
D
01
02

Figure 2. Restriction map of the variable Barn 12 region of the
Johansen strain of Oenothera hogkeri from lines D, C1 and C2. The
restriction map of the entire 3-kb Barn 12 fragment is shown (A) as
well as a 50% enlargement of the a 1.6-kb portion of this fragment (B).
Both maps were generated by the work of Kaplan (1987) and Blasko et
a1. (1988) . The fragment outlined by a filled box contains the region
of variability (striped boxes in B) in the C1, C2 and D plastome mutator
lines. C1 and C2 are named for the Cornell University origin of the
plastome mutator line: D refers to a plastome mutator line maintained
at the University of Dusseldorf, which is also variable in the Barn 12
region (Kaplan, 1987). The variable regions from these three lines
have been further characterized by DNA sequencing (Blasko et al.,
1988). This has shown that the size variability in the Earn 12 fragment
is caused by the amplification (or elimination) of a 24-bp repeat
segment within a region homologous to ORF 1708 from N1 tabacum
(Shinozaki et al., 1986). B=BamI-II, E=EcoRI, C=Clal, Bg=BglII,
Hc=HincII, R=Rsal, Hf=Hian, M=MaeIII, T=Tan, Ec=EcoRV, A=AvaII.

l7

Barn 3b :

 

 

-—

1kb

i i WW1 iilg‘j‘iw‘i 39% i

CZ

Figure 3. Restriction map of the variable Barn 3b region of the
Johansen strain of Qengjhgra 119215.23 from lines D, Cl and C2, The
restriction map of the 8-kb Bam3b fragment is shown (Kaplan (1987)
and P. O. Lim (personal communication). Black and striped segments
denote the two regions of variabiltity in this fragment (Kaplan, 1987).
Clones containing these two fragments were used as probes in
Southern hybridization experiments. C 1 and C2 are named for the
Cornell University origin of the plastome mutator line: D refers to a
plastome mutator line maintained at the University of Dusseldorf.
which is also variable in the Barn 12 region (Kaplan, 1987). B=BamI-II,
A=Ach, N=NruI, E=EcoRI, Hi=HindIII, Hc=HincII, Sp=SphI, Bg=BglII,
S=SalI, X=Xbal.

 

 

18

(1975) and Maniatis et a1. (1982) with prehybridization and
hybridizations carried out at 65°C in the absence of formamide.

19

Results

Preliminary Analysis of Restriction Patterns of pm7 chNA

Initial comparisons of restriction digests of chNA isolated from PM
and our wild-type line of the Johansen strain of 9339111122 .IJQQISQQ
indicated that several restriction fragments varied in size (Figure 4).
These differences were seen in comparisons of digests using a number
of different restriction enzymes, including BamHI, EcoRI, HaeIII,
MspI and Tan. These differences are more easily seen when probes
for specific regions of chNA are used. Variable fragments which
hybridized to the 3-kb Bam12 fragment have been characterized by
Kaplan (1987) and Blasko et al. (1988). Because of the phenotypic
difference between wild-type and 11211 plastids with respect to cyt f
(discussed further in Chapter 3), the first region of chNA used as a
probe in this series of Southern hybridization experiments was a
spinach fragment containing the cyt f (1194A) gene (Figure 5), followed
by the homologous 8-kb ngmena fragment (Barn 3a) which contains
1211A (Figure 6). Examination of Figure 6 reveals that the Barn 3a
probe hybridized to variable fragments in BamI-II- and EcoRI-digested
chNAs, but not to variable fragments in other lanes showing chNAs
digested with HaeIII, MspI or Tan. Additionally, when the Barn 3a
fragments from mutant and wild-type chNA were subcloned into
plasmid vectors and cut with endonucleases, no difi‘erences were seen
in the restriction patterns of the cloned DNAs (data not shown).
Figure 7 shows the same restricted chNAs from wild-type and

20

Smylifidization
probe: chNA

3 E H M T S K
wt pm7wtpm7wtpm7mpm7wipm7wtpm7wfpm7 lib

- 9.4
_ 4.4

-2.3
—2.0

— 0.56

 

Figure 4. Southern hybridization of pm: and C1 wild-type chNAs
probed with total wild-type chNA. Pm7 and C1 wild-type (wt)
chNAs were digested with the enzymes indicated (B=BamHI,
E=EcoRI, H=HaeIII, M=MspI, T=Tan, S=SalI, K=KpnI), separated on
a 1.2% agarose gel, transferred to nitrocellulose and probed with nick-
translated total chNA from the C1 wild-type. The wild-type chNA
was subsequently identified as the I-C1 type. Fragments which vary
between the two CpDNAS are marked to the left of each pair of lanes
with a filled circle.

21

 

Figure 5. Southern hybridization of 11121 and C1 wild-type chNAs
probed with a spinach fragment containing the DSIA gene. Fragments
of variable size can be seen in the EcoRI lanes only. The same filter
shown in Figure 4 was washed and used for hybridization with nick-
translated cloned spinach fragment Bam7.

22

 

Figure 6. Southern hybridization of 931 and C1 wild-type chNAs
probed with a Barn 3a. Fragments of variable size can be seen in the
EcoRI lanes only. The same filter shown in Figure 4 was washed and
used for hybridization with nick-translated cloned Mega, wild-
type fragment Barn 3a, which contains the petA gene.

23

. 53%an Hybridization
‘ probe: Bom3b fragment

8 E H M T S K
wt m7wtpm7wtpm7wtpm7wtpm7 mpm7wtpm7 lib

— 0.56

 

Figure 7. Southern hybridization of 91:12. and C1 wild-type chNAs
probed with a Barn 3b. Fragments of variable size can be seen in the
EcoRI, HindIII, MspI, and Tan lanes. The same filter shown in Figure

4 was washed and used for hybridization with nick-translated cloned
mm wild-type fragnent Barn 3b.

24

mutant probed with an W Bam 3b fragment. In this case. the
same variable bands in the Baml-II, Mspl and EcoRI digests which
were hybridized by the Barn 3a probe are recognized by Barn 3b. In
addition, Barn 3b also recognizes bands in the HaeIII and Tan digests
which difi'er in size between the mutant and wild-type. In each case.
the mutant fragment is approximately 200-basepairs (bp) smaller than
the corresponding fragment in the wild-type. The implication of the
results of the hybridization analysis was that Bam 3a and 3b were
adjacent fragments: thus, the variable EcoRI fragment recognized by
both the 3a and the 3b probes spans the junction of the two Barn
fragments. By further use of Southern hybridization analysis, the Barn
3b fragment was localized to the large single copy region of the
mm mm chNA molecule, neighboring the Barn 3a fragment
(Figure 8). It appeared therefore that the region which defined the
deletion in 121111 chNA mapped to an area just upstream of the 9:45

gene.

Analysis of Variable Regions in Plastome Mutator Lines

Further comparisons of chNA from various plastome mutator-induced
mutants of plastome I indicated that the mutants all fall into two
general classes with respect to their chNA restriction patterns
(Figure 9). The two classes have been named C1 and C2 after the
Cornell University origin of WW, strain Johansen, Epp's
original mutator line (Epp, 1973); the plastome I chNAs from each
of these lines are called I-C1 and I-C2 respectively. Of the seven
mutants which have been carefully examined, three belong to the I-C1

Barn 3b
psb9 9390 ath

psaA1 Ag,

m \PetA
atpE/ ... ..
«:9 ‘33- prB-

ath’ ~ :33“ ‘ \PetBD
rbcL l
psbAl

 

 

 

Bam12 \ / Bam12
rDNA rDNA

Figure 8. Physical map of W hggkgri plastome I. Adapted from
Gordon et al. (1982), the map shows the location of the two variable
regions (Barn fragments 3b and 12) and the fragment containig the
gem gene (Barn fragment 3a) in classes C 1 and C2 of our plastome
mutator line of Mega 11m strain J ohansen. Inverted repeat
regions are outlined by black arcs. Region of the ngm chNA
molecule which is inverted with respect to spinach chNA is denoted
by a stipled arc.

26

1212

23.1- 9,4
6.7
9.4- 4
Barn 3b-) -3
6.7-
4— Bam 12
4.3-
- 2.3
2.3 -
2.0_ - 2.0

 

0.6 1.2

Figure 9. Restriction digest of chNA from (green) wild-type variants
C1 and C2. CpDNA from C1 (1) and C2 (2) were digested with BamHI
and fragments were separated in the presence of ethidium bromide on
0.6% or1.2% agarose gels. Two variable fragments from the C1 and
C2 classes are identified as Bam 3b and Barn 12.

27

class, while the other four contain CpDNA which is of the I-C2 type
(indicated in Figure 10). The RFLPs which distinguish these classes
do not correlate with any particular phenotype of individual mutants.
and in fact. we have been able to recover green, photosynthetic
representatives of these two classes (C2-wt and C1-wt in Figure 10).
All molecular analyses of plastome mutator-induced mutants have been
done in comparison with the appropriate wild-type variant.

All of the plants descended from the Cornell strain and examined thus
far may be grouped into one of two major classes in terms of a 3-kb
BamHI fragment (Bam12) which is located completely within the
inverted repeat region or a second variable region defined by the 8-kb
BamHI fragment (Bam 3b)(Kaplan. 1987; Johnson et al.. 1988). A
comparison is shown in Figure 1 1 for chNAs from two plastome
mutants anl and pm; and their corresponding wild-types, lines C2
and C 1, respectively. For 2111, when BamHI and EcoRI digests of
chNA are compared to those of the chNA of the proper wild-type
(C2), no difl'erences are seen in ethidium bromide-stained gels nor in
Southern hybridization experiments using Barn 3b as a probe (Figure
11). However, the chNA from mull does differ slightly in
comparison to I-C1. A variable fragment in the plnl_l lane is indicated
with the upper arrowhead in the ethidium bromide—stained gel of
digested chNA. However, this variability is even more clearly
demonstrated in the Southern hybridization analysis of EcoRI digests
probed with Barn 3b. where the m1; fragment appears to be
intermediate in size between the corresponding fragment in the I-C 1
wild-type and either I-C2 wild-type or 91111.

28

Cornell line
Oenothera hookeri str Johansen
Epp's E15-7 plastome mutator

?

 

1 976 1 978
field field

i

i ?

 

 

assaai
@ 1.2%

 

 

 

lpm32| lpm35|

Figure 10. Pedigree of plant lines, indicating C1 and C2 classes of
plastome mutator line. Pedigree of chloroplast mutant and (green)
wild-type lines descending from Epp's Cornell Oenothera hookeri

uni/pm isolate, E-15-7. (The circled x's represent self-crosses.
According to their chNA restriction patterns. mutants in the C1 class

are boxed: mutants in the C2 class are circled.)

29

Southern Hybridization
Probe: Barn 3b fragment

rcwwimwwwww

11wtwt711wtwt711wtwt711wtwt7 kb

 

> -9.4
- 6.7
- 4.3
- 2.3
- 2.0

b i ‘

.
co RI

 

Barn HI Eco RI

Figure 1 1. Restriction digest and Southern hybridization with Barn 3b.
Chloroplast DNA was digested with EcoRI and fragements separated
on a 0.6% agarose gel and stained with ethidium bromide. Fragments
were transferred to nitrocellulose and probed with nick-translated
plasmids containing cloned Bam 3b fragment from the C1 class.
Variable fragments between the C1 and C2 classes are marked with
arrowheads. Results of the C1 and C2 variability and the pml 1 analysis
have been previously reported by Kaplan (1987).

30

Fine mapping of the Earn 12 fragment has shown that Hinfl defines
the regions of variability (Kaplan, 1987; Blasko et al.. 1988). Since
Hinfl also cuts the Bam3b fragment a number of times, it was used to
more carefully assess the variability of both of the regions in a number
of mutants isolated from this 12mm line. The left panel of Figure 12
shows Hinfl-digested chNAs from two mutants and C2 wild-type,
probed with a mixture of two subcloned fragments from the Barn 3b
region (Figure 3). The approximately 800-bp fragment which appears
to be variable between the chNAs from C1 and C2 classes is marked
with an arrowhead. There is also a slight difference in the migration
of the largest band in the m1 lane in comparison to the
corresponding band in the C2-wt lane. The right panel of Figure 12
shows the autoradiogram resulting when the same filter was probed
with a 600-bp fragment from the Earn 12 fragment containing the
region of variability (Figure 2). The most obvious difference between
the I-C1 and I-C2 chNAs is the approximately 50-bp size variability in
the smallest fragment in this panel. The larger of these variable
fragments is indicated with an arrowhead. However, upon close
examination, there is some indication of more subtle differences
existing within the I-Cz class. seen in the migration of the 120-bp and
260-bp fragments in the last two lanes of Figure 12.

31
Southern Hybridization
probe: sub 8am 3b probe: sub Bam 12

rcw—cz—r row—Ch

pm11pm7 wt pm11pm7 wt bp

-861
> ' C- l -738
4 -615
5 a I
-492
-369
- -
C. -246
-. ‘ ‘
-- - ~ "23
~ iii.

Figure 12. Southern hybridization of Hinfl-digested pml l, pm7, and
C2 wild-type chNA with Bam 3b and Bam12 probes. Pml 1 pm.
and C2 wild-type CpDNA was digested with Hinfl, separated on a 5%
polyacrylamide gel. transferred to nitrocellulose and probed with
fragments isolated from the variable region of Barn 3b (determined by
S. Kaplan and P. O. Lim. shown in Figure 3) and from Barn 12
(determined by S. Kaplan and K. Blasko. shown in Figure 2). Variable
regions between C1 and C2 types are noted with arrowheads.

 

32

W

In the course of the molecular analysis of our plastome mutator lines.
we determined that one of our mm lines contained variants of
plastome I (preliminary results reported by Sears, 1983: Sears and
Kaplan. 1984: Kaplan and Sears, 1986; Kaplan. 1987). We have
distinguished plants carrying two major chNA types by grouping
them as lines C 1 and C2, thus named after the Cornell University
origin of Epp's line of W hookeri, strain Johansen. The
variable regions in these lines appear to be the result of

insertion/ deletion events, similar to those which have been reported
between more distantly related plants (Gordon et al., 1982; Salts et al.,
1984: Schmitz and Kowallik. 1986. Doebley et al., 1987). Since we
have been able to recover wild-type representatives of these two
classes. we now know that the chNA RFLPs which are observed in a
comparison of lines Cl and C2 are not correlated with the various
plastome mutator-induced mutations. Hence. any preliminary
differences observed between "mutant" and "wild-type," were due to
the existence of chNA restriction fragment polymorphisms between
plants descended directly from Epp's original m stock.

Various examples of "heteroplasmy" (the presence of two
distinguishable types of organelle DNA in a single individual) have
occasionally been observed in a number of organisms including
W13 (Solignac et al., 1983, 1984), crickets (Harrison et al.,
1985: Rand and Harrison, 1986), sea scallops (Snyder et al., 1987)

33

and American shad (Bentzen et al., 1988). In another case.
heteroplasmy of a Holstein cow was deduced after an analysis of her
descendents showed they carried two different mtDNA types
(Hauswirth and Laipis. 1982). In some of these cases. direct
descendents of the heteroplasmic individual were heteroplasmic also
(e.g. Solignac et al..1983: Harrison et al..1985) and in other cases. the
descendents were homoplasmic. containing only one or the other type
of organelle DNA (Hauswirth and Laipis. 1982; Solignac et al., 1983).
Additionally. Moon et al. (1987) have found individual rice plants
which contain two co-existing populations of chNAs. In contrast. we
have found that our individual plant lines (both mutant and wild-type)
contain only one type of CpDNA. although we have concluded that a
close ancestor must have been heteroplasmic for the chNA
polymorphisms.

By constructing the pedigree (Figure 10) for the plastome mutants and
the green 12mm plants from which they were derived, we can
narrow the occurrence of the event(s) leading to the chloroplast DNA
variations to within two generations. The plastome mutator plants in
Professor Stubbe's collection were all derived directly from self-
crosses of Epp's Mm plant line (established from the original
m isolate, E-15-7) containing plastome type I. Some seeds from
Epp were planted in 1976 and others in the summer of 1978. Seeds
resulting from self-crosses from both field seasons were planted in
1980. The former yielded three mutants which we have analyzed
carefully: all contain chNA type C2. Two mutants were analyzed
from the latter; one represents type I-C1 while the other represents

34

type I-C2 chNA (Figure 9). From the second field accession. green
seedlings were self-crossed for maintenance of the 0111112111 strain and
all of the progeny analyzed to date (two new mutant lines and progeny
derived from at least five different green plants of the next generation)
contain CpDNA type I-C1. This homogeneity of our current m
stock could have resulted from a founder efl'ect due to the use of a
limited number of plants in the self-crosses to perpetuate the line,
and possibly from sampling only a small subset of the seeds produced
by the previous generation for the subsequent planting.

Since the original seed lots from Epp and from the 1976, 1978 and
1980 field seasons are no longer available. one can only speculate as to
the nature of the event(s) which resulted in the chNA heteroplasmy.
Various explanations may be considered. Possibly, the variations in
chloroplast DNA which define classes C1 and C2 are a consequence of
the plastome mutator activity itself. However. since we have
recovered both mutant and green plants from both the C1 and C2
plant lines. these major chNA variations are clearly not associated
with the mutation events in the mutants analyzed. In the chNA
comparisons within each line. it appears that two or possibly three
regions contain hypervariable sites, while no other fragments seem to
change. However, the variable regions found within the plastome
mutator line do not correlate with the observed phenotype of any
plastome mutator-induced mutants. Thus. one is led to the conclusion
that mutations caused by the plastome mutator are most likely point
mutations. Alternatively, although the other chNA changes seem to

involve insertion / deletion events, a relationship of the plastome

35

mutator activity with variation in these specific chNA regions can be
imagined. If, for example, the point mutations induced by plastome
mutator are the result of a defective chNA repair system, this same
defect may also increase the occurrence of other types of mutations
resulting in the hot spots of chNA variability. Furthermore. it is also
possible that the mutagenic activity of plastome mutator could be
accompanied by a recombinogenic activity which produces this
variability in chNA in the Barn 12 and Barn 3b regions.

Another possibility is that the plastome difierences could have been
due to contamination of the self-crosses with pollen from another
plant carrying a difierent plastome type. as had been suggested by
Lindenhahn et al. (1985). This is not likely because the results
presented here demonstrate that these chNAs differ only at two
specific sites, and are thus much more similar to each other than they
are to any other plastome type (Gordon et al., 1982: Epp et al.. 1987).

A more likely possibility is that the heteroplasmy may stem from the
original EMS (ethyl-methane-sulfonate) mutagenesis. Although EMS is
believed to induce predominantly point mutations in nuclear DNA
(Legator and Flamm, 1973). conceivably the EMS treatment was
responsible for the deletion] insertion events which appear to
constitute the differences between the two CpDNA variants.
Alternatively. the EMS may have caused point mutations, which in turn
could have created a rearrangement 'hot spot‘ and the subsequent
deletion/ insertion. More detailed comparisons are currently under

way in the Sears laboratory in order to further characterize the nature

36

of the chNA changes which have occurred in this plastome mutator
line of Oenothera.

The existence of chNA heteroplasmy within our plastome mutator
line underscores the necessity for using the correct wild-type in

comparisons of mutant and wild-type chNAs. Using Barn 12 and Barn
3b as probes in Southern hybridization experiments to compare C 1

and C2 wild-type chNAs to plastome mutator-induced mutants from
each of these lines initially indicated only two distinct classes of
chNA patterns within our plastome mutator line (Kaplan, 1987 ,
Johnson et al., 1988). However. results of the investigations
described here indicate the strong possibility that further variability
may indeed exist within line C1, between chNA from 1211111 and C1
in the Bam 3b region (Figure 12) and perhaps between 111111 and C2 in
the Earn 12 region (Figure 12). Less striking but still notable are the
apparent subtle difl‘erences between Hinfi-digested chNA from 9.1111
and its corresponding wild-type, C2 (Figure 12). It is not clear from

these preliminary investigations whether any further differences
within line C 1 or C2 are related to the activity of the plastome
mutator. Further investigation of these preliminary observations is
necessary in order to rule out the possibilities of partial restriction
digestion or anomalous migration of DNA fragments in the

electrophoresis conditions.

It is clear from these results that efforts to identify the site of the
primary lesion in any one particular plastome mutator-induced mutant

must employ other approaches besides chNA analysis. For example.

37

these mutants may be examined using immunological techniques to
detect the presence or absence of various important chloroplast
proteins. Various physiological assays may be carried out to determine
whether the chloroplast proteins present are indeed functional in the
mutant plastids. The degree of mutant plastid development may be
assessed by ultrastructural analysis as well as direct measurement of
the accumulation of certain chloroplast components such as
chlorophyll, carotenoids, fatty acids and lipids. The fact that the
plastome mutator system is also active in shoot-tip culture conditions
provides the opportunity for isolating revertants of plastome mutator-
induced mutants. Efl'orts by Dr. Linda Schnabelrauch to recover such
revertants of one plastome mutator-induced mutant, 9.1111» have been
successful. thus providing an additional valuable tool for use in
characterization of this particular mutant.

CHAPTER 3

Immunochemical and Molecular Analysis of the Phenotypes of pm
and a Putative Revertant of 121111

mm

Using the plastome mutator system of Qengflmza. a large number of
plastome mutants have now been isolated, following the appearance of
chlorotic sectors in plants containing an active plastome mutator
background (nmzpm). Screening of this collection. utilizing a battery
of antibodies raised against various individual chloropast proteins.
allowed the identification of a particular mutant with a unique
phenotype in cytochrome f (cyt f). Cytochrome f is a thylakoid
membrane-bound protein which functions in the cyt b6/ f complex as
an intermediate electron carrier in photosynthetic electron transport
(cytochrome structure and function reviewed by Hauska et al., 1983;
Bendall, 1982). The mutant. nmz, accumulates a protein which is
cross-antigenic with cyt f but is approximately S-kD larger than the
wild-type cyt f (Figure 13). This observation is intriguing since the
gene for cyt f (931A) is a CpDNA-encoded gene. translated on
chloroplast ribosomes as a precursor of a molecular weight which is
also approximately 5-kD larger than the mature protein. This gene
has been cloned and sequenced from a number of higher plants
including pea (Willey et al., 1984a). wheat (Willey et al., 1984b).

38

39

@ Stained lmmuno-Blot

Membrane / Soluble Membrane / Soluble
morkers-pm8-pm7-w.t.4pm8~pm7-w.t. pm8-pm7.w.t.-pm8~pm7‘w.t.

   

10-15%,5M Urea

Figure 13. Coomassie blue staining and immunoanalysis with cyt f
antibodies of mutant and wild-type chloroplast proteins. Results of an
immunological screen performed by Dr. Sears on several the plastome
mutator-induced mutants. Panel A shows a stained. denaturing
polyacrylamide gel containing membrane and soluble fractions of wild-
type (wt) and mutant (2191 and Mchloroplast proteins. To
standardize the loading of the membrane fractions. approximately
equal amounts of the or and (3 subunits of the chloroplast ATPase were
loaded in both mutant and wild-type lanes. These proteins were
separated on denaturing polyacrylamide gels. electrophoretically
transferred to nitrocelluose. probed with an anti-cyt f antibody.
followed by 125l-protein A (panel B). The arrow in the figure
indicates a high molecular weight protein which is cross-antigenic
with cyt f which is seen exclusively in the membrane fraction of pm7.
(The signal at the migration noted by the arrow in the wild-type
membrane lane in B is not reproducible and is probably due to
"spillover" from the neighboring 12ml lane in this gel.)

