Journal of Experimental Botany, Vol. 60, No. 11, pp. 3005–3010, 2009
doi:10.1093/jxb/erp158 Advance Access publication 19 May, 2009
REVIEW PAPER
The loss of DNA from chloroplasts as leaves mature: fact or
artefact?
Beth A. Rowan* and Arnold J. Bendich†
University of Washington, Department of Biology, Box 355125, Seattle, WA 98195, USA
Received 17 March 2009; Revised 21 April 2009; Accepted 22 April 2009
Abstract
In this review, the controversy regarding the preservation or degradation of chloroplast DNA (cpDNA) as chloroplasts
develop their photosynthetic capacity and leaves reach maturity is addressed. A constant amount of cpDNA during
maturity might be expected in order to support photosynthesis over the lifespan of the leaf. Nevertheless, a decline in
cpDNA during leaf development was found for all seven plant species investigated. Initial measurements showed that
Arabidopsis was similar to the other seven. The controversy arose with two recent studies concluding that the amount
of cpDNA remains constant as Arabidopsis leaves mature. These authors proposed that the observation of Arabidopsis
chloroplasts with undetectable levels of DNA was an artefact, although the most recent data support the original
findings. If the amount of cpDNA remains constant, then Arabidopsis is atypical and would not serve as a good model
for chloroplast development. It is shown that the apparently contradictory data may be attributed to methodology and
the choice of leaves to be compared. Thus, it is concluded that the controversy can be resolved, Arabidopsis can serve
as a representative model, and cpDNA degradation is a common event in chloroplast development.
Key words: Chloroplast, degradation, DNA, leaf development.
Introduction
Since the cytological discovery of DNA-containing ‘bodies’
in the chloroplast of Chlamydomonas in 1962 (Ris and Plaut,
1962), the course of research on chloroplast DNA (cpDNA)
has been marked by three major controversies. The first
involved the identification of the true cpDNA among the
DNAs obtained from chloroplast preparations. It was not
until 1972 that the use of DNase to remove contaminating
nuclear DNA from the surface of extracted chloroplasts
allowed cpDNA to be isolated in pure form (Kolodner and
Tewari, 1972). The second controversy involved the size and
linear or circular form of DNA molecules within plastids.
The use of in-gel procedures to avoid shearing the DNA
combined with the analysis of moving pictures of ethidiumstained cpDNA showed that the commonly depicted
genome-sized circle represented neither most of the cpDNA,
the template for replication, nor the segregating genetic unit
(the chromosome) in plastids (Bendich, 2004; Oldenburg and
Bendich, 2004, 2009). The third and current controversy
involves the degree to which DNA is retained as chloroplasts
develop their photosynthetic capacity and leaves reach the
period of maturity. This third controversy is the subject of
the present article.
Plastid DNA during development
Unlike eukaryotic nuclei, chloroplasts can contain many
copies of their genome. In meristematic cells, cpDNA
replication outpaces chloroplast division, leading to a net
increase in DNA per chloroplast during early seedling
development (Fujie et al., 1994; Kuroiwa, 1991). For
example, the increase is 7.5-fold and about 4-fold for wheat
and spinach, respectively (Lawrence and Possingham, 1986;
Miyamura et al., 1986). From 3–7 d after germination,
cpDNA increases in the first true Arabidopsis leaf from ;40
to ;600 genomes per chloroplast (Fujie et al., 1994). The
increase occurs when the leaf is still <0.5 mm in length,
leaves that are so small that it would be impractical to
* Present address: Department of Molecular Biology, Max Planck Institute for Developmental Biology, D-72076 Tübingen, Germany.
y
To whom correspondence should be addressed: [email protected]
ª The Author [2009]. Published by Oxford University Press [on behalf of the Society for Experimental Biology]. All rights reserved.
For Permissions, please e-mail: [email protected]
3006 | Rowan and Bendich
obtain sufficient leaf material to detect the increase using
the Southern blotting procedure.
