Plant Cell Advance Publication. Published on January 30, 2017, doi:10.1105/tpc.16.00768
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
LARGE-SCALE BIOLOGY ARTICLE
Protein Degradation Rate in Arabidopsis thaliana Leaf Growth and
Development
Lei Li, Clark J. Nelson, Josua Trösch, Ian Castleden, Shaobai Huang, A. Harvey Millar*
ARC Centre of Excellence in Plant Energy Biology, Bayliss Building M316, University of
Western Australia, 35 Stirling Highway, Crawley 6009, Western Australia, Australia
*Corresponding author: [email protected]
Short title: Protein turnover in leaf development
One-sentence summary: The degradation rate of 1228 Arabidopsis proteins were
measured, their variation assessed and the data used to calculate the protein turnover
energy costs in different leaves of the rosette.
The authors responsible for distribution of materials integral to the findings presented in this
article in accordance with the policy described in the Instructions for Authors
(www.plantcell.org) is Harvey Millar ([email protected]).
ABSTRACT
We have applied 15N labeling approaches to leaves of the Arabidopsis thaliana rosette to
characterize their protein degradation rate and understand its determinants. The progressive
labeling of new peptides with 15N and measuring the decrease in the abundance of >60,000
existing peptides over time allowed us to define the degradation rate of 1228 proteins in vivo.
We show that Arabidopsis protein half-lives vary from several hours to several months based
on the exponential constant of the decay rate for each protein. This rate was calculated from
the relative isotope abundance of each peptide and the fold change in protein abundance
during growth. Protein complex membership and specific protein domains were found to be
strong predictors of degradation rate, while N-end amino acid, hydrophobicity or aggregation
propensity of proteins were not. We discovered rapidly degrading subunits in a variety of
protein complexes in plastids and identified the set of plant proteins whose degradation rate
changed in different leaves of the rosette and correlated with leaf growth rate. From this
information, we have calculated the protein turnover energy costs in different leaves and
their key determinants within the proteome.
1
©2017 American Society of Plant Biologists. All Rights Reserved
39
INTRODUCTION
40
41
For plants to respond to the daily requirements of cellular maintenance and for their organs
42
to progress through different developmental stages, new complements of proteins need to
43
be synthesized while existing ones are degraded. Protein synthesis occurs via ribosomes in
44
the cytosol, plastid and mitochondrion. Synthesis rate is defined by the abundance and
45
availability of different mRNA (Juntawong et al., 2014), the energetic status of cells to
46
provide the amino acids and ATP for catalysis (Edwards et al., 2012), and the post-
47
translational regulation of ribosomal function (Ren et al., 2011). Cellular protein degradation
48
is achieved by the coordinated action of the proteasome on ubiquitinated proteins in the
49
cytosol (Vierstra, 2009), autophagy of cytosolic complexes and organelles through vacuole-
50
based protein degradation machinery (Araujo et al., 2011) and a network of organelle-
51
localized proteases (Janska et al., 2013; van Wijk, 2015). The rates of synthesis and
52
degradation of individual proteins define the proteomes of plant tissues and their rate of
53
change.
54
55
The combination of protein synthesis and degradation is often termed protein turnover;
56
however, there are many different definitions and assumptions made in using this term and
57
in calculating it. Protein degradation rates and calculated turnover rates can be measured on
58
a total plant protein basis by combining all polypeptides as a group and measuring rates at
59
which ‘old protein’ disappears or ‘new protein’ replaces old. These types of calculation look
60
at the amount of original protein remaining and/or the proportion of new protein to old at
61
several time points and calculate a rate of renewal. This approach typically does not
62
consider the inherent rate of degradation of the newly synthesized material. Rates calculated
63
without considering degradation of new proteins are intrinsically linked to the growth rate of
64
the tissue which can dramatically alter the pool size of the new protein component over time.
65
Protein turnover of roots has been measured this way using pulse chase
66
labelling of fast and slow growing grass species (Scheurwater et al., 2000) to define
67
degradation rates of 0.12-0.16 d-1 and a total protein half-life of 4-6 days. Using Arabidopsis
68
thaliana rosettes and 13CO2 pulse-chase labelling (and a focus on the relative isotope ratio of
69
Ala residues) degradation rates of 0.03-0.04 d-1 and a leaf protein half-life of 3.5 d were
70
calculated by combining degradation and original protein dilution through new protein
71
synthesis (Ishihara et al., 2015).
14
C-Leucine
72
73
Protein degradation and turnover can also be measured for specific proteins of interest. This
74
requires a method to identify individual proteins and to differentiate between old and new
75
polypeptides for the same protein over the timeframe of measurement. Early studies used
2
76
antibodies and generic protein biosynthesis inhibitors to do this, but given the cellular
77
consequence of protein biosynthesis inhibition, there are many pleiotropic effects in such
78
measurements in all but very short timeframes (Vogtle et al., 2009; Armbruster et al., 2010).
79
The combination of stable isotopes and peptide mass spectrometry have recently allowed
80
old and new protein populations to be differentiated without the need for protein biosynthesis
81
inhibition and have become a method of choice in new studies. The experimental
82
measurement of protein degradation rate (KD) through isotopic labeling and mass
83
spectrometry (MS) has enabled datasets of hundreds of individual protein degradation rates
84
and half-lives to be assessed in yeast, mammalian cell culture and intact animals typically
85
using single labeled amino acids (SILAC) (Price et al., 2010; Schwanhausser et al., 2011;
86
Yip et al., 2011). Plants are typically not amenable to successful SILAC labelling due to their
87
synthesis and interconversion of all protein amino acids. As a result, studies have assessed
88
the utility of various metabolic labels (2H,
89
2010; Chen et al., 2011; Li et al., 2012). Recently, metabolic labelling with inorganic
90
been the most commonly used stable isotope incorporation technique to define degradation
91
or turnover rates of organelle proteins in Arabidopsis cell culture (Nelson et al., 2013),
92
protease targets in Arabidopsis leaf mitochondria (Huang et al., 2015; Li et al., 2016), ~500
93
proteins in barley (Hordeum vulgare) leaves (Nelson et al., 2014), ~200 membrane and
94
microsomal fraction proteins from Arabidopsis roots (Fan et al., 2016), and ~500 Medicago
95
leaf and root proteins during drought and recovery from drought (Lyon et al., 2016). One
96
study has also followed selected protein degradation rates using 13CO2 in Arabidopsis leaves
97
(Ishihara et al., 2015). While the primary data from labelling in each case represent relative
98
isotope abundance (RIA) for a given peptide, different methods have been used to calculate
99
protein degradation, synthesis, and turnover rates and protein half-lives. Some of these
100
calculations considered the growth rate of tissues, the degradation rate of newly synthesized
101
proteins and the relative abundance of different proteins during labelling, while others did
102
not. Depending on the method used, the KD reported can be highly dependent on tissue
103
growth rate and/or on relative abundance of the protein over time and may not represent a
104
property of the polypeptide per se. As a result, we still have limited data on the rate of
105
protein degradation for different classes of plant proteins and little clarity about whether
106
degradation rate is a variable or constant feature of the lifecycle of a specific plant
107
polypeptide type.
13
C, and
15
N) for the purpose in plants (Yang et al.,
15
N has
108
109
Here, we aimed to define the key features of proteins that determine their experimental
110
degradation rate and measure changes in these rates for given protein sets at different
111
stages of Arabidopsis leaf growth through assembly of a larger and more precise data set
112
than used in previous studies. This approach provides a foundational dataset for protein
3
113
degradation rates in Arabidopsis and evidence of which proteins have variable degradation
114
rates as leaves grow. In the process, we have newly identified rapidly degrading subunits in
115
a variety of protein complexes in plastids, found the set of proteins for which degradation
116
rate positively or negatively correlates with leaf growth rate and considered the protein
117
turnover energy costs for total and specific proteins in different leaves of the Arabidopsis
118
rosette.
119
120
RESULTS
121
122
Changes in Arabidopsis leaf protein content at different stages of leaf development
123
When specific proteins remain in steady-state within the proteome over the time course of
124
15
125
the rate of tissue growth. To calculate this dilution effect, Arabidopsis thaliana seedlings were
126
grown in natural abundance hydroponic culture medium (99.6%
127
produced their tenth true leaf, a stage referred to as leaf production stage 1.10 (Boyes et al.,
128
2001). To compare their total protein abundance, leaf number 3, 5 and 7 from a series of
129
rosettes were collected over a period of 5 days (Supplemental Figure 1A). Proteins were
130
extracted from these leaf samples for analysis of total leaf protein content (Figure 1). This
131
allowed an assessment of the growth rate of different leaves of the rosette on a protein
132
basis. We assessed protein content by protein extraction and amido black measurement
133
(Supplemental Data set 1) and by imaging gel-separated proteins from whole leaves
134
ground into sample buffer (Supplemental Figure 1B, Supplemental Data set 1). leaf 3
135
increased its total protein content by only 3% per day, in contrast to leaves 5 and 7, which
136
grew at three and five times this rate, respectively (Figure 1). As a consequence, leaf 7
137
nearly doubled in size and total protein content over five days while leaf 3 hardly grew at all.
N progressive labelling, a single dilution effect can be calculated and applied to account for
14
N) for 21 days until they
138
139
Protein degradation rate of specific proteins in Arabidopsis leaves
140
Leaves in the Arabidopsis rosette were then sampled at the same time points as above, but
141
in this case the hydroponic medium was changed at T=0 from 99.6% 14N to 98% 15N to allow
142
progressive labelling of each leaf over the time course of 5 days. A flow diagram of the
143
experiment and the sample preparation for mass spectrometry are shown in Supplemental
144
Figure 2. Extensive LC-MS/MS analysis of these samples provided data from 61,196
145
independently quantified peptides for which we could acquire the ratio between natural
146
abundance (NA) and heavy (H) labelled peptides. Using the ratio of labelled to NA peptides
147
and fold change in protein (FCP) as each leaf grew (Figure 1), we estimated degradation
148
rate for 1228 non-redundant Arabidopsis proteins. Inclusion in this list required each protein
149
to be quantified in >=3 of 27 samples, the number of peptides quantified independently in
4
150
the 27 samples to be greater or equal to 5 and for the protein identification to be p>0.95 as
151
outlined in Methods (Supplemental Data set 2A, B). Sample calculations of Labeled Protein
152
Fraction [LPF, i.e. H/(H+NA)] are shown for a slow degrading peptide (Supplemental
153
Figure 3A) and a fast degrading peptide (Supplemental Figure 3B) in each of the three
154
leaves over the time course. The degradation rates (KD) of 1228 proteins were calculated as
155
explained in the Methods section and as shown in Supplemental Figure 3C for the example
156
of a fast and a slow degrading peptide. KD is the proportional decrease of the current protein
157
abundance per day; i.e. it is the exponential constant of the decay rate for each protein and
158
as such has units of d-1. These KD rates showed a wide distribution ranging from 0 to 2 d-1,
159
which is equivalent to half-lives for proteins of several hours to several months. These are
160
the average KD across all cell types in a leaf tissue examined and, like all quantitative
161
proteomics, any cell type-specific differences will be lost in this averaging. The median
162
relative standard error (RSE) was < 10% for measurement of KD of a specific protein
163
(Supplemental Data set 2B). In general, about 15% of the data (184 proteins) showed
164
degradation rates more than two times the KD median (~0.22 d-1), which we defined as
165
relatively fast degrading proteins; about 13% (161 proteins) showed degradation rates of
166
less than half the median (~0.055 d-1), which we defined as relatively slowly degrading
167
proteins; while the majority (72%) showed values in the four fold range between these KD
168
values (Supplemental Data set 2B).
169
170
Determination of the fraction of proteins that are labeled, represented by the LPF value, is a
171
binary differentiation between two populations of peptides. As the peptides measured
172
typically contained 8-30 Ns per peptide, only a few N atoms anywhere in the peptide would
173
place a peptide in the heavy-labelled pool. All such peptides were considered heavy-labelled
174
and the degree of
175
measurements of the incorporation of a specific amino acid within a peptide, which requires
176
a detailed knowledge of the peptide incorporation ratio in that AA and the free AA pool
177
incorporation ratio to calculate peptide synthesis levels (Ishihara et al., 2015). However,
178
despite these advantages, the LPF approach employed here can be affected by a lag before
179
new protein synthesis can be effectively distinguishable from previously present (NA)
180
proteins. The lag is overcome as soon as a 10-15% labelling level is reached in major AA
181
pools (Nelson et al., 2014). The three different leaves did show some differences in
182
labelling efficiency (Supplemental Figure 4). A regression method can be used to determine
183
the lag effect as the lag in the x-intercept value when graphing of all timepoints (Nelson et
184
al., 2014). Calculations based on this method showed leaves 3, 5 and 7 had a 5-6 h lag
185
before
186
represented 4-25% of the time interval for day 5, day 3 and day 1 data, respectively, which
15
N labelling formed no part of our calculations. This is distinct from
15
N
15
N reached this 10-15% threshold (L3 = 4.7 h; L5 = 5.9 h, L7 = 5.2 h). This
5
187
was similar to the relative standard error of our calculations for any one protein. To examine
188
if this calculable lag would systematically effect KD values, we selected 146 proteins that
189
were quantified in all samples at all timepoints and plotted their degradation rates together
190
across all three replicates, timepoints and leaf types. No significant trends were apparent to
191
indicate that this general effect would bias our data (Supplemental Figure 5). We are not
192
aware of literature that has divided leaves into cell types to show differences in
193
incorporation rate within leaves to justify the concern that lag averages are not a fair
194
reflection for an average leaf calculation.
15
N
195
196
Grouping proteins by their functional category (Figure 2A) showed that most protein groups
197
maintained the median degradation rate of about 0.11 d-1. Exceptions were proteins involved
198
in protein synthesis that degraded at closer to 0.03 d-1 and stress, secondary metabolism
199
and redox proteins that degraded at rates approaching 0.20 d-1. Individual proteins
200
degrading at very high rates were found in many functional categories and typically showed
201
rates from 0.6 up to 2.0 d-1, which was 6 to 20 times the median KD and 4 to 13 standard
202
deviations from the median (Figure 2A). The 20 fastest degrading proteins comprised a set
203
of 8 plastid proteins, 6 cytosolic proteins and 6 proteins localized elsewhere in the cell (Table
204
1). To look for evidence of differences in the rate of autophagy of cellular structures or the
205
role of intraorganellar proteases in changing the degradation rate inside specific organelles,
206
we arranged the dataset into the final subcellular location of proteins using the location
207
Bayesian classifier, SUBAcon (Tanz et al., 2013; Hooper et al., 2014). This showed a lower
208
median degradation rate of mitochondrial, plastidial and nuclear proteomes and a higher
209
median degradation rate for the proteomes of endoplasmic reticulum (ER) and the Golgi
210
apparatus. This is broadly consistent with the physical separation of organelle proteomes
211
from the cytosolic proteolysis system and the transient nature of the protein complement of
212
ER and Golgi (Figure 2B). Direct comparison of protein degradation rates in this dataset to
213
either orthologues in barley (Nelson et al., 2014) was possible because the same
214
measurement approach was used in both of these studies (Figure 2C). This comparison
215
showed that four ortholog pairs between the two species had consistently high degradation
216
rates, namely the thiamin synthesis enzymes THI1 and THIC, D1 of PSII and PTERIN
217
DEHYDRATASE (R=0.99). However, more broadly this comparison showed a relatively low
218
correlation between degradation rate of the majority of Arabidopsis and barley homologous
219
pairs (R=0.38).
220
221
Changes in Arabidopsis leaf proteomes at different stages of development and across
222
the diurnal cycle
223
To check if proteins remained in steady-state within leaf proteomes over the time course of
6
224
our measurements, the relative abundances of specific Arabidopsis proteins were assessed
225
in the samples from different timepoints by spiking natural abundance samples after protein
226
extraction with a
227
leaves kept as a reference standard. Mixed proteins were then in-solution digested by
228
trypsin for mass spectrometry analysis. A flow diagram of this experiment and the sample
229
preparation for mass spectrometry are also shown in Supplemental Figure 2. This provided
230
quantitation of 1697 non-redundant proteins. Proteins were included in this group if they
231
were quantified in >=3 of 36 samples, the number of peptides quantified independently in the
232
36 samples was greater than or equal to 5 and the protein identification had p>0.95 as
233
outlined in methods (Supplemental Data set 3A). In total, 955 proteins in this set of 1697
234
were also found in the set of 1228 proteins for which we calculated a KD value (Figure 3A,
235
Supplemental Data set 3B). Principal components analysis (PCA) of protein relative
236
abundance data from this mass spectrometry experiment showed that the three different
237
leaves in the rosette could be separated by a single principal component explaining 43% of
238
the variation (Figure 3B). Only minor variations were observed in the PCA between samples
239
of the same leaf number at different points in time. Changes in relative protein abundances
240
over time in the different leaf proteomes were compared and statistically evaluated by T-
241
tests and 1-way ANOVA (Figure 3C; Supplemental Data set 3C). This analysis indicated
242
that only a few percent of proteins significantly changed in abundance in leaf 3 across the
243
whole time course, which was close to expectation for random variation in the experiment
244
(p<0.01). Leaf 7 showed a similar pattern of percentage differences with only the samples
245
from the day 1 showing significant differences from day 5 samples. No specific proteins were
246
found to change in abundance between all timepoints in leaf 3 and leaf 7 (Figure 3C). By
247
contrast, leaf 5 showed 13-33% of the analyzed protein set differing in relative abundance by
248
day 5; however, only 0.4% (two proteins) showed consistent changes across the time course
249
(Supplemental Data set 3C). This revealed that relatively high abundance proteins of leaf 3
250
and leaf 7 were effectively in steady state during the experiment, albeit with different rates of
251
new tissue growth (Figure 1), and thus steady-state principles could be applied to establish
252
protein degradation rates for existing proteins in leaves 3 and 7 over a 5 day period. For leaf
253
5, some variation in abundance was evident but only at the day 5 timepoint. The majority of
254
data showed only 20-30% changes in abundance, for the proteins that differed at day 5,
255
which will not have a major effect on KD calculations.