40

spinach (Alt and Herrmann.. 1984), rice (Wu et al., 1986) and
W ('lyagi and Herrmann, 1986). The sequence, including the
presence of a 4—5-kD amino-terminal extension which is not found in
the mature protein, has proven to be highly conserved in many

species.

Various hypotheses can be put forth to explain this observation in
m. A number of these hypotheses involve the possibility that a
lesion may exist in the gem gene itself. Conceivably, the unusually
large cyt f protein in pm], is an unprocessed precursor of cyt f. For
example, a site of protein processing has been proposed for various
higher plant species (Willey et al., 1984a, 1984b; Alt and Herrmann,
1984; Wu et al., 1986; mg and Herrmann, 1986) by comparing the
deduced amino acid sequence of the precursor to cyt f (pre-cyt f) to
the actual N-terminal amino acid sequence of cyt f from spinach (Ho
and Krogman, 1980). If there were a lesion at this site in the 121111
gem gene, it could result in the defective processing of cyt f. Another
highly conserved region of the gem gene encoding two cysteine
residues has been suggested as a site of heme binding in cyt f (Nelson
and Neumann, 1972; Nelson and Racker, 1972; Siedow, 1980). The
covalently bound heme group in cyt f is essential for the proper
functioning of mature cyt f in photosynthetic electron transport
(Nelson and Neumann, 1972; Nelson and Racker, 1972). However, it
is not known at which point in maturation of the protein the heme is
attached. Therefore, it is possible that a lesion in M which alters
the heme binding site could have a secondary inhibitory effect on the
processing of the pre-cyt f protein.

41

An alternative explanation for the observed cyt f phenotype in DJAZ is
that perhaps a mutation is present upstream of the DSLA gene which
results in an overexpression of the gene. Such an "up" mutation could
cause the expression of the gene to outpace the activity of the
processing enzyme. Such an explanation would be consistent with the
apparently "leaky" nature of the phenotype (Figure 13), that is, the
presence of a cross-reacting protein that has a migration similar to cyt
f. Each of these possibilities involving a lesion directly in the 931131
gene were tested by DNA sequence analysis of the mutant and wild-

type genes.

Conceivably, the 40-kD protein which cross-reacts with cyt f
antibodies is not a precursor to this protein. Several other
explanations can be considered which could account for a larger
protein which is aniigenically related to cyt f. Multiple sizes of RNAs
containing the cyt f message have been found in pea (Willey, personal
communication), spinach (Alt and Herrmann, 1984) and a closely
related 9339111323 line (Tyagi and Herrmann, 1986), indicating that
the transcript of the wild-type gem gene may be processed in a
complicated manner. The 931A message may be on a polycistronic
transcript because no strong consensus transcriptional start
sequences for m exist in a number of species (Willey et al., 1983.
1984b; Alt and Herrmann. 1984. Wu et al.. 1986, 'lyagi and
Herrmann, 1986). Additionally, an ORF has been found directly
upstream from M in pea (Willey et al., 1984a), spinach (Alt and
Herrmann, 1986) and tobacco (Shinozaki et al., 1986). It is

42

conceivable that improper RNA processing of this message results in
the use of an upstream translational start site and thus leads to the
accumulation of an unusually large-sized cyt f protein in 12131.
However, all ATG codons within ZOO-hp upstream of 991A are soon
followed by a stop codon, making this alternative unlikely.

Altered splicing could also result in partial or complete inhibition of
protein processing because the protein processing site is no longer
present or not recognized in M due to an alteration in the N-
terminal sequence.

A translation product of increased size could also result from the in-
frame insertion of an approximately 100-bp element in the 921A gene.
An in-frame insertion of this size should be detectable in a restriction
analysis of the region containing the Dem gene using a number of
restriction endonucleases. A frame-shift mutation caused by a small
insertion or deletion is not likely to result in a large cyt f product due
to the many stop codons in the other two reading frames (Tyagi and

Herrmann, 1986).

Read-through of the authentic termination codon of the petA
message, due to an altered tRNA or stop codon could yield an aberrant
translation product which could be processed at the amino terminus
as usual, but would still result in the accumulation of a protein which is
cross-antigenic with cyt f but larger than the expected size. Again,
this is an unlikely explanation since there are numerous other stop

codons near the end of the gem gene.

43

In addition to proper processing of the precursor form of the protein,
proper association with heme and the other protein components of
the cytochrome complex as well as integration into the thylakoid
membrane are presumably required in order to achieve the active,
mature form of cyt f. A number of other higher plant mutants with
altered phenotypes in the cyt b6/ f complex have shown the
accumulation of one member of a complex may depend upon the
presence of other subunits of the complex (Marco and Garmier, 198 1;
Mets et al., 1983: Bendall et al.. 1986: Barkan et al., 1986; Lam and
Malkin, 1985; Lemaire et al., 1986; Olive et al., 1986). In this
context. another hypothesis arises which suggests that in the absence
of heme or another component of the cyt belf complex, both the
processing and assembly of other subunits could be aborted. These
possibilities were addressed by using a gel staining technique to
detect heme bound to protein and immunoblotting techniques to look
for the accumulation of the other components of the cyt b6] f complex:
cyt b6 and subunit IV, each encoded by chloroplast genes, and the
Reiske Fe/ S protein, which is translated on cytoplasmic ribosomes.

It is also conceivable that the larger pm], cyt f protein is a
ubiquitinated form of the mature cyt f. The difference between the
sizes of the mature and the aberrant proteins in m1 is consistent
with this possibility. Immunoanalysis with an antibody which
recognizes plant ubiquitin was used to test this hypothesis. However,
no evidence to support this possibilty was found, since the antibody
reacted only with a very high molecular weight smear (data not shown)

44

and did not detect any proteins of sizes similar to the aberrant cyt f
protein in m.

Finally, the aberrantly sized cyt f protein in 111111 may be the result of
defective processing machinery in the mutant plastids. According to
this hypothesis some element essential to chloroplast protein
processing is encoded by a plastid gene, since walls a non-
Mendelian mutant. It is possible that such a processing deficiency
could afi'ect other proteins in the chloroplast besides cyt f. To test
this, several other CpDNA-encoded proteins could be examined since
evidence indicates they are initially synthezed as larger precursors.
The post-translational processing of the D1 protein of photosystem 11
(PS 11) has been well studied (Greenbrier et al., 1978; Riesfeld et al.,
1982; Marder et al., 1984: Minami and Watanabe, 1985) and is known
to involve an acylation step as well as specific proteolytic cleavage in
order to achieve the mature protein product of approximately 32-kD
in size (Mattoo and Edelman, 1987). Early reports (Langridge, 1981)
have also proposed the existence of a precursor form of the large
subunit of ribulose bis-phosphate carboxylase/oxygenase (RuBisCO
LSU). Thus, determining the status of these other chloroplast
proteins in 111111 could indicate whether a general processing defect is
involved.

One could also ask whether the processing of nuclear-encoded
chloroplast proteins is affected in the mutant. Most proteins which
are imported from the cytoplasm contain an amino-terminal transit
peptide which is proteolytically cleaved during or after import into the

45

organelle. Furthermore, some proteins are processed a second time
within the organelle, presumably during passage across the thyakoid
membrane (Smeekens et al.,1986; Hageman et al., 1986). In order to
address the issue of post-translational processing in 11ml, a series of
immunoblotting experiments was performed using a battery of
antibodies raised against various individual chloroplast proteins,
including a collection of imported proteins as well as a number of
proteins which are translated on chloroplast ribosomes. This
approach is made possible by the fact that chloroplast genes are highly
conserved between many higher plant species and hence antibodies
raised against their gene products often cross-react.

46

Materialaanwmgds

Plant Material

Plant material was maintained as shoot-tip cultures as described in
Chapter 2. illuminated with 20 limol-m’zosec'1 photosynthetically
active radiation (PAR). All wild-type tissues and tissue containing 121111
in the A/ C nuclear background were heterozygous for the pm allele,
and thus inactive for the plastome mutator function. However, tissue
containing pmz plastids in the A/A nuclear background was
homozygous for the mutant pm allele. In this nuclear background the
potential crusts for reversion, as discussed in Chapter 2. Thus, in the
maintenance of the A/A 11ml leaf tip cultures, all tissue was

discarded which showed any evidence of reversion (appearance of

green sectors).

Spinach leaves for purification of cyt b6/ f particles were purchased

from a local farmer's market.

M13 Cloning and Dideoxy Sequencing Strateg;r

As described in Chapter 2, Barn 3a fragments containing the gem gene
encoding cyt f were isolated from wild-type and mutant chNA using a
heterologous probe from spinach (a gift from Dr. W. Bottomley, CSIRO.
Canberra) and cloned into plasmid vectors pBR322 and pUC 18/ 19
(Norrander et al., 1983). More detailed data on DNA restriction

patterns were obtained using general methods for restriction digest

47

and agarose gel electrophoresis described by Maniatis et al. (1982)
according to suppliers' specifications (Bethesda Research Laboratories
(BRL), Boehringer Mannheim, New England Biolabs).

Segments of the Bam3a fragment (Figure 14) were further cloned into
M13 vectors, according to specifications given in the BRL Instruction
Manual an M13 Cloning/Dideoxy Sequencing. M13 vectors mp 18 and
mp19 (Norrander et al., 1983: Yanisch-Perron et al., 1985) and the
bacterial host JM 107 (Yanisch-Perron et al., 1985) were employed in
the cloning and sequencing stategy originated by Sanger et al. (1977).
Figure 15 summarizes the composition of the clones which were
constructed and the sequencing strategy followed. Preparative
quantities of the double-stranded replicative form (RF) of these phage
DNAs for both the vectors and the vectors with inserts were obtained
using standard culture prodcedures for M13 phage (Messing, 1983)
and plasmid preparation protocols given by (Kahn et al., 1979). The
DNA sequence of the inserts into these vectors was determined using
both 32P— and 35S—labelled Klenow sequencing reactions on single-
stranded templates, as described in the BRL M13 Cloning

Dideoxy] Sequencing Manual. A total of 700-bp was sequenced on both
strands using the universal l7-bp primer (BRL) and two synthetic
primers prepared by Dr. C. R. Somerville (Michigan State Univ.) using
an Applied Biosystems 380A DNA Synthesizer.

48

 

 

 

é—Bam 2e Bam 3a : Barn 3b—>
E BEEKS EEHE EH E SHiNB E H
A 1 11m [III III in: J

 

B F "939391

ATG protein heme
processing binding
site site

Q'E sequenced region

 

C

 

ii A
i——-l synthetic peptide

Figure 14. Map of the Barn 3a region from the Johansen strain of
Mega 11993911 chNA. Part A shows a restriction map of the Barn
3a region from ml and its corresponding wild-type. E=EcoRI,
B=BamHI, K=KpnI, S=Sa1, H=HindlII, Hi=HincII, N=NdeI. Part B is a
10-fold enlargement of the region containing the DEJA gene and the
upstream sequences. The translation initiation codon is indicated
along with the location in the DNA sequence of the protein processing
site and the heme binding site. Part C represents the region of the
DNA sequenced in m and the wild-type.

 

49

 

 

 

 

 

 

 

 

 

 

 

 

 

heme
100-bp proggln blndlng
processing 8".
E H ATG 3““ 5
F : l I l I
wersai’ synthetic
“grim primerA
#1. AAAAAAAAAAA
WMCS‘: 'E _________
pm7-18-5 E
synthetic universal
primer 3 primer
F 1
.......... I fl‘a‘csaflnflafla
-. vvvvvvvvvv f E
pm7-18-1 E .
universal synthetic
'mer primer A
wk 2’
. AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA F E
Germ-7.9 H
synthetic universal
F 1
091-19-2

Figure 15. Strategy employed to sequence the 5' end and region
upstream of pm from m and line 02 wild-type chNA. Clones
were constructed in M13 vectors mp18 (pm7-18-5, pm7-18-1 and
Oej-18-7,9) or mp19 (Oej-19-2). This pair of vectors contains the
multiple cloning site (MCS) in opposite orientations, thus allowing the
production of single-stranded templates containing either sense
(pm7-18-5 and Oej-18-7,9) or anti-sense (pm7-18-1 and Oej-19-2)
strands of the cloned DNA. Arrows denote the direction of sequencing
proceeding from one of 3 different primers, shown as black boxes on
the arrows. Th "universal" 17 -bp primer (5'-GTAAAACGACGGCCAGT—
3') was purchased from Bethesda Research Laboratories: synthetic
primers A (5'-TCCAGTAAACCGAT'ITCAGT—3') and B (5'-
TATACATCAGTGATAGGGAAAT—S') were prepared by Dr. C.R.
Somerville (MSU) as described in the text. Wavy lines represent

vector sequence: thin straight lines denote Qangthera chNA
sequence.

50

Chloroplast Preparation and Fractionation of Membrane / Soluble
Proteins

Tissue (10-40 g leaf material) was harvested 3-4 weeks after transfer
to. fresh media. Washed chloroplasts were prepared by
homogenization and difi'erential centrifugation as in Chapter 2.
Membrane and soluble protein fractions of these chloroplast
preparations were isolated in the following manner. Plastid pellets of
approximately 100 pl volume were resuspended with vigorous
agitation in 3 volumes 50 mM Tris, pH 8.0 and centrifuged 5 minutes
in a microfuge. Supematants were saved in fresh tubes and the pellets
were resuspended as before in 3 volumes 100 mM NaCl, 50 mM The.
pH 8.0. This mixture was centrifuged as before and the resulting
supernatant pooled with the previous supernatant sample. This step
was repeated once more and the three pooled supernatants were
precipitated with 3 volumes of 50% (w/v) trichloro-acetic acid (TCA)
on ice for 30 minutes. TCA precipitates were collected by 5 minute
microfuge centrifugation. washed twice with 80% acetone and
resuspended to a final protein concentration of 0.75 rig-111'1 in SDS-
polyacrylamide gel sample buffer (50 mM Tris, pH 6.8. 0.28 M (3-
mercaptoethanol, 2% (w/v) SDS, 10% (w/v) glycerol) and made to a
final concentration of 2 mM phenol methyl sulfonyl fluoride (PMSF).
The washed chloroplast membrane pellet was resuspended in sample
buffer to a final protein concentration of 2.5 ug-ul'l. Protein
concentration was determined using the Bio-Rad protein assay (Bio-
Rad publication #81-0525), performed on aliquots of the membrane or

soluble protein preparations resuspended in water.

51

SDS-Polyacrylamide Gel Electrophoresis

Membrane and soluble chloroplast protein fractions in 50 pl (wild-
type) and 75 pl (mutant) volumes were analysed using the
discontinuous buffer system of Laemmli (1970) and 10- 15 % (w/v)
gradient SDS-polyacrylamide gels containing 5 M urea. Improved
resolution of W chloroplast polypeptides has been achieved by
using one-half the originally recommended concentration of his-
acrylamide in the gel recipes (Dr. P. Westhofl', personal
communication). Gels were stained with Commassie Blue (Sigma
Chemical Co.) to visualize the total protein pattern or
tetramethylbenzidine/ H202 (Sigma Chemical Co.) for the detection of
covalently-bound heme (Thomas et al., 1976). Except where
specifically indicated, all protein sizes discussed are apparent
molecular weights as determined by comparison to protein markers of
known sizes (Sigma Chemical Co.): bovine serum albumin (68-kD),
ovalbumin (45-kD), carbonic anhydrase (29-kD) and cytochrome c1
(12.4-kD).

Preparation and Sources of Antibodies Specific for Chloroplast
Proteins

Cyt b6/f complex was isolated from 500 g fresh spinach leaves using
octylglucoside and cholate extraction according to a procedure
published by Hurt and Hauska (1981; Gray and Phillips, 1982; Clark
and Hind, 1983). Cytochrome f was purified from this protein

52

complex by electroelution using a CBS Scientific electroeluter
apparatus as directed by the manufacturer. New Zealand white rabbits
were initially immunized as directed by Chua et al. (1982) with 500 ug
purified cyt f antigen in 1 ml 0.15 M NaCl, 10 mM Na phosphate, pH
7.0 (PBS), mixed with 1 ml complete Freund's adjuvant (Sigma
Chemical Co.): injections were repeated biweekly using incomplete
Freund's adjuvant. Sera from successive bleedings were screened by

Western blotting analysis and serum aliquots were stored at -20°C.

For use in preparation of an antibody specific for the cyt f precursor, a
synthetic peptide of the first 17 residues of unprocessed cyt f was
prepared by Dr. H. Wolfe at E. I. DuPont Co. using the amino acid
sequence deduced from the DNA sequence of W hggkeg
(Tyagi and Herrmann, 1986). This peptide was conjugated by C-
terminal cys cross—linking to a carrier protein, keyhole limpet
hemocyanin (KLH). A mixture of 0.5 - 1.0 mg peptide-KLH conjugate
and 1 ml of trehalose dimycolate (TDM) emulsion (RIBI Immune Chem
Research, Inc), reconstituted according to the supplier's
specifications, was injected biweekly into each of two white New
Zealand rabbits. Serum was collected and screened as described
above.

A number of other antisera against particular chloroplast proteins
were generously donated by various researchers. The source for each

of the antibodies is given in Table 1.

53

Table 1. Sources of antibodies used in immunoanalysis of pmz and
wild-type chloroplast proteins. Abbreviations for protein names are

given in the text.

Antigen Origin Source of Antibody
cyt f spinach E. Johnson, MSU, E Lansing
cyt f rape T. Bricker, Univ MO, Columbia
SU IV(cyt b6/fl maize A. Barkan. Univ CA. Berkeley
FeSlcyt belt) " "
DEC-33 spinach C. Jansson, MSU, E. Lansing
DEC-23 " "
DEC-16 " "
D1(PS 11) W J. Hirschberg, MSU, E. Lansing
PS II-22-kD spinach C. Yocum, Univ MI, Ann Arbor
PS lI43-kD, 47-kD tobacco C. Chia, MSU, E. Lansing
PSI " R. Malkin, Univ CA, Berkeley
LHC II-29-kD barley S. Darr. Univ IN, Bloomington
Carbonic anhydrase spinach D. Husic, M. Volakita, N. Artus,
MSU, E. Lansing
ADP-glucose spinach J. Preiss. MSU, E. Lansing
pyrophosphorylase

Ribosomal W
proteins L21, L29

J. Boynton. Duke Univ, Durham

Plasto spinach K. Keegstra, Univ WI, Madison
RubisCO small subunit maize E. Bell, MSU, E. Lansing
" alfalfa C. Schumann. MSU, E. Lansing
Ubiquitin oat R. Vierstra. Univ WI, Madison
Mitochondrial yeast T. Fox, Cornell Univ, Ithaca

cytochrome oxidase
subunit II

54

Immunoblotting Analysis

Proteins separated on SDS-PAGE were transferred to nitrocelluose
membrane filters with a 'hansphor Electrophoresis Cell (Hoefer
Scientific Instruments. San Fransisco, CA) at 4°C for 3-5 hours at 150
V using 25 mM Tris, 192 mM glycine, 20 % (v/v) methanol, pH 8.3
(Towbin et al., 1979). Filters were blocked by shaking overnight at
room temperature in 5 % (w/v) non-fat dry milk (Spartan Brand), 0.2
% Na azide (blocking bufi'er) (Kaufmann et al., 1987). Filters were
incubated in antibodies dissolved (at dilutions ranging from 1/ 100 to
1 / 1000, depending upon the antibody titer) in blocking bufi‘er for 2
hours to overnight at room temperature. Filters were washed four
times (15 minutes each) with 0.05 % Non-idet-P40 (NP40) in PBS
(phosphate-bufi‘ered saline). The nitrocelluose sheets were then
incubated 1 hour at room temperature with rocking in the PBS-NP40
solution containing 5-10 uCi of 125I-Iabelled protein A (New England
Nuclear). Filters were subsequently washed four times (15 minutes
each) with the PBS-NP40 solution, dried and exposed to Kodak X-
Omat XAR-5 film for autoradiography. Preliminary immunoblots
utilized an alkaline phosphatase-protein A detection system based
upon the procedure of Leary et al. (1983). For a number of antisera.
however, radiolabelling gave a stronger signal and thus 125I-detection
was the method of choice for these experiments.

55

Results

Cloning of Fragments from the Bam 3a Region of oral and Wild-type
chNAs

Figure 14 shows the restriction map developed for the Barn 3a region
from both the pr_n_Z mutant and the C2 wild-type. Approximately 700-
bp was sequenced from 12ml, using the fragment shown in Figure 14C
and defined by an EcoRI site within gem and an upstream EcoRI site.
The segment sequenced in the wild-type is about 50-bp smaller since
the Hindm site shown in Figure 14A was used for subcloning. Both
fragments were subcloned into M13 vectors mp 18 or mp 19; these
clones are shown in Figure 15, including the orientation of the insert
with respect to the sense strand of the gem gene. Figure 15 also
shows the sites where the synthetic primers anneal.

Analysis of Mutant and Wild-type DNA Sequences in the item Region

The DNA sequence which encodes a portion of the 1194A gene and
extends approximately 450-bp upstream from this gene in the m1
mutant and W wild-type is presented in Figure 16. Regions
encoding the proposed processing and heme binding sites for the
protein product are marked. No sequence changes are found in these
areas between p311 and its correct wild-type progenitor. line C2. but
there are differences when the sequence from these two isolates is

compared to the hggkeg Standard strain sequenced by Tyagi and

56

Figure 16. Partial sequence of petA and upstream region of chNA
from the Nicotiana tabacum and Johansen and Standard strains of
Oengtherg hgokeri. The top line shows the N. tabacum (N) sequence
(Shinozaki et al.. 1986); the second line shows the sequence for both
pm7 and wild-type I-C2 from the Johansen strain of 91 hookeri (OJ),
with nucleotide #1 representing the 5‘ end of the sequence
determined for the mutant and wild-type isolates from the Johansen
strain: the third line, which begins at nucleotide #151, contains the
available sequence from the Standard strain of Q1 hookeri (OS) as
determined by Tyagi and Herrmann (1986). Astericks above or below
the lines of sequence denote differences between the Johansen strain
and the tobacco and the Standard strain sequences. respectively.
Sequence for 268-bp from the 3' end of ORF 229 is single underlined;
sequence for approximately 300-bp from the proposed 5' end of gem
is double underlined. A putative ribosome binding site for petA (at
nucleotide #496) is underlined with a wavy line. The site at which cyt
f protein processing occurs is marked with an arrowhead. The boxed
region encodes the proposed heme binding site for cyt f.