As the leaf cells continue to expand, the amount of cpDNA
declines. For spinach, an approximately 6-fold decrease in
genomes per chloroplast was attributed to a constant amount
of cpDNA per cell partitioned among an increasing number
of chloroplasts per cell without cpDNA replication (Scott
and Possingham, 1980). The decline in cpDNA during leaf
expansion for barley and oat, however, involves cpDNA
degradation in addition to dilution, since the rate of cpDNA
decline exceeds the rate of chloroplast division (Baumgartner
et al., 1989; Hashimoto and Possingham, 1989). Degradation
of cpDNA in developing juvenile rice leaves was proposed to
act as a signal for the initiation of senescence (Sodmergen
et al., 1991). Note that the above studies focused on
immature plants, and the fate of cpDNA throughout the
lifespan of the plant was not addressed.
The controversy over constant or decreasing
amounts of cpDNA during development
A constant amount of cpDNA during maturity might be
expected in order to support photosynthesis over the
lifespan of the leaf. Nevertheless, undetectable levels of
cpDNA were found in many chloroplasts isolated from
mature Arabidopsis thaliana leaves, as determined by staining with 4#,6-diamidino-2-phenylindole (DAPI) (Rowan
et al., 2004). The amount of DNA per chloroplast declined
on average 2–7-fold long before the initiation of senescence.
We made a mistake in the Abstract of Rowan et al. (2004)
when we said that the amount of cpDNA declines until
‘most of the leaves contain little or no DNA’. We meant to
say that the amount of cpDNA declines until most of the
chloroplasts contain little or no DNA. This mistake may
have contributed to the controversy that arose with two
later studies concluding that the amount of cpDNA remains
constant as leaves mature (Li et al., 2006; Zoschke et al.,
2007). In both studies, only one method was used to
measure the cpDNA and neither used the same method as
we did in our original study. These data are now analysed in
an attempt to resolve this controversy.
Li et al. (2006) investigated the amount of cpDNA in
Arabidopsis and tobacco by visual inspection of blothybridization signals. They concluded that cpDNA
remained constant throughout development in both species,
but this conclusion was not supported by all of their data.
By our visual inspection, there is an apparent 2–3-fold
reduction in hybridization signal between the ‘young’ and
‘mature’ leaf of a 36-d-old Arabidopsis plant (Lanes 2 and 3
in Fig. 3b of Li et al., 2006). They provided no control data
for Arabidopsis to indicate equal loading of all samples or to
reveal the extent of the restriction digestion, since their
images did not include the well of the gel. We also analysed
cpDNA by blot hybridization (including the proper controls) and found that the hybridization signal for mature
tissues was 2–5-fold less than that of young tissues (Rowan
et al., 2009). For tobacco, hybridization of the chloroplast
probe with ‘promiscuous’ cpDNA in the nucleus served as
a loading control, and all signals from chloroplast-derived
cpDNA appeared to be constant by our visual inspection.
Even if we now suppose that all of their experiments
included the proper controls and did, in fact, show a constant
hybridization signal for both species, the following alternative interpretations are offered. Li et al. (2006) claim that
their leaf material ‘was produced exactly (our emphasis) as
described by Rowan et al. (2004)’. They provided the age of
the plant, but described the leaves only as ‘young’ or
‘mature’. Without knowing the size of the leaf or which leaf
was examined (and such information was not provided) it
cannot be determined whether these leaves are comparable to
ours. For tobacco, the position of the leaf along the stem was
described, but the age of the plants was not given. Thus, it
cannot be assessed whether their tobacco leaves were at
a similar developmental stage as ours, which did show
a modest decrease in DNA per chloroplast for the largest
leaf (Shaver et al., 2006).
Arabidopsis leaves undergo several rounds of endoreduplication during leaf development (Galbraith et al., 1991;
Zoschke et al., 2007; Rowan et al., 2009), and this further
confounds evaluation of the blot-hybridization data. The
DNA obtained from older leaves represents a smaller
number of cells than the same mass of DNA obtained from
young leaves, due to the increase in nuclear ploidy level
(genome copies per cell). Thus, constant hybridization
signals for cpDNA during development actually mean that
the fraction of cellular DNA represented by cpDNA is
higher for the older leaves. Since cpDNA replication ceases
after cells leave the meristem (Fujie et al., 1994), it seems
unlikely that cpDNA could possibly increase as leaves age.
