/ . Embryol. exp. Morph. 83, Supplement, 89-101 (1984)
gO,
Printed in Great Britain © The Company of Biologists Limited 1984
Subroutines in the programme of Chlamydomonas
gene expression induced by flagellar regeneration
By JEFFERY A. SCHLOSS
Department of Biology, Yale University, New Haven, CT 06511 U.S.A.
TABLE OF CONTENTS
Summary
Flagellar regeneration
A model system for studying gene expression
Flagellar regeneration requires de novo protein synthesis
Accumulation of specific mRNAs accounts for changes in protein synthesis patterns during regeneration
RNA accumulation is due largely to increased transcription
Subroutines in the programme of gene expression during flagellar regeneration
References
SUMMARY
The unicellular green alga Chlamydomonas reinhardtii possesses two anteriorflagellathat
are rapidly replaced if they are lost. A cytoplasmic pool of flagellar precursors supports
regeneration of partial length flagella, but complete regeneration requires de novo synthesis
of flagellar proteins. This increase in protein synthesis is transient and is programmed by
changes in the physical abundance of a set of RNAs. These changes were measured using
cloned cDNAs. The curves that were generated with probes for a number of different RNAs
were variations-on-the-theme of a rapid accumulation following deflagellation of the cells,
followed by similarly rapid degradation. Differences in the characteristics of the accumulation
curves suggest that several 'subroutines' for RNA metabolism appear to run concurrently
duringflagellarregeneration. A significant portion of the mRNA abundance regulation occurs
at the transcriptional level. A group of cDNAs that encode mRNAs whose abundance remains
at constant levels during regeneration provides an important internal control for the abundance and transcription measurements. Flagella contain over 200 different proteins, so the
possibility exists that the cells coordinately and rapidly alter the transcription of the same
number of genes. Chlamydomonas flagellar regeneration thus provides an opportunity to
study cellular mechanisms for coordinating the expression of a large set of genes, and correlating these changes with an easily monitored morphogenetic process.
FLAGELLAR REGENERATION
A model system for studying gene expression
The use of organisms with flagella and cilia as eukaryotic model systems for the
study of gene expression is based on the observation that regeneration of these
90
J. A. SCHLOSS
organelles following their experimental detachment induces biosynthetic alterations in the cells. The pertinent observations made on deciliated Tetrahymena
(Guttman & Gorovsky, 1979; Marcaud & Hayes, 1979; Bird & Zimmerman,
1980), multiply deciliated Strongylocentrotus embryos (Stephens, 1977; Merlino,
Chamberlain & Kleinsmith, 1978), differentiating Naegleria (Kowit & Fulton,
1974; Lai, Walsh, Wardell & Fulton, 1979), and deflagellated Chlamydomonas
(see below) are that replacement of cilia or flagella is accompanied by changes
in the synthesis of specific proteins and in the synthesis and/or accumulation of
specific mRNAs.
This article will focus on Chlamydomonas because of the unique opportunities
this cell provides for tracing the flagellar regeneration process from the level of
organelle outgrowth (Lewin, 1953; Hagen-Seyfferth, 1959; Randall etal. 1967;
Rosenbaum, Moulder & Ringo, 1969) to the synthesis and assembly of cytoplasmic precursor proteins (Rosenbaum et al. 1969; Weeks & Collis, 1976; Weeks,
Collis & Gealt, 1977; Lefebvre, Nordstrom, Moulder & Rosenbaum, 1978;
Remillard & Witman, 1982; Brunke, Collis & Weeks, 1982; L'Hernault &
Rosenbaum, 1983), to the loading of flagellar mRNAs onto polysomes (Weeks
& Collis, 1976), to the accumulation of flagellar mRNAs (Lefebvre, Silflow,
Wieben & Rosenbaum, 1980; Silflow & Rosenbaum, 1981; Minami, Collis,
Young & Weeks, 1981; Silflow et al. 1982; Brunke, Young, Buchbinder &
Weeks, 1982; Schloss, Silflow & Rosenbaum, 1984), to transcriptional regulation
(Keller, Schloss, Silflow & Rosenbaum, 1984), and finally to the structure of the
flagellar protein genes (Silflow & Rosenbaum, 1981; Brunke et al. 1982). The
choice to consider the process in this way — from the end product to the gene
instead of the more conventional gene-to-gene-product view — can be justified
on historical grounds: this is how the data were collected due to development of
increasingly sensitive technologies. Moreover, this approach forms a useful
paradigm for considering cellular control of the events involved, because it is
presumably the direction of movement of cellular signals that modulate the
response. It is these signals and their modes of action that we hope eventually to
understand.