15
N fully-labeled protein sample from a separate batch of Arabidopsis
256
257
One important factor that could explain the need for rapid degradation of some proteins is a
258
circadian or diurnal change in transcription and translation that would need to be matched by
259
the degradation rate to enable a daily rhythm in protein abundance. Expression profiling of
260
Arabidopsis leaves has previously shown that diurnal patterns of transcription can underlie
7
261
daily fluctuations in protein abundances in leaves (Mockler et al., 2007; Giraud et al., 2010;
262
Lee et al., 2010). To test if enhanced degradation rate correlates with diurnal gene
263
expression or protein abundance changes, we attempted to correlate the calculated KD with
264
the degree of change in transcript abundance of diurnally regulated genes (Mockler et al.,
265
2007) or the degree of change in Arabidopsis protein abundance following 8 hours of
266
illumination (Liang et al., 2016); however we found no significant correlation of these with KD
267
(R=0.01 and R=0.02-0.08; n=917 and 1158, respectively). By contrast, when the 1228 KD
268
values were separated into slow (n=161), intermediate (n=883) and fast (n=184) groups, as
269
outlined above, and compared to light-induced, dark-induced and non-diurnal transcripts (as
270
defined by Mockler et al., 2007), non-parametric tests showed an enrichment (p< 0.01) of
271
light-induced transcripts in the fast KD group compared to the non-diurnal or dark-induced
272
transcripts. This apparent link between light-induced diurnal expression and fast KD was
273
observed in long-day or 12h-12h experiments in both Col-0 and Ler backgrounds (Smith et
274
al., 2004; Blasing et al., 2005; Mockler et al., 2007).
275
276
Intrinsic protein properties linked to degradation rate
277
The fact that protein degradation rate of individual proteins was quite reproducible between
278
leaf types over the time course indicated that some factors intrinsic to each protein’s
279
expression, protein properties or susceptibility to modification, unfolding or proteolysis could
280
be a central controller of protein degradation rate. The N-end rule proposes that the N-
281
terminal amino acid of a protein is a key driver for turnover rate. A series of stabilizing and
282
destabilizing N-terminal AAs having already been experimentally defined in yeast and
283
mammals (Bachmair et al., 1986; Gonda et al., 1989). However, we were unable to find any
284
statistical linkage between KD and either the N-end AA of mature proteins or sets of AAs
285
considered to be stable, unstable or of average stability based on yeast and mammalian
286
data (Figure 4A, Supplemental Data set 4A). AA residues within protein sequences can
287
promote ordered aggregation, which can be a precondition for altered rates of protein
288
degradation in mammals. The TANGO algorithm can predict this aggregation propensity
289
(Fernandez-Escamilla et al., 2004). In our set of 1228 proteins, the top 100 fastest degrading
290
proteins show significantly higher protein aggregation propensity (AGG; p<0.05) and longer
291
AA length (LEN; p<0.05), but no significant difference (p=0.70) in AGG/LEN when compared
292
to the 100 slowest turnover proteins (Figure 4B, Supplemental Data set 4B). This result
293
implies that the aggregation propensity of faster degradation proteins in Arabidopsis was
294
dependent on length and there was a tendency for proteins to be longer in the set of faster
295
degrading proteins. Integral membrane proteins require different machinery for degradation
296
than soluble proteins and have different accessibility to proteases (Fleig et al., 2012; Avci
297
and Lemberg, 2015). Neither grand average of hydrophobicity (GRAVY) nor transmembrane
8
298
domain number showed overall correlations with the KD of proteins, but beta barrel structure
299
proteins had very low KD values compared to other groups of proteins (Supplemental Data
300
set 4C).
301
302
Many proteins do not operate as independent monomers in vivo but are present in strict
303
stoichiometry in protein complexes. We tested the hypothesis that degradation rates of
304
members of a complex would be more consistent with other members of the same complex
305
than the wider set of cellular proteins. The members of seven major protein complexes
306
representing the cytosol ribosome, proteasome, plastid ATP synthase, plastid ribosome,
307
photosystem I (PSI), photosystem II (PSII) and mitochondrial electron transport chain (Mt
308
ETC) were extracted from the 1228 dataset, and all other proteins in the 1228 were
309
considered in one group, designated as ‘Others’. Outlier data for specific protein complexes
310
and from the ‘Others’ group were removed for the analysis. The standard deviation (σ) was
311
calculated for KD values for each protein complex (Figure 4C, red line) with N subunits and
312
this was compared to 100,000 randomly sampled sets of the same N from the ‘Others’
313
category (blue distributions). Probability (p) values showed that in all cases, except the
314
cytosolic ribosome, the standard deviations were significantly smaller for the complexes than
315
for the random distribution sets of the same N. In a complementary approach, a pairwise
316
Silhouette (Si) clustering comparison (Rousseeuw, 1987) was performed between each
317
protein complex and ‘Others’. The ∆Si value (-2, 2) in each case showed the level of relative
318
tightness of the KD rates of the specific protein complexes compared with ‘Others’. Plastid
319
ATP synthase, PSI and proteasome had the highest ∆Si values and were thus the three
320
protein complexes with the tightest distribution in terms of the degradation rates of their
321
subunits (Supplemental Data set 4D).
322
323
A protein’s function or domain structure may also be a factor in defining its degradation rate.
324
The high degradation rate of proteins has previously been explained based on specific
325
features of a protein fold that enables rapid degradation in protein abundance, e.g. the auxin
326
degron (Nishimura et al., 2009), or the irreversible binding of inhibitors to a protein domain or
327
due to an enzymatic mechanism that damages an active site (e.g. self-hydrolyzing enzymes,
328
reactive oxygen species damage, co-substrate enzymes or catalysis with suicide substrates)
329
(Walsh, 1984; Chatterjee et al., 2011). To test systematically for domains in Arabidopsis
330
proteins linked to degradation rate, we searched for differences in protein degradation rate
331
distributions between protein sets with common protein domains. Fifty-eight non-redundant
332
protein domain sets were selected for non-parametric k-sample K-W (p<0.01) and C-I post-
333
hoc testing (Table 2, Supplemental Data set 4E). These protein domains can be divided
334
into 10 groups (A-J) based on degradation rates. The four proteins containing the SH3
9
335
domain involved in translation were the most stable proteins (median of 0.05 d-1) while the
336
four proteins (three E2 and one E1 ligases) containing the ubiquitin-conjugating enzyme
337
domain (median of 0.38 d-1) were the most unstable.
338
339
Differences in degradation rate and abundance of specific proteins in leaves of
340
varying ages
341
While overall protein degradation rates of individual proteins were relatively reproducible
342
(median RSE of all peptide data or a protein ~10%, and between leaf types in the rosette
343
RSE ~16%; Supplemental Data set 2B), there were specific proteins for which degradation
344
rate appeared to be significantly influenced by the leaf from which the protein was derived.
345
Analysis of these proteins allowed us to look for linkages between changes in protein
346
degradation rate and protein relative abundance in two very different growth scenarios.
347
Proteins in leaf 3 and leaf 7 were selected if they had significant differences in both protein
348
abundance (Figure 5A left; p<0.05; 502 pairs) and degradation rate (right; p<0.05; 433
349
pairs) and the set was separated into functional categories. Leaves 3 and 7 were chosen
350
because proteins in their proteomes best reflected a steady-state over time that could be
351
described by a simple dilution effect (Figure 3). The abundance and degradation differences
352
for these proteins were calculated as relative ∆Abundance and relative ∆KD (i.e. (Leaf 7-Leaf
353
3)/Average (Leaf 3, 7)) and are each presented as boxplots in Figure 5A. The dashed line
354
indicates the point of parity between leaf 3 and 7. Lower protein abundance in leaf 3 of 7
355
correlated with higher degradation rates for plastid proteins involved in cell organization, light
356
reactions, stress, Calvin cycle and photorespiration enzymes, while high protein abundances
357
in leaf 3 or 7 correlated with low protein degradation rate for plastid proteins involved in
358
tetrapyrrole synthesis, protein folding and protein synthesis. When we focused on
359
photosynthesis-associated proteins, and matched individual proteins across the data from
360
leaf 3, 7 and 5, we found that there was a consistent correlation between relative abundance
361
(Figure 5B) and protein degradation rate (Figure 5C) on a protein-by-protein basis across
362
different protein complexes and metabolic pathways. In these experiments, leaf 5 KD data
363
sat neatly between that of leaf 3 and leaf 7, despite some concerns about the steady-state
364
assumption rising from Figure 3. A set of KD values for 49 photosynthesis proteins were
365
also independently calculated by regression analysis across all time points, to ensure the
366
effect was not due to the method of calculation. This alternative analysis gave effectively the
367
same result i.e. degradation rates were higher in leaf 7 than leaf 3, with leaf 5 in between
368
(Supplemental Figure 6, Supplemental Data set 5A). A set of 179 unique proteins with
369
significant changes in both degradation rates and abundances between leaf 3 and 7 were
370
then analyzed using correlation analysis. Nearly 85% of these proteins showed patterns of
371
either increasing degradation rate and decreasing protein abundance, or decreasing
10
372
degradation rate and higher protein abundance (Supplemental Data set 5B).
373
374
Protein kinetic signatures of leaf growth
375
In hypertrophic mouse hearts, organ growth rate has been associated with disease and
376
differences in protein degradation rate has been used to propose the concept of ‘protein
377
kinetic signatures’ associated with cellular dysfunction (Lam et al., 2014). These kinetic
378
signatures have been defined as proteins for which a change in degradation rate correlated
379
with tissue growth rate. These rate-changing-proteins were regarded as a new type of
380
disease marker (Lam et al., 2014). We used the same principle here to determine if the
381
degradation rate of specific leaf proteins could be correlated to leaf growth rate. In our case,
382
these plant rate-changing-proteins would be biomarkers associated with plant leaf growth
383
rather than indicators of disease. By presenting protein degradation rates in different
384
functional categories as a function of the growth rates of leaves (Figure 6A-E;
385
Supplemental Data set 6A) we observed that the degradation rate of many photosynthesis-
386
related proteins positively correlated with leaf growth rate, with the exception of chaperonin
387
60 proteins (Figure 6A). The latter have their highest abundance in the fastest growing leaf
388
type, leaf number 7 (Figure 6F). For organelle ATP-dependent proteases, the degradation
389
rate of FtSH class proteases positively correlated with growth, while the degradation rate of
390
CLP, MAP and DEG classes negatively correlated with growth rate (Figure 6C). Amongst
391
the proteasome subunits and ubiquitination machinery, E3 ligases positively correlated with
392
leaf growth rate while the proteasome itself negatively correlated with leaf growth rate
393
(Figure 6E). Amongst the subunits of the plastid and cytosolic ribosomes, there was a
394
consistent negative correlation between protein degradation rate and leaf growth rate
395
(Figure 6B,D).
396
397
But what could these correlations mean? A key biochemical impact of changing degradation
398
rate as leaf growth rate changes is the alteration of the age profile of specific proteins in
399
different leaves. For example, increasing the degradation rate of ribosomal proteins as
400
leaves mature and slow their growth rate, increases the proportion of new ribosomes in the
401
mature leaf. Young growing leaves can achieve the same result by ribosome numbers
402
increasing through growth, rather than replacement. As a result, both leaves maintain similar
403
age profiles for ribosomes that are 0 to 8 days old (Supplemental Data set 6B). By contrast,
404
the slowing degradation rate of photosynthetic proteins as leaves slow their growth rate will
405
mean distinctly older age profiles for these photosynthetic enzymes in mature leaf tissues.
406
Our calculations of this effect show that ~90% of photosynthetic enzymes will be less than a
407
week old in leaf 7 compared to only ~55% of photosynthetic enzymes in leaf 3
408
(Supplemental Data set 6B).
11
409
410
Proportion of energy use in degradation and synthesis of specific protein classes in
411
different leaves
412
We also estimated total ATP costs from protein degradation and synthesis in the three leaf
413
types from the Arabidopsis rosette by summing up the costs for all the 1228 proteins in our
414
dataset. This used protein degradation rates, calculated protein synthesis rates, leaf growth
415
rate and the steady-state of the leaf proteome and coupled them with individual protein
416
lengths. We used a 5.25 ATP per AA residue cost of synthesis (including AA synthesis,
417
ribosome translation, and transport) and 1.25 ATP per AA residue cost of degradation
418
(unfolding), and estimated relative protein abundance based on modified estimates from
419
PaxDb (see Methods for details of calculations and references, see Supplemental Data set
420
7A for details). The set of 1228 proteins we have measured account for the bulk of protein
421
abundance in Arabidopsis leaves (nearly 90% of all Arabidopsis leaf protein based on ppm
422
counts from PaxDb and RBCL measurement). Assuming that cellular respiration is the
423
primary net ATP production source in leaves, the relative proportion of ATP used for protein
424
synthesis and degradation could also be determined by using leaf respiration rates of the
425
Arabidopsis rosette (Sew et al., 2013). In leaves that are not growing (e.g. leaf 3 in this
426
study), synthesis of new protein is simply replacing degraded protein and would cost 13% of
427
cellular ATP. By contrast, in leaf 7 that was rapidly growing, synthesis costs reached 38% of
428
total respiratory ATP (Figure 7A). This analysis also showed that the overall ATP cost of
429
protein degradation was ~4% of respiratory ATP synthesis across all leaf types. When the
430
protein abundance of plastids and relatively high number of higher turnover proteins in the
431
plastid were considered, nearly 70% of synthesis and degradation costs were attributable to
432
the plastid, with the cytosolic proteome representing 10% of costs and other organelles all
433
having relatively small percentage costs (Figure 7B, Supplemental Data set 7B). Within
434
functional categories, the Calvin cycle and light reactions alone represented nearly 50% of
435
all protein synthesis and degradation costs. We also used these data to highlight the set of
436
proteins in Arabidopsis that each account for more than 1% of total protein turnover costs in
437
leaves (Figure 7C-D, Table 3, Supplemental Data set 7C). The subunits of Rubisco
438
represented 13-18% of all protein costs, while representing ~17% of total cell protein
439
abundance. Some proteins were on this list largely owing to their rapid degradation rate
440
rather than their large abundance; for example, the PSII D1 subunit was expected to be a
441
high cost protein, while AOS and THI1 have not typically been considered as large energy
442
costs to the plant cell but each represented more than 1% of total ATP costs for protein
443
turnover (Figure 7C-D).
444
445
DISCUSSION
12
446
447
Key considerations in assessing plant protein degradation rates in growing leaves
448
Calculations of protein degradation require information on whether the proteome is in steady
449
state and how increases in protein abundance relate to organ or tissue growth rate. Most
450
historical studies in mammals have assumed steady-state to be the case in short timeframes
451
where no treatments were imposed (Price et al., 2010; Cambridge et al., 2011;
452
Schwanhausser et al., 2011). In plants, some reports have attempted to measure changes in
453
protein abundance of specific targets of interest (Li et al., 2012; Lyon et al., 2016), and
454
others have relied on linear regression across a time series to remove outliers and thus only
455
retain information when turnover rates have not changed (Yang et al., 2010; Nelson et al.,
456
2014; Fan et al., 2016). However, these approaches have their limitations in plants and are
457
prone to both error and to the removal of interesting biological features of interest in the
458
data. Notably, regression analysis focuses on the increasing abundance of newly
459
synthesized protein which is mainly derived from growth in plant systems and masks the real
460
kinetics of the protein degradation process. In addition, plant protein turnover studies have
461
typically provided a list of protein degradation rates, without any comparative evaluation of
462
changes in degradation during growth and development. To address these issues, we have
463
taken a series of approaches. Firstly, we undertook a quantitative proteomics analysis within
464
and between different leaves within the Arabidopsis thaliana rosette after the 10th leaf had
465
appeared and followed changes in these leaf proteomes over the 5 days required for a
466
labelling strategy to yield a substantial list of protein degradation rates. This has shown that
467
in leaves 3 and 7, the relative abundance of 97-99% of relatively high abundance proteins
468
within each leaf type did not significantly change on a per unit protein basis, even though
469
total protein per leaf increased at different rates over 5 days. The apparent steady-state
470
evident in leaves 3 and 7 was less valid in leaf 5, in which only 80-90% of the proteome
471
remained unchanged in relative abundance over the 5 days of labelling. This illustrates that
472
while quasi-steady states of proteomes can be found during plant leaf growth, these need to
473
be experimentally measured and confirmed before an analysis strategy is implemented that
474
relies of this principle. That said, we observed that the changes that were measured in leaf 5
475
were not consistent across all timepoints and the scale of the fold changes did not typically
476
alter the nesting of KD values from leaf 5 in analysis of datasets (e.g. Figure 5 and 6).
477
Secondly, we used paired data at different time points to calculate average protein
478
degradation rates rather than linear regression across all time points. This enabled more fast
479
and slow degradation rates to be measured and variable rates to be averaged across
480
timepoints and peptides, rather than these data being discarded (Li et al., 2012). Thirdly, we
481
observed that protein abundance per g FW varies in Arabidopsis leaves during development
482
(Supplemental Figure 1), hence it was important to focus our calculations of dilution of
13
15
N
14
N-
483
labelled peptides on the increase in leaf protein content rather than on size or weight of each
484
leaf. This observation is consistent with independent evidence that young leaf total protein
485
content changes over time in Arabidopsis as a proportion of leaf size (Ishihara et al., 2015).
486
487
Protein features linked to degradation rate
488
The >150-fold variation of protein degradation rates of the ~1000 relatively abundant
489
proteins
490
characteristics define the degradation kinetics of different protein types. The availability of
491
these degradation rates provided an opportunity to explore a range of possible hypotheses
492
and provide some statistical comparisons of degradation rates for different groups of
493
proteins. It is well established that the proteolysis machinery of the cell is spatially distributed
494
and the latency of proteins in different cellular locations varies, which means that not all
495
proteins will be acted upon by the same protease network or in the same timeframe (van
496
Wijk, 2015). Using subcellular location information as a basis for dissecting the dataset, we
497
observed lower protein degradation rates in metabolic organelles that are separated from the
498
cytosol by phospholipid membranes, and higher protein degradation rates in the
499
endomembrane system, which is directly involved in protein synthesis and traffics proteins to
500
the plasma membrane. This suggests that there are broad spatially-based groupings in the
501
proteome that set a median rate of degradation of proteins in different subcellular
502
compartments. However, quantitatively these patterns represent less than 2 fold of the
503
variation in median degradation rates. Diurnal fluctuation in transcription could be a marker
504
for a heightened turnover rate for specific proteins in the light or the dark. Our analysis of this
505
effect failed to see a significant correlation in the data as a whole, but we did observe that
506
proteins with light-induced transcriptional patterns were enriched in the set of proteins with
507
faster KD. This indicates that light-dependent transcriptional processes may be associated
508
with enhanced protein degradation of the proteins being synthesized, while dark-dependent
509
processes are relatively less connected.
in
Arabidopsis
leaves
raises
significant
questions
about
which
protein
510
511
Although the N-terminal residue of a polypeptide has some reported influence on turnover
512
rate for subsets of plant proteins (Apel et al., 2010; Zhang et al., 2015), the N-terminal
513
residue alone appeared to have little predictive value in our dataset to define the turnover
514
rate of unknown sets of protein of different sequences (Figure 4A). Protein aggregation
515
propensity has been linked to protein degradation rate and has been shown in other
516
eukaryotic systems to be independent of protein size (De Baets et al., 2011).