57

F1 ure16

u- HTfifiCTT mamacnca 31nnnnec11 111c1enc1 111911nnc1 5°
N ‘W

* *
n- eemnram 1caaa11cc3 *33331: -GGTTGGGM TTMTGHTTG '00
W

03- ------ 11 on an

* * * *
N -631c131c1n 131-1116111111 ear-1111011c n1nn1031ca m1ca1n1c1‘50
'W

n -001c110111 ccnccmcc norcn11c1c GBTHCHRTTT 1111111119110 20°
0.: -661c116111 ccncmrcc ne1cn11c1n GflTHCflflTTT TTHHRTHTTG
as - n 1110111111 1111911191119

*
N -oa1111cca1 rammem erc131c1cc 31cnc11619 01131111113 25°
0.) -an1r1111c01 111111311 31c aremrcncc c1cnc1161n 0109111310
as -9_me61 111111111191 3 cc 1 a are 1

seem-meann- emu-me-r-me *Itntutliwttfi 30°
-fiTTCflfiTGHH TGHCTGHTRR RGGHTCCRTT GflTfl‘l’TfiflTC TflflTCCflfl

N-
0.1-1111mm can rancmar ................................
as -engnmgnn TQfiQTGflfi --------------------------------
N
as

 

********** It 10*? t a *** "35°

-nennrec11e131n 0'er 1nca1n meannamr 13mm1c n1
0.. ----------- -- -- 1111mm eccaeancar ncmnncan
111gecn1n scenaancm 11mm econ

 

----------- --ac----Tn

-n111nc1c11 1c1n111c13 135311136633" 11311- $1331 «111131131 40°
c1naccc11 recs-cacao n Gib-soar 11c1cc1nc1 ccan1c11cc
c1mccc11 man-cccoe flCGfl- -oea1 11c1cc1nc1 ccaa1c11cc

H
H
*IIIMI **** i II E It * ¥ 45°
TCC THGRTTH 'II'TCQW TWT OTGGCTHGGG MCTRTHC fl
RGTRHHCCGH T EHGTMR THGCHGHHTT GTGGRTHGGG fi'C'l HTHCHH
fiGTHflHCCGH T CHGTMH THGCHGMTT GTGGHTHGGG H'CTHTHCHH

 

 

 

 

that!!! #101 III III IO!
N - ace -nccrnc CEAannrnn 0111011113111 1caccn1cm 161111me 500
Od-eceecccacc 1mm HT‘I GTRGHHHTTT 1cceen1cne TGHTTGGTCC
08 - ocnacccncc man 1111 01116999111 1caaen1cna Tenneegcc

 

 

 

 

w

*m * *
N - ------ area nan-131nm 1cc11111§1r meanness anEnannnc 550
0.1-01mm ----- -ncc11c Teonmnnan ancnnnc

OS- -GTGCflflflTGf-l HEART ----- -HCCT11 C TGGHTHHHGH RHGRGHTTHS

 

D
ID

 

 

 

n - 1caa1c11111 1cc31n1c3c 1caranmm 1n1c119ec1 cacncrnccn 60°
gas-Teenrc 11-111 1ccc191cac TGRTGRTRTH 1n1cn1nac1 caancmccn

-TCOHTCTRTT TCCCTHTCHC TGHTGHTHTH THTCHTHHCT CGGHCRTCCfl

V at: :1-
N - 111cnna1ec n1arcccn11 mccncaec HGGGTTHTGR anmccncon 55°
0.:- menanccc n1n1cccnn macncanc HGGGGTHTGH ana1ccncon
os- mcnnacac armcccnn macacnac HGGGGTRTGR nemccnceg

 

 

 

    
 

 

 

 

  

 

 

 

 

 

 

 

 

 

 

 

 

 

   

 

n- neaceecm G3carn ., ' as c1nn19aecc 70°
0.:- anacancra 689916 1 as 1na1nnoc
as- HHGCGHCTG GHCGIH T HG TRHTHHGC

*

1‘" -CGTGGflGHTT GHGGTTCCHC HHGCGGTHCT TCCTGHTHCT GTRTTTGHHG 75°

GGHTH GHGGTT CCC RHGCGG Fl C GGHTHCT GTHTTTGFlf-l
OS-Eg1GGflTRPT. GHGGTT CCC HHGCGG F121 8C EGGHTHCT 0TFlTTTGRR

 

03- cn011611ca 11911065

H can
05-513390311

 

 

 

 

 

 

 

 

58

Herrmann (1986). Eight single base pair changes are noted which
difi'erentiate these two strains and all of them precede the initial
methionine codon. A putative ribosome binding site is indicated by
wavy underlining. However, as was previously reported by Tyagi and
Herrmann (1986), there are no clear consensus sequences for
transcriptional regulatory sites such as a "Pribnow box." expected
approximately lO-bp upstream or a second RNA polymerase
recognition region expected approximately 35 upstream of the gem

gene.

Sequencing comparison extended farther upstream of gem for 111111
and the wild-type line C2 than has been reported for the mtg gene
from Q. hmken analysed by W and Herrmann (1986). Thus, a 260-
bp sequence was encountered which was 92% homologous to the 3'
end of ORF 229 which has been previously reported in pea (Willey et
al., 1984b). spinach (Alt and Herrmann. 1984) and tobacco (Shinozaki
et al.. 1986). Again. no differences were seen between the mutant and
the wild—type in the 3' end of ORF 229 of the intergenic region. This
region is also shown in Figure 16. The divergent bases in W
are noted by asterisks above the line of sequence. The spacing
between the peg gene and ORF 229 was 186-bp in line C2 of

Qenaihera and 231-bp in Nimitana.
Analysis of Putative cyt f Precursor-Specific Antibodies

Parallel to the molecular analysis of the gem gene, an attempt was
made to test the hypothesis that the mutant protein represented an

59

unprocessed form of pre-cyt f. A DNA sequence of the Qenglhem
11mm DflA gene was available (Tyagi and Herrmann, 1986). The
deduced amino acid sequence of the amino-terminal extension of cyt f
was analysed to choose a sequence for a synthetic peptide. Dr. H.
Wolfe (E.I. DuPont Co.) synthesized a peptide based upon the first 17
amino acids of the presumed cyt f presequence, representing 50% of
the total N-terminal extension. The final product of this synthesis
however contained an unexpected arginine residue (W olfe, personal
communication) Nonetheless, this peptide was conjugated to keyhole
limpet hemocyanin and injected into rabbits in order to produce a
"precursor-specific" poloyclonal antibody (referred to as the "anti-
peptide" antibody). Membrane proteins from wild-type and m
mutant plastids were isolated from tissue maintained in shoot-tip
culture and subjected to immunoblotting procedures using the anti-
peptide antibody. The results of this Western blotting experiment are
shown in Figure 17A. This antibody, which should recognize only the
cyt f precursor, gives a single faint signal in both the mutant and the
wild-type lanes, at a size approximately 15 kD larger than the mature
cyt f. It is likely, based upon the especially large size of this protein
and also because only a very weak signal was achieved even when the
antibody was used at a titer of 1:100, that the anti-peptide antibody
does not show specificity for the precursor but gives only a
"background" signal. Furthermore, this same nitrocellulose filter was
reprobed with the cyt f antibody so that the abberant cyt f band in the
mutant could be compared to the protein band recognized by the anti-
peptide antibody. Figure 173 shows that these bands do not

6O

Immunoblot
anti-peptide anti-cyt 1
wt pm7 wt pm7 kD

 

Figure 17. Immunoblot of wild-type and mm? total chloroplast
proteins probed with antibodies against synthetic cyt f precursor-
conjugate and cyt f. The arrowhead indicates a cross-reacting
polypeptide present in pm7 and wild-type (wt). Reprobing of the
filter from panel A with cyt f antibodies shows that the protein
recognized by anti-peptide immunoglobilin does not comigrate with
the aberrant cyt f in pm7.

61

comigrate, that is. the two antibodies do not recognize the same

proteins.

An alternative explanation for the lack of recognition of a protein in
11311 which is the size expected for pre-cyt f is that perhaps the
suggested presequence (Tyagi and Herrmann. 1986) is not the
authentic N-terminal sequence of the cyt f precursor. The presequnce
proposed by 'lyagi and Herrmann (1986) is based on the agreement
between the size of the in 111:9, translation product of the M
message and the predicted size of cyt f with this additional N-terminal
extension. Additionally, a putative ribosome binding site is present 8-
bp upstream of the AUG codon proposed by mg and Herrmann
(1986). Nonetheless, it is possible that wild-type cyt f is the
translation product of a message in which the codon encoding the
known N-terminal residue of the mature protein (Iyr) is spliced to
mRNA from a region even farther downstream than the proposed
(contiguous) presequence. In 11ml, altered mRNA splicing could
disrupt the region recognized by the protease and protein processing
activity may be affected. This could result in a mutant cyt f which is
larger than the mature product. but not detected by the antibody
prepared against the synthetic peptide from the presumed
presequence of 'lyagi and Herramnn (1986) such as was seen in this

experiment.

Ubiquitin (UBQ) is a small protein which becomes bound to a variety of
proteins as a signal for their degradation (Finley and Varshausky,
1985). The size of UBQ (2-3-kD) is similar to the increased size of the

62

aberrant cyt f in 111111. Although UBQ has not been found to be present
in chloroplasts. the possibilty that the aberrant cyt f in pm7 is a UBQ-
conjugate was investigated using an antibody raised against oat UBQ
(Shanan et al., 1987). The only cross-reacting material detected was
a smaear of high molecular weight (greater than 80-kD) bands from
total protein preparations from both mutant and wild-type (data not

shown) .

Analysis of Heme Staining Activity

Denatured chloroplast proteins from the mutant and wild -type were
separated electrophoretically on polyacrylamide gels along with the
purified proteins of the spinach cyt b6] f complex and analysed by
heme staining. The mutant does not show heme staining activity in
the region of the gel where either the mature or the putative pre-cyt f
is expected. There is also no stain in the mutant lanes at a size which
corresponds to cyt be. (This is not unexpected however since the
TMBZ procedure does not reliably stain non-covalently bound heme
proteins. such as cyt b6.) Occasionally heme-staining bands were
observed in lanes having proteins from the mutant (Figure 18) and
sometimes also in lanes containing proteins from the incompatible
wild-type (data not shown). However, none of these bands
comigrated with mature cyt f (38-40-kD in our gel system) nor do they
migrate at a size expected for the cyt f precursor (43-45-kD).

63

 

Heme Stained Gel
-68
O‘ifl- d -45
P—v .
"H6 '—‘"""
O
-29
e
o - -12.4

Figure 18. Heme-stained SDS-polyacrylamide gel (lo-15% acrylamide
gradient, 4 M urea) of mutant and wild-type chloroplast proteins.
Pairs of wild-type (wt) and pm lanes represent independent
preparations of chloroplast proteins. Filled circles indicate the
location of heme-staining proteins in the wt: open circles indicate
heme-staining proteins in the mutant. A purified cyt b6/ f preparation
from spinach is included for reference. The positions of the mature
cyt f and its putative precursor in W are labelled "m" and "",p
respectively.

64

Immunoanalysis of Wild-type and Mutant Chloroplast Proteins

Immunological screening with a variety of different antibodies revealed
additional alterations for several other chloroplast proteins. Each
antibody was used to probe either both membrane and soluble
fractions of chloroplast proteins isolated from mutant and wild-type
tissues or. in a few cases. the membrane fractions only. Mutant lanes
were loaded with approximately 1.5-fold more protein than wild-type
lanes in the gels used for this immunoblotting analysis (Figure 19). All
western blots were performed on mutant and wild-type proteins in
both A/A and A/ C nuclear backgrounds. The same results were
observed in both cases and therefore only the A/ C data are presented

here.

Figure 20 is a composite figure showing the results of several
immunoblotting experiments probing plastid membrane proteins with
antibodies which are specific for three proteins of the cyt b3/ f
complex. Cyt f appears to be the only protein in this group which is
qualitatively altered in 111111. Cyt f antibodies reproducibly recognize a
protein in ml which migrates at the size expected for the mature cyt
f protein. Earlier analyses also showed that the fourth member of this
complex, cyt be, is also present, in wild-type quantities, in the mutant
chloroplast membrane fraction (Sears, personal communication)
Figure ZOB,C shows that the two other proteins of the cyt b6/f
complex (the subunit IV and the Reiske FeS protein) are present.

Like the cyt f protein, subunit IV is translated on chloroplast
ribosomes, but with no terminal extension. The FeS protein, however,

65

Stained Gel
Mem Sol

wt pm7 wt W7

 

Figure 19. Coomassie-stained gel of membrane and soluble fractions of
mutant and wild-type chloroplast proteins. A 10-15% SDS-
polyacrylamide gel containing 5 M urea was used to separate mutant
and wild-type (wt) membrane and soluble proteins.

66

Immunoblot
anti-cytf anti-FeS anti-SU IV
pm7 kD

    

M pm7 wt

 

'17

 

Figure 20. Immunoblot of mm? and wild-type membrane fractions
probed with antibodies against components of the cyt b5/f complex.
Panel A shows that am? plastids accumulate a protein which cross-
reacts with spinach cyt f antibodies at a size which is consistent with
that of pre-cyt f (44—kD). Both and mutant and wild-type (wt) plastids
accumulate proteins which cross-react with spinach cyt f (A), maize
Fe/S (B) and maize SU IV (C) at the predicted mature sizes. (SU IV is
not synthesized a high molecular weight precursor [Heinemeyer et al.,
1984].)

67

is translated on cytoplasmic ribosomes as a precursor with a proposed
transit peptide of 68 amino acid residues (Titggen et al., 1986;
Steppuhn et al., 1987). Although both proteins appear to be reduced
in quantity in m1, there are no indications of any significant
accumulation of unprocessed precursor to the FeS protein in either

121111 or the wild-type plastids.

Western blotting also shows that the 33-kD subunit of the oxygen
evolving complex (OEC-33) is found at the expected mature size and
does not appear to be lower in quantity in the mutant in comparison to
the wild-type (Figure 21). In contrast. the remaining two imported
proteins associated with this complex show qualitative differences in
the mutant compared to the wild-type. Antibodies raised against the
23-kD and the 16-kD components of the spinach OEC (OEC-23 and
OEC-16) cross-react with proteins of approximate sizes 28-kD and 20-
kD, respectively, in the mutant. while recognizing proteins of about
23-kD and 16-kD in the wild-type (Figure ZIB,C). In other systems, a
protein of about 32-kD recognized by the OEC-16 antibody has been
suggested to be a dimer (Dr. C. Jansson, personal communication).
However, the absence of a 16-kD monomer in 121111 suggests that the
cross-reacting 32-kD protein is not a dimer of 16-kD units.

Antibody analysis of a third group of chloroplast proteins, including a

number of those associated with either of the two photosystems in the
thylakoid membrane, is shown in Figure 22. The D1 protein of the PS
11 complex is the one other chloroplast protein known to be translated

on chloroplast ribosomes as a precursor (Greenbrier et al., 1978:

68

Immunoblot
anuosc-sa antl-OEC-23 anti-osc-te
wt pm7 ‘ wt pm7

 

 

Figure 21. Immunoblot of 9191 and wild-type membrane fractions
probed with antibodies against components of the OEC subcomplex of
PS 11.

69

Riesfeld et al.. 1982; Marder et al.. 1984: Minami and Watanabe,
1985). A Dl-specific antibody detected no mature-sized product (32-
kD) in the mutant. but instead a possible degradation product around
23-kD (panel A). Panel B shows an immunoanalysis using a monoclonal
antibody which was made against the imported 29-kD (minor)
apoprotein of the barley light harvesting chlorophyll complex (LHC)
for PS 11 (Darr et al.. 1986). The analogous protein in W
appears to be accumulated in lesser amounts in the mutant than in the
wild-type, but at the expected size for the mature protein product.

The chloroplast-translated 43-and 47 -kD chlorophyll-binding PS 11
proteins. as well as the 22-kD non-chlorophyll-binding core protein of
PS 11 (Ljungberg et al.. 1984, 1986: Ghanotakis et al.. 1987a) and a 28-
kD component of PS 11, proposed to be a chlorophyll-binding core
protein (Ghanotakis et al., 1987b), are all recognized by a single
antibody preparation made against the purified 22-kD protein from
spinach (panel C). Inspection of panel C in this figure indicates that
the only one of these proteins appears to be either altered or deficient
in the mutant - the 28-kD component of PS 11. This protein. as well as
the 22-kD PS 11 protein, are each thought to be products of nuclear
genes and imported into the chloroplast (Dr. C. Yocum, personal
communication). The upper two proteins in panel C probably
represent the W analogs of the spinach 47-kD and 43-kD
chlorophyll proteins of PS II, which are recognized by this antibody
(Dr. C. Yocum, personal communication). A protein of approximately
45-kD surprisingly appears to be greatly increased in quantity in 12211

70

Immunoblot

anti-Di mil-LHC ll-29 anti-PS ll-22 anti-PSI
wtpm? wt‘vpm7wtpm7wtpm7 kD

 

-45

.. .0
.~ 0

A B C D

Figure 22. Immunoblot of pm7 and wild-type membrane fractions
probed with antibodies against components of PS 11 and PS 1. Open
circles indicate the absence in M of a protein at that position: filled
circles indicate the presence of a protein in 12131 at that position.
Circles on the left margin are indicators for panels A and B; circles on
the right margin are indicators for panels C and D.

71

in this gel but this quantitative relationship was not consistently
observed in other immunoblots with this antibody.

Also depicted in this figure is the result of immunoblotting analysis
with polyclonal antibodies for the spinach P700 chlorophyll-binding
reaction center protein of the photosystem I (PS I). the gene for
which is encoded on chNA (panel D). Like the 29-kD LHC-II, this
protein appears to be reduced in quantity in the mutant but, migrates
at the mature. wild-type size. Further deficiencies in other
unidentified cross-reacting proteins in pml may represent the lack of
additional chlorophyll-binding subunits of PS I.

All of the proteins discussed thus far are ultimately located in the
thylakoid membrane itself or in the lumen of the thylakoid membrane.
Figure 23 shows the results of an analysis of a number of
cytoplasmically-translated chloroplast proteins which, following
import into the chloroplast, are eventually destined for the stroma.
Two of these proteins, the L21 subunit of the chloroplast ribosomal
complex (Schmidt et al., 1984, 1985) and the soluble stromal enzyme,
carbonic anhydrase (Jacobsen et al.. 1975: Reed. 1979; Tsuzuki et al.,
1985), both appear to be present but at reduced levels in the mutant.
Both subunits of another enzyme normally found in the stroma. ADP-
glucose pyrophosphorylase (PYRO) (Krishnan et al., 1986) were also
found at the expected mature sizes and did not appear to be greatly
lowered in abundance in the mutant.

72

Immunoblot

antl-PYRO anti-CA

wt pm7 M me kD
- _ - 68

  
  

-45

-12.4

Figure 23. Immunoblot of pmz and wild-type membrane (A) or soluble
(B,C) fractions probed with antibodies against a ribosomal protein and
two stromal proteins. Both mutant (pm7) and wild-type (wt) plastids
accumulate proteins which cross-react with antibodies against

hl n chloroplast ribosomal protein L21 (A), spinach ADP-
glucose pyrophosphorylase (PYRO, B), and spinach carbonic anhydrase
(CA, C) at the predicted mature sizes.

73

A summary of the combined results of these immunological screens is
provided in Table 2. This table indicates the location of the genes for
these proteins in nuclear or chloroplast DNA and also shows the final
suborganellar destination of the proteins themselves. As well, the
range of sizes for the mature proteins in a variety of plants and similar
"generalized" sizes for the precursor proteins, if appropriate, are

given.

Immunoanalysis of a Putative Revertant of 121111

A possible revertant of the 121111 plant line was isolated by Dr. Linda
Schnabelrauch by propagating tissue containing the mutated pmz
plastome in a homozygous mm nuclear background and in leaf-tip
culture until green revertant sectors were observed. These sectors
were separated and subcloned until sufficent material was accumulated
to allow protein purification from the putative revertant. (The
revertant is referred to as "putative" because, although the tissue is
green, it does not appear to be fully photosynthetically competent. 1
Although roots can be induced, attempts to transfer the plant material
to medium lacking sucrose or to a soil medium for growth in the
greenhouse have not been successful (Sears, personal communication).
Figure 24 presents the results of an immunoanalysis of these proteins
with three of the antibodies which showed alterations in the analysis
of the m1 mutant: anti-cyt f. anti-OEC-23 and anti-PS 11-22 (Figures
19, 20 and 21, respectively). Cytochrome f is a representative of a
class of chNA-encoded proteins made as precursors for which the

processing is altered in will: OEC-23 is a member of a class of

74

Table 2. Summary of immunological analysis of 11ml plastid proteins.
Gene locations in chNA or nuclear DNA (nucDNA) are given. The
final location of the protein product in the plastid is also specified.
Protein sizes refer to range of apparent molecular weights known for
these proteins from various plant species. Proteins found in the
mutant as cross-reacting polypeptides which are larger in size than
the mature wild-type proteins are indicated by "".> The presence of
mature-sized cross-reactive material in 121111 is signified by "".+ The
total absence of any signal at the mature size or greater is shown by "-.
Three antibodies were tested but did not cross-react with 91m
wild-type or pm7; the results of these analyses are denoted by a " ?."
In some cases. mutant proteins were localized to the membrane (M)
or the soluble (S) fraction of the chloroplast. The analyses of cyt b6,
RubisCO LSU (large subunit) and chloroplast ATP synthase (cpATPase)
subunits were included in a preliminary analysis performed by Dr.
Sears. Protein abbreviations are given in the text. Where possible,
comprehensive reviews are given as references.

Protein Gene Protein Precursor Mature Refer- m1 m1

 

location location size size ence a pheno- loca-
type tion
cyt f chNA thylakoid 38-43 33-38 1 > M
cyt b6 chNA thylakoid -- 20-23 1 + -
b6/f-SU IV chNA thylakoid -- l 5 1 + M
b6/f-FeS nucDNA thylakoid 25-26 1 9 2 + M
OEC-33 nucDNA lumen 38-39 33-34 2 + S/ M
OEC-23 nucDNA lumen 33(28) 23 2 > M/ S
OEC- 16 nucDNA lumen 26(?) 1 6- 1 7 2 > M/ S
PS II-D 1 chNA thylakoid 34 32 1 - M(wt)
PS 11-5 1 chNA thylakoid -- 49-5 1 1 + M
PS II-44 chNA thylakoid -- 4 1 -46 1 + M
PS II-22 nucDNA? thylakoid 32? 22 3 + M/ S
PS II-28 nucDNA? thylakoid ? 28 3 - M / S(wt)
LHC II-29 nucDNA thylakoid 32-33 28-29 2 + M
PS I chNA thylakoid -- 69-70 1 + M
cpATPase-achNA thylakoid -- 58-59 1 + -
cpATPase-B chNA thylakoid -- 56-57 1 + -
PC nucDNA lumen 1 7 1 1 2 ? ?
RubiscoLSU chNA stroma 53-58? 52-56 1 + S
RubiscoSSUnucDNA stroma 20-2 1( 1 8) 1 4- 1 6 4 ? ?
CA nucDNA stroma ? 32-33? 5 + M/ S
PYRO nucDNA stroma 73/ 76 5 1 / 54 6 + S
rp-L2 l nucDNAcp-ribosome 1 7 .4( 16.3) 14.6 7 + . M
rp-L29 nucDNAcp-ribosome 14.4 12.2 7 ? ?

a 1. Crouse et al., 1985; 2. Titggen et al., 1987: 3. Dr. C. Yocum,
pers. com.; 4. Schmidt and Mishkind, 1986; 5. Dr. M. Volakita,
pers com.; 6. Krishnan et al., 1986: 7. Schmidt et al., 1984,1985.

75

Immunoblot

pm+lpm pm+lpm pm/pm pm/pm
p7m pm7 pm7-rev

-27

anti PSll-22 ‘
-21

I
i
1 ' 44
anti cyt f . -40

C

-28

- -
antiOEC-23
.‘123

Figure 24. Immunoblot of mm? and wild—type chloroplast proteins in
stable and unstable plastome mutator nuclear backgrounds probed
with antibodies against PS II-22, cyt f and OEC—23. Plastid proteins
were isolated from wild-type (wt, lane 1) or mutant (lanes 2,3) in the
stable (pm_+/m) nuclear background (lanes 1.2) or the unstable
(pm/pm) nuclear background (lane 3). Lane 4 contains plastid
proteins isolated from a putative (green) revertant of pm7 in the
unstable (pm/pm) nuclear background. Immunoblots were probed
with antibodies against PS-22 (A), cyt f (B), or OEC-23 (C). In all cases
the revertant material shows a protein pattern similar to that of the
wt.