A corollary of the ability to explore the biosynthetic steps leading to organelle
morphogenesis is that, because the organelle in question is biochemically
heterogeneous, this should be an excellent system in which to ask whether particular elements of gene structure are important in controlling the expression of
a large number of protein-coding templates. A typical eukaryotic flagellum contains nine outer doublet microtubules, each composed of a complete A-tubule
and a partial B-subfibre, surrounding two complete central microtubules. In
addition to microtubules, several other structures such as dynein arms, radial
spokes, nexin links, and the central sheath are also present, regularly arrayed
along the length of the flagellum. It is not surprising, then, that when flagella are
isolated and analysed on polyacrylamide gels, they are found to contain, in
addition to the major microtubule proteins alpha- and beta-tubulin, a large
Programmed gene expression during flagellar regeneration
91
number of other polypeptides. In Chlamydomonas, the ability to dissect flagella
using functional, biochemical, morphological, and genetic approaches has led to
the mapping of a number of these polypeptides to specific structures. For example, pfl4 cells are unable to swim and contain a mutation in a single locus
(Ebersold, Levine, Levine & Olmsted, 1962). Electron microscopy demonstrates that the flagella of these cells are fully formed except for the lack of radial
spoke structures. This correlates with the absence of several proteins, discerned
by comparing polyacrylamide gels of flagella isolated from wild-type and pfl4
cells (Witman, Fay & Plummer, 1976; Piperno, Huang & Luck, 1977). Similar
analyses, along with flagellar fractionation studies, have led to the assignment to
specific flagellar structures of 60 of the over 200 different flagellar polypeptides
that may be visualized on polyacrylamide gels (Witman, Carlson, Berliner &
Rosenbaum, 1972; Witman, Plummer & Sander, 1978; Huang, Piperno & Luck,
1979; Piperno, Huang, Ramanis & Luck, 1981; Adams, Huang, Piperno &
Luck, 1981; Luck, 1984). It is these proteins that are synthesized during flagellar
regeneration and that provide the interesting possibility of studying cellular
mechanisms for coordinately regulating the expression of a large number of
genes.
Flagellar regeneration requires de novo protein synthesis
When Chlamydomonas cells are deflagellated by mechanical shear (Rosenbaum et al. 1969) or pH shock (Witman et al. 1972) rapid flagellar outgrowth
begins immediately and then continues at a decelerating rate until flagella reach
full length in 1 to 3 h. Both vegetative cells and gametes are capable of regenerating flagella. If cells are deflagellated in the presence of inhibitors of protein
synthesis, partial regeneration occurs. While cells normally regenerate 12-15 /im
long flagella, vegetative cells in cycloheximide regenerate only approximately
6/im long flagella, while cycloheximide-treated gametes regrow flagella only
about 3/mi long (Rosenbaum et al. 1969; Lefebvre et al. 1978). These experiments demonstrate that the cells contain a cytoplasmic pool of assembly competent flagellar precursors sufficient for assembling approximately half- or onefifth-length flagella, depending on the culture conditions. Another estimate of
the size of this pool was made by the completely independent criterion of
immunochemical determination of beta-tubulin content. These experiments
show that deflagellated cell bodies contain just over half as much beta-tubulin
protein as is present in a pair of flagella (Piperno & Luck, 1977), in close agreement with the vegetative cell assembly-competent-pool estimate derived from
the inhibitor experiment.
If the cytoplasmic pool of flagellar precursors can supply components for
regeneration of only partial length flagella, then deflagellation of these cells must
induce a programme of flagellar protein synthesis to provide for the complete
regeneration that normally occurs. Induction of protein synthesis has been
demonstrated by pulse-labelling cells with [35S]sulphate before deflagellation (to
92
J. A. SCHLOSS
establish a synthetic rate 'baseline') and at successive intervals during regeneration, and analysing the labelled whole-cell proteins by polyacrylamide gel
electrophoresis (Weeks et al. 1911 \ Lefebvre et al. 1978; Silflow et al. 1982;
Remillard & Witman, 1982). These studies demonstrate that the synthetic rates
of alpha- and beta-tubulin and several other polypeptides, some of which may
be identified as flagellar components by comigration with bona fide flagellar
proteins, are elevated in deflagellated cells, relative to their basal rates in nondeflagellated cells. Changes in the synthetic rates of these proteins occur over a
background of constant synthetic rates, before and during regeneration, of the
majority of the polypeptides visualized on the electropherograms.