Tobacco leaves, however, do not undergo endoreduplication during development (our unpublished observations)
and constant hybridization signals would reflect a constant
amount of cpDNA during development.
As discussed above, cpDNA first increases and then
decreases during leaf development and maturity. For
Arabidopsis, the increase occurs very early during development of the first leaf, when the leaf is less than 1.5 mm long
(about 15% expanded; Fujie et al., 1994; Rowan et al.,
2004). For tobacco, the increase occurs until the leaf is 30%
expanded (146 cm2; Shaver et al., 2006). Figure 1 is
a schematic representation of the changes in the average
DNA amount per plastid as a function of leaf development
for maize, Arabidopsis, and tobacco. Limited sampling of
developmental stages could lead to the erroneous conclusion that cpDNA remains constant throughout development. For example, a young leaf before cpDNA replication
and a mature leaf in which cpDNA is in decline can have
about the same amount of cpDNA, on average (compare
positions I and III for maize and Arabidopsis or positions II
and IV for tobacco). A constant amount of cpDNA would
also be observed if the period of decline had already passed
(positions IV and V for each plant).
Zoschke et al. (2007) concluded that the amount of
cpDNA remains constant throughout development in
Arabidopsis using data from real-time quantitative PCR
Loss of cpDNA as leaves mature | 3007
Fig. 1. Schematic representation of changes in the amount of
cpDNA per plastid during development in three plant species.
Increase in cpDNA amount due to cpDNA replication occurs very
early in development in maize (red line), followed by a rapid
decline. For Arabidopsis (blue line), the increase in cpDNA occurs
slightly later and the decline in cpDNA amount is much later. For
tobacco (grey line), cpDNA increases more gradually and the
decline is less severe. The Roman numerals indicate stages of leaf
development. I–III represent expanding leaves (LDS <1, see text),
and IV and V represent expanded leaves (LDS >1).
(qPCR) and nuclear ploidy. The ratio of cpDNA copies to
nuclear DNA copies was assessed for each developmental
stage using qPCR. This ratio was then multiplied by the
average ploidy of nuclei obtained from leaves of the same
developmental stage to calculate the number of chloroplast
genomes for an ‘average’ cell. We also found that the
amount of DNA per ‘average’ cell did not vary during
development using this method (Rowan et al., 2009). The
ratio of chloroplast-to-nuclear DNA decreases as leaves
age, but the average ploidy increases. There is a 2–3-fold
decrease in the chloroplast-to-nuclear DNA ratio, but the
increase in the average nuclear ploidy is also about 2–3-fold.
Multiplying these two parameters results in no change in
the calculated number of chloroplast genomes for an
‘average’ cell during leaf development. This ‘average’ cell is
a mathematical construct that may not be an accurate
representation of any cell (Levsky and Singer, 2003).
As discussed above, cpDNA increases during initial leaf
development. Why, then, does the amount of cpDNA
appear constant throughout leaf development if the qPCR/
nuclear ploidy method is used? It is possible that the
developmental stages selected by Zoschke et al. (2007)
missed the increase. Although the authors provided detailed
information regarding the developmental stage of the their
leaves, there is still some ambiguity. The authors equate
their ‘Ros20’ leaves to the 7-d-old and 9-d-old first rosette
leaves used in Rowan et al. (2004). Although these leaves
were, in fact, similar in size to ours (1–2 mm), they may not
have been at the same developmental stage. They did not
indicate which rosette leaves were used for these samples.
Their plants were 20-d-old, so it is unlikely that they used
the first rosette leaf because the first rosette leaf is typically
fully expanded to 8–10 mm (not 1–2 mm) after 20 d (Rowan
et al., 2007). Both the fourth and fifth rosette leaves of 20-dold plants are typically 1–2 mm in length. These leaves
reach approximately 15 mm in length when fully expanded.
It is possible that their leaf samples were not equivalent to
those used in Rowan et al. (2004), considering that 1–2 mm
long first rosette leaves are 10–20% expanded, but 1–2 mm
long fifth rosette leaves are only 7–13% expanded.