517
Arabidopsis dataset, aggregation propensity and protein size both correlated with turnover
518
rate when the extremes of the fast and slow degradation distribution were compared, but
519
when the higher aggregation propensity of large proteins was accounted for, there was no
14
In our
520
longer a relationship with degradation rate (Figure 4B). This suggests that while aggregation
521
propensity could account for some differences in degradation rate, as shown in mammals, it
522
is mainly due to protein length rather than amino acid composition of proteins in Arabidopsis.
523
Subunits of protein complexes might be expected to turn over at similar rates, and this was
524
largely confirmed when six out of seven physical complexes or protein systems were found
525
to show significantly tighter distributions of degradation rates compared to random samplings
526
of proteins(Figure 4C). This means that co-membership of a complex could be a useful
527
predictive tool for predicting unknown protein degradation rates.
528
families and domain groups were shown by this analysis as the major factors that could be
529
used to predict degradation rates (Table 2). Nearly 40-fold differences in degradation rate
530
were observed for proteins in different domain classes.
Specific protein fold
531
532
Protein stability and the correlations with protein function
533
One essential goal of protein degradation studies is to help elucidate the biological functions
534
of specific proteins based on their variable turnover rates, and to this end efforts have been
535
made in correlation analysis of protein turnover and protein biological functions across
536
different organisms (Schwanhausser et al., 2011; Christiano et al., 2014; Karunadharma et
537
al., 2015). Consistent with evidence in yeast (Christiano et al., 2014) ribosomal subunits are
538
some of the most stable and long-lived proteins in Arabidopsis (Figure 2A). This means
539
ribosomal proteins can be used for a long time period after their biogenesis, with their
540
abundance controlled post-translationally by protein degradation at late stages of
541
development. However, even in these long-lasting protein sets, the relative degradation rate
542
distribution for the cytosolic ribosome had a higher standard deviation than those of other
543
protein complexes (Figure 4C). These non-uniform degradation rates in the cytosolic
544
ribosome may be related to the specialized role of structural subunits in ribosome and
545
altered stages of assembly which has been proposed in a ribosome turnover study in human
546
cells (Doherty et al., 2009).
547
548
Within photosynthetic complexes (Figure 5C) it was evident that three proteins had high
549
degradation rates, well beyond the normal distribution for their complexes. While D1
550
(Atcg00020) is well known for its high degradation rate at the reaction center of PSII (Melis,
551
1999; Takahashi and Badger, 2011; Lambreva et al., 2014), the discovery that PetD
552
(Atcg00730) of the cytochrome b6f complex and PIFI (At3g15840) of the NDH complex had
553
comparable degradation rates was not anticipated. These three proteins are core
554
components required for electron transport functions of PSII, cytochrome b6f and NDH
555
complexes, respectively. Interestingly, PetD is subject to proteolysis in highly purified
556
cytochrome b6f complex in vitro even in the presence of protease inhibitors, by an unknown
15
557
mechanism (Zhang and Cramer, 2004). PIFI has been shown to be essential for NDH
558
complex-mediated chlororespiration in Arabidopsis and PIFI knockout mutants exhibit a
559
greater sensitivity to photoinhibition (Wang and Portis, 2007). Their fast degradation rate in
560
vivo might reveal some functional similarity of these proteins in their respective protein
561
complexes. D1 and PetD proteins are both chloroplast encoded. D1 is at the plastoquinone
562
binding site of PSII and is directly involved in coordinating and protecting the metal cluster
563
responsible for producing oxygen and protons from H2O (Umena et al., 2011; Lambreva et
564
al., 2014; Wei et al., 2016).Cyanobacterial PetD forms the p-side of the cytochrome b6f
565
complex which accepts protons from plastosemiquinone and defines a route for H+ transfer
566
in the complex (Hasan et al., 2013). Also, the transcripts for D1 and PetD proteins are both
567
found to be up-regulated in the adg1-1 tpt-2 double mutant, which shows rapid
568
photoinhibition under high light (Schmitz et al., 2014). D1 protein is widely recognized as a
569
major target in PSII in the photodamage and photoprotection cycle (Melis, 1999; Takahashi
570
and Badger, 2011). PetD and PIFI, like the D1 subunit, might represent new targets for light
571
or other active electron and proton transfer pathway-associated damage and repair
572
processes.
573
574
575
Protein abundance and turnover changes in young leaves with rapid chloroplast
576
division
577
We observed that plastid-localized light-reaction and Calvin cycle proteins showed lower
578
protein abundance coupled with lower protein stability in leaf 7 (Figure 5), while ribosomal
579
proteins showed higher stability and accumulated in the fast-growing leaf 7 (Figure 5A). This
580
supports the proposal that regulatory mechanisms for accumulation of ribosomal proteins in
581
fast-growing stages in plants is independent of transcript abundance but more likely due to
582
stabilization of the assembled ribosome complexes (Baerenfaller et al., 2012; Ponnala et al.,
583
2014). Increased levels of synthesis, folding and degradation proteins could provide high
584
protein synthesis and degradation capability in younger leaves. This may in turn explain
585
faster turnover of a range of metabolic protein sets in younger leaves, such as Calvin cycle
586
and photorespiratory enzymes and light reaction complex subunits (Figure 5A,C).
587
588
Analysis of the relative protein degradation changes with leaf growth rate (Figure 6)
589
highlighted that while the degradation rate for the majority of the photosynthetic apparatus
590
components positively correlated with growth (Figure 6A), those of CPN60A and CPN60B
591
negatively correlated with growth. These chaperonins are needed for plastid division (Suzuki
592
et al., 2009), so their increased stabilization and abundance (Figure 6F) are consistent with
593
the more frequent chloroplast division events that occur in young leaves (Osteryoung and
16
594
Pyke, 2014). The selective degradation changes observed in younger leaves could also be
595
caused by specific changes in proteolysis and translational control when leaves are growing
596
at a faster rate. Analysis of relative protein degradation rates versus leaf growth rate (Figure
597
6) highlighted that while turnover of protein synthesis components negatively correlated with
598
growth (Figure 6B,D), there were specific classes of organelle proteases whose turnover
599
rate positively correlated with growth. The FtsH protein set highlighted in Figure 6 are the
600
same proteins that are highlighted in the protein domain set FtsH (IPR016135), which was
601
one of the fastest degrading domain groups in our analysis (Table 2). These proteases have
602
direct roles in the turnover of the photosynthetic apparatus and in plastid thylakoid formation
603
(Zaltsman et al., 2005; Kato et al., 2012; Rowland et al., 2015). It has been reported that
604
mRNA for FtsH1, FtsH2, FtsH5 and FtsH8 generally increases in young developing leaves
605
(Zaltsman et al., 2005). High mRNA levels for these proteases provides a fast synthesis
606
potential, which is consistent with our finding that FtsH1, FtsH2, FtsH5 and FtsH8 exhibit
607
rapid rates of degradation in younger leaves mostly without large changes in relative
608
abundance (Figure 6C, Supplemental Data set 8).
609
610
Plastid turnover is controlled, however, not only by internal proteases but also by the
611
proteasome and vacuolar quality control pathways mediated by E3 (Ling et al., 2012;
612
Woodson et al., 2015). Ubiquitination plays a role in control of plastids via a RING-type
613
ubiquitin E3 ligase, SP1, which mediates the ubiquitination and degradation of TOC
614
components that are required for developmental transition of plastids (Ling et al., 2012;
615
Huang et al., 2013). Most recently, the plant U-box E3 ligase PUB4 was found to mediate the
616
ubiquitination and selective degradation of damaged chloroplast through globular vacuoles
617
(Woodson et al., 2015). While the proteasome subunit degradation rate was low in young
618
leaves, the degradation rates of three E3 ligases we identified (one RING-type At1g57800,
619
two F-box At4g39756 and At5g01720) were higher in younger leaves (Figure 6E). Enhanced
620
degradation of selective E3 ligases in younger leaves could thus also help maintain fast
621
chloroplast biogenesis in younger leaves. However, it is still unclear whether this is achieved
622
by promotion of developmental transition and/or by selective breakdown of damaged
623
plastids.
624
625
Energy costs of leaf protein turnover
626
It has been widely accepted that active transport and protein turnover are the two highest
627
cost processes in cell biology (Scheurwater et al., 2000). Calculations of energy costs for
628
protein turnover depend on knowing the synthesis and degradation rates of proteins, as well
629
as the length and abundance of each protein. This study provided the rate data, length is
630
straightforward to determine, and abundance was estimated used the PaxDb estimates for
17
631
Arabidopsis leaves (pax-db.org/). Expressing this ATP demand for protein synthesis and
632
degradation as a proportion of available cellular ATP requires a cellular ATP synthesis rate
633
calculation. While mitochondria are the primary ATP source in all eukaryotes, chloroplasts
634
can also produce ATP through photophosphorylation in plants. Early studies showed that
635
isolated intact chloroplast protein synthesis could be driven by light or added ATP while
636
etioplast protein synthesis could be driven only by added ATP (Siddell and Ellis, 1975; Ellis,
637
1981). This demonstrated that photophosphorylation and ATP imported through transporters
638
could both be energy sources for chloroplast protein synthesis. However, several more
639
recent calculations concluded that in planta additional ATP needs to be imported into
640
chloroplast through translocators to meet the demands of CO2 assimilation and
641
photorespiration without even considering the energy cost of other biosynthetic processes
642
including protein synthesis (Flugge et al., 2011; Kramer and Evans, 2011). This leaves
643
chloroplasts dependent on import of respiration-derived ATP for most biosynthetic processes
644
including protein synthesis. The importance of ATP import for chloroplast maintenance is
645
also evident by reports that restriction of ATP import through NTT translocators can lead to
646
dwarf and necrotic leaf phenotypes (Reinhold et al., 2007) and can disrupt hormonal
647
signaling (Schmitz et al., 2010). Our calculations are based on the caveat that respiration is
648
the only net source of ATP available for protein synthesis in plants. The dominance of the
649
plastid proteome as a major energy cost through these calculations supports the importance
650
of ATP import into chloroplasts based on the insufficiency of ATP supply from
651
photophosphorylation.
652
653
Our data show that on a leaf basis the degradation costs were effectively static at ~4%,
654
while synthesis rates varied considerably with the rate of leaf growth. Protein synthesis and
655
degradation together cost ~16% of the total leaf oxidative phosphorylation-dependent ATP
656
production in developed leaves with slow growth rates and ~42% in developing leaves with
657
fast growth rates. This analysis suggests the protein turnover energy budget is one of the
658
highest maintenance costs in growing Arabidopsis leaves. The Calvin cycle and the protein
659
machinery of the light reactions are the highest cost functions in the growing Arabidopsis leaf
660
(Figure 7, Table 3). These categories also share the top five high energy cost individual
661
proteins. Interestingly, photorespiration proteins in plastids, mitochondria and peroxisomes
662
are all among the top 20 proteins in terms of energy cost (Supplemental Data set 7A).
663
Changes in the degradation rates or abundance of high energy cost proteins could cause
664
significant fluctuations of the total energy budget and alter the ability of cells to reach a new
665
balance between energy supply and consumption. The availability of a direct calculation of
666
energy cost for specific proteins under different growth rates in an individual leaf will help to
667
explain energy constraints in changing the abundance of specific proteins.
18
668
669
670
METHODS
671
672
Hydroponic growth of Arabidopsis
673
Arabidopsis thaliana accession Columbia-0 plants were grown under 16/8-h light/dark
674
conditions with cool white T8 tubular fluorescent lamps 4000K 3350 lm (Osram, Germany)
675
with intensity of 100–125 μmol m-2 s-1 at 22 °C. The hydroponic protocol was as described
676
previously (Waters et al., 2012) and used a modified Hoagland solution (2 mM CaCl2, 6 mM
677
KNO3, 0.5 mM NH4NO3, 0.5 mM MgSO4, 0.25 mM KH2PO4, 0.05 mM KCl and 0.04 mM Fe-
678
EDTA) supplemented with micro elements (25 μM H3BO3, 2 μM MnCl2, 2 μM ZnSO4, 0.5 μM
679
CuSO4, 0.15 μM CoCl2 and 0.25 μM (NH4)6Mo7O24) and 2.6 mM MES, and the pH was
680
adjusted to 5.8-6.0. Seeds were planted on the surface of stone wool stuffed in 1.5-ml black
681
tubes with the bottoms cut out to sit in 24-well floater tubes racks containing 160 ml growth
682
medium. The seeds were vernalized under 4 °C for 2-3 days before being transferred to the
683
growth chambers. Half-strength growth medium was used for the first week. A single plant
684
was placed in every tube and four tubes in each floater tube rack (Supplemental Figure
685
1A). The growth medium was changed every 7 days.
686
687
Unlabelled Arabidopsis plants were grown for 21 days until they reached leaf production
688
stage 1.10 (T0) (Boyes et al., 2001) in natural abundance medium as noted above. Leaves
689
3, 5 and 7 from three biological replicates were collected at T0 (21 days), T1 (22 days), T3
690
(24 days) and T5 (26 days). Thirty-six unlabelled leaf whole protein samples were extracted
691
and protein concentration from each leaf was determined by an amido black assay.
692
15
693
For progressive
694
production stage 1.10 (T0) (Boyes et al., 2001) in natural abundance medium. The growth
695
medium was then discarded and the growth racks rinsed four times with fresh medium
696
without nitrogen (no KNO3 or NH4NO3) to ensure the old solution was washed out. A total of
697
160 ml of
698
and the plants were grown for 1, 3 and 5 days (T1, 3 and 5) before collecting leaf number 3,
699
5 and 7 for total protein extraction (Supplemental Figure 1A). Four leaves numbered 3, 5 or
700
7 from plants in one rack were pooled as a biological replicate. Three biological replicates
701
were collected at each time point. Twenty-seven
702
unlabelled samples (T0) were collected and stored separately for later analysis.
15
N labelling, plants were also grown for 21 days until they reached leaf
N medium (6 mM K15NO3, 0.5 mM
15
NH415NO3) was added for every four plants
15
N labelled (T1, 3 and 5) and nine
703
704
To obtain a fully labeled
15
N protein reference standard,
19
15
N medium (with 6 mM K15NO3, 0.5
15
NH415NO3) was used to replace the natural abundance nitrogen in the medium and
705
mM
706
plants were grown from seed in this medium for 26 days (T5).
707
number 3, 5 and 7 was analyzed separately and each showed 98%
708
(determined from median of 1500-2000 peptides). Fully
709
5 and 7 samples were pooled and used as
710
subsequent experiments.
15
N labelling efficiency in leaf
15
N incorporation
15
N labeled total proteins from leaf 3,
15
N fully labelled proteins reference for
711
712
Fold change in protein abundance (FCP) following leaf production stage 1.10
713
Aliquots of the thirty-six unlabelled leaf samples (T0, 1, 3 and 5) noted above were also used
714
for fresh weight (FW) and protein content measurements (Supplemental Figure 1). Liquid
715
nitrogen snap frozen samples (0.1 g) were vortexed with Qiagen tissue lysis beads (5 mm)
716
and boiled in 1×Sample buffer (2% [w/v] SDS, 62.5 mM Tris-HCl, 10% [v/v] glycerol, 5% [v/v]
717
mercaptoethanol, bromophenol blue) for 5 minutes and then centrifuged at 10, 000 x g for 10
718
minutes. The supernatant was collected and separated with SDS-PAGE on Bio-Rad
719
Criterion precast gels (10–20% [w/v] acrylamide, Tris-HCl, 1 mm, 18-well comb gels).
720
Electrophoresis was performed at 300 V for 15-20 min. Proteins were visualized by colloidal
721
Coomassie Brilliant Blue G250 staining. ImageJ was used to quantify protein content in each
722
sample (Supplemental Figure 1B). FCP was determined by combining fresh weight and
723
protein content changes (Supplemental Data set 1).
724
725
Determining changes in specific protein abundance over 5 days using a fully labeled
726
15
727
A total of 50 µg of each the thirty-six unlabeled leaf protein samples noted above was mixed
728
individually with 50 µg of the fully
729
Each sample was separated into 96 fractions by high pH HPLC separation and further
730
pooled into 12 fractions. A total of 432 fractions from thirty-six samples were analyzed by
731
mass spectrometry. Filtered samples (5 µl each) were loaded onto a C18 high-capacity nano
732
LC chip (Agilent Technologies) using a 1200 series capillary pump (Agilent Technologies) as
733
described previously. Following loading, samples were eluted from the C18 column directly
734
into a 6550 series quadrupole time-of-flight mass spectrometer (Agilent Technologies) with a
735
1200 series nano pump using the following buffer B (0.1% FA in acetonitrile) gradient: 5-35%
736
in 35 min, 35-95% in 2 min and 95-5% in 1 min. Parameter settings in the mass
737
spectrometer were as described previously (Nelson et al., 2014).
N protein reference standard
15
N-labelled reference and digested in solution by trypsin.
738
739
740
741
Agilent .d files were processed by an in-house script written in Mathematica for partial
15
labelling quantification (Nelson et al., 2014) modified to calculate the ratio between
14
peptide (L) and fully
15
N
N
N-labelled reference peptide (H). L and H formed two separate
20
742
populations by a non-negative least square (NNLS) approach as described previously (Guan
743
et al., 2011; Nelson et al., 2014). All filters used for partial
744
al., 2014) and two more filters were applied: firstly, a minimum 80%
745
required to disregard quantifications with high noise level between L and H; secondly, a
746
minimum signal-to-noise ratio of 5 was used to disregard high neighboring noise contribution
747
to L and H. 432 Agilent .d files were further processed by TPP to get protein probabilities for
748
identification. Only proteins with p>0.95 (FDR less than 0.6%) were further reported.
749
Examples for CPN60A (At2g28000), HPT108 (At3g63190) and PGK1 (phosphoglycerate
750
kinase 1, At3g12780) in leaf 3, 5 and 7 are shown in Supplemental Figure 7.