76

nuclear-encoded proteins made as precursors and then imported into
the chloroplast. the processing of which is also affected in 121111; PS
II-28 is a representative of a class of nuclear-encoded proteins which
is absent in 111111. All three antisera tested gave similar results in that
the protein pattern seen in the mutant analysis has returned to a wild-
type pattern in the putative revertant. In contrast to previous results
(Figure 21) with the OEC-23 antiserum. the mutant lanes appear to
contain a small amount of mature- sized protein. This would not be
surprising in the homozygous background, as it is possible that some
reversion may have already occurred. However, this protein was not
reproducibly observed in either background and probably represents
spillover from the neighboring wild-type or revertant lanes in this

particular gel.

77

W

Initial analysis of 11211 showed an accumulation of a polypeptide which
was about 5-kD larger than but also cross-antigenic with cyt f (Figure
13). A number of mutational lesions could give such a result, but in all
cases. at least part of the cyt f reading flame must be intact since the
protein is recognized by cyt f antibodies. This protein may represent
an unprocessed precursor to cyt f. due to a lesion in the gene itself or
due to a defect in the chloroplast protein processing machinery. For
example. a lesion in the upstream regulatory region of the gem gene
could potentially create an 'up' mutation. resulting in the
overexpression of the gene. Alternatively, the lesion could be a large
in-flame insertion in the gene or an insertion or deletion which fuses
cyt f to another reading flame.

A number of lines of evidence including the lack of any strong
transcriptional start sites and the presence of an ORF approximately
200-bp upstream suggest that the gem gene may be transcribed as a
polycistronic message. Altered RNA processing of this message and
the aberrant use of an upstream initiation codon could conceivably lead
to a larger than normal PLEA gene product. The lesion in 91:11 could
also be modified so as to allow read-through of the stop codon which
normally terminates translation of the pm message. for example.
through the use of an altered tRNA. This could result in a protein of a
size greater than wild-type. The molecular analyses described in this
chapter were designed to explore the possibility that the primary

78

lesion in 9.1111 was located in or near 953A. the gene encoding cyt f.
Subsequent analysis of the mutant phenotype at the level of various
proteins addressed the specificity of the m1 lesion for cyt f.

The possibility that a sizable insertion or deletion could have created a
major alteration in the reading frame of DSIA was addresssed through
the use of restriction endonucleases. Analysis of chNA restriction
patterns flom the mutant and wild-type Bam 3a region showed no
difi'erences between the two regions (Figures 14) and thus, no
evidence of major insertions or deletions in the pm gene.
Furthermore, the existence of many termination codons in the two
alternate reading flames (Tyagi and Herrmann. 1986) renders the
possibilities a flame-shifi mutation or a change in the stop codon
unlikely.

Translation from an upstream initiation codon resulting in a pm gene
product of increased size is improbable since the potential upstream
start codons are each closely followed by a stop codon, and thus, would
not yield a protein product of the size seen in 11ml. Furthermore. as
mentioned above, premature initiation of translation would have to
maintain the normal reading flame of cyt f in order to give cross-
antigenic products. It is also possible that altered splicing of the
upstream portion of the the gem message could subsequently affect
the protein processing site and lead to the accumulation of a large
mutant cyt f protein. The merits of this possibility are addressed
below in the context of the analysis of the m1 revertant.

79

The results discussed thus far have addressed the question of whether
various factors afi'ecting Dem gene expession. acting in either a 'cis' or
a 'trans' fashion, could be responsible for the accumulation of an
unusually large 11ng gene product in 111111. Each of these factors,
including insertion or deletion events in the 031A gene and alterations
to components of the chloroplast translational apparatus, such as an
altered tRNA which permits read-through of the normal termination
codon may be ruled out based on the results of the restriction analysis
of the region containing the gem gene and the examination of the
sequence for the 99291112123 pm gene itself. The remaining viable
hypotheses all involve the possibility that the aberrantly-sized cyt f m
the mutant is an unprocessed precursor of cyt f. Perhaps the
precursor is not processed as a result of a mutation which directly
alters the DNA encoding the processing site or the heme-binding site
of cyt f. Alternatively, factors other than the gene encoding cyt f
could be altered and consequentially interfere with proper processing
of cyt f. These may be other components of the cyt be/ f complex.
such as the heme group or other protein subunits. or perhaps the
processing apparatus itself. If some factor other that the gem gene is
responsible for the altered cyt f in 121111. then the processing of other
proteins besides cyt f may also be altered. This, it is worthwhile to
investigate whether the processing of other chloroplast protein
precursors is also affected in 111111.

Initial investigations focused on the confirmation of the identity of the
large mutant cyt f protein as a cyt f precursor. Unfortunately, the
attempts to prepare an antibody which specifically recognizes the

80

precursor region of cyt f by immunizing rabbits with a conjugate of
KLH and a synthetic peptide derived from the cyt f presequence were
unsuccessful. It is possible that the synthetic peptide created flom
the deduced amino acid sequence flom wild-type mm cyt f did
not maintain the necessary epitope(s) which are found in this region
in the "intact" precursor. Furthermore. amino acid analysis had shown
that the synthetic peptide either contained an error or was a mixture
of two peptides. since an unexpected arginine residue was found to be
present. Addition of this highly basic residue may have significantly
affected structure and maintenance of the epitope(s) and thus the
peptide antibody may not recognize the cyt f precursor. In spite of the
neagtive results with this antibody. the close correlation in the sizes of
the abberant cyt f in the mutant and the proposed size of the
W precursor. based on the deduced amino acid sequence of
the wild-type W cyt f sequence would argue that they are
identical. Furthermore, the existence of a putatve ribosome binding
site at the correct position, (Tyagi and Herrmann. 1986) favors the
identification of the large cyt f protein in the mutant as an
unprocessed precursor to cyt f. (The possibility remains that altered
mRNA splicing may be involved. This will be discussed further below.)
The primary gene product of M was determined to be about 35-kD
by the immunoprecipitation of m jam-translated poly A“ mRNA flom
pea (Willey et al.,1984a), wheat (Willey et al.,1984b) and spinach (Alt
and Herrmann. 1984). This compares to a mature cyt f of
aproximately 3l-kD in each of these plants (Willey et al.,1984a,b; Alt
and Herrmann, 1984), indicating that about 4-kD of pre-cyt f is
removed during processing. This size closely agrees with the 4-kD

81

difl'erence in size between the cyt f species in 111111 and wild-type
W. Since the gem sequences are colinear (and
approximately 90% homologous) with the gem sequences from the
aforementioned plants. the size of the large cyt f product is consistent
with the proposal that it is a precursor to cyt f.

A sequence change in the area encoding the protein processing site.
the heme binding site or an upstream regulatory region could
conceivably be responsible for the accumulation of a precursor in 111111.
To test this hypothesis. critical regions of 11:15 were sequenced flom
both the mutant and wild-type chNA. The results shown in Figure 16
indicate that m is not likely to be the site of the primary lesion.
since no differences were found between the mutant and the wild-type
genes in the region encoding the processing site and the heme-
binding site, ruling out the possibility that protein processing is
abnormal because of an alteration in the processing site or the site at
which the heme group is covalently bound. The lack of any DNA
differences as far as 450-bp upstream of the 1111A gene argues against
the possibility that a change in the 5' regulatory region results in
overexpression in 121111. Since no good transcriptional start sites were
found in this region in the mutant or the wild-type, it is possible that
such an 'up' promoter mutation could be located 5' to the upstream
ORF or could even be in a trans-acting regulator. However, results of
the immunoanalysis discussed below make the possibility of
transcriptional overexpression of the gem gene in 9311 improbable as
the primary defect and thus. the point mutation in 111111 is most likely
elsewhere in the chloroplast genome. The results of this sequence

82

analysis are sumcient to rule out several of the hypotheses proposed
for the aberrant cyt f phenotype in 111111. Since no differences were
found between the mutant and wild-type 931A sequences in the
regions encoding the processing site or the heme binding site, if
altered processing of pre-cyt f occurs in the mutant it is not due to
changes in these sites. As well. the absence of mutations within the
region approximately 500-bp upstream of the gene gives no evidence
for altered regulation of 12:15 expression.

Besides proper protein processing, the maturation of cyt f requires
the covalent binding of a heme group. DNA sequencing showed that
no differences existed in the DNA encoding the heme binding site of
cyt f. Results of heme staining of plastid proteins indicated that wild-
type plastids in the compatible A/A nuclear background (Stubbe.

1959) contain heme proteins at the expected size for cyt f and (much
more faintly) at the predicted size for cyt be, which contains a non-
covalently-bound heme group (Haehnel, 1984). In the mutant, heme
was present but not bound to proteins of sizes expected for either
mature cyt f nor precursor cyt f (Figure 18). Wild-type plastids in a
less compatible A/ C nuclear background (Stubbe, 1959) sometimes
contained the aberrantly-sized heme proteins as well (data not shown).
Trivial explanations could explain the aberrant sizes of these heme-
staining bands in 111111. One possibility is that the chloroplast
preparations were contaminated with other organelles which contain
heme-staining proteins. For example, mitochondrial cytochromes or
peroxisomal catalases would, if present, produce a positive TMBZ stain
(Thomas et al.. 1976). However, it is unlikely that the relatively low

83

speed centrifugation (6000 x g. 5 minutes) used to pellet the
chloroplasts would pellet the significantly smaller mitochondria and
peroxisomes. In fact, 25,000 x g (30 minutes) is required to pellet
mitochondria flom W (Dr. B Sears, personal communication).
Additionally, Western blotting experiments with an antibody raised
against yeast cytochrome oxidase subunit II and succinate
dehydrogenase activity assays showed no evidence of mitochondrial
contamination in 213.1 plastid preparations (data not shown). whereas
enriched spinach mitochondrial preparations gave positive results.
Furthermore a native gel stain specific for the presence of catalase.
failed to indicate peroxisomal contamination of m1 plastids when
compared to the positive staining of purified catalase (Sigma Chemical
Co.) (data not shown). These particular heme-staining bands were
probably the result of occasional incomplete solubilization of the
mutant or incompatible wild-type protein samples or perhaps the
association of flee heme with more abundant chloroplast proteins.
Having ruled out the possibiltity that the aberrant heme-staining bands
were due to contamination of the plastid flaction, I have concluded
that heme is indeed present in the mutant plastids. although it is not
associated with proteins of sizes expected for mature cyt f nor its

precursor.

There are other reports of examples of abberant sized cytochrome
proteins in chloroplasts. In one case (Bhaya and Castelfranco. 1986),
pea chloroplasts were fed radioactive precursors of heme biosynthesis
which were incorporated into a 43-kD protein. These results were
interpreted to indicate the existence of a novel chloroplast

84

cytochrome, as opposed to a cyt f precursor. since no cross-reactivity
occurred with a cyt f antibody (Dr. P. Castelflanco, personal
communication). Allred and Staehelin (1986a) also reported the
presence of abnormally sized heme-staining bands in pea which they
attributed to pools of cyt f which remained unassociated with the cyt
b6/f complex. In this case, the size observed was too small to be the
unprocessed precursor of cyt f (Willey et al., 1984a). Possibly pea
chloroplasts process cyt f to an intermediate prior to the processing
which produces a mature product. However, the emstence of a totally
unprocessed cyt f appears to be an observation unique to the m1

mutant of Qenaihera hookeri.

In contrast to the complex protein patterns resulting flom non-
discriminating protein stains such as Coomassie Blue, immunoanalysis
provides a means by which one may specifically determine which
proteins are affected in the mutant chloroplasts. In this case not only
the presence or absence of a protein in 111111 was informative. but also
whether any precursor forms of the proteins accumulated in the
mutant. The results of the Western blotting experiments, summarized
in Table 2. allow the classification of these chloroplast proteins into
three categories, those which were present at expected mature sizes
in both the mutant and the wild-type, those which were absent in the
mutant chloroplasts and those which were found in the mutant at sizes
larger than expected for the mature wild-type polypeptides. Table 2
also shows that there were proteins in each category which are known
to be the products of chNA genes as well as proteins which are
known to be imported from the cytoplasm. Since the immunoblotting

85

analyses were performed on material which had been separated into
membrane and soluble protein flactions, further conclusions may be
drawn as to the intraorganellar location of the mature or precursor-
sized proteins in M. All proteins analysed which were present in
ml were found in the flaction in which they would be expected to be
found in the wild-type plastids. In some cases, as with carbonic
anhydrase and all of the imported subunits of PS II, proteins were
found in both membrane and soluble flactions. This indicated that
either cross-contamination of the two fractions had occurred. proteins
were located in both flactions in the mutant or perhaps they were only
loosely associated with the membranes and were easily 'lost' to the
soluble flaction. The latter possibility is more likely since other

membrane proteins were found only within the 'membrane flaction.‘

In contrast to cyt f, the proteins in the mutant which cross-react with
the OEC-23 and OEC-16 antibodies are intermediate in size between
the mature size and the presumed primary precursor size. One
explanation for these aberrantly sized proteins in m1 is that they
represent intermediate processing products. Similarly, in mg, a
non-Mendelian mutant of tobacco. proposed intermediate processing
products of the OEC-34 and the OEC-23 proteins have been shown to
accumulate (Chia and Amtzen, 1986). However, 1111;; appears to
process OEC-16 and cyt f normally. cDNA clones for genes encoding
each of the three proteins in this sub-complex of PS 11 have since been
isolated flom spinach (Tittgen et al.. 1986) and the DNA and deduced
amino acid sequences for all three have been determined (Tyagi et al..
1987: Jansen et al.. 1987), along with the actual N-terminal amino

86

acid sequences for OEC-23 and OEC-16 (Vater et al.. 1986). Amino-
terminal transit peptides of approximately 5-kD for OEC-34 and lO-kD
for OEC-23 and OEC-16 are indicated by the sequence analyses, as
well as by the sizes of the proteins immunoprecipitated flom 1;; ELILQ
translation products for these proteins (Westhoff et al.. 1985).

The most unexpected results of the immunological analysis of 11311 are
those which show that three difi'erent chloroplast protein precursors
accumulate in the mutant. In contrast. the results showing reduced
amounts of other chloroplast proteins are less surprising in such a
chlorotic mutant. Similarly reduced levels of a number of chloroplast
proteins have quite often been seen in other non-green plastome
mutator-induced mutants (Sears. personal communication). In
addition, the D 1 protein, which is one of the two proteins missing in
ml, is known to turn over rapidly in chloroplasts (Mattoo et al.,
1984, Wettem and Galling, 1985: Greenberg et al.. 1987). One could
have possibly anticipated the rapid degradation of any unprocessed
precursor of this protein (conceivably, the faint cross-reactive material
seen in the mutant in Figure 22A). A similar speculation could be
made for the other chloroplast protein which appears to be
completely depleted in 12ml, the 28-kD chlorophyll-binding core
protein of PS 11. however much less is known about the stability of this
protein. It is consistent with the chlorophyll deficiency of this mutant
that a chlorophyll-binding protein is absent flom the plastids. On the
other hand, other chlorophyll-binding proteins (LHC-II-29, PS I, etc)

are still present but in lesser amounts in the mutant than in the wild-

type-

87

A number of the proposed hypotheses for the cyt f phenotype can be
ruled out by the results of these immunological analyses. The
observation that altered protein patterns of both a chloroplast-
encoded gene product (cyt fl and a chloroplast protein imported from
the cytoplasm (OEC-23) have retumed to wild-type patterns in a
putative revertant of m (Figure 24) links these traits to a single
mutation which affects the processing of particular subset of
chloroplast proteins. Furthermore, the presence of the PS 11 22-kD
protein in plastids of the revertant suggests that the 11ml mutation
also afi‘ects the accumulation of other chloroplast proteins. Therefore.
the lesion in m must affect some process which is not unique to cyt
f. This reasoning. together with the gem DNA sequence analysis.
suggests that the aberrant accumulation of a number of putative
chloroplast protein precursors in pm], is not due to a lesion in the
gem gene itself. As well, the possibility that. the altered cyt f
phenotype is due to improper processing or translational termination
of the gem message is unlikely, since the mRNAs for the OEC
polypeptides are cytoplasmically processed and translated, whereas
cyt f is completely synthesized within the chloroplast.

The hypothesis that maturation of cyt f could have been afi'ected
because of the absence of other subunits in the cyt b6/ f complex is also

not supported by these immunoblotting results. All other members of
the cyt b6] f complex were present in mi, although the quantity of
these proteins may have been significantly reduced in the mutant.
However, even if the quantitative reduction in subunit IV and the FeS

88

protein were important to the processing of cyt f, it is not clear how
or why this should affect the processing of OEC-23 and OEC-l6.
Furthermore. the third member of the PS II oxygen-evolving sub-
complex. OEC-33. was present in apparently normal amounts. These
results do not completely rule out the possibilty that one of the
proteins which does appear to be totally deficient in 91111 is critical to
the processing of other chloroplast proteins. However, such an
explanation requires that the complete development of the chloroplast
thylakoid membrane is a highly intergrated process. and the
disruption of one critical component of this membrane may have a

deleterious efi‘ect on other components of the maturation process.

Thus, the remaining hypothesis seems the most likely: some essential
element of the processing function for chloroplast proteins must be
defective in m and furthermore this processing function is shared
by certain chloroplast and nuclear gene products. This element could
be a component of the processing machinery itself or some other
factor which regulates the maturation of this subset of chloroplast
proteins. The question remains as to what features are shared by
these three proteins (cyt f, OEC-23 and OEC-l6) so that they display
such a related response to the mutational lesion in mnz despite the
difi'erences in gene locations, protein functions and ultimate
destinations in the chloroplast.

These three proteins (cyt f. OEC-23 and OEC-16). are similar to a
number of other chloroplast proteins in that they must enter or

traverse the thylakoid membrane in order to reach the sites where

89

they ultimately function in photosynthetic electron transport. It is
possible that a thylakoid-bound protease similar to the one recently
proposed to process a PC intermediate (Hageman et al.. 1986) may
serve to process this subset of thylakoid and lumen proteins. The
simplest explanation for the apparent abberant processing in 111111 is a
deficiency in this protease activity in the mutant. Alternatively.
perhaps this protease activity is sensitive to the status of the
membrane in which it probably resides and functions.

A thylakoid-located protease activity has also been proposed for the
processing of the D1 protein (Marder et al. 1984). In contrast to the
other proteins discussed here. this processing is thought to occur
flom the carboxy-end. However. little information is available for the
biochemical nature of this protease.

Like D1 (Sayre et al.. 1986), cyt f must move flom the chloroplast
ribosomes to become at least partially embedded in the thylakoid
membrane where it associates with other components of its protein
complex (Schmid et al., 1976; Cox and Andersson, 1981; Shaw and
Henwood. 1985: Manfield and Andersson, 1985; Allred and
Staehelin. 1985). According to a structural model based on the
analysis of the deduced amino acid sequence for cyt f (Willey et al..
1984a.b: Alt and Herrmann, 1984: Wu et al.. 1986: Mg and
Herrmann. 1986), the majority of the protein resides in the thylakoid
lumen with only one membrane-spanning region and a short C-
terminal stretch located in the stroma. Other models based on
primary sequence analysis for OEC-23 and OEC-16 suggest that these

90

extrinsic proteins are ultimately locatedin the lumen of the thyakoid
membrane (Jansen et al.. 1987). Therefore, it is conceivable that
these three proteins are all subject to processing within the thylakoid

membrane.

In this context. it is worth noting that along with the precursor-sized
cyt f which accumulated in the mutant membranes. a significant
amount of mature-sized cyt f was reproducibly found in in the
membrane flaction of the m1 plastids. This suggests that the m
phenotype may be 'leaky' with respect to the processing of cyt f.
perhaps because the cyt f precursor in 111111 is still recognized by the
protease. although the processing activity is less efficeint than in wild-
type plastids. The observation that both the presumed precursor and
the presumed mature protein are located in the membrane fraction of
m is strong support for the suggestion that pre-cyt f is processed by
a membrane-localized protease. It is also clear that processing is not

an essential pre-requisite for the insertion of cyt f into a membrane.

Extensive analysis of protein uptake and processing by organelles has
shown that transit peptides. and presumably sequences and/ or
structures within them, play a critical role in correct targeting, uptake
and subsequent processing of various imported proteins (Neupert and
Schatz, 1981: Sabatini et al., 1982: Hay et al.. 1984: Schmidt and
Mishkind, 1986). Although little homology is seen in primary amino
acid sequences for complete transit peptides for many imported
chloroplast proteins, a common amino acid flamework for a

"generalized" transit peptide has been proposed (Karlin-Neumann and

91

Tobin, 1986). These authors suggest that short "blocks" of shared
homologr at the beginning. middle and end of the presequence
represent regions which are important to proper protein uptake,
processing and / or localization. Other researchers have reported little
evidence of homology among various chloroplast and mitochondrial
transit peptides (Schatz and Butow, 1983; Coleman and Robinson.
1986; Hurt et al., 1986).

An alternative model for transit peptide function has been developed
based upon the apparent conservation of amino acid structural
"domains." In chloroplasts and mitochondria, the N -terminal region of
the transit peptide often contains a short stretch of predominantly
basic residues. In a number of imported chloroplast proteins. amino
acid sequences are capable of forming B-pleated sheet structures at or
near the site proposed for the initial cleavage event (von Heijne. 1983;
1984 a.b,c; Raise et al., 1988). This event results in a processing
product which is intermediate in size between the precursor and the
mature protein. and would be the first stage in a two-step processing
system. In addition. each transit peptide sequence contains an
internal stretch of 15-20 hydrophobic residues in the C-terminal half
of the presequence, immediately proximal to the terminal processing
site (van Loan and Schatz, 1987; Singer et al., 1987). This sequence
has been proposed to function as a "stop-transfer" signal for transport
across the inner mitochondrial membrane and the thlyakoid
membrane in chloroplasts. According to this theory, this hydrophobic
stretch of amino acids could serve to anchor the protein in the

membrane, as the precursor is being threaded through the membrane.

92

Alternative theories assume that the precursor moves across the
membrane in a globular form, through membrane channels,
independent of interaction between the membrane and the

hydrophobic regions of the transit peptide.

A two-domain model for two-step processing of proteins which are
imported into the chloroplast has been proposed, based upon recent
work with the chloroplast protein plastocyanin (PC) (Smeekens et al..
1986; Hageman et al., 1986). This model shares a number of
features with models for cytoplasmically translated mitochondrial
proteins. To reach its final destination. the initial precursor to PC
(pre-PC) has to cross a total of three chloroplast membranes: the
outer and inner membrane of the chloroplast envelope and the

thylakoid membrane.

The model proposes that the initial processing step occurs during or
following the uptake of the precursor (mediated by the 'chloroplast
import domain') via a specific protease located in the stroma. This
step results in an intermediately-processed protein which still retains
a flaction of the presequence. the "thylakoid transfer domain"
(Hageman et al.. 1986). Subsequently, the complete processing of the
intermediate to the mature-sized protein is accomplished by a second
protease located in the thylakoid or perhaps in the lumen itself
(Hageman et al.. 1986). This model does not address the issue of
where the addition of particular prosthetic groups such as the copper

in PC occurs.