In addition, the method of calculation of chloroplast
genomes per ‘average’ cell using qPCR and average nuclear
ploidy could obscure cell-to-cell variation with respect to
participation in cpDNA replication and degradation. Many
leaf cells contain plastids that do not differentiate into
chloroplasts. DNA in these plastids does not amplify
dramatically during early leaf development, as it does in
developing chloroplasts of mesophyll cells (Miyamura et al.,
1990). A constant, low level of cpDNA is maintained in nongreen cells, while the amount of cpDNA in green cells (the
cells from which most of our isolated chloroplasts originated)
changes drastically. The effect of the cpDNA decline during
maturity may be obscured when an averaging method is used
because the cells participating in the decline are only
a fraction of all cells (Rowan et al., 2009).
In summary, the data of Li et al. (2006) and Zoschke
et al. (2007) may not conflict with those of Rowan et al.
(2004) and Shaver et al. (2006). Differences in the selection
of developmental stages and the methods used may explain
the different conclusions reached in these studies.
Alternative interpretations of the cpDNA
decline
Li et al. (2006) suggested that the undetectable levels of
cpDNA observed by Rowan et al. (2004) could have arisen
as an artefact of the isolation process because the DNase
used to remove extraorganellar DNA preferentially degraded the DNA of chloroplasts from mature leaves. While
DNase can artefactually reduce the amount of DNA per
chloroplast for tobacco and Medicago truncatula (Shaver
et al., 2006), our original data (Rowan et al., 2004) did
show, in fact, that DNase has no effect on the Arabidopsis
chloroplasts we studied. We have since adopted a high-salt
protocol for isolating chloroplasts that avoids the use of
DNase and still observe a developmental decline in cpDNA
amount (Rowan et al., 2007). Thus, it is unlikely that the
observation of chloroplasts with undetectable levels of
DNA in Arabidopsis resulted from DNase treatment during
isolation.
Zoschke et al. (2007) also suggested that our observation
of chloroplasts with undetectable DNA resulted from an
artefact of the isolation process. Chloroplasts from older
leaves are expected to contain larger starch grains than
those from younger leaves and might therefore be more
likely to rupture during isolation. Since the DAPI-DNA
staining of chloroplasts within leaf sections also becomes
less intense during leaf development (Rowan et al., 2009),
3008 | Rowan and Bendich
the cpDNA decline occurs in vivo and is not an artefact of
chloroplast isolation.
Additional hypotheses for how the original data from
Rowan et al. (2004) might not be due to degradation of
cpDNA during development have to be considered. In
chloroplasts of mature leaves, the autofluorescence of chlorophyll could quench the DAPI signal or DNA might be
present as small fragments incapable of generating a signal.
Neither of these hypotheses is supported by our data, which
showed that variation in DNA content among plastids as
measured by qPCR corresponds to variation in fluorescence
from DAPI (using fluorescence microscopy) or the more
sensitive fluorophore SYTO 42 (using flow cytometry)
(Rowan et al., 2009). The possibility also has to be considered
that the decline we observed for Arabidopsis is due to dilution
of a constant amount of cpDNA by chloroplast division, as
observed for some other plant species. However, a decline in
average DNA amount per plastid was found when chloroplast numbers are not increasing, indicating that dilution is
not responsible for the reduced DNA content of mature
chloroplasts (Rowan et al., 2009). The amount of cpDNA per
plastid declined 2–3-fold on average, which is similar to the
2–7-fold decline in average DNA amount per plastid for the
broader developmental range of tissues examined in our
original report (Rowan et al., 2004, 2009).
Viability in the absence of DNA
Cells and organelles may survive despite the complete loss
of genetic material. As red blood cells develop from their
haematopoietic precursors, they can lose their nuclear DNA
in mammals, but not in birds or amphibians. The mean
lifespan for red blood cells in humans is ;50 d (Cohen
et al., 2008). In the mesozoan Dicyema japonicum, mtDNA
copy number increases during early embryogenesis, then
decreases as larvae develop (Awata et al., 2005). In mature
larvae, mtDNA was detected in germ cells using in situ
hybridization, but ‘no signals were detected in peripheral
cells, reflecting an extremely low copy number or the
complete absence of mtDNA’. Mitochondria of adult
somatic cells were similarly found to retain little or no
mtDNA. Thus, organelles and cells may function for
extended periods without DNA.