15
N labelling were kept (Nelson et
15
N enrichment was
751
752
Protein abundance was represented as ratio to reference. Protein abundance comparison of
753
the same protein from leaves 3 and 7 was presented as relative ∆Abundance (i.e. (leaf 7
754
abundance–leaf 3 abundance)/Average (leaf 3 and leaf 7 abundance)). The processed
755
quantification data in provided in Supplemental Data set 9 and the peptide identification
756
data for each timepoint in Supplemental Data sets 10-13.
757
758
Protein Separation, Digestion, Mass Spectrometry Analysis and Determination of
759
Peptide 15N Incorporation Ratio
760
The twenty-seven 15N progressively labeled and three unlabelled leaf samples (0.1-0.2 g)
761
were snap frozen in liquid nitrogen and homogenized using Qiagen tissue lysis beads (5
762
mm) by vortex. A total plant protein extraction kit (PE0230-1KT, Sigma Chemicals) was used
763
to extract total proteins. The final pellet of total protein was dissolved in Solution 4 and then
764
reduced and alkylated by tributylphosphine (TBP) and iodoacetamide (IAA) as described in
765
the Sigma manual. The suspension was centrifuged at 16,000 g for 30 min and the
766
supernatant was assay for protein concentration by amido black quantification as described
767
previously (Li et al., 2012). Protein (100 µg) in solution from each sample was then mixed
768
with equal volume of 2×sample buffer (4% SDS, 125 mM Tris, 20% glycerol, 0.005%
769
Bromophenol blue and 10% mercaptoethanol, pH6.8) before being separated on a Biorad
770
protean II electrophoresis system with a 4% (v/v) polyacrylamide stacking gel and 12% (v/v)
771
polyacrylamide separation gel. Proteins were visualized by colloidal Coomassie Brilliant Blue
772
G250 staining.
773
774
The SDS-PAGE gel lane for each sample was cut into 12 fractions. The resulting 360 gel
775
pieces were twice decolorized by de-stain solution (10 mM NH4NO3, 50% Acetonitrile) for 45
776
min. Dried gel pieces from one fraction were digested with 375 ng trypsin at 37 °C overnight
777
and then extracted as described previously (Nelson et al., 2014). The digested peptide
778
solutions were dried by vacuum centrifugation at 30 °C, re-suspended in 20 µl loading buffer
21
779
(5% Acetonitrile, 0.1% Formic Acid) and filtered by Millipore 0.22 micron filters before being
780
loaded for HPLC separation. A total of 5 µl of filtered samples were loaded onto a C18 high-
781
capacity nano LC chip (Agilent Technologies) using a 1200 series capillary pump (Agilent
782
Technologies) as described previously. Following loading, samples were eluted from the C18
783
column directly into a 6550 series quadrupole time-of-flight mass spectrometer (Agilent
784
Technologies) with a 1200 series nano pump using the following buffer B (0.1% FA in
785
Acetonitrile) gradient: 5-35% in 35 min, 35-95% in 2 min and 95-5% in 1 min. Parameter
786
settings in the mass spectrometer were as described previously (Nelson et al., 2014).
787
788
Agilent .d files were converted to mzML using the Msconvert package (version 2.2.2973)
789
from the Proteowizard project, and mzML files were subsequently converted to Mascot
790
generic files using the mzxml2 search tool from the TPPL version 4.6.2. Mascot generic file
791
peak lists were searched against an in-house Arabidopsis database comprising ATH1.pep
792
(release 10) from The Arabidopsis Information Resource (TAIR) and the Arabidopsis
793
mitochondrial and plastid protein sets (33621 sequences; 13487170 residues) (Lamesch et
794
al., 2012), using the Mascot search engine version 2.3 and utilizing error tolerances of 100
795
ppm for MS and 0.5 Da for MS/MS; “Max Missed Cleavages” set to 1; variable modifications
796
of oxidation (Met) and carbamidomethyl (Cys). We used iProphet and ProteinProphet from
797
the Trans Proteomic Pipeline (TPP) to analyze peptide and protein probability and global
798
false discovery rate (FDR) (Nesvizhskii et al., 2003; Deutsch et al., 2010; Shteynberg et al.,
799
2011). The reported peptide lists with p=0.8 have FDRs of <3% and protein lists with p=0.95
800
have FDRs of <0.5%. Quantification of LPFs (labeled protein fraction) were accomplished by
801
an in-house script written in Mathematica as described previously (Nelson et al., 2014). The
802
data for determination of LPF for peptides by
803
Supplemental Data set 14 and the peptide identification data in Supplemental Data set
804
15.
15
N progressive labelling are provided in
805
806
Calculation of KD and KS
807
Protein degradation is a first order and stochastic process, and an exponential rate (KD) is
808
used to represent the proportional decrease of current protein abundance per day. Protein
809
synthesis is a zero order process, and is thus described as a proportional increase of its
810
starting pool (KS/A) per day (Claydon and Beynon, 2012; Li et al., 2012). ‘A’ is the starting
811
abundance for a specific protein at the beginning of the labelling period (T0). Because of the
812
stochastic nature of degradation, newly synthesized proteins experience the same
813
proportional decrease as existing proteins. As a result, calculation of gross synthesis rates
814
requires the measurement of newly synthesized proteins and the addition of the proportion
815
of newly synthesized proteins lost by degradation. Protein degradation rate KD and synthesis
22
816
rate KS/A for specific proteins were calculated from the progressive labeling data using a
817
combination of Fold Change in Protein (FCP) and the labeled peptide fraction (LPF)
818
following a method described previously (Li et al., 2012).
819
820
KD calculation:
=−
821
ln
∙ (1 −
)
KS calculation:
=
−
1−
∙
∙
∙
822
823
The fresh weight of leaves with a protein content correction was first used as an approximate
824
value of Fold Change in Protein Abundance (Measured FCP) and this was applied to
825
calculate degradation rates (KD). However, it is challenging to measure FCP and protein
826
turnover simultaneously from the same samples, and sample-to-sample variation in
827
measuring them separately affects the calculation of protein degradation rates. In addition,
828
different leaves were found to show different nitrogen uptake and assimilation capability
829
(Masclaux-Daubresse et al., 2010) and could thus lead to different lag effects in
830
labelling. Younger leaves appeared to show faster nitrogen assimilation rates into amino acid
831
pools compared to relatively older leaves, which is evidenced by the higher
832
in the same peptide of younger leaves. To avoid these effects on degradation rate
833
calculation and comparisons across leaves, a median polish strategy, which has been widely
834
used for multiple experiment normalization (Lim et al., 2007), was employed to pre-process
835
the data. The median LPF value in each sample is a good indicator of FCP for each
836
experimental replicate. Based on a time point to time point protein turnover rate calculation
837
(Li et al., 2012), a calculated FCP for each sample can be deduced (Calculated FCP).
15
N
15
N enrichment
∙
=
1−
(
)
838
The comparison of calculated FCP and measured FCP showed a strong correlation (k=1.00,
839
R=0.95), but calculated FCP recognizes subtle growth changes in individual samples and
840
gives more precise calculations of KD and KS values. Calculated FCP was applied for further
841
calculations. Protein degradation comparison of the same protein from Leaves 3 and 7 is
842
presented as a relative ∆KD value (i.e. (L7 KD–L3 KD)/Average (L3 and 7 KD)).
843
844
Defining protein orthologues in barley
845
Protein sequences were obtained from MIPS PlantsDB (Nussbaumer et al., 2013) for barley
846
and orthologues for those proteins with quantified degradation rates in barley (Nelson et al.,
23
847
2014) were then acquired through BLASTP provided by The Arabidopsis Information
848
Resource (TAIR) release 10. The highest scoring Arabidopsis orthologue was selected for
849
each protein in barley given its E-value was below 10-5. Correlation of protein degradation
850
rates between Arabidopsis and barley was assessed by parametric (Pearson) and
851
nonparametric (Spearman and Kendal) tests.
852
853
Energy cost and production for protein turnover
854
For a specific protein, energy cost for turnover was calculated to depend upon absolute
855
protein abundance (PA), amino acid length (AAL); protein degradation rate (KD) and
856
synthesis rate (KS). Protein degradation cost was based on 1-1.5 ATP per residue in the
857
proteasome degradation pathway (Peth et al., 2013). Protein synthesis cost was based on 5-
858
5.5 ATP per residue in ribosome translation, protein transport and amino acid biosynthesis
859
(Piques et al., 2009; Kaleta et al., 2013). The energy cost in ATP for degradation and new
860
synthesis of a specific protein was calculated by 1.25∙PA∙AAL∙KD/5.25 PA∙AAL∙KS (µmol∙d-1).
861
Whole protein content was measured as fresh weight of leaf tissues (~2% protein per FW).
862
Whole amino acid absolute abundance can be calculated based on whole protein content
863
and average amino acid molecular weight. The gross AA residue content of proteins of each
864
leaf number was calculated by averaged values from T0-5: leaf 3 (9.59 µmol), leaf 5 (18.69
865
µmol), leaf 7 (18.97 µmol) and overall (47.24 µmol). Relative protein abundances for the
866
1228 proteins in ppm were acquired from the Arabidopsis leaves abundance datasets
867
aggregated in the PaxDb database (Piques et al., 2009; Gfeller et al., 2011; Baerenfaller et
868
al., 2012; Wang et al., 2012) with the exception that the numbers for the Rubisco large
869
subunit (RBCL), which is well outside the normal distribution of values in PaxDb, were
870
replaced by a number from experimental evidence, based on the fact that 40% of total
871
soluble protein in Arabidopsis is Rubisco (Eckardt et al., 1997) and that only 50-60% of leaf
872
protein is soluble. This means RBCL represents 23.2% of total leaf protein by mass or
873
16.5% by molar ratio in Arabidopsis, replacing PaxDb numbers for RBCL of 8.8% by mass
874
and 6.6% by molar ratio.
875
876
The ATP production rate of respiration is based on experimentally determined respiration
877
rates presented as oxygen consumption 100 nmol.min-1.g-1 in different Arabidopsis leaves
878
(Sew et al., 2013). ATP production rates in different leaves can be deduced based on 1 O2-
879
4.5 ATP: leaf 3 (34.20 µmol·d-1), leaf 5 (67.05 µmol·d-1), leaf 7 (69.86 µmol·d-1) and overall
880
(171.11 µmol·d-1). Calculations of energy cost and production for protein turnover are
881
detailed in (Supplemental Data set 7A).
882
24
883
Accession Numbers
884
All accession numbers used are from the TAIR10 annotation of the Arabidopsis genome
885
(www.arabidopsis.org).
886
887
888
Supplemental data sets 1-15 can be accessed in the Dryad repository: Protein degradation
rate in Arabidopsis thaliana leaf growth and development, doi:10.5061/dryad.q3h85
889
Raw MS data for
890
PXD004550.
891
Raw MS data for the relative abundance of proteins between leaves and time points can be
892
downloaded via ProteomeXchange: PXD004549.
15
N progressive labelling can be downloaded via ProteomeXchange:
893
894
Supplemental Data
895
Supplemental Figure 1. Arabidopsis leaf growth rates and leaf protein content.
896
897
Supplemental Figure 2. A flow diagram of tissue sampling and mass spectrometry for
898
progressive labelling and
899
5 and 7 of the Arabidopsis rosette over time.
15
N
15
N spiked analysis of protein abundance experiments in leaves 3,
900
901
Supplemental Figure 3. Determination of the labeled protein fraction (LPF) for peptide and
902
protein degradation rate measurements.
903
15
904
Supplemental Figure 4. The
N enrichment level in the heavy labelled
905
population in each leaf at each timepoint.
15
N peptide
906
907
Supplemental Figure 5. The protein degradation (KD) rates of the 146 proteins that were
908
present in all 27 samples analyzed, before and after the median-polish normalization.
909
910
Supplemental Figure 6. A set of 49 degradation rates (KD) for photosynthesis proteins
911
involved in different protein complexes and functional categories calculated by regression in
912
leaf 3, 5 and 7.
913
914
Supplemental Figure 7. The relative protein abundance of CPN60A (At2g28000), HPT108
915
(At3g63190) and PGK1 (PHOSPHOGLYCERATE KINASE 1, At3g12780) in leaves 3, 5 and
916
7.
917
918
Supplemental Data set 1. Growth rates of Arabidopsis leaf 3, leaf 5 and leaf 7.
919
25
920
Supplemental Data set 2. Protein degradation rates of 1228 proteins and statistical analysis
921
across samples.
922
923
Supplemental Data set 3. Ratios of specific protein abundances relative to a
924
labelled reference.
fully15N
925
926
Supplemental Data set 4. The correlation of intrinsic properties of proteins with their protein
927
degradation rates.
928
929
Supplemental Data set 5. Comparison of protein degradation rate (KD) and protein
930
abundance between selected proteins in leaf 3 and leaf 7.
931
932
Supplemental Data set 6. Comparison of protein degradation rate (KD) and leaf growth
933
rates in leaf 3, leaf 5 and leaf 7.
934
935
Supplemental Data set 7. Calculation of ATP costs for protein synthesis and degradation in
936
leaf 3, leaf 5 and leaf 7.
937
938
Supplemental Data set 8. Comparison of protein degradation rate and protein abundance
939
for selected proteins of interest.
940
941
Supplemental Data set 9. Determination of protein abundance fold change relative to a
942
reference in 36 individual leaf samples.
15
N
943
944
Supplemental Data sets 10-13. The peptide identification data for each time point for
945
peptides at T=0,1,3 and 5 days in the spiked in 15N reference experiment.
946
947
Supplemental Data set 14. Determination of LPF for peptides by
948
in 27 individual leaf samples.
15
N progressive labelling
949
950
Supplemental Data set 15. The peptide identification data for
951
peptides at all time points.
15
N progressively labelled
952
953
Acknowledgements:
954
This work was supported by the facilities of the Australian Research Council Centre of
955
Excellence Program (CE140100008). AHM was funded by an ARC Future Fellowship
956
(FT110100242). The authors declare that they have no conflict of interest.
26
957
958
Author contributions:
959
LL performed most of the experiments and data analysis. CJN helped design experiments,
960
and advised on and performed some of the mass spectrometry and data analysis. JT and IC
961
provided computational support, scripts and aided analysis of the data. AHM, LL and SH
962
designing experiments, analyzed and discussed results. AHM and LL wrote the manuscript.
963
All authors commented on the results and the manuscript.
964
27
965
Figure Legends
966
967
Figure 1. Individual growth rates of leaves in the Arabidopsis rosette and the
968
consequent fold change in protein over time. Leaves 3, 5 and 7 from Arabidopsis plants
969
grown in hydroponic culture were collected over a time course (T0, 1, 3 and 5 days)
970
beginning with 21-day-old plants that had developed 10 leaves (leaf production stage 1.10).
971
Typical images of leaves 3 to 8 from these plants at the beginning of the experiment are
972
shown. The growth of leaves over the subsequent 5 days were determined by following the
973
fold change in protein per leaf at each time point compared to T0 for leaf 3 (green), leaf 5,
974
(yellow) and leaf 7 (red). An average relative growth rate (d-1) was determined from both
975
changes in fresh weight and protein content by scanning images of gels (see Supplemental
976
Figure 1, Supplemental Data set 1).
977
978
Figure 2. Degradation rates of 1228 Arabidopsis proteins reported as averages based
979
on functional categories or localization inside cells, and in comparison to barley
980
orthologues.
981
(A) Average Arabidopsis leaf protein degradation rates box-plotted by functional categories
982
and their localization in cells. Functional categories were acquired from MapCave and
983
localization information from SUBAIII. Only major functional categories and localization
984
groups (n>10) are shown. Median, quartiles and standard deviations are shown. Outliers are
985
shown as hollow dots and extreme outliers are highlighted as solid squares and annotated.
986
The five major localizations are highlighted with colors: Cytosol (Cyt-Blue), Nucleus (Nc-
987
Red), Extracellular (Et-Cyan), Mitochondrion (Mt-Yellow) and Plastid (Pt-Green). (B) Protein
988
degradation rates of the 1228 proteins box-plotted based on localization within the cell.
989
Abbreviations as in (A), and dual targeting proteins (DUAL), Vacuole (Vc), plasma
990
membrane (PM), endoplasmic reticulum (ER), peroxisomal (Px) and Golgi. (C) Protein
991
degradation rates of a subset of 288 Arabidopsis proteins for which orthologues in barley
992
had published rates. One spot represents a protein turnover rate in barley (x-axis) and
993
Arabidopsis (y-axis). Four outliers (2SD above median, in red dashed line rectangle) are
994
shown (Pearson correlation, R=0.99). The remaining 284 proteins (pointed out by black
995
dashed line rectangle) are shown by black dots (Pearson correlation, R=0.38).
996
997
Figure 3 Changes in relative abundance estimates for individual proteins in leaves of
998
the Arabidopsis rosette during 5 days of leaf growth. (A) The intersection of the number
999
of specific proteins with a measured KD value and a protein relative abundance assessment.
1000
(B) Principle components analysis of specific protein abundances from mass spectrometry
1001
analysis. Abundance data for 955 proteins (KD values recorded in Figure 2) were used for
28
1002
the analysis. The clustering of leaf 3 (green), leaf 5 (yellow) and leaf 7 (green) is shown in
1003
the grey shading. The full data are provided in Supplemental Data set 3A and B. (C) A
1004
matrix of paired statistical comparisons of protein abundance between day 0, day 1, day 3
1005
and day 5 time points for each leaf is presented with the percentage of the proteins from (B)
1006
that significantly changed in abundance (t-test p<0.01) reported (full data in Supplemental
1007
Data set 3C). Outside the matrix is the percentage of proteins measured that were found to
1008
change in abundance (t-test P<0.01) across the entire time course in each leaf.
1009
1010
Figure 4. Intrinsic properties of Arabidopsis proteins that correlate with protein
1011
degradation rate. (A) Comparison of protein degradation rate with experimental evidence
1012
of the mature N-terminal amino acid residue. Only protein sets in which more than 5
1013
members had the same N-terminal amino acid residue were examined, a set of 376
1014
proteins in total. A non-parametric Kruskal-Wallis k-samples test showed no significant
1015
difference among the 12 N-terminal groups. However, the K, S and D groups showed
1016
significantly different degradation rate distributions from T by a Conover-Iman post-hoc test.
1017
No statistically significant differences were found when degradation rates of stable AAs (A,
1018
V, M, G, T, P), unstable AAs (K, F, D, R, Q) and average stability AAs (all others AAs) were
1019
compared (Supplemental Data set 4A). These groups were defined based on the
1020
consensus of yeast and mammal N-end rules.