Accumulation of specific mRNAs accounts for changes in protein synthesis patterns during regeneration
Polyribosomes isolated from regenerating cells direct the run-off translation
of more tubulin than do polysomes from control, nonregenerating cells (Weeks
& Collis, 1976). Is tubulin mRNA stored in control cells for subsequent loading
onto polysomes? Experiments in which RNA was isolated from control and
regenerating Chlamydomonas and then translated in vitro suggest that this is not
the case; rather the amount of mRNAs encoding tubulin and numerous other
flagellar proteins increases in response to deflagellation (Lefebvre et al. 1980).
To measure changes in mRNA abundance directly, recombinant DNAs complementary to the mRNAs have been cloned and selected. Using Chlamydomonas alpha- and beta-tubulin cDNAs (Silflow & Rosenbaum, 1980) or chicken
alpha-tubulin cDNA and Chlamydomonas beta-tubulin cDNA (Minami et al.
1981; Brunke et al. 1982) to probe RNA transfers, a dramatic increase in tubulin
mRNA has been demonstrated in regenerating cells. Accumulation of tubulin
mRNAs, over their control cell levels, is detectable within 2 min after deflagellation (Fig. 1). The mRNA levels continue to increase, peaking about an hour after
deflagellation, and then undergo a similarly rapid decrease. After flagella have
reached full length, in about 2 h, tubulin mRNAs have returned to basal levels.
Chlamydomonas cells contain two alpha- and two beta-tubulin mRNAs, which
correspond to the four tubulin genes of the organism, and which are coordinately
regulated during flagellar regeneration (Fig. 1 and Silflow & Rosenbaum, 1981;
Brunke et al. 1982). The alpha-tubulin mRNAs exhibit considerable homology,
as do the beta-tubulin mRNAs, but there is no evidence for homology between
alpha- and beta-tubulin mRNAs sequences. This cross hybridization accounts
for the visualization of both bands in the RNA transfer experiments shown in
Fig. 1, although each filter was probed with a cDNA corresponding to a single
alpha-tubulin or beta-tubulin mRNA.
Although the tubulin mRNAs appear coordinately regulated on RNA transfers, there are differences in the accumulation kinetics of alpha- and beta-tubulin
messages. In the experiment shown in Fig. 2, RNA was purified from samples
of a synchronous gametic culture before and at closely spaced times after
Programmed gene expression during flagellar regeneration
93
29362293a
206916311426293622932069-
16311426ndf
15
30
46
60
80
100 130
Fig. 1. Relative concentrations of alpha- and beta-tubulin mRNAs during flagellar
regeneration. RNA isolated from vegetative cells before (ndf) and at 2,15, 30, 46,
60,80,100, and 130 min after deflagellation was resolved on denaturing agarose gels
(2|Ug total RNA /lane) and transferred to nitrocellulose filters. The filters were
probed with 32P-labelled pcflO-2 (top panel) and pcf9-12 (bottom panel), which are
full-length alpha- and beta-tubulin cDNA clones, respectively. Autoradiography of
thefiltersreveals that tubulin mRNAs are present in control, non-deflagellated cells,
accumulate for about an hour, and then return to control cell levels after regeneration is complete. The sizes (in numbers of nucleotides) of marker fragments are
shown on the left. (Methods for this and subsequentfiguresare described in Schloss
etal. 1984.)
deflagellation. The RNA was applied in 'dots' to nitrocellulose filters, probed
with [32P]labelled cloned tubulin cDNAs, and hybridization was assayed by
scintillation counting (Schloss et al. 1984). Alpha- and beta-tubulin mRNAs
reach maximal levels, 11- to 12-fold higher than their control cell levels, at
EMB 83S
94
J. A. SCHLOSS
80
100
Minutes
120
140
160
180
Fig. 2. Different accumulation kinetics for alpha- and beta-tubulin mRNAs. Alphatubulin (•) and beta-tubulin (A) mRNAs were measured in RNA samples isolated
before and after deflagellation of gametes. The values are plotted as a ratio of the
amount of each mRNA present at each time point during regeneration relative to the
amount of that mRNA present in control cells (before deflagellation). Thus a value
of 1 indicates that an RNA species is present at the same level as in control cells.