Avoiding controversy in the future
There is a general lack of consistency in how leaf ‘age’ is
reported. Descriptions of leaf development range from
‘young’ and ‘mature’ to measurements of leaf size. Research
on all aspects of leaf development will be improved by the
establishment of a standard method for reporting the age of
leaves. Leaf size alone is not sufficient to determine the age of
the leaf because different leaves may reach different maximum
sizes. The plastochron index and leaf plastochron index can be
useful for comparing the development of individual plants or
leaves of a species (Erickson and Michelini, 1957; Larson and
Isebrands, 1971). These methods give a good estimate of leaf
development relative to a reference leaf. Since the reference
leaf size is arbitrarily assigned, however, it is difficult to
describe the true developmental status of the leaf, especially
if leaves on an individual plant reach different maximum
sizes or if leaf development is compared among plant species
with different maximum leaf sizes. For monocot leaves,
development is perhaps best expressed as a function of
distance from the basal meristem. This method is not suitable
for dicot leaves because the location of the meristematic
tissue is less distinct.
We propose defining leaf age as a function of leaf
expansion and time. Leaves are assigned a ‘Leaf Development Stage’ (LDS) using the following equation:
LDS ¼ ðS=MSÞþD
where S is the size (area or length) of the leaf, MS is the
maximum size that particular leaf reaches in the growth
conditions specified, and D is the number of days after the
leaf has reached full expansion.
LDS calculations for Arabidopsis are illustrated. A 1 mm
Arabidopsis first rosette leaf would have an LDS of 0.1
because S is 1, MS is 10, and D is 0 because the leaf has not
reached full expansion. A 1 mm fifth rosette leaf would have
an LDS of 0.07 because S is 1, MS is 15, and D is 0. A 10
mm first rosette leaf on a 30-d-old plant would have an LDS
of 11 because S is 10, MS is also 10, and D is 10, since
maximum expansion is reached at about day 20 (Rowan
et al., 2007). This is a useful and convenient method for
expressing the age of a leaf because LDS values less than 1
represent leaves that are still expanding, while LDS values
greater than 1 represent fully expanded leaves. (A leaf with
an LDS of 10 should not be considered to be 100-fold more
developed than a leaf with an LDS of 0.1. The LDS values
scale linearly with development until maximum expansion is
reached (LDS¼1). LDS values greater than 1 indicate the
passage of time since the completion of leaf expansion and
do not scale linearly with development.) Figure 2 shows how
LDS is useful to compare developmental status between two
leaves that reach different maximum sizes.
In addition to a common system for defining leaf age,
future controversy may also be avoided by exercising caution
when comparing data obtained from averaging methods to
data obtained using methods that measure variation among
individuals. It should be made clear whether cpDNA amount
refers to an average amount or to an amount per chloroplast,
per cell, or a fraction of total cellular DNA.
Conclusion
Like the two previous controversies in cpDNA research, the
controversy over whether the amount of cpDNA declines or
remains constant during development can be resolved.
Differences in the leaf material and methods employed in
different studies may explain the conflicting results. The
alternative explanations for the original finding that DNA
per chloroplast declines during development (Rowan et al.,
Loss of cpDNA as leaves mature | 3009
Acknowledgements
We thank Dr Delene Oldenburg for helpful discussions and
for critically reading this manuscript.
References
Awata H, Noto T, Endoh H. 2005. Differentiation of somatic
mitochondria and the structural changes in mtDNA during development of the dicyemid Dicyema japonicum (Mesozoa). Molecular
Genetics and Genomics 273, 441–449.
Baumgartner BJ, Rapp JC, Mullet JE. 1989. Plastid transcription
activity and DNA copy number increase early in barley chloroplast
development. Plant Physiology 89, 1011–1018.