93

A certain potential for regulatory control is provided for in the model
as the thylakoid-located protease activity appears to be specific for the
intermediate and does not recogonize the primary precursor of PC in
111 21119. experiments (Hageman et al.. 1986). Preliminary analysis of
this protease also indicates that the protein is extractable flom the
thylakoid with gentle detergent treatment and once freed from the
membrane, the water soluble protein retains its specific proteolytic
activity (Hageman et al., 1986). The example of PC processing is
especially applicable to the results presented here since like the OEC
polypeptides, it is also ultimately directed to the lumen of the
thylakoid (Haehnel, 1 984).

A prediction of the amino acid residue(s) implicated in the
intermediate processing events of OEC-23 and OEC-16 has been made
based upon the potential ability of these "cleavage domains" to form [5-
pleated sheet structures (Jansen et al., 1987). The theoretical sizes
proposed for the intermediate processing products are in close
agreement with the sizes of the proteins in ml which cross-react
with OEC-23 and OEC-16 antibodies. Figure 25 shows the deduced
amino acid sequence for the transit peptides of spinach OEC-23 and
OEC-16 (Jansen et al., 1987) and the presequence of cyt f . The OEC-
33 presequence fl‘yagi et al., 1987) is included for comparison since
it is evidently processed properly in 91111. The comparisons are
presented with the N-terminal residue of the mature sequence of cyt f
(+1) aligned with the N-terminal residues of the mature sequences of
the imported proteins. Thus, the sequences of the transit peptides

94

flom the imported proteins extend farther than does the presequence
of cyt f. Visual inspection of these sequences (Figure 25) suggests that
with the possible exception of the arginine residue directly preceding
the conserved hydrophobic region and the alanine residue at - 1. no
extensive homologr exists between these sequences. Computer-
assisted searches for homologr between cyt f and these imported
chloroplast proteins were performed (Microgenie Software Systems,
Beckman). These allowed consideration of differences in. length of the
presequences, and also of residues which are not identical but rather
conserved with respect to charge or polarity. However, the results of
these analyses indicated that even in the more restricted cases. cyt f,
OEC-23 and OEC-16 are no more similar to each other than they are
to OEC-33.

A possible pattern is evident however, if one compares the
hydrophilicity plots of the C-terminal portions of the OEC-23 and
OEC-16 transit peptides with that of the presequence of cyt f (Figure
26). Comparison of these plots reveals that all of these peptides share
a hydrophobic stretch near the C-terminal of the peptides. This
hydrophobic sequence has been proposed to serve as a possible signal
for transport across the thylakoid membrane (Singer et al.. 1987).
Although the general structural similarity is intriguing, again, it does
not distinguish these three proteins flom other proteins which also
contain N-terminal transit peptide extensions.

Antibodies raised against spinach PC did not cross-react with any
angmgm proteins which migrated at sizes expected for the prePC or

95

Figure 25. Amino acid sequences of the N-terminal extensions of the
precursors of cyt c1, cyt f. and the components of the OEC
subcomplex of PS II. Sequences were deduced from the published
DNA sequences for yeast cyt c1 (Sadler et al.. 1984). spinach OEC-23.
OEC-16 (Jansen et al.. 1987). spinach OEC-33 (Tyagi et al.. 1987) and
Oenothera cyt f (this work). The presequences are aligned at the
presumed site of the terminal processing event (residue -1).
Hydrophobic stretches are double-underlined: charged resudues
preceding the hydrophobic stretches are marked with an asterisk:
conserved alanine residues at -1 are boxed: regions capable of forming

B-pleated sheet structures are single underlined.

96

 

 

Figure 25.

----84 -83 -82 -81 -80 -?0 -?8 -7? ~76 -75 -74 -73 -72 -7l -70 -00 '60 -6?
cm: '

CYTF

cacao: mikSEH-W-LEU-Hls-Hls-Hls-WILE-Stfl-

 

III”6--flET4IJFOLN4IJFflET4IJFfifififiiflhfiUfidlflhLEU4lifflLVdflJkflEfldllkflLfl-
0EC3341fl3flLfl4IJPGEH4uflP0LN4IJF8EHifliFWHH4HiiiEUHGU+4WIF¢H34UW§4llrfiLfi-

-OflI-65 -64-4N3-£E!-0fl -60 ‘50-4fli-{fil-50 ~55'4it-4El-52 ~51-4Il-40 -48
flET4flEk8EH4iiFLEU4fl5t1NSfiHIPJHPfiIJFflLHfiHIPJHB—

 

 

-6£fidHIFflLfldIJFOLVdUI¥flLVdfl5FflLfl4IJFOLNfliIPWVfidIJFfiLflfifllrfiifl-Hjfitfls-
-UfibfiJflFGLUdldflfiEfidJlbflLfl-HJk6ER4INL6Efi4iiFflBBdUflbGEfidlfiflPfiO‘flfFWHfl-
i4iEFflfl04iiFJHH4uaF0LN4JabflflBdiEFWHR4IJFflBH4WI:CVS%J£P£UR¥*EP0U¥4¥IF

4?-46-45-44-43-42-4l40-39-38-37-3-35-34-33-32-31-m-29
mmvs-m-m—sm—WMWWHWM-

 

 

'23 '27 -26 -25 '24 ‘23 '22 '21 '20 '19 '19 '17 '16 '15 '14 '13 '12 'll '10
“Jim-WW L‘W
'SEH-THP'lLE-L93-LVS-GLU-lLE-THH-HHGEQEH-lLE-SER-LEU-SEH-LEU-HET-lLE-TVR-ILE-
'4iEFGEH4IAhUHD4§§F€HG4¥U§LEU4IJF£EU—flHRdfilrLEU-Hi?flLVfliJFfiLfifllJ¥UflL-
‘GLU4IJFGLU‘niFfiEfldmifflfldhflJFflEffiJikflLV4*EkUthiJFflLfldlnflLEU4IJF623-
'411FflUfi4iikL9943EFUHL4mflhfiLfl‘niriflafiLEU4‘JFflLYdJflPflLfldUflPfiLfi‘flfFflEfl-

 

4-0-7-6-5-4-3-2-1+i+2+e+4+s+e+7+e+9+ro
-TVfl-HJi-flSP-SEB-LEU- - - - . ”F «Emma-amusem-
_na-ssn—rts ”wrufiWv-mu-
av-sen—rvs—m-w - - - E
W 3 '~‘ “ tum-mam

’2

' TVRdMJFGLUleFiLfidMiFUflLflflikflLV-

97

Figure 26. Hydrophilicity plots for the presequences of cyt f and the
C-terminal portions of the presequences for components of the OEC
subcomplex of PS II. Hydrophilicity values (Hopp and Woods, 1981)
have been plotted against the fraction of the length of the amino acid
sequence of this portions of the deduced sequences for cyt f (this
work)(residues -33 through-1), OEC-23 (residues -42 through -
l)(Jansen et al.. 1987), OEC-16 (residues -39 through -1)(Jansen et
al., 1987) and OEC-33 (residues -40 through -1)(Tyagi et al.. 1987).

Figure 26.

 

I U I f . I .
u...-3-ééiwéirtizitttwpbiéiéélééLszfiél
nulrmlmrhn unlimlmrlrm imllmlnujim rurlmrlrmluu unlmrluulim litlJJlllliillllili

J.

eeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee

 

 

eeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee

 

 

...........................................................

 

 

 

 

 

 

I

q—d ...........................................................

E

'E

‘1

I

d V

E

I

5

rd

E

: . ........ i‘c' .......................
ifilfifltil'tfl'ITfT

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.0 079 L0

fraction of length

CYT F

OEC 23

OEC 18

DEC 33

99

mature PC in either mutant or wild-type plastids. Antibodies specific
for D1, as shown in Figure 22A. recognize (very faintly) only presumed
degradation products in 121111. It is not known if this degradation of
D1 was preceded by precursor processing. So no conclusions may be
drawn as to whether the previously reported thylakoid proteases
which process pre-Dl (Mattoo and Edelman. 1987) and the PC
intermediate (Hageman et al.. 1986) are the same as those which
appear to be deficient in activity in m. However. previous reports of
such protease activities in the thylakoid are consistent with the
suggestion that the apparent lack of processing of certain precursors
in m could occur because this processing is dependent upon
proteases normally found in the thylakoid. which in turn may not be
properly developed in this mutant. However. 11ml does process some
other imported proteins to their correct mature sizes. As noted above.
these proteins include the OEC-33, which is eventually both
functionally and structurally associated with OEC-23 and OEC- 16 in
the thylakoid lumen (Andersson, 1986; Ghanotakis et al., 1987) and
the L21 chloroplast ribosomal protein. which undergoes a two-step
processing in W (Schmidt et al.. 1984; 1985). In order
to reconcile these results. allowance must be made for different
classes of precursors to be processed by different mechanisms. quite
possibly by different proteases with different activity requirements.
Cyt f is the only known thylakoid protein encoded by a chloroplast
gene which is post-translationally processed from the N-terminal end.
Perhaps the maturation of cyt f depends on a protease which also
processes certain imported proteins. Regulation conferring specificity
may be achieved by simply utilizing different proteases which

100

recognize particular precursor substrates. Alternatively, processing of
a number of precursors might rely upon the activity of a single
protease which has different affinities for different precursors in
different suborganellar locations. in the presence of different
regulating elements or at different active sites on the enzyme. The
simplest of these possiblities which is consistent with the results of
the m1 analysis is illustrated in Figure 27. This model proposes that
pre-cyt f and the processing intermediates of OEC-23 and OEC-16,
although originally synthesized in difi'erent subcellular compartments.
may utilize a common processing function in the thylakoid membrane
for final maturation. This model is based on the one proposed by
Hageman et al., (1986), with the additional suggestion that chNA-
encoded precursors may share processing functions with imported
chloroplast proteins. Specifically, the shared function could be the
second step in this processing scheme. which presumably facilitates
the passage into or through the thylakoid membrane of intermediately
processed imported proteins or CpDNA-encoded precursors. Neither
_ this nor the original model address the issue of the stability of the
peptide fragments processed away from the mature proteins.
Furthermore, the severity of a lesion affecting this second processing
step may difi‘er depending on where the gene product originated or
where it is destined to reside. Regulation such as this could explain
why the CpDNA-encoded. thylakoid protein cyt f is also present at the
size expected for the mature product while there are no mature-sized
proteins present for the nuclear DNA-encoded, lumen proteins. OEC-

23 and OEC-16.

101

 

 

 

   
 

     

 

 

 

Thylakoid-
bound protease ¢ NH

"18er
2 OEC

 

 

Stromat
protease

    
  

'\ .\\\\\\'.I

W /
OEC-int

 

 

pro-OEC

Figure 27. Model for processing of cyt f, OEC-23, and OEC-16 in the
chloroplast. The proposed processing pathway of the OEC proteins is
similar to that described for plastocyanin (Hageman et al., 1986;
Smeekens et al., 1986), as discussed in the text. The additional
feature of this model is the dependence of the processing of pre-cyt f
and the intermediate (int) forms of OEC-23 and OEC-16 on a common
thylakoid -bound protease.

102

The possibility remains that although DNA sequence analysis shows
that the heme-binding site of cyt f if normal in 11131, and heme-
staining of plastid proteins indicates that the mutant is capable of
making heme, this heme may be altered so that it is unable to be
properly bound by chloroplast cytochromes, including cyt f. If the
heme moiety were altered in such a way, perhaps cyt b559, a
component of PS 11 with a proposed function in water oxidation
(Pakrasi et al., 1988) could also be missing its non-covalently bound
heme. This could subsequently affect the assembly of PS 11 and
perhaps the processing of components of the OEC sub-complex of PS
II. Antibodies specific for cyt b559 were not tested in these
experiments. and the presence of a non-covalently bound heme is
generally poorly detected by the T‘MBZ/ H202 heme stain. so this
proposal remains only speculative.

The suggestion that heme could act as a regulator of processing is
reminiscent of the observation that cyt c 1 is one of a number of
imported mitochondrial proteins which are post-translationally

processed in at least two steps (van Loon et al., 1986: Hartl et al.
1987). This is of particular interest since cyt c1 is the mitochondrial

"analog" of cyt f in the chloroplast. Although cyt c1 is a nuclear gene
product (Wakabayashi et al.. 1980; Sadler et al., 1984) while the gene
for cyt f is encoded in chNA, Figure 28 shows that the mature
proteins share some structural similarity (Willey et al., 1983: Alt and
Herrmann, 1983). The most striking similarities are seen in the two
hydrophobic stretches: one proximal to the terminal cleavage sites (I)

and one near the carboxy-terminus of the mature protein (11) (Figure

103

Hydrophilicity Plot

 

 

 

 

 

 

+4

3 hgdrophmc
4.3-: ...........................................................
.25.
...:.
+0-3 ~ CYT F
.'E
-251
-3.E

: ....................... hydriiéhiibié .......................
'4111111111111111—1111

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0

fraction of length
+4
hydrophilrc

+3 ...........................................................
+2
w cv1c1

 

l +
N .-
lillllillllllilllllllLLliuLlllLlllll

l
(0

o ..... i'c .......................
r g 1 l I l 1 p r 1 1 1 i p 1 1 I 1 i
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0

fraction of length

 

 

-4

 

Figure 28. Hydrophilicity plots for the presequences of cyt f and cyt
c1. Hydrophilicity values (Hopp and Woods, 1981) have been plotted
against the flaction of the length of the amino acid sequence for
Qengthm cyt f (Tyagi and Herrmann. 1986) and yeast cyt c1 (Sadler
et al., 1984). "I" and "11" indicate two hydrophobic stretches of likely

significance.

104

28). Furthermore extensive functional similarity also exists since both
proteins act as intermediate electron carriers through their associated
heme (Bendall, 1982: Hauska et al.. 1983). Hartl et al. (1987)
proposed that mechanisms similar to those used to process cyt c1 may
act on precyt f in chloroplasts. Figure 25 also shows the complete
amino acid sequence of cyt c1 (Sadler et al.. 1984) Thus the cyt c1,
cyt f, and the OEC precursor polypeptides can be compared. Again.
only a stretch of hydrophobic residues at the carboxy-terminus of the
presequence. preceded by a charged residue is common between cyt

c1 and the other sequences.

Another pertinent aspect of the cyt c 1 analogy involves a yeast mutant
which is deficient in heme biosynthesis. This mutant accumulates the
intermediate processing product of cyt c 1, indicating that heme
attachment is necessary for the second step of the processing of this
protein (Lin et al., 1982: Gasser et al.. 1982: Ohashi et al.. 1982).
However. this yeast mutant is not exactly analogous to 11ml since 12111
plastids are capable of synthesizing heme. Furthermore, respiratory
electron transport, which also depends on heme appears to be normal
in 12311 since heterotrophic growth of the plants in culture suggests
that mitiochondrial electron transport is fully functional. Finally, there
is recent evidence that. in maize. mitochondrial and chloroplast
hemes may share a biosynthetic pathway (Scheegurt and Beale. 1986).
This would also favor the conclusion that 121111 is capable of
synthesizing heme in the plastid although it does not eliminate the
possibility that the binding of this heme to chloroplast cytochromes
may be inhibited in some way.

105

The following chapter presents a further chacterization the
physiological activity and structure of 111111 plastids. Further
examination of membrane structure is especially pertinent since
significant amounts of mature-sized cyt f were seen in the membrane
flaction of the mutant plastids. If some proper processing of cyt f is
occurring in 9311 and the resulting mature-sized product is found in a
membrane fraction, it is worthwhile to investigate the nature of these
chloroplast membrane flactions in m. The results of analyses
designed to determine whether levels of any of several known
chloroplast components were critically altered in {1211 will be
discussed in terms of whether any of these components may serve to

regulate protein processing and chloroplast development in 111111.

CHAPTER 4

Physiological Analysis: Structure. Composition and Activity of Mutant
and Wild-type Plastids

Inimduetian

Immunological analyses of proteins flom 111111 and wild-type plastids
have revealed evidence which suggests a major alteration in
chloroplast protein processing in this plastome mutator-induced
mutant of 9mm. Proteins affected include one encoded by a
chloroplast gene (cyt f) as well as the nuclear gene products. OEC-23
and OEC-l6. The suggestion that the processing of these three
proteins in 111111 may be inhibited by a lesion in some shared feature of
the proteolytic machinery is supported by the observation that all
three proteins appear to be normally processed in a putative revertant

of 111111.

Experiments described in this chapter are concerned with further
characterization of m mutant plastids. including the status of other
non-protein components of the chloroplast. A number of such aspects
of thylakoid development may be addresssed experimentally, including
an analysis of the extent of photosynthetic electron transport present
in the mutant, the ultrastructural development of the mutant plastid
and the biochemical analysis of various components of mature
chloroplasts such as chlorophyll, carotenoids, fatty acids and lipids.

106

107

The results of these experiments are then discussed in the context of
various possible connections between chloroplast protein processing,

accumulation of chloroplast components and the degree of thylakoid
deveIOpment in 111111.

108

Wade
Plant Material

Plant materials containing C1 wild-type plastids or C2 p191 plastids in
the A/A nuclear background and either C2 wild-type or pug, mutant
chloroplasts in the A/ C nuclear background were maintained as shoot-
tip cultures as described in Chapter 2. Tissue containing C2 wild-type
plastids in the A/A nuclear background has only recently become
available and was used in the wild-type ultrastructural analysis. Shoot-
tip cultures were illuminated with either 2, 20 or 100 rrmol-m'Z-S'1
photosynthetically active radiation (PAR). as specified for each
particular parameter measured. For all analyses discussed in this
chapter, tissue which had been maintained in culture for 3-4 weeks

since the previous transfer was used.

Photosynthetic Activity Measurements

2:111 and wild-type plastids were analysed for photosynthetic electron
transport activity. Whole chain and PS I-dependent photosynthetic
rates were determined by measuring the methyl viologen-mediated
Mehler reaction as oxygen uptake using a water-jacketed oxygen
electrode (Hansetech) (Chia et al., 1986). Chloroplasts (25 or 50 pg
chlorophyll. measured in 80% acetone, as given by MacKinney, 1941)

were added to the assay mixture containing 50 mM Tricine, pH 7 .8,
10 mM NaCl. 5 mM MgCl2, 100 mM sorbitol, 1 mM NH4Cl, 0.1 1.1M M

109

gramicidin, 0.5 mM Na azide, and 1 mg (3000 units) mr-l superoxide
dismutase and illuminated with 1000 umol-m'z-S'1 PAR. Whole chain
activity was measured using water as an electron donor to PS 11 and
0.5 mM methyl viologen (MeV) as an artificial electron acceptor from
PS 1. Values for PS I-dependent activity were analysed as oxygen
consumption derived flom oxygen electrode assays using N ,N ,N ',N‘
tetramethyl-p-phenylenediamine (TMPDH2) reduced with 2.5 mM
ascorbate as an artificial electron donor to PS I and 0.5 mM MeV as an
artificial acceptor in the presence of 1 uM 3-(3,4-dichlor0phenyl)-1,1-
dimethyl urea (DCMU). DCMU-sensitive PS II-mediated electron
transport activities were determined spectrophotometrically with an
Hitachi 100-60 as earlier described (Steinback et al., 1979) using
water as a source of electrons and 40 M DPIP (2,6-
dichlorophenolindolphenol) as an artificial electron acceptor flom PS
11.

Methods for Ultrastructural Analysis

12ml leaf tissue was prepared for thin-sectioning and ultrastructural
analysis by fixation in 3% (v/v) glutaraldehyde in Na cacodylate buffer
(pH 7.2) for 1 hour. post-fixation in 2% (w/v) 0504 in the same buffer,
dehydration through a series of graded ethanol solutions and
embedding in Spurr's resin (Spurr, 1969). Blocks containing
embedded 12ml samples were sent to Dr. M. Epp (ARCO) for further
processing as described in Epp and Parthasarathy (1987). Wild-type
tissue was embedded and sectioned for electron microscopic analysis 4

by Dr. K. Klomperans according to standard procedures.

110

Determination of Chlorophyll Content

Extraction of chlorophyll for quantitative analysis was performed on
0.2 - 0.5 g samples of wild-type and 111111 leaves (three replicates each)
in dimethyl formamide (DMF) under green safe-light conditions. as
directed by Moran and Porath (1980). Absorbance values obtained
from spectral analysis with a Perkin-Elmer spectrophotometer were
quantified using equations presented by Moran (1982).

Determination of Carotenoid Content

Total plant carotenoids were isolated by extraction of 0.3 - 0.7 g of
mutant or wild-type leaf tissue (two replicates each) in 10 ml ice-cold
acetone in the presence of Na bicarbonate (1g/ 10g sample) and 1%
(w/v) butylated hydroxytoluene (BHT) for approximately 5 minutes in
reduced light conditions (Britten. 1985). Extracts were filtered
through Whatman type 1 filter paper and concentrated to known
volumes under a stream of N2 gas. Total carotenoid levels were
determined by spectral analysis with an Hitachi Model 1 10 using a
general 1% extinction coefficient for carotenoids of 2500 at 450 nm
(Davies, 1976). Any remaining sediment was removed by forcing
samples through Sep Pac 18 cartridges (Millipore) before injecting 50
pl aliquots into a C-18 long column (4.6 x 20 cm, Varian, MCH-4-n-
cap) for separation of individual carotenoids by reverse phase high
pressure liquid chromatography (HPLC) (V arian 5000 LC) using a
linear gradient of 90% aqueous methanol to 100% ethylacetate over

111

20 minutes at 1 ml/min (Casadero et al., 1983). Peaks were detected
and quantified by using a Varian electron detector.

Determination of Fatty Acid Content

Methyl ester derivatives of mutant and wild-type fatty acids flom
leaves were prepared according to Browse et al. (1986a), employing a
1 hour treatment of 0.2 - 0.5 g leaf samples in methanolic-HCI at 80°C.
followed by phase separation via the addition of 1 volume 0.9 % (w/v)
NaCl and 1 volume hexane spiked with 17:0 fatty acid methyl ester as
a standard. Three replicate extractions were made for each mutant
and wild-type genotype. Fatty acid methyl esters in the organic phase
were quantitated by gas chromatography and flame ionization
detection (Browse et al., 1986a).

Measurement of Lipid Synthesis

Rates of lipid synthesis were determined for wild-type and 111111
tissues by following the incorporation of Na 14*C-acetate into plant
lipids according to methods given by Browse et al. (1986b). For the
'16:3' plant, W5 111311333. a 2 hour labelling period is sufficient
(Browse et al., 1986b). However, because Qengthega is an '18:3' plant
which uses the eukaryotic pathway of lipid synthesis (see Heinz and
Roughan, 1983 for review), labelling of two replicate mutant or wild-
type shoot-tip cultures was allowed to proceed for 48 hours under
moderate light conditions of 50 pmol-m‘Z-sec'1 PAR. Following this
period of uptake. tissues were frozen in liquid N2 and leaf lipids

112

extracted with chloroformzmethanol (1:1) as described in Browse et al.
(1986b). Individual lipids were separated on one-dimensional thin
layer chromatography (TLC) plates (Baker Si 250 PA, 200 pm)
impregnated with (NH4)2$O4 (Khan and Williams. 1977). TLC plates
were autoradiographed to locate lipid spots. which were then
quantified by liquid scintillation as described in Browse et al. (1986b).

113

Results

Photosynthetic Electron Transport Activity

To date, it has been dificult to measure significant levels of electron
transport in subcellular flactions of many W tissues (Hallier
and Heber 1977 ; Hallier et al., 1978). Despite extensive attempts at
modification of the organelle isolation procedures (for example. the
inclusion of polyvinyl pyrrolidone [Schneider and Hallier. 1970] in
homogenization bufi'ers) and varying assay protocols, the best wild-
type levels reached were approximately 50 percent of rates optimally
achieved with other more conventional plant tissues (Haehnel, 1984)
(Table 3) . In these experiments. the 21:11 plastid membrane
preparations consistently showed no detectable electron transport

activity.