1021
distribution graph of protein aggregation propensity (AGG) of each protein, the AA length of
1022
each protein (LEN), and the ratio of AGG/AA length between the top 100 most stable and
1023
bottom 100 most unstable proteins from the set of 1228 proteins with degradation rates.
1024
Statistical significance was estimated by the Kolmogorov-Smirnov test. (C) The standard
1025
deviation (σ) of the protein degradation rates of the protein subunits of 7 major protein
1026
complexes compared to randomly sampled distributions of σ for the same sized group of
1027
proteins from the set of 1228.
1028
Proteasome, Pt ATP synthase, Pt Ribosome, PSI, PSII and Mt ETC; all other proteins were
1029
considered to be one group ‘Others’. The standard deviation (σ) was calculated for each
1030
specific protein complexes with N subunits and then compared to the distribution of 100,000
1031
random samplings of groups of size N from ‘Others’. The red line shows the σ value of a
1032
specific protein complex while the blue distribution shows a histogram of σ values from the
1033
100,000 random samplings of each value of N. The probability (p) value is the proportion of
1034
100,000 random samplings of N that show a smaller σ than the specific protein complex.
(B) A Cumulative Relative Frequency
The protein complexes were: Cytosol Ribosome,
1035
1036
Figure 5. Comparisons of the relative protein abundance and degradation rate of
1037
specific proteins that differed in their measurements between leaf 3 and leaf 7. (A)
1038
Proteins with significant changes (T-test, p<0.05) in abundance (502 pairs, left) or
29
1039
degradation rate (433 pairs, right) between leaf 3 and leaf 7 are presented. The changes in
1040
relative abundance and degradation rate are shown as relative ∆Abundance and ∆KD by
1041
comparing leaf 7 to leaf 3 data. Major functional categories (as in Figure 2) that comprised
1042
at least 5 proteins for abundance or degradation data are displayed. All remaining proteins
1043
were combined in the group ‘Others’. Outliers are presented as hollow dots. The dashed line
1044
shows the boundary for no change in abundance and degradation between leaf 3 and leaf 7.
1045
(B) Relative ∆Abundance was compared for leaf 3, 5 and 7 for photosynthesis protein
1046
complexes and functional categories (110 proteins). The proteins on the X-axis are in the
1047
same order for each leaf comparison. Error bars show standard errors of relative
1048
∆Abundance from each leaf separately, all comparisons are relative to leaf 3. (C) Protein
1049
degradation rates (KD) was compared for leaves 3, 5 and 7 for photosynthesis protein
1050
complexes and functional categories (116 proteins). The proteins on the X-axis are in the
1051
same order for each leaf comparison. Average KD and standard errors calculated from data
1052
for each leaf separately are shown. The proteins in (B) and (C) are not in the same order but
1053
are shown in rank order within each complex to show the consistent trend in relative
1054
∆Abundance and protein degradation rates (KD) across the three leaves.
1055
1056
1057
Figure 6. Correlation of Arabidopsis protein degradation rates with leaf growth rate.
1058
Protein degradation rates (KD) in leaves 5 and 7 were normalized to corresponding values
1059
for the same proteins in leaf 3. Ratios of KD to leaf 3 KD values for (A) 92 photosynthesis
1060
proteins; (B) 11 plastid ribosome proteins; (C) plastid proteases including CLPs (CLPP2,
1061
CLPP5, CLPP6, CLPC, CLPR1, CLPR2, CLPR4, CLPR 6, Clp), DEG (DEGP1, DEGP 2)
1062
and MAP, and FTSHs (FTSH1, FTSH2, FTSH5, FTSH8); (D) 12 cytosol ribosome proteins
1063
and 10 initiation and elongation proteins; and (E) Proteasome E2 and E3 ligase proteins,
1064
are shown plotted against the leaf growth rates of leaves 3, 5 and 7. All proteins shown had
1065
positive or negative Pearson correlations between the ratio of KD and leaf growth rate of
1066
R>0.9 or R<-0.9, respectively. The red dashed line is a boundary for negative and positive
1067
correlations.
1068
compared with leaf 3 (*p<0.05). Error bars are SE of all data from each leaf separately.
(F) The relative protein abundance of CPN60 A&B in leaves 5 and 7 is
1069
1070
1071
Figure 7. ATP energy budgets for protein degradation and synthesis in Arabidopsis
1072
leaves. (A) The proportion of cellular ATP used for protein degradation (blue), protein
1073
synthesis (yellow) and other aspects of cell maintenance (grey). Energy budgets are
1074
presented as the averages across leaves 3, 5 and 7, and in each leaf individually. (B) The
1075
proportion of ATP used for protein degradation (blue) and synthesis (yellow) in maintaining
30
1076
the proteomes of major cellular organelles and structures (when ATP cost >1% of total ATP
1077
use).
1078
functional categories of proteins (when ATP cost >2% of total ATP use). (D) The fifteen
1079
individual proteins with the highest ATP cost to maintain in the Arabidopsis proteome; only
1080
four of these were from the two highest cost functional categories (dotted lines). Subcellular
1081
location, functional categories and specific high cost proteins were sorted by the proportion
1082
of energy cost for protein synthesis (yellow). Respiration was presumed to be the primary
1083
net ATP production source for protein synthesis and its value is based on experimental data
1084
from the Arabidopsis rosette.
(C) ATP used for protein degradation (blue) and synthesis (yellow) by major
1085
1086
1087
1088
1089
31
1090
1091
1092
1093
1094
Table 1 The twenty most-rapidly degrading proteins in Arabidopsis leaves. Protein
accession number, short name, location in the cell and functional category are shown for
each protein. The protein’s degradation rate (KD) and its standard error (SE) are shown
along with the number of biological samples in which it was observed (Brep No.), the number
of peptide spectra used to calculate the KD and the calculated half-life of the protein.
Protein
Location
Functional Category
KD
-1
(d )
SE
Brep
No.
Pep
No.
Halflife (d)
31
0.36
At5g54770
THI1
Plastid
co-factor and vitamin metabolism
1.93
0.16
12
At1g79930
HSP91
Cytosol
stress
1.88
0.49
6
6
0.37
0.11
7
12
0.43
At4g17090
BAM3
Plastid
major CHO metabolism
1.62
At3g19710
BCAT4
Cytosol
AA synthesis
1.15
0.09
5
8
0.60
0.31
4
5
0.64
At3g22200
POP2
Mito
AA synthesis
1.09
AtCg00020
D1
Plastid
light reaction
1.08
0.15
11
36
0.64
0.05
7
8
0.66
At5g19140
AILP1
Plastid
metal handling
1.05
At1g23290
RPL27A
Cytosol
protein synthesis
0.97
0.05
5
5
0.72
0.09
10
28
0.78
At2g29630
THIC
Plastid
co-factor and vitamin metabolism
0.89
At3g15840
PIFI
Plastid
light reaction
0.87
0.10
8
8
0.80
0.14
7
11
0.82
At1g72930
TIR
Cytosol
stress
0.84
At1g11910
APA1
Vacuole
protein degradation
0.78
0.06
12
13
0.89
At3g20050
TCP-1
Cytosol
protein folding
0.77
0.34
5
7
0.90
0.03
9
9
0.91
At5g19940
PAP
Plastid
cell organisation
0.76
At4g14030
SBP1
Cytosol
metal handling
0.76
0.14
6
8
0.91
0.32
7
7
0.92
At1g22930
TCP-11
Nucleus
not assigned
0.75
At3g02110
SCPL25
Extracellular
protein degradation
0.73
0.02
9
9
0.94
0.31
4
5
0.95
At4g15545
NA
Nucleus
not assigned
0.73
At1g42960
NA
Plastid
not assigned
0.69
0.38
6
7
1.00
misc
0.69
0.05
11
15
1.00
At5g62350
Invertase
Extracellular
1095
1096
1097
32
1098
Table 2 Average degradation rates (KD) of groups of proteins containing the same protein
1099
domain. Shown are 58 protein domains (with at least 4 protein members per domain) that
1100
each had a tight average KD distribution (Median/SD>2). Non-parametric k-sample K-W
1101
analysis (p<0.01) and a C-I post-hoc test was used to divide the protein domains into groups
1102
A-J. Domains are listed by increasing degradation rates (all data in Supplemental Data set
1103
4E). The number of protein members (No.), KD minimum, maximum and median for each
1104
protein domain are shown.
-1
Protein Domains
Translation protein SH3-like domain
-1
-1
PD Code
No.
KD (d )
Min
KD (d )
Max
KD (d )
Median
IPR008991
4
0.024
0.070
0.047
A
0.076
0.069
A-B
Group
Glyceraldehyde-3-phosphate dehydrogenase, type I
IPR006424
4
0.051
Ribosomal protein L12 family
IPR027534
4
0.041
0.086
0.068
A-B
0.093
0.064
A-B
Bifunctional inhibitor/plant lipid transfer helical domain
IPR016140
4
0.054
PsbQ-like domain
IPR023222
4
0.050
0.095
0.070
A-B
0.096
0.076
A-B
NB-ARC
IPR002182
4
0.055
Zinc finger, PHD-type
IPR001965
4
0.060
0.093
0.080
A-B
0.104
0.084
A-B
Zinc finger, RING/FYVE/PHD-type
IPR013083
6
0.016
Serine/threonine-protein kinase, active site
IPR008271
4
0.025
0.104
0.080
A-B
0.116
0.084
A-B
Mitochondrial substrate/solute carrier
IPR018108
5
0.044
PsbP family
IPR002683
11
0.046
0.140
0.086
A-B
0.173
0.079
A-C
Chlorophyll a/b binding protein domain
IPR023329
14
0.040
Tetratricopeptide repeat
IPR019734
4
0.067
0.104
0.087
A-D
0.115
0.088
A-D
Protein kinase, ATP binding site
IPR017441
5
0.025
Transketolase-like, pyrimidine-binding domain
IPR005475
4
0.069
0.117
0.081
A-D
0.137
0.080
A-D
Riboflavin synthase-like beta-barrel
IPR017938
4
0.064
Protein kinase-like domain
IPR011009
8
0.025
0.124
0.096
A-D
ATPase, alpha/beta subunit, nucleotide-binding domain
IPR000194
5
0.050
0.112
0.102
A-E
0.151
0.084
A-E
Galactose mutarotase-like domain
IPR011013
5
0.069
Aminoacyl-tRNA synthetase, class II (G/ P/ S/T)
IPR002314
6
0.068
0.135
0.093
B-E
0.135
0.107
B-E
Isocitrate/isopropylmalate dehydrogenase domain
IPR019818
4
0.037
Proteasome, subunit alpha/beta
IPR001353
10
0.072
0.136
0.093
B-E
0.151
0.091
B-E
Glycoside hydrolase-type carbohydrate-binding, subgroup
IPR014718
4
0.084
Malate dehydrogenase, active site
IPR001252
5
0.025
0.181
0.123
B-E
0.178
0.103
B-E
Ribosomal protein S5 domain 2-type fold
IPR020568
8
0.034
HAD hydrolase, subfamily IA
IPR006439
4
0.069
0.152
0.101
B-F
0.135
0.115
B-F
Isopropylmalate dehydrogenase-like domain
IPR024084
5
0.037
Short-chain dehydrogenase/reductase, conserved site
IPR020904
4
0.058
0.148
0.116
B-G
0.202
0.129
C-G
PDZ domain
IPR001478
5
0.047
D-isomer specific 2-hydroxyacid dehydrogenase domain
IPR029752
5
0.005
0.176
0.160
D-G
0.179
0.132
D-G
Aminotransferase, class I/classII
IPR004839
8
0.055
Threonyl/alanyl tRNA synthetase, SAD
IPR012947
4
0.097
0.168
0.119
D-H
0.204
0.128
D-H
Dehydrogenase, E1 component
IPR001017
4
0.071
Calycin
IPR012674
4
0.082
0.210
0.128
D-H
6
0.093
0.224
0.114
D-H
Fructose-bisphosphate aldolase, class-I
IPR000741
33
Glutathione S-transferase, C-terminal
IPR004046
6
0.062
0.230
0.159
E-H
0.219
0.160
E-H
Clp, N-terminal
IPR004176
4
0.065
AMP-dependent synthetase/ligase
IPR000873
4
0.070
0.222
0.146
E-H
0.271
0.160
E-H
ATP-grasp fold, subdomain 1
IPR013815
7
0.063
14-3-3 protein, conserved site
IPR023409
7
0.096
0.185
0.129
E-H
Aldo/keto reductase
IPR001395
4
0.100
0.172
0.135
E-I
0.220
0.131
F-I
Thioredoxin domain
IPR013766
12
0.094
Immunoglobulin-like fold
IPR013783
5
0.103
0.171
0.152
F-I
0.227
0.153
F-I
Glutaredoxin
IPR002109
6
0.076
Pyridine nucleotide FAD/NAD(P)-binding domain
IPR023753
11
0.070
0.296
0.170
F-I
0.220
0.169
G-I
Thioredoxin
IPR005746
9
0.103
Alpha-D-phosphohexomutase superfamily
IPR005841
5
0.101
0.298
0.153
G-I
0.261
0.161
G-I
ATPase, AAA-type, conserved site
IPR003960
8
0.056
Tubulin/FtsZ, C-terminal
IPR008280
4
0.061
0.276
0.208
G-I
0.266
0.164
G-I
Alcohol dehydrogenase superfamily, zinc-type
IPR002085
8
0.097
Rossmann-like alpha/beta/alpha sandwich fold
IPR014729
13
0.077
0.376
0.201
H-I
0.196
0.179
H-J
Immunoglobulin E-set
IPR014756
5
0.141
Ferredoxin [2Fe-2S], plant
IPR010241
4
0.130
0.267
0.194
H-J
0.292
0.198
H-J
Aconitase/3-isopropylmalate dehydratase domain
IPR001030
4
0.134
Pyruvate/Phosphoenolpyruvate kinase-like domain
IPR015813
6
0.172
0.271
0.197
I-J
0.221
0.218
I-J
UspA
IPR006016
5
0.192
Peptidase, FtsH
IPR005936
4
0.162
0.261
0.238
I-J
4
0.351
0.388
0.377
J
Ubiquitin-conjugating enzyme/RWD-like
IPR016135
1105
1106
34
1107
Table 3 ATP cost of protein degradation and synthesis for major functional categories of leaf
1108
proteins. Major functional categories were divided into seven groups. Minor functional
1109
categories and non-assigned proteins were grouped together as ‘Other’. Functional
1110
categories were sorted by the proportional cost of protein synthesis. The number of proteins
1111
in each group, KD mean and standard deviation (SD) are shown.
-1
-1
No.
KD (d )
Mean
KD (d )
SD
Deg Cost (%)
Syn Cost (%)
Calvin cycle
27
0.11
0.09
25.2
30.1
light reaction
84
0.13
0.15
16.5
16.3
photorespiration
11
0.12
0.05
2.7
2.4
AA synthesis
50
0.19
0.22
3.6
3.2
secondary metabolism
33
0.18
0.13
1.8
2.0
TCA
31
0.12
0.05
2.1
2.0
hormone metabolism
15
0.25
0.18
2.5
2.0
glycolysis
24
0.14
0.05
2.0
1.9
co-factor and vitamin metabolism
10
0.38
0.60
2.3
1.6
major CHO metabolism
17
0.25
0.36
1.3
1.1
lipid metabolism
29
0.13
0.09
1.1
1.1
tetrapyrrole synthesis
16
0.12
0.07
1.1
1.0
nucleotide metabolism
26
0.13
0.11
1.0
1.0
C1-metabolism
12
0.14
0.04
1.1
0.9
ETC
21
0.12
0.08
0.3
0.3
minor CHO metabolism
10
0.15
0.07
0.2
0.2
protein synthesis
76
0.08
0.14
4.0
4.2
protein degradation
81
0.17
0.14
3.3
2.8
protein folding
24
0.12
0.17
0.7
0.8
protein targeting
18
0.13
0.08
0.5
0.5
protein AA activation
20
0.15
0.12
0.5
0.4
protein PTMs
16
0.15
0.08
0.2
0.2
redox
52
0.15
0.08
3.3
3.1
misc.
53
0.16
0.11
2.8
2.5
stress
43
0.19
0.30
2.1
2.1
signalling
26
0.14
0.08
0.7
0.6
cell organisation
22
0.19
0.16
1.5
1.3
cell cycle
14
0.14
0.10
0.9
1.0
Functional Categories
Photosynthesis (48.9%)
Metabolism (18.4%)
Protein (8.9%)
Stress and Signalling (8.3%)
Cell (2.6%)
35
20
0.17
0.10
0.4
0.3
RNA binding
13
0.15
0.10
1.3
1.2
RNA regulation
33
0.12
0.07
1.0
1.0
DNA
11
0.13
0.10
0.1
0.1
transport
31
0.12
0.06
1.4
1.3
metal handling
11
0.33
0.31
0.5
0.3
Other (9.1%)
248
0.15
0.12
9.7
9.1
cell wall
RNA and DNA (2.2%)
Transport and Metal Handling (1.6%)
1112
1113
1114
36
1115
Reference
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
Apel, W., Schulze, W.X., and Bock, R. (2010). Identification of protein stability determinants
in chloroplasts. Plant J 63, 636-650.
Araujo, W.L., Tohge, T., Ishizaki, K., Leaver, C.J., and Fernie, A.R. (2011). Protein degradation
- an alternative respiratory substrate for stressed plants. Trends Plant Sci 16, 489498.
Armbruster, U., Zuhlke, J., Rengstl, B., Kreller, R., Makarenko, E., Ruhle, T., Schunemann,
D., Jahns, P., Weisshaar, B., Nickelsen, J., and Leister, D. (2010). The Arabidopsis
thylakoid protein PAM68 is required for efficient D1 biogenesis and photosystem II
assembly. Plant Cell 22, 3439-3460.
Avci, D., and Lemberg, M.K. (2015). Clipping or Extracting: Two Ways to Membrane Protein
Degradation. Trends in cell biology 25, 611-622.
Bachmair, A., Finley, D., and Varshavsky, A. (1986). In vivo half-life of a protein is a function
of its amino-terminal residue. Science 234, 179-186.
Baerenfaller, K., Massonnet, C., Walsh, S., Baginsky, S., Buhlmann, P., Hennig, L., HirschHoffmann, M., Howell, K.A., Kahlau, S., Radziejwoski, A., Russenberger, D.,
Rutishauser, D., Small, I., Stekhoven, D., Sulpice, R., Svozil, J., Wuyts, N., Stitt, M.,
Hilson, P., Granier, C., and Gruissem, W. (2012). Systems-based analysis of
Arabidopsis leaf growth reveals adaptation to water deficit. Mol Syst Biol 8, 606.