Fig. 2. The leaf development stage (LDS) method for expressing
the age of leaves. The value for LDS expresses the developmental
stage of a leaf as a function of leaf expansion and time (see text).
The LDS values over time are shown for two different leaves. LDS
values less than 1 represent expanding leaves, and values greater
than 1 reflect the time since maximum expansion. The leaf shown
in (A) reaches a maximum size of 10 mm. The leaf shown in (B)
reaches a maximum size of 25 mm. Leaf size is measured as the
length of the leaf blade. At Time 1, both leaves are 6 mm long, but
they are not at the same developmental stage. At Time 2, both
leaves have just reached full expansion. At Time 3, both leaves
have been fully expanded for 4 d.
2004) are not supported by the most recent data (Rowan
et al., 2009). It is therefore concluded that DNA per
chloroplast does, in fact, decline as leaves mature.
DNA per chloroplast declines 2–7-fold in Arabidopsis
(Rowan et al., 2004, 2007, 2009), 2–3-fold in pea, M.
truncatula, and tobacco, and up to 10-fold in maize
(Oldenburg et al., 2006; Shaver et al., 2006). For maize and
M. truncatula, the light environment profoundly influences
the amount of cpDNA: plants grown in constant darkness
retain cpDNA during leaf development (Oldenburg et al.,
2006; Shaver et al., 2008). For maize, genes that affect
chloroplast biogenesis also affect the persistence of cpDNA
(Oldenburg et al., 2006).
A reluctance to accept cpDNA loss without compromising plant fitness may prolong the third and current
controversy. Yet, even single-celled algae shed cpDNA.
Chloroplasts of Acetabularia lose DNA during the vegetative phase of their life cycle (Coleman, 1979; Woodcock
and Bogorad, 1970), and Chlamydomonas reinhardtii
cpDNA is degraded in response to phosphorous limitation
without substantially affecting viability (Yehudai-Resheff
et al., 2007) and in the zygote during mating (Nishimura
et al., 2002). For mature leaves of multicellular plants, the
loss of cpDNA would not affect reproduction because
gametes arise from the shoot apical meristem, not the leaf.
Despite the large body of existing evidence, the third
controversy will only be laid to rest with the realization
that cpDNA degradation is a common event in chloroplast
development.
Bendich AJ. 2004. Circular chloroplast chromosomes: the grand
illusion. The Plant Cell 16, 1661–1666.
Cohen RM, Franco RS, Khera PK, Smith EP, Lindsell CJ,
Ciraolo PJ, Palascak MB, Joiner CH. 2008. Red cell life span
heterogeneity in hematologically normal people is sufficient to alter
HbA1c. Blood 112, 4284–4291.
Coleman AW. 1979. Use of the fluorochrome 4#,6-diamidino-2phenylindole in genetic and developmental studies of chloroplast DNA.
Journal of Cell Biology 82, 299–305.
Erickson RO, Michelini FJ. 1957. The plastochron index. American
Journal of Botany 44, 297–305.
Fujie M, Kuroiwa H, Kawano S, Mutoh S, Kuroiwa T. 1994.
Behavior of organelles and their nucleoids in the shoot apical meristem
during leaf development in Arabidopsis thaliana L. Planta 194,
395–405.
Galbraith DW, Harkins KR, Knapp S. 1991. Systemic endopolyploidy in Arabidopsis thaliana. Plant Physiology 96, 985–989.
Hashimoto H, Possingham JV. 1989. DNA levels in dividing and
developing plastids in expanding primary leaves of Avena sativa.
Journal of Experimental Botany 40, 257–262.
Kolodner R, Tewari KK. 1972. Molecular size and conformation of
chloroplast deoxyribonucleic acid from pea leaves. Journal of
Biological Chemistry 247, 6355–6364.
Kuroiwa T. 1991. The replication, differentiation, and inheritance of
plastids with emphasis on the concept of organelle nuclei. International
Review of Cytology 128, 1–61.
Larson PR, Isebrands JG. 1971. The plastochron index as applied
to developmental studies of cottonwood. Canadian Journal of Forest
Research 1, 1–11.