Ultrastructural Analysis

_Qengjhm wild-type and mutant tissue proved to be quite amenable to
ultrastructural and biochemical analysis however. £1112 tissue flom
leaf tip cultures was embedded and sent to Dr. M. Epp (ARCO Plant
Cell Research Institute) for electron microscopic analysis. This
analysis revealed a striking lack of developed internal membrane
structure (Figure 29A.B). For comparison, Figure 29C shows an
9539mm type I-C2 wild-type chloroplast in the A/A nuclear
background also grown in shoot-tip culture (tissue prepared and

114

Table 3. Photosynthetic electron transport activity in angtherg wild-
type and M mutant chloroplasts. "Whole chain" assays were
performed using an oxygen electrode with water or diphenyl carbazol
(DPC) as an electron donor to PS II and methyl viologen (MeV) as an
artificial electron acceptor from PS I. Values for PS I-dependent
(dichloromethyl urea [DCMU] insensitive) activity were derived from
oxygen electrode assays using reduced tetramethyl phenamine
diamine (TMPDH2) as an artificial electron donor to PS I and MeV as
an artificial acceptor. DCMU-sensitive PS II-dependent activities were
determined spectrophotometrically with water or DPC as electron
donors and DPIP (dichlorophenylindoyl phenol) as an artificial
electron acceptor for PS 11. Rates presented represent the range of
results for several independent experiments.

Genotype Whole Chain PS II-dependent PS I-dependent
(umol 02/ (pmol DCPIP/ (umol 02/
mg chl/hr) mg chl/hr) mg chl/hr)

wt 37 -94 1 16-285 37

m1 0 o o

 

Figure 29. Electron micrographs of plastids from leaf-tip cultures of
mutant (A. B) and wild-type (wt. C). Arrows in panel A point to small
vesicles: the arrow in panel B points to an elongated membrane-like
structure. Magnification of all panels is appoximately 15,000x.
Ultrastructural data for the pm7 plastids have been previously
published (Epp and Parthasarthy. 1987).

116

sectioned by Dr. K. Klomparens. Michigan State Univiversity). Some
but not all mutant sections contained prominant layered structures
which have been interpreted as isolated "granal stacks" (Epp and
Parthasarthy. 1987) and numerous small vesicles (black arrows in
Figure 29A). A few sections also contained structures resembling
single elongated membranes (indicated by a white arrow in Figure
298). However, in light of the lack of electron transport activity and
the deficiency or complete lack of a number of thylakoid membrane
proteins. it seems unlikely that these structures are fully developed or
fully fimctional granal or stromal lamellae. Alternatively, the "stacked"
structures may represent protein aggregates. The micrographs do
however show the presence of a double outer membrane surrounding
the plastid. probable chloroplast ribosomal units. plastoglubuli and a
large number of vesicular structures. A similar lack of thylakoid
development is observed in the ultrastructure of m plastids when
the mutant plastids are maintained as chlorotic sectors in variegated
greenhouse-grown plants (Epp and Parthasarthy. 1987). In this case.
there were fewer plastoglobuli present and larger single vesicles but

still little evidence of thylakoid development.
Analysis of Chlorophyll and Carotenoid Accumulation

In order to study a number of more specific aspects of chloroplast
development in pmz. chlorophyll and carotenoid levels were
determined for both wild-type and mutant tissues which had been
maintained in shoot-tip culture under a range of light levels. Type I-
C2 mutant and wild-type plastids were each examined in two different

117

nuclear backgrounds. resulting in either the A/A-I compatible

plastome-genome combination or the A/C-I incompatible combination.

Figure 30 confirms what was easily seen by simple visual inspection of
the plants, that is, that when grown under a variety of light conditions,
the mutant accumulates significantly less chlorophyll on a fresh weight
basis than does the wild-type (p<0.0l). Chlorophyll concentrations, as
well as values for all of the other parameters measured were subjected
to statistical analysis using an analysis of variance (Table 4). This
analysis allows one to make comparisons between mutant and wild-
type chloroplasts. comparisons between the two nuclear backgrounds.
and comparisons among the light regimes. For chlorophyll levels,
Table 4 shows that the nuclear background has a significant effect and
furthermore. chlorophyll levels in both the mutant and the wild-type
appear to be sensitive to the higher light conditions. This
photosensitivity seems to be especially acute in the mutant where
chlorophyll levels approach the limits of detection.

One main function proposed for carotenoid pigments in plants is to
provide protection from photo-damage under high light conditions
(Casadoro et al., 1983: Oelmuller and Mohr, 1985; Mayfield et al..
1986a. 1986b). Figure 31 presents values for the concentration of
total carotenoids found in mutant and wild-type material. As tested by
the ANOVA shown in Table 4, any differences between mutant and
wild-type or between the two nuclear backgrounds are not statistically
significant (p>O.l). Furthermore. when four major plant carotenoids -

fi-carotene, neoxanthin, violoxanthin and lutein - were separated and

118

 

'7 1400 n
1200 d
1000 -

3001

600‘

 

 

 

Chlorophyll Content (pg-gFW

2 20 100
Light Level (umol-m 'z-s '

Figure 30. Chlorophyll content of tissue containing wild-type and pm7
plastome type I plastids in A/A and A/ C nuclear backgrounds. Plants
were grown in 3 different light regimes for 3 weeks and chlorophyll
content was then determined. Bars represent the mean 3; SD (n=3).

1w

Table 4. Results of statistical analysis using split plot analysis of
variance tests. ANOVA treatment (Snedecor and Cochran. 1967) of
pigment, fatty acid and lipid data from Mg wild-type and p191
mutant plastids in A/A and A/ C nuclear backgrounds. C = chloroplast
genome. N = nuclear genome, L = light, “ = p < 0.1. ” = p < 0.05,
W" = p < 0.01

Chlorophyll C 1 1 1 .47 ""' CxN 1.24
N 34.80“ CxL 0.16
L 15. 12 NxL 0.09
CxNxL 0 19
6&6665616 ""“E """"""" 6 'i"2""""'c'§EN """""" 6 .62“-
N 0 18 CxL 0.01
1 31 NxL 0.12
CxNxL 0.1 l
16?o’f£fit375€16"""c """"""" 1" 66.367“ ””””” 6 Yfi ““““““ 6 .66"
N 2. 15 C x L 0.03
L 5.20“ N x L 0.10
C x N x L 0.00
163661333616 ""2: """"" 5'5'71'1'5'"""é'£ N ““““““““ 6 363""
N 0. 10 C x L 0.10
L 2 1 .90‘“ N x L 0.06
C x N x L 0.01
163'1’E't't375c'id"""c"""’"BEE—""'"c‘fiv """""" 6 562'"
N 0.18 C x L 0.01
1.31 N x L 0.12
C x N x L 0.1 1
i‘éi'z‘fa't‘a‘aaam‘e """"""" 5672'5‘7"""'é‘£‘1'v """""""" 6 :65."
N 0.71 C x L 0.00
L 2.27 N x L 0.12
C x N x L 0. l 1
16:.3'526'5616 ""'c """"""" 1 '1766‘"""E'£‘N """"""" 6 362""
N 0.1 1 C x L 0.03
L 28.03“" N x L 0.25
C x N x L 0.00
’"fi’dfi'""""’c"""""o'.66 """""" 6 in? """"" 6 '16-"
N 0.00
""p6""""""""""6'1'6 """""" 6 £16 """" 6 "'72"-

120

Table 4. (con'td)

DGD C 0.54 CxN 0 00

N 0.34

-___S_L____-______C______._-__fé..5_ -------- 6 £1? -------- 5 '3'?"-
N 2.86

""156” __._._.__.__.-_C """"""" 2' 9'36 """"""" CxN _________ 5 77'9""
N 0.18

""56 "“"_"'6"""""177'§ """""" 6 ifi"'"—"i'2'61'"
N 15.19

""51""--""""""""7’6§ ''''''' 6 £1? """"" 6 66""

20
.0.
a:
u

121

 

)

 

 

 

Carotenoid Content (ug-gFW

2 20 100
- -1
Light Level (umol-rn 2~s )

Figure 31. Total carotenoid content of tissue containing wild-type and
pmz, plastome type I plastids in A/A and A/ C nuclear backgrounds.
Plants were grown in 3 different light regimes for 3 weeks and total

carotenoid content was then determined. Bars represent the mean 1
SD (n=2).

122

analysed. no major difi'erences were seen for any of these specific
carotenoids in plant tissue grown at high light levels (Figure 32). In
fact, the only obvious difi‘erences in carotenoids at any light level were
the decrease in the percent B-carotene and the increase in the
percent neoxanthin in mutant tissue grown in low light. Since these
values are each expressed as a percent of total carotenoid. they are not
independent measurements and cannot be statistically analysed.
However. in neither case is the one type of carotenoid totally deficient
and thus no low light-sensitive lesion in carotenoid synthesis is
indicated. Hence. there was no evidence indicating a lesion in mg] in

the biosynthesis or accumulation of these specific carotenoids.

Analysis of Fatty Acid Accumulation and Lipid Synthesis

Another aspect of chloroplast development which was measured in
both wild-type and m1 mutant tissue is the degree of accumulation of
fatty acids and the ability of the mutant to synthesize lipids from these
fatty acids. Mutant and wild-type tissues were maintained in shoot-tip
culture under light levels ranging from 2 - 100 umol-m'Z-s‘l. Fatty
acids in crude extracts were derivatized to methyl esters, extracted
with hexane. separated and quantified by gas-liquid chromatography.
The comparisons of fatty acid composition for mutant and wild-type
are shown in Figure 33. For three of the five fatty acids examined,
16:0. 18:0, and 18:2, significant difi'erences in accumulation between
mutant and wild-type were observed(p<0.001, p<0.05 and p<0.05,

respectively) .

123

Figure 32. Major carotenoids in tissue containing wild-type and mu?
plastome type I plastids in A/A and A/ C nuclear backgrounds. Plants
were grown in 3 different light regimes for 3 weeks and the presence
of major carotenoids was then determined. Bars represent the mean
:1; SD (n=2). High, medium and low light refers to 100. 20 and 2

umol-m'Z-s‘1 PAR. respectively.

124

 

 

 

         

  

 

 

 

 

 

 

7 7
w. m m m I??? b - 44/44,. m m ...w%%%%% b
unis . m .m ............ . ...
......... a.mw m :. gazaa;ak:<xa .u w
n H w n m
g M .
Azzazazazazzzzzzazaz mm . ....... um
........ t o M m ,_, .... t 0
U n O .M n
m m m
0 t. m
w m e.
M We m
q.<-q-.-qq..q4 1.14‘41-qqnqiq
m w w m m m m o m w w w m m .o. o
4.92.0 s: gasses 22228 323 $823828 3:228

 

Figure 32.

125

Figure 33. Fatty acid content of tissue containing wild-type and um?
plastome type I plastids in A/A and A/ C nuclear backgrounds. Plants
were grown in 3 different light regimes for 3 weeks and chlorophyll
content was then determined. Bars represent the mean 1; SD (n=3).
High, medium and low light refers to 100, 20 and 2 umol-m'Z-S'1
PAR. respectively.

126

 

 

7/////////J

  

.....................
l y... .. I ,1. .. .. .. .l 1.. I. ... .l .I .l ...l I

  

18:3

r¢¢¢¢¢¢¢

  

18:1 18:2

18:0
Fatty Acid - High Light

 

16:0

 

. 4 q . .
O.
0
1

A _- 2:99: 2280 22 as“.

Figure 33.

 

18:0 18:1 18:2 18:3

Fatty Acid - Medium Light

16:0

 

 

 

15.0 '
10.0 '1

A 7 2:93 2280 22 be“.

 

 

 

7////////////////////////////A

.......................

 

n...
.6
A72:

0.0 '1

99: 22:8 22 em...

18:0 18:1 18:2

16:0

Fatty Acid - Low Light

127

The extent to which these fatty acids were potentially converted into
lipids in the mutant was addressed by applying 14C-labelled acetate to
shoot-tip cultures of m and wild-type tissues. then allowing 48
hours for the synthesis into various lipids. Figure 34 illustrates the
amount of radioactivity incorporated into monogalactosyldiacylglycerol
(MGD). phosphatidylglycerol (PG). digalactosyldiacylglycerol (DG),
sulphoquinovosyldiacylglycerol (SL), phosphatidylglycerol (PG).
phosphatidylcholine (PC) and phosphatidylethanolamine (PE).
Analysis of variance revealed no significant differences between levels
of lipid synthesis in the mutant and the wild-type (p>0. l) or in
comparisons between the A/A and the A/ C nuclear backgrounds

(p>0. 1). Again, these results indicate that there is no biosynthetic
lesion in the mutant with respect to lipid synthesis.

128

 

  

50 I AAlwt
[:1 ACth

40 I AAlpm7
AClme

    
        

 

      
   

% of Total Label

\\\\\\~o

-._'~._\\._\;‘.§..~.§.§ xx)“ \“~.‘\._\._‘-. \. x x x

S\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\“ I
.\\\\\\\§\§§\§\\\\\\\\\

.\\\\\\\\\\\
t\\\\\\\\\\\ '

 

MGDPGDGDSL PE PC Pl
Lipid

Figure 34. Lipid synthesis in tissue containing wild-type and m
plastome type I plastids in A/A and A/ C nuclear backgrounds. Plants
were fed 14C-acetate for 48 hours in a light regime of 50 umol-m‘2-S'

éPA(R b)efore the samples were harvested. Bars represent the mean i
D n=2 .

129

Discussion

Examples of proteins which are not properly processed or those
which are not accumulated in the plastids of the plastome mutator-
induced mutant ml were discussed in the preceding chapter.
Results of investigations discussed in Chapter 4 reveal that in addition
to differences in .chloroplast proteins, little internal membrane
structure was found in the mutant plastids and not surprisingly, no
detectable electron transport activity was present in the m1 plastids.
Futhermore, a lack of chlorophyll was found in the mutant tissue in
comparison to wild-type. as well as significant differences in levels of
three fatty acids - 16:0, 18:0 and 18:2 fatty acids. However, these 1
analyses did not reveal any evidence of significant differences in total
carotenoid levels or in the synthesis of seven chloroplast lipids.

The pigment. fatty acid and lipid measurments were also performed
on two difierent nuclear genotypes in combination with plastome type
I chloroplasts. The A/A nuclear background is more compatible with
plastome I (Stubbe, 1959) and chloroplasts are present. The A/ C
nuclear background. however, is less compatible with plastome type I,
and tissue containing this plastome-genome combination is pale
green, containing plastids which are not fully developed (Drillisch,
1975). Comparisons of chlorophyll, carotenoid, fatty acid and lipid
levels in A/A I and A/ C I tissues revealed significant differences in
chlorophyll concentrations only (p<0.05)

130

The observation of a lack of activity in whole chain electron transport
assays is not unexpected for a mutant that is not photosynthetically
competent. However. despite the accumulation of a number of
components of the various protein complexes. specifically the cyt b6/ f
complex and PS I. there was no evidence of partial electron transport
through either photosystem. This result is reasonable since the heme
staining described in Chapter 3 showed that the cytochrome complex

components lacked bound heme to act as an electron carrier.

Consistent with this lack of physiological activity was the clear absence
of developed thylakoid membrane structure in the mutant plastids as
shown by the electron microscopic analysis of M leaf tissue. It
seems more probable that this deficiency is the result of aborted
development in m plastids as opposed to the senescence of
developed chloroplasts in these cells. This argument is based upon
the observation that the mutant tissue does not go through stages of
greening before becoming chlorotic but rather appears pale yellow
from very early stages of "development" in shoot-tip culture. Epp and
Parthasarthy (1987) have shown that m1 plastids lack developed
thylakoids when they are maintained as chlorotic sectors in wild-type
greenhouse grown plants. Hence. the lack of thylakoid membrane
development is not a consequence of maintaining the mutant in shoot-
tip culture. as has been suggested for aberrant plastid development
under various other tissue culture conditions (Parthier. 1979; Colijn et
al.. 1982).

131

The ultrastructural analysis suggested the usefulness of a more
detailed examination of specific chloroplast components. These
analyses were performed using mutant and wild-type plastids in two
different nuclear backgrounds. The incompatible genotype
combinations were included because m1 plastids were only available
in an A/A nuclear background which was also homozygous for the nu;
gene. and thus unstable in terms of plastome mutator activity.
Including both nuclear backgrounds in these experiments allows the
analysis of one type of plastome- genome incompatibility as well as the
comparison of mutant and wild-type phenotypes. The design of the
statistical analysis also makes use of the two nuclear backgrounds as
another set of replicates for the comparison of the mutant and wild-
type chloroplasts.

This more detailed analysis of the levels of various chloroplast
components in m included the measurement of chlorophyll and
carotenoid pigments. as well as an examination of the fatty acids and
lipids found in the mutant. Figure 34 shows that the mutant is not
simply deficient in the ability to synthesize chlorophyll. Instead. m2
displays symptoms of photobleaching under the relatively higher light
condition of 100 umol-m'Z-s‘l. Photobleaching. or the loss of
chlorophyll pigments under conditions of high light. is a characteristic
of many carotenoid-deficient plants (Casadoro et al.. 1983: Oelmi‘iller
and Mohr, 1985: Mayfield et al.. 1986a. 1986b). The lack of any
signifigant difference in total carotenoid accumulation (Figure 3 l) and
the presence of four specific carotenoid pigments (Figure 32)
demonstrate that the photobleaching seen in M is probably not a

132

consequence of a substantial loss of photo-protective carotenoid
pigments. Since a significant degree of photobleaching is observed in
the wild-type as well as the mutant tissue. it seems unlikely that this
photosensitivity is the major factor responsible for the absence of

thylakoid development in the mutant.

Other major plastid components which are critical to the development
of mature wild-type chloroplasts are the fatty acids (made in the
plastid for the entire plant cell - Roughan and Slack. 1982. 1984) and
the lipids which incorporate these fatty acids. Dome et al. (1982)
reported major alterations in lipid composition in a barley mutant
Wang) which also displays altered internal membrane structure
and the absence of chloroplast ribosomes. The W mutant
showed a drastic reduction in the ratio of MGD to DGD. This
reduction was attributed to the absence of the thylakoid membranes.
the major site of MGD accumulation in chloroplasts (Roughan and
Slack. 1982. 1984: Douce and Joyard. 1980).

In order to determine whether a major deficiency in fatty acid or lipid
biosynthesis accompanied the aberrant internal membrane structure
in 91111. levels of steady state fatty acid accumulation and newly
synthesized lipids were measured. Although three types of fatty acids
showed significant difi'erences between mutant and wild-type. in the
cases of the 18:0 and 18:2 fatty acids. relative concentrations are not
consistent between the two nuclear backgrounds. In the case of the
16:0 fatty acids in tissue grown under high light conditions. although
analysis of variance suggests that the difference between mutant and

133

wild-type levels are highly significant (p<0.001). the mutant plastids
are still able to accumulate 60-80% of the 16:0 fatty acid found in the
wild-type chloroplasts.

These chloroplast-synthesized fatty acids are necessary substrates for
the synthesis of major plant lipids (Roughan and Slack. 1982. 1984;
Douce and J oyard. 1980). Synthesis of four lipids found primarily in
the chloroplast. MGD. DGD. PC. and SL (Roughan and Slack. 1982.
1984; Douce and Joyard. 1980) was found to be quite similar in m
and wild-type tissue. The remaining lipids analysed (PE. PC and P1)
are found outside of the chloroplast (Roughan and Slack. 1982. 1984;
Douce and J oyard. 1980) and were also found to have equivalent rates
of synthesis in mutant and wild-type (Figure 34). The lipid analysis of
the (lengthen wild-type and m mutant described here was of
synthesis. measured by assessing the rate of l4*C--acetate labelling of
plant lipids. rather than a quantitative steady-state measurement as
was done by Dorrie et al. (1982). This may account for the difference
in lipid levels in two mutants both of which lack thylakoids. However.
it is important to note that the barley 3111951113133 mutant is also
deficient in plastid ribosomes (BOrner et al.. 1976). a condition not
present in m. Hence. the conclusion which may be drawn from
these results is that the mechanisms for synthesizing chloroplast-
specific lipids (MGD. PG. DGD and SL) as well as for exporting fatty
acids to the cytoplasm. synthesizing extrachloroplastic lipids (PE. PC
and PI). are intact in the mutant. Furthermore. the absence of a
significant difference in mutant and wild-type lipid synthesis suggests

134

that the minimal differences observed in fatty acids (necessary for
lipid synthesis) are not fundamental to the altered phenotype of pm7.

Earlier analyses have characterized the extent of possible alterations in
chloroplast protein processing in 2311. Subsequently. experiments
discussed in this chapter were designed to determine the degree to
which other non-protein components of the thylakoid membrane
accumulated in the m plastids. It does not seem likely that
thylakoid membrane development would progress normally in the
absence of fully processed precursors to a number of subunits for
various protein complexes. In fact. the lack of processing is
accompanied by a complete deficiency in at least two other chloroplast
proteins which are components of PS II. as well as decreases in other
polypeptides of cyt b6] f and PS I. A lack of processing oould possibly
block membrane development. coupled with the loss of a major
constituent of the membrane such as chlorophyll. Hence. one
hypothesis suggests that a gene encoding some component of this
processing function is the site of the primary lesion in m. Without
proper processing activity in the chloroplast. thylakoid membrane
development may be aborted.

Alternatively. the polypeptide precursors which accumulate in 2311 (as
well as other precursors) may be dependent on thylakoid-bound
proteases for the final processing to mature-sized proteins. The state
and abundance of thylakoids could thus affect protease abundance and
activity. Thus. the lesion in 2311 could have a primary effect on
thylakoid development. for example by inhibiting the accumulation of

135

chlorophyll to some essential threshold level. As a result. the
thylakoid-bound processing machinery responsible for the maturation
of a particular set of chloroplast proteins may not be able to function
properly. This secondary efi'ect would have to be somewhat specific
however, as other proteins do seem to be correctly processed in m
and other plastome mutator-induced mutants which do not
accumulate chlorophyll do appear to process pre-cyt f (Sears. personal

communication) .

Results of experiments described here do not allow the discrimination
between these two possible hypotheses. Thus. it is not possible to
specify a ”cause and effect" relationship between chloroplast protein
processing and thylakoid membrane development. However. it is
clear that. despite the altered processing and accumulation of a
number of chloroplast proteins and an extensive 'lack of chlorophyll
accumulation and thylakoid development. a number of other
chloroplast components remain present in m1 plastids at relatively
normal levels.

136

Chapter 5. Summary and Conclusions

One major conclusion which may be drawn from the results of the
research presented in this thesis is that the processing of a particular
subset of chloroplast proteins can be blocked by a single mutaional
event. In the case of m1. a plastome mutator-induced mutant of
W mm some but not all of the chloroplast proteins
encoded in either the plastome or the nuclear genome displayed
phenotypes which most likely resulted from faulty protein processing
in the mutant. The proposed deficiency in processing of cyt f. OEC-23
and OEC-16 in 9311 may be the result of a lesion in a proteolytic
enzyme common to the processing of these three proteins or
alternatively. it could be a more indirect lesion in some factor which
controls or influences this activity. How closely the lack of internal
membrane structure found in 21111 is related to the phenotype of
altered processing is not clear.