Blasing, O.E., Gibon, Y., Gunther, M., Hohne, M., Morcuende, R., Osuna, D., Thimm, O.,
Usadel, B., Scheible, W.R., and Stitt, M. (2005). Sugars and circadian regulation make
major contributions to the global regulation of diurnal gene expression in
Arabidopsis. The Plant cell 17, 3257-3281.
Boyes, D.C., Zayed, A.M., Ascenzi, R., McCaskill, A.J., Hoffman, N.E., Davis, K.R., and
Gorlach, J. (2001). Growth stage-based phenotypic analysis of Arabidopsis: a model
for high throughput functional genomics in plants. Plant Cell 13, 1499-1510.
Cambridge, S.B., Gnad, F., Nguyen, C., Bermejo, J.L., Kruger, M., and Mann, M. (2011).
Systems-wide proteomic analysis in mammalian cells reveals conserved, functional
protein turnover. J Proteome Res 10, 5275-5284.
Chatterjee, A., Abeydeera, N.D., Bale, S., Pai, P.J., Dorrestein, P.C., Russell, D.H., Ealick, S.E.,
and Begley, T.P. (2011). Saccharomyces cerevisiae THI4p is a suicide thiamine thiazole
synthase. Nature 478, 542-546.
Chen, W.P., Yang, X.Y., Harms, G.L., Gray, W.M., Hegeman, A.D., and Cohen, J.D. (2011). An
automated growth enclosure for metabolic labeling of Arabidopsis thaliana with 13Ccarbon dioxide - an in vivo labeling system for proteomics and metabolomics
research. Proteome Sci 9, 9.
Christiano, R., Nagaraj, N., Frohlich, F., and Walther, T.C. (2014). Global proteome turnover
analyses of the Yeasts S. cerevisiae and S. pombe. Cell Rep 9, 1959-1965.
Claydon, A.J., and Beynon, R. (2012). Proteome dynamics: revisiting turnover with a global
perspective. Mol Cell Proteomics 11, 1551-1565.
De Baets, G., Reumers, J., Delgado Blanco, J., Dopazo, J., Schymkowitz, J., and Rousseau, F.
(2011). An evolutionary trade-off between protein turnover rate and protein
aggregation favors a higher aggregation propensity in fast degrading proteins. PLoS
Comput Biol 7, e1002090.
Deutsch, E.W., Mendoza, L., Shteynberg, D., Farrah, T., Lam, H., Tasman, N., Sun, Z.,
Nilsson, E., Pratt, B., Prazen, B., Eng, J.K., Martin, D.B., Nesvizhskii, A.I., and
Aebersold, R. (2010). A guided tour of the Trans-Proteomic Pipeline. Proteomics 10,
37
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1150-1159.
Doherty, M.K., Hammond, D.E., Clague, M.J., Gaskell, S.J., and Beynon, R.J. (2009).
Turnover of the human proteome: determination of protein intracellular stability by
dynamic SILAC. J Proteome Res 8, 104-112.
Eckardt, N.A., Snyder, G.W., Portis, A.R., Jr., and Orgen, W.L. (1997). Growth and
photosynthesis under high and low irradiance of Arabidopsis thaliana antisense
mutants with reduced ribulose-1,5-bisphosphate carboxylase/oxygenase activase
content. Plant physiology 113, 575-586.
Edwards, J.M., Roberts, T.H., and Atwell, B.J. (2012). Quantifying ATP turnover in anoxic
coleoptiles of rice (Oryza sativa) demonstrates preferential allocation of energy to
protein synthesis. J Exp Bot 63, 4389-4402.
Ellis, R.J. (1981). Chloroplast Proteins - Synthesis, Transport, and Assembly. Annual Review of
Plant Physiology and Plant Molecular Biology 32, 111-137.
Fan, K.T., Rendahl, A.K., Chen, W.P., Freund, D.M., Gray, W.M., Cohen, J.D., and Hegeman,
A.D. (2016). Proteome Scale-Protein Turnover Analysis Using High Resolution Mass
Spectrometric Data from Stable-Isotope Labeled Plants. J Proteome Res.
Fernandez-Escamilla, A.M., Rousseau, F., Schymkowitz, J., and Serrano, L. (2004).
Prediction of sequence-dependent and mutational effects on the aggregation of
peptides and proteins. Nat Biotechnol 22, 1302-1306.
Fleig, L., Bergbold, N., Sahasrabudhe, P., Geiger, B., Kaltak, L., and Lemberg, M.K. (2012).
Ubiquitin-dependent intramembrane rhomboid protease promotes ERAD of
membrane proteins. Molecular Cell 47, 558-569.
Flugge, U.I., Hausler, R.E., Ludewig, F., and Gierth, M. (2011). The role of transporters in
supplying energy to plant plastids. J Exp Bot 62, 2381-2392.
Gfeller, A., Baerenfaller, K., Loscos, J., Chetelat, A., Baginsky, S., and Farmer, E.E. (2011).
Jasmonate controls polypeptide patterning in undamaged tissue in wounded
Arabidopsis leaves. Plant Physiol 156, 1797-1807.
Giraud, E., Ng, S., Carrie, C., Duncan, O., Low, J., Lee, C.P., Van Aken, O., Millar, A.H.,
Murcha, M., and Whelan, J. (2010). TCP transcription factors link the regulation of
genes encoding mitochondrial proteins with the circadian clock in Arabidopsis
thaliana. Plant Cell 22, 3921-3934.
Gonda, D.K., Bachmair, A., Wunning, I., Tobias, J.W., Lane, W.S., and Varshavsky, A. (1989).
Universality and structure of the N-end rule. J Biol Chem 264, 16700-16712.
Guan, S.H., Price, J.C., Prusiner, S.B., Ghaemmaghami, S., and Burlingame, A.L. (2011). A
Data Processing Pipeline for Mammalian Proteome Dynamics Studies Using Stable
Isotope Metabolic Labeling. Molecular & Cellular Proteomics 10. M111.010728
Hasan, S.S., Yamashita, E., Baniulis, D., and Cramer, W.A. (2013). Quinone-dependent
proton transfer pathways in the photosynthetic cytochrome b6f complex. Proc Natl
Acad Sci U S A 110, 4297-4302.
Hooper, C.M., Tanz, S.K., Castleden, I.R., Vacher, M.A., Small, I.D., and Millar, A.H. (2014).
SUBAcon: a consensus algorithm for unifying the subcellular localization data of the
Arabidopsis proteome. Bioinformatics 30, 3356-3364.
Huang, S., Nelson, C.J., Li, L., Taylor, N.L., Stroher, E., Peteriet, J., and Millar, A.H. (2015).
INTERMEDIATE CLEAVAGE PEPTIDASE55 Modifies Enzyme Amino Termini and Alters
Protein Stability in Arabidopsis Mitochondria. Plant Physiol 168, 415-427.
Huang, W., Ling, Q., and Jarvis, P. (2013). The ubiquitin-proteasome system regulates
chloroplast biogenesis. Commun Integr Biol 6, e23001.
38
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
Ishihara, H., Obata, T., Sulpice, R., Fernie, A.R., and Stitt, M. (2015). Quantifying protein
synthesis and degradation in Arabidopsis by dynamic 13CO2 labeling and analysis of
enrichment in individual amino acids in their free pools and in protein. Plant Physiol
168, 74-93.
Janska, H., Kwasniak, M., and Szczepanowska, J. (2013). Protein quality control in organelles
- AAA/FtsH story. Biochim Biophys Acta 1833, 381-387.
Jayapal, K.P., Sui, S., Philp, R.J., Kok, Y.J., Yap, M.G., Griffin, T.J., and Hu, W.S. (2010).
Multitagging proteomic strategy to estimate protein turnover rates in dynamic
systems. J Proteome Res 9, 2087-2097.
Juntawong, P., Girke, T., Bazin, J., and Bailey-Serres, J. (2014). Translational dynamics
revealed by genome-wide profiling of ribosome footprints in Arabidopsis. Proc Natl
Acad Sci U S A 111, E203-212.
Kaleta, C., Schauble, S., Rinas, U., and Schuster, S. (2013). Metabolic costs of amino acid and
protein production in Escherichia coli. Biotechnol J 8, 1105-1114.
Karunadharma, P.P., Basisty, N., Chiao, Y.A., Dai, D.F., Drake, R., Levy, N., Koh, W.J., Emond,
M.J., Kruse, S., Marcinek, D., Maccoss, M.J., and Rabinovitch, P.S. (2015).
Respiratory chain protein turnover rates in mice are highly heterogeneous but
strikingly conserved across tissues, ages, and treatments. FASEB J 29, 3582-3592.
Kato, Y., Sun, X., Zhang, L., and Sakamoto, W. (2012). Cooperative D1 Degradation in the
Photosystem II Repair Mediated by Chloroplastic Proteases in Arabidopsis. Plant
Physiol. 159:1428-39
Kramer, D.M., and Evans, J.R. (2011). The importance of energy balance in improving
photosynthetic productivity. Plant Physiol 155, 70-78.
Lam, M.P., Wang, D., Lau, E., Liem, D.A., Kim, A.K., Ng, D.C., Liang, X., Bleakley, B.J., Liu, C.,
Tabaraki, J.D., Cadeiras, M., Wang, Y., Deng, M.C., and Ping, P. (2014). Protein
kinetic signatures of the remodeling heart following isoproterenol stimulation. J Clin
Invest 124, 1734-1744.
Lambreva, M.D., Russo, D., Polticelli, F., Scognamiglio, V., Antonacci, A., Zobnina, V.,
Campi, G., and Rea, G. (2014). Structure/function/dynamics of photosystem II
plastoquinone binding sites. Curr Protein Pept Sci 15, 285-295.
Lamesch, P., Berardini, T.Z., Li, D., Swarbreck, D., Wilks, C., Sasidharan, R., Muller, R.,
Dreher, K., Alexander, D.L., Garcia-Hernandez, M., Karthikeyan, A.S., Lee, C.H.,
Nelson, W.D., Ploetz, L., Singh, S., Wensel, A., and Huala, E. (2012). The Arabidopsis
Information Resource (TAIR): improved gene annotation and new tools. Nucleic Acids
Research 40, D1202-1210.
Lee, C.P., Eubel, H., and Millar, A.H. (2010). Diurnal changes in mitochondrial function reveal
daily optimization of light and dark respiratory metabolism in Arabidopsis. Molecular
& Cellular Proteomics 9, 2125-2139.
Li, L., Nelson, C.J., Solheim, C., Whelan, J., and Millar, A.H. (2012). Determining degradation
and synthesis rates of arabidopsis proteins using the kinetics of progressive 15N
labeling of two-dimensional gel-separated protein spots. Mol Cell Proteomics 11,
M111 010025.
Li, L., Nelson, C., Fenske, R., Trösch, J., Pružinská, A., Millar, A.H., and Huang, S. (2016).
Changes in specific protein degradation rates in Arabidopsis thaliana reveal multiple
roles of Lon1 in mitochondrial protein homeostasis. Plant J. (in press,
10.1111/tpj.13392)
Liang, C., Cheng, S., Zhang, Y., Sun, Y., Fernie, A.R., Kang, K., Panagiotou, G., Lo, C., and Lim,
39
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
B.L. (2016). Transcriptomic, proteomic and metabolic changes in Arabidopsis thaliana
leaves after the onset of illumination. BMC plant biology 16, 43.
Lim, W.K., Wang, K., Lefebvre, C., and Califano, A. (2007). Comparative analysis of
microarray normalization procedures: effects on reverse engineering gene networks.
Bioinformatics 23, i282-288.
Ling, Q., Huang, W., Baldwin, A., and Jarvis, P. (2012). Chloroplast biogenesis is regulated by
direct action of the ubiquitin-proteasome system. Science 338, 655-659.
Lyon, D., Castillejo, M.A., Mehmeti-Tershani, V., Staudinger, C., Kleemaier, C., and
Wienkoop, S. (2016). Drought and recovery: independently regulated processes
highlighting the importance of protein turnover dynamics and translational
regulation in Medicago truncatula. Mol Cell Proteomics. 15:1921-37
Masclaux-Daubresse, C., Daniel-Vedele, F., Dechorgnat, J., Chardon, F., Gaufichon, L., and
Suzuki, A. (2010). Nitrogen uptake, assimilation and remobilization in plants:
challenges for sustainable and productive agriculture. Ann Bot 105, 1141-1157.
Melis, A. (1999). Photosystem-II damage and repair cycle in chloroplasts: what modulates
the rate of photodamage ? Trends Plant Sci 4, 130-135.
Mockler, T.C., Michael, T.P., Priest, H.D., Shen, R., Sullivan, C.M., Givan, S.A., McEntee, C.,
Kay, S.A., and Chory, J. (2007). The DIURNAL project: DIURNAL and circadian
expression profiling, model-based pattern matching, and promoter analysis. Cold
Spring Harbor symposia on quantitative biology 72, 353-363.
Nelson, C.J., Li, L., Jacoby, R.P., and Millar, A.H. (2013). Degradation rate of mitochondrial
proteins in Arabidopsis thaliana cells. J Proteome Res 12, 3449-3459.
Nelson, C.J., Alexova, R., Jacoby, R.P., and Millar, A.H. (2014). Proteins with high turnover
rate in barley leaves estimated by proteome analysis combined with in planta isotope
labeling. Plant Physiol 166, 91-108.
Nesvizhskii, A.I., Keller, A., Kolker, E., and Aebersold, R. (2003). A statistical model for
identifying proteins by tandem mass spectrometry. Anal Chem 75, 4646-4658.
Nishimura, K., Fukagawa, T., Takisawa, H., Kakimoto, T., and Kanemaki, M. (2009). An
auxin-based degron system for the rapid depletion of proteins in nonplant cells. Nat
Methods 6, 917-922.
Nussbaumer, T., Martis, M.M., Roessner, S.K., Pfeifer, M., Bader, K.C., Sharma, S.,
Gundlach, H., and Spannagl, M. (2013). MIPS PlantsDB: a database framework for
comparative plant genome research. Nucleic acids research 41, D1144-1151.
Osteryoung, K.W., and Pyke, K.A. (2014). Division and dynamic morphology of plastids.
Annu Rev Plant Biol 65, 443-472.
Peth, A., Nathan, J.A., and Goldberg, A.L. (2013). The ATP costs and time required to
degrade ubiquitinated proteins by the 26 S proteasome. J Biol Chem 288, 2921529222.
Piques, M., Schulze, W.X., Hohne, M., Usadel, B., Gibon, Y., Rohwer, J., and Stitt, M. (2009).
Ribosome and transcript copy numbers, polysome occupancy and enzyme dynamics
in Arabidopsis. Mol Syst Biol 5, 314.
Ponnala, L., Wang, Y., Sun, Q., and van Wijk, K.J. (2014). Correlation of mRNA and protein
abundance in the developing maize leaf. The Plant journal : for cell and molecular
biology 78, 424-440.
Price, J.C., Guan, S., Burlingame, A., Prusiner, S.B., and Ghaemmaghami, S. (2010). Analysis
of proteome dynamics in the mouse brain. Proc Natl Acad Sci U S A 107, 1450814513.
40
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
Reinhold, T., Alawady, A., Grimm, B., Beran, K.C., Jahns, P., Conrath, U., Bauer, J., Reiser, J.,
Melzer, M., Jeblick, W., and Neuhaus, H.E. (2007). Limitation of nocturnal import of
ATP into Arabidopsis chloroplasts leads to photooxidative damage. Plant J 50, 293304.
Ren, M., Qiu, S., Venglat, P., Xiang, D., Feng, L., Selvaraj, G., and Datla, R. (2011). Target of
rapamycin regulates development and ribosomal RNA expression through kinase
domain in Arabidopsis. Plant Physiol 155, 1367-1382.
Rousseeuw, P.J. (1987). Silhouettes - a Graphical Aid to the Interpretation and Validation of
Cluster-Analysis. Journal of Computational and Applied Mathematics 20, 53-65.
Rowland, E., Kim, J., Bhuiyan, N.H., and van Wijk, K.J. (2015). The Arabidopsis Chloroplast
Stromal N-Terminome: Complexities of Amino-Terminal Protein Maturation and
Stability. Plant Physiol 169, 1881-1896.
Scheurwater, I., Dunnebacke, M., Eising, R., and Lambers, H. (2000). Respiratory costs and
rate of protein turnover in the roots of a fast-growing (Dactylis glomerata L.) and a
slow-growing (Festuca ovina L.) grass species. J Exp Bot 51, 1089-1097.
Schmitz, G., Reinhold, T., Gobel, C., Feussner, I., Neuhaus, H.E., and Conrath, U. (2010).
Limitation of nocturnal ATP import into chloroplasts seems to affect hormonal
crosstalk, prime defense, and enhance disease resistance in Arabidopsis thaliana.
Mol Plant Microbe Interact 23, 1584-1591.
Schmitz, J., Heinrichs, L., Scossa, F., Fernie, A.R., Oelze, M.L., Dietz, K.J., Rothbart, M.,
Grimm, B., Flugge, U.I., and Hausler, R.E. (2014). The essential role of sugar
metabolism in the acclimation response of Arabidopsis thaliana to high light
intensities. J Exp Bot 65, 1619-1636.
Schwanhausser, B., Busse, D., Li, N., Dittmar, G., Schuchhardt, J., Wolf, J., Chen, W., and
Selbach, M. (2011). Global quantification of mammalian gene expression control.
Nature 473, 337-342.
Sew, Y.S., Stroher, E., Holzmann, C., Huang, S., Taylor, N.L., Jordana, X., and Millar, A.H.
(2013). Multiplex micro-respiratory measurements of Arabidopsis tissues. New
Phytol 200, 922-932.
Shteynberg, D., Deutsch, E.W., Lam, H., Eng, J.K., Sun, Z., Tasman, N., Mendoza, L., Moritz,
R.L., Aebersold, R., and Nesvizhskii, A.I. (2011). iProphet: multi-level integrative
analysis of shotgun proteomic data improves peptide and protein identification rates
and error estimates. Molecular & cellular proteomics : MCP 10, M111 007690.
Siddell, S.G., and Ellis, R.J. (1975). Protein synthesis in chloroplasts. Characteristics and
products of protein synthesis in vitro in etioplasts and developing chloroplasts from
pea leaves. Biochem J 146, 675-685.