Maximal accumulation of alpha-tubulin mRNA occurs earlier than maximal accumulation of beta-tubulin mRNA. The abundance of a constitutive mRNA (0, pcf813, see also Fig. 3) remains constant through the experiment. In this experiment,
flagella were 14-4(im long in control cells and regenerated to 13-6 /im long by
190 min.
different times after deflagellation. The accumulation curves are complex, with
two peaks (at 25 and 60 min for alpha and at 45 and 80 min for beta) separated
by a decrease in abundance (at 45 and 60min for alpha and beta, respectively).
The differences between the accumulation kinetics are corroborated by the
observations that (1) the abundance fluctuations occur at different times for
alpha- and beta-tubulins, (2) similar curves were obtained for the same RNA
experiment using a number of different tubulin cDNA probes (not shown), and
(3) an RNA (pcf8-13) whose abundance remains constant during regeneration
(Schloss et al. 1984) shows no fluctuation in this experiment (Fig. 2). Furthermore, differences in accumulation curves for alpha- and beta-tubulin mRNAs
have been observed in two independent experiments (Schloss et al. 1984 and
Programmed gene expression during flagellar regeneration
95
unpublished observations). The significance of these observations for understanding control of tubulin gene expression remains to be determined.
Protein-labelling experiments and translation of purified mRNA in vitro suggest that many mRNAs other than tubulin undergo significant abundance
changes during flagellar regeneration. To measure these changes, probes for
some of these mRNAs were prepared (Schloss et al. 1984). Poly(A) + RNA from
regenerating cells was reverse transcribed and inserted into a plasmid vector.
The resulting recombinant DNA bank was screened to identify cDNAs whose
homologous RNAs increase in abundance following deflagellation. Cloned
cDNAs whose corresponding RNAs remain at constant abundance before and
during regeneration were also selected. The latter serve as important controls for
a number of experiments.
The regulation patterns for the newly selected clones were studied (Schloss et
al. 1984). As in the experiments described above, RNA was isolated before and
after deflagellation of a synchronous gamete culture and the RNA was applied
to nitrocellulose filters. The cDNA probes were 32P-labelled by nick translation,
incubated with the RNA dots, and the extent of hybridization was determined
by scintillation counting. In addition to tubulins, eight other cDNAs have been
characterized that detect different mRNAs that accumulate following deflagellation (Fig. 3A). All of the RNAs are detected in control cells. The abundance of
some of these mRNAs (corresponding to cDNA clones pcf6-2, 6-187, 9-26, and
6-100) changes after deflagellation only by a factor of two or three, while others
(pcf3-21) undergo a 15-fold increase. Some of the RNAs attain maximal abundance at 30min in this experiment, others do so at 50min, while yet others
probably peak sometime between 30 and 50min. Most of the RNAs approach
control cell levels by the time flagella regenerate to full length (Fig. 3E).
The differences in timing of maximal accumulation for various RNAs are
probably not simply a function of size. For example, pcf6-8 peaks in abundance
earlier than does pcf6-100, yet these RNAs are 2680 and 1150 nucleotides long,
respectively (Schloss et al. 1984). The possibility remains, however, that the
original transcript for pcf6-100 is longer than that for pcf6-8.
Three additional cDNA clones reveal an entirely different pattern of mRNA
regulation. These RNAs increase two-to-four-fold in abundance within lOmin
after deflagellation, and by 30min have fallen to or below their levels in control
cells (Fig. 3B). Thus the changes in abundance for these RNAs are completed
while another set of RNAs (Fig. 3A) are still accumulating. At least two
possibilities exist concerning the identities of these mRNAs. They may encode
stress-induced proteins. A subset of the heat-shock proteins of Chlamydomonas
is induced by deflagellation (G. May & J. Rosenbaum, manuscript in preparation), but the kinetics of this response have not been well characterized. They
may alternatively be radial spoke protein mRNAs. Remillard & Witman (1982)
demonstrated that radial spoke proteins are synthesized most rapidly during the
first 15 min after gamete deflagellation. Attempts to identify the protein products
96
J. A. SCHLOSS
of these three mRNAs by hybrid selection and translation are in progress.