Another conclusion from this work is that despite the clear lack of
development of mature thylakoid structures in this mutant. a
surprising number of chloroplast components. including various
chloroplast proteins. pigments. fatty acids and lipids. are present at
relatively normal levels in the m2 plastids. A few proteins are absent.
chlorophyll levels are quite low and the abundance of a few fatty acids
is minimally altered. Possibly one of the missing or deficient
components in 121111 is the signal to which the processing function

responds.

137

A more extensive analysis of 121131 is necessary to accurately describe
the defect resulting in the accumulation of a number of aberrantly
large chloroplast proteins. Cytochrome f is not the only gene product
for which a large cross-antigenic product is accumulated in M and
subsequently found only as a mature-sized protein in the 9191
revertant. Therefore. it seems unlikely that the primary lesion lies in
the p313 gene. although it is conceivable that a lesion which alters cyt
f processing could have a secondary efi'ect on the processing of other
chloroplast proteins. The sequencing data discussed in Chapter 3
ruled out a lesion in the processing site itself and the 5' upstream
region. However. it should be noted that the identification of the
precursor-specific N-terminal extension is only tentatively based on
the presence of a presumed ribosome binding site upstream of the
proposed translation start site and the shared homology of the
presumed deduced presequence with that from other plants which
have been examined using m m translation analysis (Alt and
Herrmann, 1984. Willey et al.. 1984a.b). In fact. on Northern blots. a
complex pattern of RNAs hybridize to a 2% probe (Tyagi and
Herrmann. 1986). Conceivably, RNA splicing could add message
encoding the N-terminal portion of the precursor to the 0.6.1.8
transcript. A mutation in pml in this part of the protein could alter
the ability of a protease to recognize the processing site. The
existence of a non-adjacent segment which encodes the cyt f
precursor N-terminus could explain why the anti-peptide antibodies
did not react against the larger protien in m1. Since the peptide
antigen was synthesized to match the amino acid sequence deduced

138

from the reading frame immediately 5' to the mature cyt f protein
sequence. According to this hypothesis. the cyt f N-terminal sequence
could actually be quite different.

This possibility could be experimentally addressed by employing the
technique of $1 nuclease protection analysis (Berk and Sharp. 1977)
hybridizing an end-labelled single-stranded DNA probe. encoding the
presumed transcriptional start site. to the message (Weaver and
Weissman. 1979). Digestion of this hybrid with $1 nuclease would
allow the identification of the RNA transciption initiation site for MA-
This in turn would show whether transcript processing is involved in
9:15 expression. If processing were indicated. the mutant transcript
size and sequence could be analysed for evidence of any faulty mRNA

processing.

Results of in 211329. translation of wild-type Oenothera messages for cyt
f. OEC-23 and OEC-16 would allow the determination of the sizes of
these primary gene products for these proteins. This could reinforce
the identification of the aberrant proteins in M as unprocessed cyt f
and intermediately processed OEC-23 and OEC- 16. Ultimately. the
most rigorous manner in which to verify the identity of these mutant
proteins as incompletely processed precursors is to purify these
proteins from m tissue. using a combination of immunoamnity
chromatography (Kristiansen. 1978) and gel purification. and subject
the purified material to N-terminal microsequencing.
Microsequencing procedures employing a gas phase sequenator

(Hewick et al.. 1981: Aebersold et al.. 1986) are well-suited for use

139

with small amounts of proteins and the number of residues of of

interest here.

Screening of mutant and wild-type chloroplast proteins could also
reveal additional critical differences between mutant and wild-type
which are related to the m phenotype. It is also possible that such a
difi‘erential screening of mutant and wild-type chloroplast proteins.
either by in mm transcription/ translation of chNAs or by in m
radiolabelling of newly-synthesized proteins, may also provide a means
to discover the processing enzyme which may be deficient in M. It
should be remembered. however. that at least two other plastid
proteins. the functions of which are not thought to be related to
protein processing. were observed to be missing in pm], and this
could make the interpretation of the differential screening results
diflcult. Furthermore. it is also possible that a nonfunctional protease

may still be present even if there is no activity seen.

In conclusion. it is most likely that 931A. the plastid gene encoding cyt
f. is probably not itself the site of the primary lesion in m. Instead.
the point mutation in and indicated by the restriction analysis of 121111
and the wild—type chNA probably lies somewhere else in the
chloroplast genome. Thus. there is an unidentified plastome gene
which encodes a trans-acting factor which affects processing of a
subset of both imported and chloroplast-synthesized chloroplast
proteins.

140

A search for the primary lesion in and chNA and hence the potential
identification of the plastome gene involved in chloroplast protein
processing could utilize the following strategr. DNA from mutant and
wild-type plastids would be digested with a number of separate
restriction endonucleases. melted and allowed to reanneal as a
mixture of wild-type and mutant chNA fragments. The resulting
fragments would represent wild-type reannealed to wild-type strands.
mutant to mutant strands or hybrids of wild-type and mutant strands.
The hybrid fragments which contain the point mutation should be
sensitive to S 1 nuclease at the site of the point mutation (Shenk et al.
1975; Vogt. 1981). Therefore when chNA treated in this manner is
analyzed electrophoretically. two new fragments should be found in
comparison to controls using wild-type or mutant chNA alone.
Digested fragments of reasonable size (thus the justification for
employing more than one restriction enzyme) could then be isolated
and cloned for further sequencing analysis in order to locate the point
mutation. The possibility that the mismatch found is due to the
hypervariability in certain chNA fragments of Oenothera (Blasko et
al.. 1988) may be addressed by also performing this analysis on newly
germinated and purified m1 tissue. This mutant material will have
undergone fewer rounds of chNA replication than the currently used
tissue which has been maintained in shoot-tip culture for seven years.
Presumably. this "newer" tissue will be less exposed to events in
chNA replication or repair which may be involved in the occurrance
of hypervariability.

141

The 91111 mutant may prove to be useful for further study of
chloroplast protein processing. The accumulated precursors may be
more extensively localized using techniques of immunocytochemistry
or protease-sensitivity of right-side-out compared to inside-out
membrane preparations (Akerlund et al.. 1982). Organelle uptake
studies may be used to follow kinetics of protein maturation as well as
the degree of degradation (a feature of the relationship between
chloroplast development and protein accumulation which was not
addressed in this study). Finally. the m1 mutant may serve as a
valuable source from which to purify precursors or intermediates
which are not naturally available from other systems. Purified
preparations of intermediates may be subjected to amino acid
sequence analysis in order to clearly determine cleavage sites. Such
intermediates (or fusion proteins synthesized based upon the
knowledge of the sequence of these intermediates) may also be
utilized in biochemical studies using reconstitution of subcellular
fractions to further investigate the mechanism of and requirements for
chloroplast protein processing. Such reconstitution experiments
could involve both homologous and heterologous combinations of the
m1 precursors with chloroplast membrane and soluble fractions from
W and other plants in the presence of various amounts of

cofactors. energy sources and so forth.

Such future experimental possibilities represent valuable outgrowths
from the analysis of the role of chloroplast gene products in
chloroplast development using a mutational approach.

BIBLIOGRAPHY

BIBLIOGRAPHY

Aebersold R H. D B Teplow. L E Hood. 8 B H Kent (1986)
Electroblotting onto activated glass. J Biol Chem 261:4229-4238

Ahne F. A M Merlos-Lange. B F Lang. K Wolf (1984) The

mitochondrialgenome of the fission yeast W
m. Curr Genet 8:517-524

Akerlund H-E. C J ansson. B Andersson (1982) Reconstitution of
photosynthetic water splitting inside-out thylakoid vesicles and
identification of a participating polypeptide. Biochim Biophys Acta
68 1:1 - 10

Allred D R. L A Staehelin (1985) Lateral distribution of the
cytochrome be/ f and coupling factor ATP synthetase complexes of
chloroplast thylakoid membranes. Plant Physiol 78: 199-202

Allred D R. L A Staehelin (1986) Spatial organization of the
cytochrome bs-f complex within chloroplast thylakoid membranes.
Biochim Biophys Acta 849:94-103

Alt J . R G Herrmann (1984) Nucleotide sequence of the gene for pre-
apocytochrome f in the spinach plastid chromosome. Curr Genet
8:55 1 -557

Alt J . P Westhofi'. B B Sears. N Nelson. E Hurt. G Hauska. R G
Herrmann (1983) Genes and transcripts for the polypeptides of the
cytochrome b6 / f complex from spinach thylakoid membranes.
EMBO J 2:979-986

AndersSon B (1986) Proteins participating in photosynthetic water
oxidation 13 L A Staehelin. C J Amtzen. eds. Photosynthesis. III.
Photosynthetic Membranes and Light Harvesting Systems.
Encyclopedia of Plant Physiology New Series. Vol 19. Springer-
Verlag. New York. pp 447-456

Backer J . F Foury (1985) Repair properties in yeast mitochondrial
mutators. Curr Genet 10:7-13

Barkan A, D Miles. W C Taylor (1986) Chloroplast gene expression in
nuclear. photosynthetic mutants of maize. EMBO J 5:1421-1427

142

143

Bendall D S (1982) Photosynthetic cytochromes of oxygenic
organisms. Biochim Biophys Acta 683:1 19-151

Bendall D S. M Sanguansermsri. J Girard-Bascou. P Bennoun (1986)
Mutations of thmggmgngg rgighgggii afi'ecting the cytochrome LLf
complex. FEBS Lett 203:31-35

Bentzen P. W C Leggett. G G Brown (1988) Length and restriction site
heteroplasmy in the mitochondrial DNA of American shad (gm

W- Genetics 118:509-518

Berk A J . P A Sharp (1977) Sizing and mapping early adenovirus
mRNAs by gel electrophoresis of 51 endonuclease digested hybrids.
Cell 12:721-733

Bhaya D. P Castelfranco (1986) Synthesis of a putative c-type
cytochrome by intact. isolated pea chloroplasts. Plant Physiol
8 1:960-964

Blasko K. S A Kaplan. K G Higgins. R Wolfson. B B Sears (1988)
Variation in copy number of a 24-base pair tandem repeat in the
chloroplast DNA of m 11.221923 strain Johansen. Curr Genet.
submitted

BOrner T. B Schumann. R Hagemann (1976) Biochemical studies on a
plastid ribosome-deficient mutant of m Eugen. In T
Bficher. W Neupert. W Sebald. S Werner. eds. Genetics and
Biogenesis of Chloroplasts and Mitochondria. Elsevier/ North
Holland Biomedical Press. Amsterdam. pp 41-48

Barrier T. B B Sears (1986) Plastome mutants. Plant Mol Biol Report
4:69-92

Britton G (1985) General carotenoid methods. In JI-I Law. HC Rilling.
eds. Steroids and Isoprenoids Part B. Methods in Enzymology.
Vol 111. Academic Press. New York. pp 113-149

Browse J. P McCourt. C R Somerville (1986a) Fatty acid composition
of leaf lipids determined after combined digestion and fatty acid
methyl ester formation from fresh tissue. Anal Biochem 152: 141-
145

Browse J. N Warwick. C R Somerville. C R Slack (1986b) Fluxes
through the prokaryotic and eukaryotic pathways of lipid synthesis
in the 'l6:3' plant W Indiana. Biochem J 235:25-31

Casadoro G. G Heyer-Hansen. C G Kannangara. S P Gough (1983) An
analysis of temperature and light sensitivity in timing mutants of
barley. Carlsberg Res Commun 48:95-129

HM

Chia C P. C J Arntzen (1986) Evidence for two-step processing of
nuclear-encoded chloroplast proteins during membrane assembly.
J Cell BiollO3z725-73l

 

Chia C P. J H Duesing. J L Watson. R Guy. C J Arntzen (1986)
Characterization of cytoplasmic mutants of Ma tabacum with
altered photosynthetic function. Curr Genet 10:469-479

Chiu W-L. W Stubbe. B B Sears (1988) Plastid inheritance in
W: organelle genome modifies the extent of biparental
plastid transmission. Curr Genet 13:181-189

Chua N-H. S G Bartlett. M Weiss (1982) Preparation and
characterization of antibodies to chloroplast proteins. In M
Edelman. R B Hallick. N-H Chua. eds. Methods in Chloroplast
Molecular Biology. Elsevier Biomedical Press. Amsterdam. pp
1063- 1080

Chua N-H. G W Schmidt (1978) Post-translational transport into intact
chloroplasts of a precursor to the small subunit of ribulose-1.5-
bisphosphate carboxylase. Proc Natl Acad Sci USA 75:61 10-61 14

Clark R D. G Hind (1983) Isolation of a five-polypeptide cytochrome
b-f complex from spinach chloroplasts. J Biol Chem 258: 100348-
10354

Cleland R (1972) 9mm: Cytogenetics and Evolution. Academic
Press. New York p 370

 

Colman A. C Robinson (1986) Protein import into organelles:
hierarchical targeting signals. Cell 46:321-322

Colijn C M. P Sijmons. J N M Mol. A J Kool. H J J Nijkamp (1982)
Light and benzylaminopurine induce changes in ultrastructure and
gene expression in plastids of Rename, hybrid; cell cultures. Curr
Genet 6:129-135

Cox E C (1976) Bacterial mutator genes and the control of
spontaneous mutation. Ann Rev Genet 10:135- 156

Cox R P. B Andersson (1981) Lateral and transverse organization of
cytochromes in the chloroplast thylakoid membrane. Biochem
Biophys Res Commun 103:1336- 1342

Crouse E J . J M Schmitt. H-J Bohnert (1985) Chloroplast and
cyanobacterial genomes. genes and RNAs: a compilation. Plant Mol
Biol Report 3:43-89

145

Curtis S E. M T Clegg (1984) Molecular evolution of chloroplast DNA
sequences. Mol Biol Evol 1:291-301

Darr S C. S C Somerville. C J Arntzen (1986) Monoclonal antibodies
to the light-harvesting chlorophyll a/b protein complex of
photosystem II. J Cell Biol 103:733-740

Davies B H (1976) Carotenoids. In The chemistry and biochemistry of
plant pigments. Ed 2. Academic Press. New York. pp 154—155

Del Giudice L. D R Massardo. F Manna. K Wolf (1986) Isolation and
characterization of a conditional mutant in W
W producing rho° petites at the nonpermissive
temperature. Curr Genet 1212201-204

Doebley J F. D P Ma. W T Renfroe (1987) Insertion/deletion
mutations in the Zea chloroplast genome. Curr Genet 1 1:617-624

Doherty A. J C Gray (1979) Synthesis of cytochrome f by isolated pea
chloroplasts. Eur J Biochem 98:87-92

Dome A-J. J-P Carde. J Joyard. T BOrner. R Douce (1982) Polar lipid
composition of a plastid ribosome-deficient barley mutant. Plant
Physiol 69: 1467- 1470

Drillisch. M (197 5) Vergleichhende Untersuchungen an den A-
Genotypen von W. PhD Thesis. University of Dusseldorf.
Dusseldorf

Epp M D (1973) Nuclear gene-induced plastome mutations in
WW 1. Genetic analysis. Genetics 75:465-483

Epp M D. M V Parthasarthy (1987) Nuclear gene-plastome mutations

in 931191119413 hmkgn . II. Phenotype description with electron
microscopy. Amer J Bot 74:143-151

Epp M D. J L Erion. B B Sears. W Stubbe (1987) The plastome mutator
of Qenoihsra continues to function as originally described. Mol
Gen Genet 206:515-518

Finley D. A Varshavsky (1985) The ubiquitin system: fimctions and
mechanisms. TIBS 10:343-347

Foury F. J Kolodynski (1983) 1211 mutation blocks recombination
between mitochondrial rho+ and rh0' genomes having tandemly

arrayed repeat units in W W. Proc Natl Acad
Sci USA 80:5345-5349

146

Foury F. E Van Dyck (1985) A PIF-dependent recombinogenic signal in
the mitochondrial DNA of yeast. EMBO J 4:3525-3530

Galliard T (1974) Techniques for overcoming problems of lipolytic
enzymes and lipoxygenases in the preparation of plant organelles.
In S Fleischers. L Packer eds. BIomembranes. Part A. Methods in
Enzymology. vol 31. Academic Press. New York. pp 520-528

Gasser S M. A Ohashi. G Daum. P C Bdhni. J Gibson. G A Reid. T
Yonetani. G Schatz (1982) Imported mitochondrial proteins
cytochrome b2 and cytochrome c1 are processed in two steps.
Proc Natl Acad Sci USA 79:267-271

Genga A. L Bianchi. L Foury (1986) A nuclear mutant of
W mm deficient in mitochondrial DNA
replication and polymerase activity. J Biol Chem 261:9328-9332

Ghanotakis D F. C M Waggoner. N R Bowlby. D M Demetriou. G T
Babcock. C F Yocum (1987) Comparitive structural and catalytic
properties oxygen-evolving photosystem II preparations. Photosyn
Res 14:191-199

Glickman B. P Van den Elsen. M Radman (1978) Induced
mutagenesis in dam‘ mutants of Whig 9911: A role for 6-
methyladenine residues in mutation avoidance. Mol Gen Genet
163:307-3 12

Gordon K H J. E J Crouse. H J Bohnert. R G Herrmann (1982)
Physical mapping of differences in chloroplast DNA of the five wild-
type plastomes in W subsection Euoenothera. Theor Appl
Genet 61:281-296

Gray J C. A L Phillips (1982) Isolation of cytochrome f and a
cytochrome b-f complex. In M Edelman. R B Hallick. N-H Chua.
eds. Methods in Ch‘ .oroplast Molecular Biolog, Elsevier Biomedical
Press. Amsterdam pp 917-932

 

Grebanier A E. D M Coen. A Rich. L Bogorad (1978) Membrane
proteins synthesized but not processed by isolated maize
chloroplasts. J Cell Biol 78:734-746

Greenberg B M. V Gaba. A K Mattoo. M Edelman (1987) Identification
of a primary mm degradation product of the rapidly-turning—over
32 kd protein of photosystem II. EMBO J 6:2865-2869

Hageman J . C Robinson. S Smeekens. P Weisbeek (1986) A thylakoid
processing protease is required for complete maturation of the
lumen protein plastocyanin. Nature 324:567-569

147

Hallier U W. U W Heber (1977) Cytochrome f deficient plastome
mutants of mm. Plant Cell Physiol :257-273

Hallier U W. J M Schmitt. U Heber. S S Chaianova. A D Volodarsky
(1978) Ribulose-l.5-bisphosphate carboxylase-deficient plastome
mutants of W. Biochim Biophys Acta 504:67-83

Hartl F-U. J Ostermann. B Guiard. W Neupert (1987) Successive
translocation into and out of the mitochondrial matrix: targeting of
proteins to the intermembrane space by bipartite signal peptide.
Cell 51:1027-1037

Harrison R G. D M Rand. W C Wheeler (1985) Mitochondrial DNA size
variation within individual crickets. Science 228: 1446- 1448

Hauska G. E Hurt. N Gabellini. W Lockau (1983) Comparative aspects
of quinol-cytochrome c/plastocyanin oxidoreductases. Biochim
Biophys Acta 726:97-133

Hauswirth W W. P J Laipis (1982) Mitochondrial DNA polymorphism
in a maternal lineage of Holstein cows. Proc Nat Acad Sci 79:4686-
4690

Hay R. P BOhni. S Gasser (1984) How mitochondria import proteins.
Biochim Biophys Acta 779:65-87

Heinemeyer W. J Alt. R G Herrmann (1984) Nucleotide sequence of
the clustered genes for apocytochrome b6 and subunit 4 of the
cytochrome b/ f complex in the spinach plastid chromosome.
Curr Genet 8:543-549

Heinz E. P G Roughan (1983) Similarities and differences in lipid
metabolism of chloroplasts isolated from 18:3 and 16:3 plants.
Plant Physiol 72:273-279

Herrmann R G (1982) The preparation of circular DNA from plastids.
In M Edelman. R B Hallick. N-H Chua (eds) Methods in Chloroplast
Molecular Biology. Elsevier Biomedical Press. Amsterdam pp 259-
280

Herrmann R G. P Westhofi‘. J Alt. P Winter. J Tittgen. C Bisanz. B B
Sears. N Nelson, E Hurt. G Hauska. A Viebrock. W Sebald (1983)
Identification and characterization of genes for polypeptides of the
thylakoid membrane. In 0 Cifi'eri, L Dure III eds. Structure and
function of plant genomes. Plenum Press. New York pp 143-153

148

Herrmann R G. P Wesfliofi. J Alt. P Winter. J Tittgen. N Nelson (1985)
Thylakoid membrane proteins and their genes. In L van Vloten-
Doting. G S P Groot. T C Hall eds. Molecular form and function of
the plant genome. Plenum Press. New York pp 233-256

Hewick R M. M W Hunkapiller. L E Hood. W J Dreyer (1981) A gas-
liquid solid phase peptide and protein sequenator. J Biol Chem
256:7990-7997

Ho K K. D W Krogmann (1980) Cytochrome f from spinach and
cyanobacteria. J Biol Chem 255:3855-3861

Hopp T P. K R Woods (1981) Prediction of protein antigenic
determinants from amino acid sequences. Proc Natl Acad Sci USA
78:3824-3828

Hudspeth M E S. D S Shumerd. K H Tatti. L I Grossman (1980) Rapid
purification of yeast mitochondrial DNA in high yield. Biochim
Biophys Acta 610:221-228

Hurt E. G Hauska (1981) A cytochrome f/bs complex of five
polypeptides with plastoquinol-plastocyanin-oxidoreductase activity
from spinach chloroplasts. Eur J Biochem 117:591-599

Hurt E C. A P G M van Loon (1986) How proteins find mitochondria
and intramitochondrial compartments. TIBS l 1:204-207

Jacobsen B S. F Fong, R L Heath (1975) Carbonic anhydrase of
spinach. Plant Physiol 55:468-474

Jansen T. C Rother. H Reinke. K Beyreuther. C Jansson, B Andersson.
R G Herrmann (1987) Nucleotide sequence of cDNA Clones
encoding the complete '23 kDa' and '16 kDa' precursor proteins
associated with the photosynthetic oxygen-evolving complex from
spinach. FEBS Lett 216:234-240

Jensen K H. D L Herrin. F G Plumley. G W Schmidt (1986) Biogenesis
of photosystem II complexes: 'hanscriptional. translational. and
posttranslational regulation. J Cell Biol 103:1315-1325

Johnson E M. S A Kaplan. K Blasko. R Wolfson. B B Sears (1988)
Qennthera plastome mutator: mutation frequencies and chloroplast
DNA polymorphisms. Mol Gen Genet. submitted

Karlin-Nemann G. E M Tobin (1986) Transit peptides of nuclear-
encoded chloroplast proteins share a common amino acid
framework. EMBO J 5:9-13

149

Kaplan S A (1987) Characterization of chloroplast DNA from wild-type
and mutant plastids of mm 11mm. str J ohansen. M Sc
Thesis. Michigan State Univ. East Lansing

Kaplan S A. B B Sears (1986) Plgsgomg mutator gene induces
alterations in chloroplast DNA. Genetics 1 13:s22.