Smith, S.M., Fulton, D.C., Chia, T., Thorneycroft, D., Chapple, A., Dunstan, H., Hylton, C.,
Zeeman, S.C., and Smith, A.M. (2004). Diurnal changes in the transcriptome
encoding enzymes of starch metabolism provide evidence for both transcriptional
and posttranscriptional regulation of starch metabolism in Arabidopsis leaves. Plant
physiology 136, 2687-2699.
Suzuki, K., Nakanishi, H., Bower, J., Yoder, D.W., Osteryoung, K.W., and Miyagishima, S.Y.
(2009). Plastid chaperonin proteins Cpn60 alpha and Cpn60 beta are required for
plastid division in Arabidopsis thaliana. BMC Plant Biol 9, 38.
Takahashi, S., and Badger, M.R. (2011). Photoprotection in plants: a new light on
photosystem II damage. Trends Plant Sci 16, 53-60.
Tanz, S.K., Castleden, I., Hooper, C.M., Vacher, M., Small, I., and Millar, H.A. (2013). SUBA3:
41
1350
1351
1352
1353
1354
1355
1356
1357
1358
1359
1360
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
1375
1376
1377
1378
1379
1380
1381
1382
1383
1384
1385
1386
1387
1388
1389
1390
1391
1392
1393
a database for integrating experimentation and prediction to define the SUBcellular
location of proteins in Arabidopsis. Nucleic Acids Res 41, D1185-1191.
Umena, Y., Kawakami, K., Shen, J.R., and Kamiya, N. (2011). Crystal structure of oxygenevolving photosystem II at a resolution of 1.9 A. Nature 473, 55-60.
van Wijk, K.J. (2015). Protein maturation and proteolysis in plant plastids, mitochondria, and
peroxisomes. Annu Rev Plant Biol 66, 75-111.
Vierstra, R.D. (2009). The ubiquitin-26S proteasome system at the nexus of plant biology.
Nat Rev Mol Cell Biol 10, 385-397.
Vogtle, F.N., Wortelkamp, S., Zahedi, R.P., Becker, D., Leidhold, C., Gevaert, K., Kellermann,
J., Voos, W., Sickmann, A., Pfanner, N., and Meisinger, C. (2009). Global Analysis of
the Mitochondrial N-Proteome Identifies a Processing Peptidase Critical for Protein
Stability. Cell 139, 428-439.
Walsh, C.T. (1984). Suicide substrates, mechanism-based enzyme inactivators: recent
developments. Annu Rev Biochem 53, 493-535.
Wang, D., and Portis, A.R., Jr. (2007). A novel nucleus-encoded chloroplast protein, PIFI, is
involved in NAD(P)H dehydrogenase complex-mediated chlororespiratory electron
transport in Arabidopsis. Plant Physiol 144, 1742-1752.
Wang, M., Weiss, M., Simonovic, M., Haertinger, G., Schrimpf, S.P., Hengartner, M.O., and
von Mering, C. (2012). PaxDb, a database of protein abundance averages across all
three domains of life. Mol Cell Proteomics 11, 492-500.
Waters, M.T., Nelson, D.C., Scaffidi, A., Flematti, G.R., Sun, Y.K., Dixon, K.W., and Smith,
S.M. (2012). Specialisation within the DWARF14 protein family confers distinct
responses to karrikins and strigolactones in Arabidopsis. Development 139, 12851295.
Wei, X., Su, X., Cao, P., Liu, X., Chang, W., Li, M., Zhang, X., and Liu, Z. (2016). Structure of
spinach photosystem II-LHCII supercomplex at 3.2 A resolution. Nature 534, 69-74.
Woodson, J.D., Joens, M.S., Sinson, A.B., Gilkerson, J., Salome, P.A., Weigel, D., Fitzpatrick,
J.A., and Chory, J. (2015). Ubiquitin facilitates a quality-control pathway that
removes damaged chloroplasts. Science 350, 450-454.
Yang, X.Y., Chen, W.P., Rendahl, A.K., Hegeman, A.D., Gray, W.M., and Cohen, J.D. (2010).
Measuring the turnover rates of Arabidopsis proteins using deuterium oxide: an
auxin signaling case study. Plant J 63, 680-695.
Yip, C.Y., Harbour, M.E., Jayawardena, K., Fearnley, I.M., and Sazanov, L.A. (2011). Evolution
of respiratory complex I: "supernumerary" subunits are present in the alphaproteobacterial enzyme. J Biol Chem 286, 5023-5033.
Zaltsman, A., Feder, A., and Adam, Z. (2005). Developmental and light effects on the
accumulation of FtsH protease in Arabidopsis chloroplasts--implications for thylakoid
formation and photosystem II maintenance. Plant J 42, 609-617.
Zhang, H., and Cramer, W.A. (2004). Purification and crystallization of the cytochrome b6f
complex in oxygenic photosynthesis. Methods Mol Biol 274, 67-78.
Zhang, H., Deery, M.J., Gannon, L., Powers, S.J., Lilley, K.S., and Theodoulou, F.L. (2015).
Quantitative proteomics analysis of the Arg/N-end rule pathway of targeted
degradation in Arabidopsis roots. Proteomics 15, 2447-2457.
42
5
6
7
15
2.5
2.0
1.5
8
d -1
4
0.
Fold Change in Protein
3
-1
0.03 d-1
d
.08
0
1.0
0.5
0.0
0 1 3 5
0 1 3 5
0 1 3 5
Time Course (d)
Figure 1. Individual growth rates of leaves in the Arabidopsis rosette and the consequent fold change in protein over time.
Leaves 3, 5 and 7 from Arabidopsis plants grown in hydroponic culture were collected
over a time course (T0, 1, 3 and 5 days) beginning with 21-day-old plants that had
developed 10 leaves (leaf production stage 1.10). Typical images of leaves 3 to 8 from
these plants at the beginning of the experiment are shown. The growth of leaves over the
subsequent 5 days were determined by following the fold change in protein per leaf at
each time point com-pared to T0 for leaf 3 (green), leaf 5, (yellow) and leaf 7 (red). An
average relative growth rate (d-1) was determined from both changes in fresh weight and
protein content by scan-ning images of gels (see Supplemental Fig 1, Supplemental Data
set 1).
A
Functional Category
No.
At1g23290 RPL27A
Cyt protein synthesis
33
Cyt glycolysis
17
Cyt secondary metabolism 14
Cyt stress
17
Cyt misc
15
Cyt protein degradation
12
Cyt redox
11
Cyt /Nc protein degradation 16
Et misc
14
Mt ETC
20
Mt TCA
20
Nc RNA regulation
16
Pt protein synthesis
37
Pt protein folding
14
Pt lipid metabolism
16
Pt calvin cycle
27
Pt nucleotide metabolism 13
Pt lightreaction
84
Pt protein targeting
11
Pt AA synthesis
34
Pt tetrapyrrole synthesis
15
24
Pt redox
Pt misc
11
Pt protein degradation
25
Pt major CHO metabolism 13
Pt secondary metabolism 13
Others
686
At1g79930 HSP91
At5g62350 Invertase
At5g13650 elongation factor
Atcg00020 D1
A3g48560 CSR1
At4g17090 BAM3
At5g54770 THI1
0.0
0.5
1.0
KD (d-1)
1.5
2.0
C
B
Mt
128
2.0
Pt
557
Nc
82
1.8
Et
65
Vc
10
R=0.99
THI1
Arabidopsis KD (d-1)
1.6
DUAL 59
1.4
1.2
1.0
Cyt
245
PM
31
ER
13
Px
35
0.4
Golgi 3
0.2
0.0
0.5
KD (d-1)
1.0
0.8
0.6
D1
THIC
pterin dehydratase
R=0.38
0.0
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6
Barley KD (d-1)
Figure 2. Degradation rates of 1228 Arabidopsis proteins based on averages in functional
categories or localization inside cells, and in comparison to barley orthologues.
(A) Average Arabidopsis leaf protein degradation rates box-plotted by
functional categories and their localization in cells. Functioal categories were acquired from MapCave
and localization information from SUBAIII. Only major functional categories and localization groups
(n>10) are shown. Median, quartiles and standard deviations are shown. Outliers are shown as hollow
dots and extreme outliers are highlighted as solid squares and annotated. The five major localizations are
highlighted with colors: Cytosol (Cyt-Blue), Nucleus (Nc-Red), Extracellular (Et-Cyan), Mitochondrion (MtYellow) and Plastid (Pt-Green). (B) Protein degradation rates of the 1228 proteins box-plotted based on
localization within the cell. Abbreviations as in (A), and dual targeting proteins (DUAL), Vacuole (Vc),
plasma membrane (PM), endoplasmic reticulum (ER), peroxisomal (Px) and Golgi.
(C) Protein degradation rates of a subset of 288 Arabidopsis proteins for
which orthologues in barley had published rates. One spot represents a protein turnover rate in Barley (xaxis) and Arabidopsis (y-axis). Four outliers (2SD above median, in red dashed line rectangle) are shown
(Pearson correlation, R=0.99). The remaining 284 proteins (pointed out by black dashed line rectangle)
are shown by black dots (Pearson correlation, R=0.38).
A
Relative
Abundance Set
y axis
300
PC 2
16.8%
200
B
Degradation
Rate (KD) Set
955
5
1228
5
3
-100
3
0
1
Leaf 5
100
Leaf 3
1697
5
100
200
0
1
x axis
PC1
3 42.7%
-100
Leaf 7
-200
0
300
1
C
Leaf 3
Leaf 5
Leaf 7
Days
0
0.0
1
2.8
3
1.5
2.8
5
1.7
3.2
0.0
0
1
3
0.4
0.2
5
2.1
1.7
29.0
33.2
12.7
0
1
3
0.0
0.2
5
0.6
0.2
1.8
14.1
0.4
0
1
3
5
Days
Figure 3. Changes in relative abundance estimates for individual proteins in leaves of the
Arabidopsis rosette during 5 days of leaf growth. (A) The intersection of the number of
specific proteins with a measured KD value and a protein relative abundance assessment. (B)
Principle components analysis of specific protein abundances from mass spectrometry analysis.
Abundance data for 955 proteins (KD values recorded in Figure 2) were used for the analysis.
The clustering of leaf 3 (green), leaf 5 (yellow) and leaf 7 (green) is shown in the grey shading.
The full data are provided in Supplemental Data set 3A and B. (C) A matrix of paired statistical
comparisons of protein abundance between day 0, day 1, day 3 and day 5 time points for each
leaf is presented with the percentage of the proteins from (B) that significantly changed in
abundance (t-test p<0.01) reported (full data in Supplemental Data set 3C). Outside the matrix is
the percentage of proteins measured that were found to change in abundance (t-test P<0.01)
across the entire time course in each leaf.
A
KD (d-1)
1.0
0.5
0.0
10 64 26 12 152 34
5
5
8
28
8
24
K
Q
P
L
G
R
T
S
V
D
A
E
119
39
Average Unstable
239
Stable
Cumulative Relative Freq
B
C
1.0
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0 0
p<0.05
2000 4000 6000 8000
Aggregation Propensity
1.0
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.00
Stable
Unstable
p<0.05
500 1000 1500 2000 2500
Length
1.0
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
p=0.70
0
5
10
15
AGG/LEN
20
No.of Samples
PS I
PS II
Mt ETC
Cyt Ribosome Proteasome Pt ATP synthase Pt Ribosome
6000
p<0.01
p<0.01
p=0.20
p<0.01
p<0.01
p=0.05
p<0.05
5000
4000
3000
2000
1000
0
0.0 0.1 0.2 0.0 0.1 0.2 0.0 0.1 0.2 0.0 0.1 0.2 0.0 0.1 0.2 0.0 0.1 0.2 0.0 0.1 0.2
σ (N=24)
σ (N=16)
σ (N=7)
σ (N=31)
σ (N=17)
σ (N=33)
σ (N=20)
Figure 4. Intrinsic properties of Arabidopsis proteins that correlate with protein degradation rate.
(A) Comparison of protein degradation rate with experimental evidence of the mature N-terminal amino
acid residue. Only protein sets in which more than 5 members had the same N-terminal amino acid
residue were examined, a set of 376 proteins in total. A non-parametric Kruskal-Wallis k-samples test
showed no significant difference among the 12 N-terminal groups. However, the K, S and D groups
showed significantly different degradation rate distributions from T by a Conover-Iman post-hoc test. No
statistically significant differences were found when degradation rates of stable AAs (A, V, M, G, T, P),
unstable AAs (K, F, D, R, Q) and average stability AAs (all others AAs) were compared (Supplemental
Data set 4A). These groups were defined based on the consensus of yeast and mammal N-end rules.
(B) A Cumulative Relative Frequency distribution graph of protein aggregation propensity (AGG) of
each protein, the AA length of each protein (LEN), and the ratio of AGG/AA length between the top 100
most stable and bottom 100 most unstable proteins from the set of 1228 proteins with degradation
rates. Statistical significance was estimated by the Kolmogorov-Smirnov test. (C) The standard
deviation (σ) of the protein degradation rates of the protein subunits of 7 major protein complexes
compared to randomly sampled distributions of σ for the same sized group of proteins from the set of
1228. The protein complexes were: Cytosol Ribosome, Proteasome, Pt ATP synthase, Pt Ribosome,
PSI, PSII and Mt ETC; all other proteins were considered to be one group ‘Others’. The standard
deviation (σ) was calculated for each specific protein complexes with N subunits and then compared to
the distribution of 100,000 random samplings of groups of size N from ‘Others’. The red line shows the
σ value of a specific protein complex while the blue distribution shows a histogram of σ values from the
100,000 random samplings of each value of N. The probability (p) value is the proportion of 100,000
random samplings of N that show a smaller σ than the specific protein complex.
Leaf 7 vs Leaf 3
More
Less
A
Leaf 7 vs Leaf 3
Functional Category
Stable
Unstable
Pt cell organisation
Pt lightreaction
Cyt stress
Pt calvin cycle
Pt protein targeting
Pt major CHO metabolism
Others
Pt misc
Pt protein degradation
Pt redox
Pt AA synthesis
Pt protein folding
Pt protein synthesis
Cyt protein synthesis
Pt tetrapyrrole synthesis
-2
B
-1
0
1
Relative ΔAbundance
2
-2
-1
0
1
Relative ΔKD
2
2.0
Relative ΔAbundance
1.5
PS II
PS I
Cyt b6f
Chlororespiration
1.0
Photorespiration
Calvin Cycle
0.5
0.0
-0.5
-1.0
ATP
synthase
-1.5
Electron NADH
Carrier
DH
-2.0
Leaf 3
C
1.6
Leaf 5
Atcg00020 D1
1.2
Atcg00730
PetD
1.0
KD (d-1)
Chlororespiration
Cyt b6f
1.4
0.8
Leaf 7
PS II
At3g15840
PIFI
Photorespiration
PS I
0.6
ATP
synthase
Calvin Cycle
Electron NADH
Carrier DH
0.4
0.2
STN7
0.0
Figure 5. Comparisons of the relative protein abundance and degradation rate of specific proteins that differed in
their measurements between leaf 3 and leaf 7. (A) Proteins with significant changes (T-test, p<0.05) in abundance (502
pairs, left) or degradation rate (433 pairs, right) between leaf 3 and leaf 7 are presented. The changes in relative abundance
and degradation rate are shown as relative ∆Abundance and ∆KD by comparing leaf 7 to leaf 3 data. Major functional
categories (as in Figure 2) that comprised at least 5 proteins for abundance or degradation data are displayed. All remaining
proteins were combined in the group ‘Others’. Outliers are presented as hollow dots. The dashed line shows the boundary for
no change in abundance and degradation between leaf 3 and leaf 7. (B) Relative ∆Abundance was compared for leaves 3, 5
and 7 for photosynthesis protein complexes and functional categories (110 proteins). The proteins on the X-axis are in the
same order for each leaf comparison. Error bars show standard errors of relative ∆Abundance from each leaf separately, all
comparisons are relative to leaf 3. (C) Protein degradation rates (KD) was compared for leaves 3, 5 and 7 for photosynthesis
protein complexes and functional categories (116 proteins). The proteins on the X-axis are in the same order for each leaf
comparison. Average KD and standard errors calculated from data for each leaf separately are shown. The proteins in (B) and
(C) are not in the same order but are shown in rank order within each complex to show the consistent trend in relative
∆Abundance and Protein degradation rates (KD) across the three leaves.
B
A
4.0
PSBF
3.0
2.0
1.0
0.0
CPN60A&B
0.03
0.08
Growth Rates (d-1)
C
FTSH1,2
5&8
1.0
0.0
CLPs
MAP&DEGs
0.03
0.6
0.4
0.2
0.0
0.03
0.08
Growth Rates (d-1)
1.0
0.08
Growth Rates (d-1)
0.15
Cyt Ribosome, Initiation & Elongation
0.8
0.6
0.4
0.2
0.0
0.03
0.15
E
0.08
Growth Rates (d-1)
0.15
F
Proteasome, E2 and E3
2.8
E3
2.2
1.6
1.0
0.4
0.03
0.08
0.15
Growth Rates (d )
-1
Ratio of protein abundance
to reference
Ratio of KD to Leaf 3
0.8
D
1.5
0.5
Pt Risosome
1.0
0.15
Pt Proteases
2.0
Ratio of KD to Leaf 3
Ferredoxin 1
Ratio of KD to Leaf 3
Photosynthesis
Ratio of KD to Leaf 3
Ratio of KD to Leaf 3
5.0
2.5
CPN60 A
CPN60 B
2.0
1.5
*
*
1.0
0.5
0.0
Leaf 3
Leaf 5
Leaf 7
Figure 6. Correlation of Arabidopsis protein degradation rates with leaf growth rate. Protein degradation rates (KD) in
leaves 5 and 7 were normalized to corresponding values for the same proteins in Leaf 3. Ratios of KD to leaf 3 KD
values for (A) 92 photosynthesis proteins; (B) 11 plastid ribosome proteins; (C) plastid proteases including CLPs
(CLPP2, 5&6; CLPC, CLPR1, 2, 4&6 and Clp), DEG (DEGP1&2) and MAP, and FTSHs (FTSH1, 2, 5&8); (D) 12 cytosol
ribosome proteins and 10 initiation and elongation proteins; and (E) Proteasome E2 and E3 ligase proteins, are shown
plotted against the leaf growth rates of leaves 3, 5 and 7. All proteins shown had positive or negative Pearson correlations
between the ratio of KD and leaf growth rate of R>0.9 or R<-0.9, respectively. The red dashed line is a boundary for
negative and positive correlations. (F) The relative protein abundance of CPN60 A&B in leaves 5 and 7 is compared
with leaf 3 (*p<0.05). Error bars are SE of all data from each leaf separately.