The synthetic rates of most of the proteins in the cells are not altered during
regeneration. Several 'constitutive' cDNA probes were selected. The abundance
measurements for each of these mRNAs vary from control cell levels by less than
25 % during flagellar regeneration (Fig. 3C).
The cDNA probes described to this point represent poly(A) + RNAs and each
hybridizes to a single band in RNA transfer experiments (Schloss et al. 1984).
Several cDNA clones have been identified that hybridize to a very large number
of different-sized poly(A) + RNAs. One example of such a clone is shown in Fig.
30
60
90 120
Regeneration time (min.)
C
30
60
90 120
Regeneration time (min.)
30
60
90 120
Regeneration time (min.)
Fig. 3. Accumulation curves for regulated and constitutive RNAs. RNA samples
isolated from control and regenerating gametes were dotted onto nitrocellulose
filters and probed with labelled cDNAs. Hybridization was assayed by scintillation
counting. The data are expressed as described in the legend to Fig. 2. (The beginning
point for each curve is arbitrarily placed on the ordinate to provide a graphic display
in which curves to not overlap.) The name of the cloned cDNA used to measure each
RNA is given to the right of each curve. Panels A, B and D contain data for RNAs
that show a number of differently shaped accumulation curves (see text). Results for
constitutive RNAs are shown in panel C. Each cDNA clone represents a different
RNA species except for clones 6-87 and 6-100, which contain portions of the same
RNA sequence. Flagellar length measurements for this experiment are shown in
panel E. Control cell flagellar length averaged 13-5 pan. Parts of this figure are
redrawn from Schloss et al. (1984).
Programmed gene expression during flagellar regeneration
97
4. Some of these RNAs are regulated during flagellar regeneration. The accumulation kinetics for these RNAs differ from those of RNAs described above. Also
present in control cells, these RNAs drop slightly below control cell levels soon
after deflagellation and then increase dramatically between 10 and 50min (Fig.
3D). They are lost from the cells, slowly at first and then more rapidly. Little
more is known about these RNAs at this time. Experiments are in progress to
determine their cytoplasmic localization and the organization of the genes that
encode them.
RNA accumulation is due largely to increased transcription
A set of RNAs can accumulate because transcription of the genes encoding
them increases or because the turnover rates of the RNAs themselves decrease.
To determine whether deflagellation induces changes in transcription, nuclei
were isolated from control and deflagellated Chlamydomonas cells, and transcription from these nuclei was carried out in vitro in the presence of labelled
RNA precursors. This experimental approach is based on the observation that
358322931631-
779517-
C
10 30
50
70
90
120
C
10
30
50 70
90
Fig. 4. A cloned cDNA that is homologous to a broad class of RNAs. RNA samples
from control and regenerating gametes (same experiment as Fig. 3) were resolved
on denaturing agarose gels (2jUg total RNA/lane), transferred to nitrocellulose
filters, and probed with pcf5-14 (left panel) and an alpha-tubulin cDNA clone (right
panel). Pcf5-14 contains a 1400 base pair cDNA insert and hybridizes to RNAs
covering a large range of molecular sizes. The size range of homologous RNAs does
not change during regeneration, although the hybridization intensity varies considerably (see Fig. 3D). Probing the same RNAs with a tubulin cDNA results in a discrete
hybridization band in each gel lane. The sizes of marker fragments appear on the left.
120
98
J. A. SCHLOSS
isolated nuclei maintain template specificity and are permeable to nucleotides,
so that transcripts may be labelled (Smith & Huang, 1976).
Products of nuclear transcription in vitro from control and regenerating cell
nuclei were assayed by hybridization to cloned cDNAs. Regenerating cell nuclei
are more active in transcribing alpha- and beta-tubulin RNAs than are control
cell nuclei. Yet the intensity of hybridization to constitutive cDNAs (eg. pcf2-40)
is the same for in vitro transcripts from both sets of nuclei (Keller et al. 1984).
These studies have recently been expanded to include several different cDNA
clones that are homologous to RNAs that accumulate during regeneration (L.
Keller & J. Schloss, unpublished observations). These experiments suggest that
much of the RNA accumulation that occurs during regeneration is the result of
increased transcription of a specific set of genes.