Lawrence ME, Possingham JV. 1986. Microspectrofluorometric
measurement of chloroplast DNA in dividing and expanding leaf cells
of Spinacia oleracea. Plant Physiology 81, 708–710.
Levsky JM, Singer RH. 2003. Gene expression and the myth of the
average cell. Trends in Cell Biology 13, 4–6.
Li W, Ruf S, Bock R. 2006. Constancy of organellar genome copy
numbers during leaf development and senescence in higher plants.
Molecular Genetics and Genomics 275, 185–192.
Miyamura S, Kuriowa T, Nagata T. 1990. Multiplication and
differentiation of plastid nucleoids during development of chloroplasts
3010 | Rowan and Bendich
and etioplasts from proplastids in Triticum aestivum. Plant and Cell
Physiology 31, 597–602.
Miyamura S, Nagata T, Kuroiwa T. 1986. Quantitative fluorescence
microscopy on dynamic changes of plastid nucleoids during wheat
development. Protoplasma 133, 66–72.
Nishimura Y, Misumi O, Kato K, Inada N, Higashiyama T,
Momoyama Y, Kuroiwa T. 2002. An mt(+) gamete-specific nuclease
that targets mt(–) chloroplasts during sexual reproduction in
C. reinhardtii. Genes and Development 16, 1116–1128.
Rowan BA, Oldenburg DJ, Bendich AJ. 2009. A multiple-method
approach reveals a declining amount of chloroplast DNA during
development in Arabidopsis. BMC Plant Biology 9, 3.
Scott NS, Possingham JV. 1980. Chloroplast DNA in expanding
spinach leaves. Journal of Experimental Botany 31, 1081–1092.
Shaver JM, Oldenburg DJ, Bendich AJ. 2006. Changes in
chloroplast DNA during development in tobacco, Medicago truncatula,
pea, and maize. Planta 224, 72–82.
Oldenburg DJ, Bendich AJ. 2004. Most chloroplast DNA of maize
seedlings in linear molecules with defined ends and branched forms.
Journal of Molecular Biology 335, 953–970.
Shaver JM, Oldenburg DJ, Bendich AJ. 2008. The structure of
chloroplast DNA molecules and the effects of light on the amount of
chloroplast DNA during development in Medicago truncatula. Plant
Physiology 146, 1064–1074.
Oldenburg DJ, Bendich AJ. 2009. Chloroplasts. In: Kriz AL, Larkins
BA, eds. Molecular genetic approaches to maize improvement, Vol.
63. Berlin, Heidelberg: Springer-Verlag, 325–326.
Sodmergen, Kawano S, Tano S, Kuroiwa T. 1991. Degradation of
chloroplast DNA in second leaves of rice (Oryza sativa) before leaf
yellowing. Protoplasma 160, 89–98.
Oldenburg DJ, Rowan BA, Zhao L, Walcher CL, Schleh M,
Bendich AJ. 2006. Loss or retention of chloroplast DNA in maize
seedlings is affected by both light and genotype. Planta 225, 41–55.
Woodcock CLF, Bogorad L. 1970. Evidence for variation in the
quantity of DNA among plastids of Acetabularia. Journal of Cell
Biology 44, 361–375.
Ris H, Plaut W. 1962. Ultrastructure of DNA-containing areas in
the chloroplast of Chlamydomonas. Journal of Cell Biology 13,
383–391.
Yehudai-Resheff S, Zimmer SL, Komine Y, Stern DB. 2007.
Integration of chloroplast nucleic acid metabolism into the phosphate
deprivation response in Chlamydomonas reinhardtii. The Plant Cell 19,
1023–1038.
Rowan BA, Oldenburg DJ, Bendich AJ. 2004. The demise of
chloroplast DNA in Arabidopsis. Current Genetics 46, 176–181.
Rowan BA, Oldenburg DJ, Bendich AJ. 2007. A high-throughput
method for detection of DNA in chloroplasts using flow cytometry.
Plant Methods 3, 5.
Zoschke R, Liere K, Börner T. 2007. From seedling to mature plant:
Arabidopsis plastidial genome copy number, RNA accumulation and
transcription are differentially regulated during leaf development. The
Plant Journal 50, 710–722.