Kaplan S A. K Blasko. E M Johnson. P O Lim. R Wolfson. B B Sears
(1987) W plastome mutator line is heteroplasmic for
chloroplast DNA restriction fragment length polymorphisms. Cold
Spring Harbor Meeting on Molecular Biology of Mitochondria and
Chloroplasts. Abstracts p 65

Kaufmann S H. C M Ewing. J H Shaper (1987) The erasable Western
blot. Anal Biochem 161:89-95

Khan M. R Kolter. C Thomas. D Figurski. R Meyer. E Remaut. D R
Helinski (1979) Plasmid cloning vehicles derived from plasmids
ColEI. F. R6K and RK2. In R Wu. ed. Recombinant DNA. Methods
in Enzymology. Vol 68. Academic Press. New York. pp 268-280

Khan M-U. J P Williams (1977) Improved thin-layer chromatographic
method for the separation of major phospholipids and glycolipids
from plant lipid extracts and phosphatidyl glycerol and
bis(monoacylglyceryl) phosphate from animal lipid extracts. J
Chromotogr 140:179-185

Kimball R F (1979) Relationship between repair processes and
mutation induction in bacteria. In W M Generoso. M D Shelby. F J
de Serres eds. DNA Repair and Mutagenesis in Eukaryotes. Plenum.
New York. pp 1-23

Krishnan H B. C D Reeves. T W Okita (1986) ADPglucose
pyrophosphorylase is encoded by different mRNA transcripts in leaf
and endosperm of cereals. Plant Physiol 8 1:642-645

Kristiansen T (1978) Matrix-bound antigens and antibodies In O
Hofl'man-Ostenhof. eds. Affinity Chromatography. Pergamon Press,
Oxford. pp191-201

Kutzelnigg H. W Stubbe (197 4) Investigations on plastome mutants in
W I. General considerations. Sub-Cell Biochem 3:73-89

Laemmli. U K (1970) Cleavage of structural proteins during the
assembly of the head of bacteriophage T4. Nature 227:680-695

Lam E. R Malkin (1985) Characterization of a photosynthetic mutant
of Lanna lacking the cytochrome be-f complex. Biochim Biophys
Acta 810:106-109

 

~ 150

Langridge P (1981) Synthesis of the large subunit of spinach ribulose
bisphosphate carboxylase may involve a precursor polypeptide.
FEBS Lett 123:85-89

Leary J J. D J Brigati. D C Ward (1 983) Rapid and sensitive
colorimetric method for visualizing biotin labelled DNA probes
hybridized to DNA or RNA immobilized on nitrocellulose: bioblots.
Proc Natl Acad Sci USA 80:4045-4049

Legator M S. W G Flamm (1973) Environmental mutagenesis and
repair. Ann Rev Biochem 42:683-708

Lemaire C. J Girard-Bascou. F-A Wollman. P Bennoun (1986) Studies
on the cytochrome b6/ f complex. 1. Characterization of the

complex subunits in W mums. Biochim BIOPhYS
Acta 851:229-238

Lindenhahn M M. M Mettzlafi'. R Hagemann (1985) Remarkable
restriction patterns between green and white branches of 'plastome
mutator' plants of Qqngmm W . Mol Gen Genet 200505-
505

Ljungberg U. H-E Akerlund. B Andersson (1986) Isolation and
characterization of the lO-kDa and 22-kDa polypeptides of higher
plant photosystem 2. Eur J Biochem 158:477-482

Ljungberg U. H-E Akerlund. C Jansson, B Andersson (1984)
Identification of polypeptides associated with the 23 and 33 kDa
proteins of photosynthetoc oxygen evolution. Biochim Biophys Acta
767: 145- l 52

Loomis W D (1974) Overcoming problems of phenolics and quinones
in the isolation of plant enzymes and organelles. In S Fleischers. L
Packer eds. Biomembranes PartA. Methods in Enzymology. vol 3 1.
Academic Press. New York. pp528-544

MacKinney G (1941) Absorbtion of light by chlorophyll solutions. J
Biol Chem 140:315-322

Maniatis T. E F Fritsch. J Sambrook (1982) Molecular cloning. A
laboratory manual. Cold Spring Harbor Laboratory Press. New York
p 545

Mansfield R W. J M Anderson (1985) Transverse organization of
components within the chloroplast cytochrome b-563 / f complex.
Biochim Biophys Acta 809:435-444

151

Marder J B. P Goloubinofi'. M Edelman (1984) Molecular architecture
of the rapidly metabolized 32-kilodalton protein of photosystem II.
J Biol Chem 259:3900-3908

Maroc J . J Garnier (1981) Gel electrophoresis of chloroplast
membranes of mutants of thamydnmgnas [mm which have
impaired photosystem II function and lack photosynthetic
cytochromes. Biochim Biophys Acta 637:473-480

Mattoo A K. M Edelman (1987) Intramembrane translocation and
posttranslational palmitoylation of the chloroplast 32-kDa
herbicide-binding protein. Proc Natl Acad Sci USA 84:1497-1501

Mattoo A K. H Hofi'man-Falk. J B Marder. M Edelman (1984)
Regulation of protein metabolism: coupling of photosynthetic
electron transport to In 1119 degradation of the rapidly metabolized
32-kilodalton protein of the chloroplast membranes. Proc Natl
Acad Sci USA 81:1380-1384

Mayfield S P. T Nelson. W C Taylor (1986a) The fate of chloroplast
proteins during photooxidation in carotenoid-deficient maize
leaves. Plant Physiol 82:760-764

Mayfield S P. T Nelson. W C Taylor. R Malkin (1986b) Carotenoid
synthesis and pleiotropic efi'ects in carotenoid-deficient seedlings
of maize. Planta 169:23-32

McCourt P. L Kunst. J Browse. C R Somerville (1987) The efi‘ects of
reduced amounts of lipid unsaturation on chloroplast ultrastructure
and photosynthesis in a mutant of W- Plant Physiol
84: 353-360

Messing J (1983) New M13 vectors for cloning. In R Wu. A
Grossman, K Moldave. eds. Recombinant DNA. Methods in
Enzymology. Vol 101. Academic Press. New York. pp 20-78

Metz J G. D Miles. A W Rutherford (1983) Characterization of nuclear
mutants of maize which lack the cytochrome f / b-563 complex.
Plant Physiol 73:452-459

Michaelis P (1969) Uber Plastiden-Restitutionen (Ruckmutationen).
Cytologia 34 (suppl): 1-1 15

Minami E, A Watanabe (1985) Detection of a precursor polypeptide of
the rapidly-synthesized 32.000 dalton thylakoid protein in spinach
chloroplasts. Plant Cell Physiol 26:839-846

152

Moon E. T-H Kao. R Wu (1987) Rice chloroplast DNA molecules are
heterogeneous as revealed by DNA sequences of a cluster of genes.
Nuc Acids Res 15:611-630

Moran R (1982) Formulae for determination of chlorophyllous
pigments extracted with N. N -dimethylformamide Plant Physiol
69: 1376- 138 1

Moran R. D Porath (1980) Chlorophyll determination in intact tissues
using N,N-dimethylformamide. Plant Physiol 65:478-479

Nakata T. I Takabe (1971) Plating of isolated tobacco mesophyll
protoplasts on agar medium. Planta 99:12-20

Nelson N . J Neumann (1972) Isolation of a cytochrome be-f particle
from chloroplasts. J Biol Chem 247:1817-1824

Nelson N. E Racker (1972) Partial resolution of the enzymes catalyzing
photophosphorylation. J Biol Chem 247 :3848-3853

Neupert W. G Schatz (1 98 1) How are proteins imported into
mitochondria. TIBS 6:1-4

Norrander J. T Kempe. J Messing (1983) Construction of improved
M13 vectors using oligodeoxynucleotide-directed mutagenesis.
Gene 26:101-106

Oelmfiller R. H Mohr (1985) Carotenoid composition in milo
(Sarghmn 151123.112) as affected by phytochrome and chlorophyll.
Planta 164:390-395

Ohashi A. J Gibson. I Gregor. G Schatz (1982) Import of proteins into
mitochondria. J Biol Chem 257:13042-13047

Ohyama K. H Fukuzawa. T Kohchi. H Shirai. T Sano. S Sano. K
Umesono. Y Shiki. M Takeuchi. Z Chang. S Aota. H Inokuchi. H
Ozeki (1986) Chloroplast gene organization deduced from complete
sequence of liverwort Mmhanna pnbnngmha chloroplast DNA.
Nature 322:572-574

Olive J . O Vallon. F-A Wollman. M Recouvreur. P Bennoun (1986)
Studies on the cytochrome b6] f complex. 11. Localization of the
complex in the thylakoid membranes from spinach and
thamxdsmsnas mm by immunocytochenfistry and freeze-
fracture analysis of b6/ f mutants. Biochim Biophys Acta 851:239-
248

153

Ortiz W. R Malkin (1985) Topographical studies of the subunits of the
thylakoid cytochrome be-f complex. Biochim Biophys Acta
808: 164-170

Pakrasi H B. J G K Williams. C J Arntzen (1988) Targeted mutagensis
of the ME and ME genes blocks photosynthetic electron
transport: evidence for a functional role of cyt b559 in photosystem
II. EMBO J 7:325-332

Palmer J D (1985) Evolution of chloroplast and mitochondrial DNA in
plants and algae. kl R J MacIntyre ed. Molecular Evolutionary
Genetics. Plenum Publ. New York pp 13 1 -239

Parthier B (1979) The role of phytohormones (cytokinins) in
chloroplast development. Biochem Physiol Pflanzen 174: 173-214

Phillips A L. J C Gray (1984) Synthesis of components of the
cytochrome b;f complex by isolated pea chloroplasts. Eur J
Biochem 138:591-595

Possingham J V. M E Lawrence (1983) Controls to plastid division.
Int Rev Cytol 84:1-56

Potrykus I (1970) Mutation und Rfickmutation extrachromosomale
vererbter Plastidenmerkmale von Emma. Z Pflanzenzuch 63:24-
40

Radman M. R E Wagner Jr. B W Glickman. M Meselson (1979) DNA
methylation. mismatch correction and genetic stability. In M
Alacevic ed. Progress in Environmental Mutagenesis. Elsevier-
North Holland. Amsterdam pp 121-130

Rand D M. R G Harrison (1986) Mitochondrial DNA transmission
genetics in crickets. Genetics 114:955-970

Redei G P. S B Plurad (1973) Hereditary structural alterations of
plastids induced by a nuclear gene in W. Protoplasma
77:36 1 -380

Reed M L (1979) Intracellular localization of carbonate dehydratase
(carbionic anhydrase) in leaf tissue. Plant Physiol 63:2 16-2 17

Reisfeld A. A K Mattoo. M Edelman (1982) Processing of a
chloroplast-translated membrane protein In um. Eur J Biochem
1245:125-129

Rigby P W J. M Drekrnan. C Rhodes. P Berg (1977) Labeling
deoxyribonucleic acid to high specific activity In vim; by nick
translation with DNA polymerase I. J Mol Biol 1 13:237-251

154

Rochaix J D (1985) Genetic organization of the chloroplast. Intl Rev
Cytol 93:57-91

Rotse D. F Thelier. S J Horvath. J M Tomich. J H Richards. D S
Allison. G Schatz (1988) Amphiphilicity is essential for
mitochondrial preseqence function. EMBO J 7 :649-655

Roughan P G. C R Slack (1982) Cellular organization of glycerolipid
metabolism. Ann Rev Plant Physiol 33:97 -132

Roughan G. R Slack (1984) Glycerolipid synthesis in leaves. TIBS
9:383-386

Sabatini D D. G Kreibich. T Morimoto. M Adesnik (1982) Mechanisms
for the incorporation of proteins in membranes and organelles. J
Cell Biol 92:1-22

Sadler I. K Suda. G Schatz. F Kaudewitz. A Haid (1984) Sequencing of
the nuclear gene for yeast cytochrome c1 precursor reveals an
unusually complex amino-terminal presequence. EMBO J 9:2137-
2 1 43

Salts Y. R G Herrmann. N Peleg. U Lavi. S Izhar. R Frankel. J S
Beckmann (1984) Physical mapping of plastid DNA variation
among eleven W species. Theor Appl Genet 69:1-14

Sanger F. S Nicklen. A R Coulson (1977) DNA sequencing with chain
terminating inhibitors. Proc Natl Acad Sci USA 74:5463-5467

Sayre. R T. B Andersson. L Bogorad (1986) The topology of a
membrane protein: the orientation of the 32 kd Qb-binding
chloroplast thylakoid membrane protein. Cell 47 :601-608

Schatz G. R A Butow (1983) How are proteins imported into
mitochondria? Cell 32:316-318

Schmid G H. A Radunz. W Menke (1977) Localization and function of
cytochrome f in the thylakoid membrane. Z Naturforsch 32cz271-
280

Schmidt G W. M L Mishkind (1986) The transport of proteins into
chloroplasts. Ann Rev Biochem 55:879—9 12

Schmidt R J . N W Gillham. J E Boynton (1985) Processing of the
precursor to a chloroplast ribosomal protein made in the cytosol
occurs in two steps. one of which depends on a proteins made in
the chloroplast. lMol Cell Biol 5: 1093-1099

 

155

Schmidt R J. A M Myers. N W Gillham. J E Boynton (1984)
Chloroplast ribosomal proteins of W synthesized in
the cytoplasm are made as precursors. J Cell Biol 98:201 1-2018

Schmitz U K. K V Kowallik (1986) Polymorphism and gene
arrangement, among plastomes of ten 391191221111 species. Plant Mol
Biol 7:115-127

Schneegurt M A. S I Beale (1986) Biosynthesis of protoheme and
heme a from glutamate in maize. Plant Physiol 81:965-971

Schneider v, U w Hallier (1970) Polyvinylpyrrolidon als Schutzstofi‘
bei der Untersuchung gerbstoffgehemmter Enzymreaktionen.
Planta 94:134-139

SchOtz F (1954) Uber Plastiden konkurrenz bei W . Planta
43: 183-240

Schatz F (1974) Untersuchungen uber die Plastidenkonkurrenz bei
anm IV. Der Einfluss des Genoms auf der
Durchsetzungsfahigkeit der Plastiden. Biol Zbl 93:41-64

Sears B B (1983) Genetics and evolution of the chloroplast. In P
Gustafsson. ed. Stadler Symposium XV. Univ Missouri Press.
Columbia. pp 1 19-139

Sears B B. Kaplan S A (1984) Nuclear gene induces changes in
chloroplast DNA. Genetics 107:s96

Sederofi' R R (1984) Structural variation in mitochondrial DNA. Adv
Genet 22: l- 108

Shanklin J. M Jabben. R Vierstra (1987) Red light-induced formation
of ubiquitin-phytochrome conjugates: Identification of possible
intermediates of phytochrome degradation. Proc Natl Acad Sci
USA 84:359-363

Shaw P J . J A Henwood (1985) Immuno-gold localization of
cytochrome f. light-harvesting complex. ATP synthase and
ribulosel.5-bisphophate carboxylase/oxygenase. Planta 165:333-
339

Shenk T E. C Rhodes. P W J Rigby. P Berg (1975) Biochemical
method for mapping mutational lesions in DNA with S 1 nuclease: .
the location of deletions and temperature-sensitive mutations in
Simian Virus 40. Proc Natl Acad Sci USA 72:989-993

156

Shinozaki K. M Ohme. M Tanaka. T Wakasugi. N Hayashida. T
Matsubayashi. N Zaita. J Chunwongse. J Obokata. K Yamaguchi-
Shinozaki. C Ohto. K Torazawa. B Y Meng. M Sugita. H Deno. T
Kamogashira. K Yamada. J Kusuda. F Takaiwa. A Kato. N Tohdoh. H
Shimada. M Sugiura (1986) The complete nucleotide sequence of
the tobacco chloroplast genome: its gene organization and
expression. EMBO J 5:2043-2049

Shumway L K. T E Weier (1967) The chloroplast structure of 19an
maize. Amer J Bot 54:773-780

Siedow J N. L E Vickery. G Palmer (1980) The nature of the axial
ligands of spinach cytochrome f. Arch Biochem Biophys 203: 101-
107

Singer S J. P A Maher. M P Yafi’e (1987) On the transfer of integral
proteins into membranes. Proc Natl Acad Sci USA 84:1960-1964

Smeekens S. C Bauerle. J Hageman. K Keegstra. P Weisbeek (1986)
The role of the transit peptide in the routing of precursors toward
different chloroplast compartments. Cell 46:365-375

Snedecor G W. W G Cochran ( 1967) Statistical Methods. Ed 6. Univ
Iowa Press. Ames pp 369-375

Snyder M. A R Fraser. J LaRoche. K E Gartner-Kepkay. E Zouros
(1987) Atypical mitochondrial DNA from the deep sea scallop

W W . Proc Nat Acad Sci 84:7595-7599

Solignac M. M Monnerot. J -C Mounolou (1983) Mitochondria DNA

heteroplasmy in Wang. Proc Nat Acad Sci
80:6942-6946

Solignac M. J Genermont. M Monnerot. J-C Mounolou (1984)
Genetics of mitochondria in M90113: mtDNA inheritance in
heteroplasmic strains of I2. mannnana. Mol Gen Genet 197: 183-
188

Somerville C R. S C Somerville. W L Ogren (1981) Isolation of
photosynthetically active protoplasts and chloroplasts from
Wang. Plant Sci Lett 21 8.9-96

Southern E M (1975) Detection of specific sequences among DNA
fragments separated by gel electrophoresis. J Mol Biol 98:503-517

Steinback K E. J J Burke. C J Arntzen (1979) Evidence for the role of
surface-exposed segments of the light-harvesting complex in
cation-mediated control of chloroplast structure and function. Arch
Biochem Biophys 195:356-361

157

Steppuhn J . C Rather. J Hermans. T Jansen. J Salnikow. G Hauska. R
G Herrmann (1987) The complete amino-acid sequence of the
Reiske FeS-precursor protein from spinach chloroplast deduced
from cDNA analysis. Mol Gen Genet 210:171-177

Stubbe W (1957) Dreifarbenpanaschierung bei Oenothera. I.
Entrnischung von drei in der Zygote vereinigten Plastidomen. Ber
Deutsch Bot Gesell 70:221-227

Stubbe W (1959) Genetische analyse des Zusammenwirken von Genom
und Plastom bei W. Z Verebungsl 90:288-298

Stubbe W. R G Herrmann (1982) Selection and maintenance of
plastome mutants and interspecific genome / plastome hybrids from
. In M Edelman. R B Hallick. N-H Chua. eds. Methods in
Chloroplast Molecular Biology, Elsevier Biomedical Press.
Amsterdam pp 149-165

Thomas P E. D Ryan. W Levin (1976) An improved staining procedure
for the detection of the peroxidase activity of cytochrome P-450 on
sodium dodecyl sulfate polyacryamide gels. Anal Biochem 75: 168-
1 76

Thompson D. V Walbot. E H Coe Jr (1983) Plastid development in
191311 and mutator-afiected maize plants. Amer J Bot 70:940-950

Tittgen J. J Hermans. J Steppuhn. T Jansen. C Jansson, B Andersson.
R Nechhushtai. N Nelson. R G Herrmann (1986) Isolation of cDNA
clones for fourteen nuclear-encoded thylakoid membrane proteins.
Mol Gen Genet 204:258-265

Towbin H. T Staehelin. J Gordon (1979) Electrophoretic transfer of
proteins from polyacrylamide gels to nitrocellulose sheets:
procedure and applications. Proc Natl Acad Sci USA 76:4350-4354

Tsuzuki M. S Miyachi. G E Edwards (1985) Localization of carbonic
anhydrase in mesophyll cells of terrestrial C3 plants in relation to
C02 assimilation. Plant Cell Physiol 26: 881-891

W. A K. J Hermans. J Steppuhn. C J ansson. F Vater. R G Herrmann
(1987) Nucleotide sequence of cDNA clones encoding the
complete "33 kDa" precursor protein associated with the
photosynthetic oxygen-evolving complex from spinach. Mol Gen
Genet 207:288-293

Tyagi A K. R G Herrmann (1986) Location and nucleotide sequence of

the pre-apocytochrome f gene on the 9mm Men plastid
chromosome (Euoenothera plastome I). Curr Genet 10:481-486

158

van Loon A P G M. A W Brandli. G Schatz (1986) The presequences of
two imported mitochodrial proteins contain information for
intracellular and intramolecular sorting. Cell 44:801-8 12

van Loon A P G M. G Schatz (1987) Transport of proteins to the
mitochondrial inner membrane space: the 'sorting' domain of the
cytochrome c1 presequence is a stop-transfer sequence specific for
the mitochondrial inner membrane. EMBO J 6:2441-2448

Vater J. J Salnikow. C Jansson (1986) N-terminal sequence and
secondary structure analysis of extrinsic membrane proteins in the
water-splitting complex of spinach. FEBS Lett 203:230-234

Vogt V M (1980) Purification and properties of $1 nuclease from
. In L Grossman. K Moldave. eds. Nucleic Acids Part 1.
Methods in Enzymolog. Vol 65. Academic Press. New York. pp
248-255

von Heijne G (1983) Patterns of amino acids near signal-sequence
cleavage sites. Eur J Biochem 133:17-2 1

von Heijne G (1986a) Mitochondrial targeting sequences may form
amphiphilic helices. EMBO J 5:1335-1342

 

von Heijne G (1986b) A new method for predicting signal sequence
cleavage sites. Nucleic Acid Res 14:4683-4690

von Heijne G (1986c) Towards a comparitive anatomy of N-terminal
topogenic protein sequences. J Mol Biol 189:239-242

Wakabayashi S. H Matsubara. C H Kim. K Kawai. T E King (1980) The
complete amino acid sequence of bovine heart cytochrome c 1.
Biochem Biophys Res Commun 97: 1548- 1554

Walbot V. E H Coe (1979) Nuclear gene Igjnp, conditions a
programmed change to ribosome-less plastids in Zea mm. Proc
Natl Acad Sci USA 76:2760-2764

Walker G C (1985) Inducible DNA repair systems. Ann Rev Biochem
54:425-457

Weaver R F. C Weissman (1979) Mapping of RNA by a modification of
the Berk-Sharp procedure: the 5' termini of 158 B globin mRNA

precursor and mature 108 B globin mRNA have identical map
coordinates. Nuc Acid Res 7:1175-1 183

Westhofi‘ P. C Jansson. L Klein-Hitpafi. R Berzbom. C Larsson. S G
Bartlett (1985) Intracellular coding sites of polypeptides

159

associated with photosynthetic oxygen evolution of photosystem II.
Plant Mol Biol 4:137-146

Wettem M. G Galling (1985) Degradation of the 32-kilodalton

thylakoid-membrane polypeptide of thgmydomgnas reinhargi Y-l.
Planta 166:474-482

Willey D L. A D Aufiret. J C Gray (1984a) Structure and topology of
cytochrome f in pea chloroplast membranes. Cell 36:555-562

WilleyD L. CJ Howe.ADAufi'ret. C MBowman. TADyer.JC Gray
(1984b) Location and nucleotide sequence of the gene for
cytochrome f in wheat chloroplast DNA. Mol Gen Genet 194:416-
422

Willey D L. A K Huttly. A L Phillips. J C Gray (1983) Localization of the
gene for cytochrome f in pea chloroplast DNA. Mol Gen Genet
189:85-89

Wu N. J C COté. R Wu (1986) Nucleotide sequence of the rice
cytochrome f gene and the presence of sequence variation near this
gene. Gene 50:271-78

Yanisch-Perron C. J Vieira. J Messing (1985) New M13 host strains
and the complete sequences of Ml3mp and pUC vectors. Gene
33: 103- 1 19

Zurawski G. M T Clegg (1987) Evolution of higher-plant chloroplast
DNA-encoded genes: implications for structure-function and
phylogenetic studies. Ann Rev Plant Physiol 38:391-4 18

"iiiiiiiiiiiiiii