Nuclear transcription may provide an experimental system in which to test for
molecules that signal the nucleus to alter its pattern of gene expression. Such a
signal presumably arises in the cytoplasm and is related to flagellar length or
some other as yet undetermined physiological parameter. By adding cytoplasmic
fractions to isolated nuclei, it may be possible to reconstitute changes in transcription in vitro. These experiments will probably require nuclei that are capable of initiating new transcripts. Recent experiments provide encouraging
evidence that reinitiation occurs in the isolated nuclei (Keller et al. 1984; and
unpublished observations).
Experiments carried out in parallel with the isolated nuclei studies involve
labelling of RNA synthesized by the cells in vivo. Cells are pulsed with
[32P]orthophosphate before and after deflagellation, and then RNA is isolated
and hybridized with cloned cDNAs of interest. In agreement with the results
described above, these experiments demonstrate an increase in the transcription
rate of alpha- and beta-tubulin RNAs, but no change in the rate of transcription
of the constitutive RNA homologous to pcf2-40 (E. Baker, J. Schloss & J.
Rosenbaum, manuscript in preparation). The change in tubulin RNA transcription is sufficient to account for the majority of the tubulin RNA accumulation,
but is also consistent with the possibility that there may be changes in RNA
stability. Direct measurements of RNA stability are in progress.
Subroutines in the programme of gene expression during flagellar regeneration
The results described above suggest that the changes in gene expression that
occur in response to deflagellation of Chlamydomonas are more complex than
just a coordinate increase in transcription of a collection of genes. Different
RNAs undergo different patterns of accumulation, suggesting that subsets of the
large group of genes whose transcription changes following deflagellation of the
cells may respond to a variety of regulatory molecules. The observation that different RNAs peak in abundance at different times is particularly interesting in
light of the suggestion that proteins that assemble into a particular flagellar structure are synthesized simultaneously, but proteins that assemble into different
Programmed gene expression during flagellar regeneration
99
structures may be synthesized during various phases of the regeneration response
(Remillard & Witman, 1982). Thus it will be very interesting to determine
whether DNA sequence elements outside of the protein-coding regions of these
genes are shared by genes that are 'coordinately' regulated during the flagellar
response, and whether genes in different subsets actually have different regulatory sequences. The availability of probes for constitutively expressed sequences
will permit us to eliminate from consideration as putative specific flagellar
regulatory sequences, those regions of the DNA that may be shared by Chlamydomonas genes because they are involved in the basic mechanics of transcription.
If in addition to regulating transcription, the cells also regulate RNA stability,
then additional regulatory molecules and receptors must be invoked. Preliminary data from the in vivo RNA-labelling experiments described above, and
differences between the kinetics of abundance decreases for the various RNAs
for which probes are now available, may be interpreted as indicating that specific
groups of RNAs are subject to different degradation pathways, and that factors
involved in these pathways may themselves be subject to regulated expression.
Such speculation may be useful in devising testable hypotheses consistent with
the experimental data.
The studies on flagellar regeneration that are described here provide the basis
for continuing investigation in what has turned out to be a very tractable developmental model system. By combining the sensitivity of recombinant DNA technology with progress that has been made using genetic, biochemical, morphological,
and functional approaches to studying Chlamydomonas flagella, we expect in the
next few years to test theories that have broad relevance to understanding the
basic programmes of animal and plant development.
The author thanks Joel Rosenbaum for the opportunity to participate in his laboratory's
ongoing studies offlagellarregeneration, and Ellen Baker and Laura Keller for permission to
include some of their results prior to publication. The work was supported by U.S. Public
Health Service awards to J. Rosenbaum. The author was a Muscular Dystrophy Association
Fellow and a U.S. Public Health Service Fellow during parts of the work.
REFERENCES
G. M. W., HUANG, B., PIPERNO, G. & LUCK, D. J. L. (1981). Central-pair
microtubular complex of Chlamydomonasflagella:polypeptide composition as revealed by
analysis of mutants. /. Cell Biol. 91, 69-76.
BIRD,R. C. & ZIMMERMAN, A.M. (1980). Induction of tubulin synthesis during cilia regeneration in growing Tetrahymena. Expl Cell Res. 128, 199-205.
BRUNKE, K. J., COLLIS, P. S. & WEEKS, D. P. (1982). Post-translational modification of
tubulin dependent on organelle assembly. Nature 297, 516-518.
BRUNKE, K. J., YOUNG, E. E., BUCHBINDER, B. U. & WEEKS, D. P. (1982). Coordinate
regulation of the four tubulin genes of Chlamydomonas reinhardi. Nucleic Acids Res. 10,
1295-1310.
EBERSOLD, W. T., LEVINE, R. P., LEVINE, E. E. & OLMSTED, M. A. (1962). Linkage maps in
Chlamydomonas reinhardi. Genetics 47, 531-543.
GUTTMAN, S. D. & GOROVSKY, M. A. (1979). Cilia regeneration in starved Tetrahymena: an
inducible system for studying gene expression and organelle biogenesis. Cell 17, 307-317.
ADAMS,
100
J. A. SCHLOSS
M. (1959). Zu Kenntnis der Geisseln und der Chemotaxis von Chlamydomonas eugametos (Moewus). Planta 53, 376-401.
HUANG, B., PIPERNO, G. & LUCK, D. J. L. (1979). Paralyzed flagella mutants of Chlamydomonas reinhardtii defective for axonemal doublet microtubule arms. /. biol. Chem. 254,
3091-3099.
KELLER, L. R., SCHLOSS, J. A., SILFLOW, C. D. & ROSENBAUM, J. L. (1984). Transcription of
alpha- and beta-tubulin genes in vitro in isolated Chlamydomonas reinhardi nuclei. J. Cell
Biol. 98, 1138-1143.
KOWIT, J. D. & FULTON, C. (1974). Programmed synthesis of tubulin for the flagella that
develop during cell differentiation in Naegleria gruberi. Proc. natn. Acad. Sci., U.S.A. 71,
2877-2881.
LAI, E. Y., WALSH, C , WARDELL, D. & FULTON, C. (1979). Programmed appearance of
translatable flagellar tubulin mRNA during cell differentiation in Naegleria. Celt 17,
867-878.
LEFEBVRE, P. A., NORDSTROM, S. A., MOULDER, J. E. & ROSENBAUM, J. L. (1978). Flagellar
elongation and shortening in Chlamydomonas. IV. Effects of flagellar detachment,
regeneration, and resorption on the induction offlagellarprotein synthesis. /. Cell Biol. 78,
8-27.
LEFEBVRE, P. A., SILFLOW, C. D., WIEBEN, E. D. & ROSENBAUM, J. L. (1980). Increased levels
of mRNAs for tubulin and other flagellar proteins after amputation or shortening of
Chlamydomonasflagella.Cell 20, 469-477.
LEWIN, R. (1953). Studies on theflagellarof algae. II. Formation offlagellaby Chlamydomonas in light and darkness. Ann. N.Y. Acad. Sci. 56, 1091-1093.
L'HERNAULT, S. W. & ROSENBAUM, J. L. (1983). Chlamydomonas alpha-tubulin is posttranslationally modified in theflagelladuringflagellarassembly. J. Cell Biol. 97, 258-263.
LUCK, D. J. L. (1984). Genetic and biochemical dissection of the eucaryoticflagellum./. Cell
Biol. 98, 789-794.
MARCAUD, L. & HAYES, D. (1979). RNA synthesis in starved deciliated Tetrahymena pyriformis. Eur. J. Biochem. 98, 267-273.
MERLINO, G. T., CHAMBERLAIN, J. P. & KLEINSMITH, L. J. (1978). Effects of deciliation on
tubulin messenger RNA activity in sea urchin embryos. /. biol. Chem. 253, 7078-7085.
HAGEN-SEYFFERTH,
MINAMI, S. A., COLLIS, P. S., YOUNG, E. E. & WEEKS, D. P. (1981). Tubulin induction in C.
reinhardii: requirement for tubulin mRNA synthesis. Cell 24, 89-95.
G., HUANG, B. & LUCK, D. J. L. (1977). Two-dimensional analysis of flagellar
proteins from wild-type and paralyzed mutants of Chlamydomonas reinhardtii. Proc. natn.
Acad. Sci., U.S.A. 74, 1600-1604.
PIPERNO, G., HUANG, B., RAMANIS, Z. & LUCK, D. J. L. (1981). Radial spokes of Chlamydomonas flagella: polypeptide composition and phosphorylation of stalk components. J.
Cell Biol. 88, 73-79.
PIPERNO, G. & LUCK, D. J. L. (1977). Microtubular proteins of Chlamydomonas reinhardtii.
An immunochemical study based on the use of an antibody specific for the beta-tubulin
subunit. /. biol. Chem. 252, 383-391.
RANDALL, J., CAVALIER-SMITH, T., MCVITTIE, A., WARR, J. R. & HOPKINS, J. M. (1967).
Developmental and control processes in the basal bodies and flagella of Chlamydomonas
reinhardii. Devi Biol. Suppl. 1, 43-83.
REMILLARD, S. P. & WITMAN, G. B. (1982). Synthesis, transport, and utilization of specific
flagellar proteins during flagellar regeneration in Chlamydomonas. J. Cell Biol. 93,
615-631.
ROSENBAUM, J. L., MOULDER, J. E. & RINGO, D. L. (1969). Flagellar elongation and shortening in Chlamydomonas. The use of cycloheximide and colchicine to study synthesis and
assembly offlagellarproteins. J. Cell Biol. 41, 600-619.
SCHLOSS, J. A., SILFLOW, C. D. & ROSENBAUM, J. L. (1984). mRNA abundance changes
duringflagellarregeneration in Chlamydomonas reinhardtii. Molec. Cell. Biol. 4,424-434.
PIPERNO,
SILFLOW, C. D., LEFEBVRE, P. A., MCKEITHAN, T. W., SCHLOSS, J. A., KELLER, L. R. &
ROSENBAUM, J. L. (1982). Expression offlagellarprotein genes duringflagellarregeneration
in Chlamydomonas. Cold Spring Harbor Symp. Quant. Biol. XLVI, 157-169.
Programmed gene expression during flagellar regeneration
101
C. D. & ROSENBAUM, J. L. (1981). Multiple alpha- and beta-tubulin genes in
Chlamydomonas and regulation of tubulin mRNA levels after deflagellation. Cell 24,
81-88.
SMITH, M. M. & HUANG, R. C. C. (1976). Transcription in vitro of immunoglobulin kappa light
chain genes in isolated mouse myeloma nuclei and chromatin. Proc. natn. Acad. Sci.,
U.S.A. 73, 775-779.
STEPHENS, R. E. (1977). Differential protein synthesis and utilization during cilia formation
in sea urchin embryos. Devi Biol. 61, 311-329.
WEEKS, D. P. & COLLIS, P. S. (1976). Induction of microtubule protein synthesis in Chlamydomonas reinhardii during flagellar regeneration. Cell 9, 15-27.
SILFLOW,
WEEKS, D. P., COLLIS, P. S. & GEALT, M. A. (1977). Control of induction of tubulin synthesis
in Chlamydomonas reinhardi. Nature 268, 667-668.
G. B., CARLSON, K., BERLINER, K. & ROSENBAUM, J. L. (1972). Chlamydomonas
flagella. I. Isolation and electrophoretic analysis of microtubules, matrix, membranes, and
mastigonemes. J. Cell Biol. 54, 507-539.
WITMAN, G. B., FAY, R. & PLUMMER, J. (1976). Chlamydomonas mutants: evidence for the
roles of specific axonemal components in flagellar movement. In Cell Motility (ed. R.
Goldman, T. Pollard & J. Rosenbaum), pp. 969-986. Cold Spring Harbor Laboratory.
WITMAN, G. B., PLUMMER, J. & SANDER, G. (1978). Chlamydomonasflagellarmutants lacking
radial spokes and central tubules. Structure, composition, and function of specific axonemal
components. J. Cell Biol. 76, 729-747.
WITMAN,
NOTE ADDED IN PROOF: The DNA sequences of all four Chlamydomonas tubulin gene
promoters and 5' coding regions have now been published and reveal several structures with
possible regulatory significance (K. Brunke, J. Anthony, E. Sternberg & D. Weeks, 1984.
Molec. Cell. Biol. 4,1115-1124). DNA sequence determinations for the entire coding regions
of the two beta-tubulin genes reveal that the proteins they encode are identical. Three intervening sequences occupy identical positions in these genes and the third is highly conserved
between the two genes (J. Youngblom, J. Schloss & C. Silflow. 1984. Molec. Cell. Biol. In
press).