A
Overall
71%
35
4%
4%
13%
Leaf 3
Degradation
Synthesis
Others
25%
3%
C
B
4%
Leaf 5
84%
19%
77%
Leaf 7
58%
38%
Plastid
Cytosol
Mitochondrion
Peroxisome
Extracellular
PM
Nucleus
Vacuole
ATP use by Functional Categories (%)
30
25
20
15
10
5
ATP use for Subcellular Proteome Maintenance (%)
0 10 20 30 40 50 60 70 80
0
calvin cycle
lightreaction
protein synthesis
AA synthesis
redox
protein degradation
misc
photorespiration
stress
secondary metabolism
TCA
hormone metabolism
glycolysis
N-metabolism
Co-factor and vitamine metabolism
D
Highest ATP Cost Individual Proteins (%)
0
RBCL
RCA
ATPB
ATPA
THI1
DRT112
LOX2
RABE1b
GLU1
PMDH2
FBA2
D1
GLDP1
AOS
PSBP-1
5
10
15
Figure 7. ATP energy budgets for protein degradation and synthesis in Arabidopsis leaves. (A) The proportion of cellular ATP used for
protein degradation (blue), protein synthesis (yellow) and other aspects of cell maintenance (grey). Energy budgets are presented as the
averages across leaves 3, 5 and 7, and also in each leaf individually. (B) The proportion of ATP used for protein degradation (blue) and
synthesis (yellow) in maintaining the proteomes of major cellular organelles and structures (when ATP cost >1% of total ATP use).
(C) ATP used for protein degradation (blue) and synthesis (yellow) by major functional categories of proteins (when ATP cost >2% of
total ATP use). (D) The fifteen individual proteins with the highest ATP cost to maintain in the Arabidopsis proteome; only four of these were
from the two highest cost functional categories (dotted lines). Subcellular location, functional categories and specific high cost proteins were
sorted by the proportion of energy cost for protein synthesis (yellow). Respiration was presumed to be the primary net ATP production
source for protein synthesis and its value is based on experimental data from the Arabidopsis rosette.
20
Parsed Citations
Apel, W., Schulze, W.X., and Bock, R. (2010). Identification of protein stability determinants in chloroplasts. Plant J 63, 636-650.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Araujo, W.L., Tohge, T., Ishizaki, K., Leaver, C.J., and Fernie, A.R. (2011). Protein degradation - an alternative respiratory substrate
for stressed plants. Trends Plant Sci 16, 489-498.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Armbruster, U., Zuhlke, J., Rengstl, B., Kreller, R., Makarenko, E., Ruhle, T., Schunemann, D., Jahns, P., Weisshaar, B., Nickelsen,
J., and Leister, D. (2010). The Arabidopsis thylakoid protein PAM68 is required for efficient D1 biogenesis and photosystem II
assembly. Plant Cell 22, 3439-3460.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Avci, D., and Lemberg, M.K. (2015). Clipping or Extracting: Two Ways to Membrane Protein Degradation. Trends in cell biology 25,
611-622.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Bachmair, A., Finley, D., and Varshavsky, A. (1986). In vivo half-life of a protein is a function of its amino-terminal residue. Science
234, 179-186.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Baerenfaller, K., Massonnet, C., Walsh, S., Baginsky, S., Buhlmann, P., Hennig, L., Hirsch-Hoffmann, M., Howell, K.A., Kahlau, S.,
Radziejwoski, A., Russenberger, D., Rutishauser, D., Small, I., Stekhoven, D., Sulpice, R., Svozil, J., Wuyts, N., Stitt, M., Hilson, P.,
Granier, C., and Gruissem, W. (2012). Systems-based analysis of Arabidopsis leaf growth reveals adaptation to water deficit. Mol
Syst Biol 8, 606.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Blasing, O.E., Gibon, Y., Gunther, M., Hohne, M., Morcuende, R., Osuna, D., Thimm, O., Usadel, B., Scheible, W.R., and Stitt, M.
(2005). Sugars and circadian regulation make major contributions to the global regulation of diurnal gene expression in
Arabidopsis. The Plant cell 17, 3257-3281.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Boyes, D.C., Zayed, A.M., Ascenzi, R., McCaskill, A.J., Hoffman, N.E., Davis, K.R., and Gorlach, J. (2001). Growth stage-based
phenotypic analysis of Arabidopsis: a model for high throughput functional genomics in plants. Plant Cell 13, 1499-1510.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Cambridge, S.B., Gnad, F., Nguyen, C., Bermejo, J.L., Kruger, M., and Mann, M. (2011). Systems-wide proteomic analysis in
mammalian cells reveals conserved, functional protein turnover. J Proteome Res 10, 5275-5284.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Chatterjee, A., Abeydeera, N.D., Bale, S., Pai, P.J., Dorrestein, P.C., Russell, D.H., Ealick, S.E., and Begley, T.P. (2011).
Saccharomyces cerevisiae THI4p is a suicide thiamine thiazole synthase. Nature 478, 542-546.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Chen, W.P., Yang, X.Y., Harms, G.L., Gray, W.M., Hegeman, A.D., and Cohen, J.D. (2011). An automated growth enclosure for
metabolic labeling of Arabidopsis thaliana with 13C-carbon dioxide - an in vivo labeling system for proteomics and metabolomics
research. Proteome Sci 9, 9.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Christiano, R., Nagaraj, N., Frohlich, F., and Walther, T.C. (2014). Global proteome turnover analyses of the Yeasts S. cerevisiae
and S. pombe. Cell Rep 9, 1959-1965.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Claydon, A.J., and Beynon, R. (2012). Proteome dynamics: revisiting turnover with a global perspective. Mol Cell Proteomics 11,
1551-1565.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
De Baets, G., Reumers, J., Delgado Blanco, J., Dopazo, J., Schymkowitz, J., and Rousseau, F. (2011). An evolutionary trade-off
between protein turnover rate and protein aggregation favors a higher aggregation propensity in fast degrading proteins. PLoS
Comput Biol 7, e1002090.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Deutsch, E.W., Mendoza, L., Shteynberg, D., Farrah, T., Lam, H., Tasman, N., Sun, Z., Nilsson, E., Pratt, B., Prazen, B., Eng, J.K.,
Martin, D.B., Nesvizhskii, A.I., and Aebersold, R. (2010). A guided tour of the Trans-Proteomic Pipeline. Proteomics 10, 1150-1159.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Doherty, M.K., Hammond, D.E., Clague, M.J., Gaskell, S.J., and Beynon, R.J. (2009). Turnover of the human proteome:
determination of protein intracellular stability by dynamic SILAC. J Proteome Res 8, 104-112.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Eckardt, N.A., Snyder, G.W., Portis, A.R., Jr., and Orgen, W.L. (1997). Growth and photosynthesis under high and low irradiance of
Arabidopsis thaliana antisense mutants with reduced ribulose-1,5-bisphosphate carboxylase/oxygenase activase content. Plant
physiology 113, 575-586.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Edwards, J.M., Roberts, T.H., and Atwell, B.J. (2012). Quantifying ATP turnover in anoxic coleoptiles of rice (Oryza sativa)
demonstrates preferential allocation of energy to protein synthesis. J Exp Bot 63, 4389-4402.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Ellis, R.J. (1981). Chloroplast Proteins - Synthesis, Transport, and Assembly. Annual Review of Plant Physiology and Plant
Molecular Biology 32, 111-137.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Fan, K.T., Rendahl, A.K., Chen, W.P., Freund, D.M., Gray, W.M., Cohen, J.D., and Hegeman, A.D. (2016). Proteome Scale-Protein
Turnover Analysis Using High Resolution Mass Spectrometric Data from Stable-Isotope Labeled Plants. J Proteome Res.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Fernandez-Escamilla, A.M., Rousseau, F., Schymkowitz, J., and Serrano, L. (2004). Prediction of sequence-dependent and
mutational effects on the aggregation of peptides and proteins. Nat Biotechnol 22, 1302-1306.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Fleig, L., Bergbold, N., Sahasrabudhe, P., Geiger, B., Kaltak, L., and Lemberg, M.K. (2012). Ubiquitin-dependent intramembrane
rhomboid protease promotes ERAD of membrane proteins. Molecular Cell 47, 558-569.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Flugge, U.I., Hausler, R.E., Ludewig, F., and Gierth, M. (2011). The role of transporters in supplying energy to plant plastids. J Exp
Bot 62, 2381-2392.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Gfeller, A., Baerenfaller, K., Loscos, J., Chetelat, A., Baginsky, S., and Farmer, E.E. (2011). Jasmonate controls polypeptide
patterning in undamaged tissue in wounded Arabidopsis leaves. Plant Physiol 156, 1797-1807.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Giraud, E., Ng, S., Carrie, C., Duncan, O., Low, J., Lee, C.P., Van Aken, O., Millar, A.H., Murcha, M., and Whelan, J. (2010). TCP
transcription factors link the regulation of genes encoding mitochondrial proteins with the circadian clock in Arabidopsis thaliana.
Plant Cell 22, 3921-3934.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Gonda, D.K., Bachmair, A., Wunning, I., Tobias, J.W., Lane, W.S., and Varshavsky, A. (1989). Universality and structure of the N-end
rule. J Biol Chem 264, 16700-16712.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Guan, S.H., Price, J.C., Prusiner, S.B., Ghaemmaghami, S., and Burlingame, A.L. (2011). A Data Processing Pipeline for Mammalian
Proteome Dynamics Studies Using Stable Isotope Metabolic Labeling. Molecular & Cellular Proteomics 10. M111.010728
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Hasan, S.S., Yamashita, E., Baniulis, D., and Cramer, W.A. (2013). Quinone-dependent proton transfer pathways in the
photosynthetic cytochrome b6f complex. Proc Natl Acad Sci U S A 110, 4297-4302.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Hooper, C.M., Tanz, S.K., Castleden, I.R., Vacher, M.A., Small, I.D., and Millar, A.H. (2014). SUBAcon: a consensus algorithm for
unifying the subcellular localization data of the Arabidopsis proteome. Bioinformatics 30, 3356-3364.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Huang, S., Nelson, C.J., Li, L., Taylor, N.L., Stroher, E., Peteriet, J., and Millar, A.H. (2015). INTERMEDIATE CLEAVAGE
PEPTIDASE55 Modifies Enzyme Amino Termini and Alters Protein Stability in Arabidopsis Mitochondria. Plant Physiol 168, 415-427.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Huang, W., Ling, Q., and Jarvis, P. (2013). The ubiquitin-proteasome system regulates chloroplast biogenesis. Commun Integr Biol
6, e23001.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Ishihara, H., Obata, T., Sulpice, R., Fernie, A.R., and Stitt, M. (2015). Quantifying protein synthesis and degradation in Arabidopsis
by dynamic 13CO2 labeling and analysis of enrichment in individual amino acids in their free pools and in protein. Plant Physiol
168, 74-93.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Janska, H., Kwasniak, M., and Szczepanowska, J. (2013). Protein quality control in organelles - AAA/FtsH story. Biochim Biophys
Acta 1833, 381-387.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Jayapal, K.P., Sui, S., Philp, R.J., Kok, Y.J., Yap, M.G., Griffin, T.J., and Hu, W.S. (2010). Multitagging proteomic strategy to estimate
protein turnover rates in dynamic systems. J Proteome Res 9, 2087-2097.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Juntawong, P., Girke, T., Bazin, J., and Bailey-Serres, J. (2014). Translational dynamics revealed by genome-wide profiling of
ribosome footprints in Arabidopsis. Proc Natl Acad Sci U S A 111, E203-212.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Kaleta, C., Schauble, S., Rinas, U., and Schuster, S. (2013). Metabolic costs of amino acid and protein production in Escherichia
coli. Biotechnol J 8, 1105-1114.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Karunadharma, P.P., Basisty, N., Chiao, Y.A., Dai, D.F., Drake, R., Levy, N., Koh, W.J., Emond, M.J., Kruse, S., Marcinek, D.,
Maccoss, M.J., and Rabinovitch, P.S. (2015). Respiratory chain protein turnover rates in mice are highly heterogeneous but
strikingly conserved across tissues, ages, and treatments. FASEB J 29, 3582-3592.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Kato, Y., Sun, X., Zhang, L., and Sakamoto, W. (2012). Cooperative D1 Degradation in the Photosystem II Repair Mediated by
Chloroplastic Proteases in Arabidopsis. Plant Physiol. 159:1428-39
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Kramer, D.M., and Evans, J.R. (2011). The importance of energy balance in improving photosynthetic productivity. Plant Physiol
155, 70-78.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Lam, M.P., Wang, D., Lau, E., Liem, D.A., Kim, A.K., Ng, D.C., Liang, X., Bleakley, B.J., Liu, C., Tabaraki, J.D., Cadeiras, M., Wang, Y.,
Deng, M.C., and Ping, P. (2014). Protein kinetic signatures of the remodeling heart following isoproterenol stimulation. J Clin
Invest 124, 1734-1744.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Lambreva, M.D., Russo, D., Polticelli, F., Scognamiglio, V., Antonacci, A., Zobnina, V., Campi, G., and Rea, G. (2014).
Structure/function/dynamics of photosystem II plastoquinone binding sites. Curr Protein Pept Sci 15, 285-295.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Lamesch, P., Berardini, T.Z., Li, D., Swarbreck, D., Wilks, C., Sasidharan, R., Muller, R., Dreher, K., Alexander, D.L., GarciaHernandez, M., Karthikeyan, A.S., Lee, C.H., Nelson, W.D., Ploetz, L., Singh, S., Wensel, A., and Huala, E. (2012). The Arabidopsis
Information Resource (TAIR): improved gene annotation and new tools. Nucleic Acids Research 40, D1202-1210.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Lee, C.P., Eubel, H., and Millar, A.H. (2010). Diurnal changes in mitochondrial function reveal daily optimization of light and dark
respiratory metabolism in Arabidopsis. Molecular & Cellular Proteomics 9, 2125-2139.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Li, L., Nelson, C.J., Solheim, C., Whelan, J., and Millar, A.H. (2012). Determining degradation and synthesis rates of arabidopsis
proteins using the kinetics of progressive 15N labeling of two-dimensional gel-separated protein spots. Mol Cell Proteomics 11,
M111 010025.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Li, L., Nelson, C., Fenske, R., Trösch, J., Pružinská, A., Millar, A.H., and Huang, S. (2016). Changes in specific protein degradation
rates in Arabidopsis thaliana reveal multiple roles of Lon1 in mitochondrial protein homeostasis. Plant J. (in press,
10.1111/tpj.13392)
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Liang, C., Cheng, S., Zhang, Y., Sun, Y., Fernie, A.R., Kang, K., Panagiotou, G., Lo, C., and Lim, B.L. (2016). Transcriptomic,
proteomic and metabolic changes in Arabidopsis thaliana leaves after the onset of illumination. BMC plant biology 16, 43.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Lim, W.K., Wang, K., Lefebvre, C., and Califano, A. (2007). Comparative analysis of microarray normalization procedures: effects on
reverse engineering gene networks. Bioinformatics 23, i282-288.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Ling, Q., Huang, W., Baldwin, A., and Jarvis, P. (2012). Chloroplast biogenesis is regulated by direct action of the ubiquitinproteasome system. Science 338, 655-659.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Lyon, D., Castillejo, M.A., Mehmeti-Tershani, V., Staudinger, C., Kleemaier, C., and Wienkoop, S. (2016). Drought and recovery:
independently regulated processes highlighting the importance of protein turnover dynamics and translational regulation in
Medicago truncatula. Mol Cell Proteomics. 15:1921-37
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Masclaux-Daubresse, C., Daniel-Vedele, F., Dechorgnat, J., Chardon, F., Gaufichon, L., and Suzuki, A. (2010). Nitrogen uptake,
assimilation and remobilization in plants: challenges for sustainable and productive agriculture. Ann Bot 105, 1141-1157.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Melis, A. (1999). Photosystem-II damage and repair cycle in chloroplasts: what modulates the rate of photodamage ? Trends Plant
Sci 4, 130-135.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Mockler, T.C., Michael, T.P., Priest, H.D., Shen, R., Sullivan, C.M., Givan, S.A., McEntee, C., Kay, S.A., and Chory, J. (2007). The
DIURNAL project: DIURNAL and circadian expression profiling, model-based pattern matching, and promoter analysis. Cold Spring
Harbor symposia on quantitative biology 72, 353-363.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Nelson, C.J., Li, L., Jacoby, R.P., and Millar, A.H. (2013). Degradation rate of mitochondrial proteins in Arabidopsis thaliana cells. J
Proteome Res 12, 3449-3459.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Nelson, C.J., Alexova, R., Jacoby, R.P., and Millar, A.H. (2014). Proteins with high turnover rate in barley leaves estimated by
proteome analysis combined with in planta isotope labeling. Plant Physiol 166, 91-108.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Nesvizhskii, A.I., Keller, A., Kolker, E., and Aebersold, R. (2003). A statistical model for identifying proteins by tandem mass
spectrometry. Anal Chem 75, 4646-4658.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Nishimura, K., Fukagawa, T., Takisawa, H., Kakimoto, T., and Kanemaki, M. (2009). An auxin-based degron system for the rapid
depletion of proteins in nonplant cells. Nat Methods 6, 917-922.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Nussbaumer, T., Martis, M.M., Roessner, S.K., Pfeifer, M., Bader, K.C., Sharma, S., Gundlach, H., and Spannagl, M. (2013). MIPS
PlantsDB: a database framework for comparative plant genome research. Nucleic acids research 41, D1144-1151.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Osteryoung, K.W., and Pyke, K.A. (2014). Division and dynamic morphology of plastids. Annu Rev Plant Biol 65, 443-472.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Peth, A., Nathan, J.A., and Goldberg, A.L. (2013). The ATP costs and time required to degrade ubiquitinated proteins by the 26 S
proteasome. J Biol Chem 288, 29215-29222.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Piques, M., Schulze, W.X., Hohne, M., Usadel, B., Gibon, Y., Rohwer, J., and Stitt, M. (2009). Ribosome and transcript copy
numbers, polysome occupancy and enzyme dynamics in Arabidopsis. Mol Syst Biol 5, 314.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Ponnala, L., Wang, Y., Sun, Q., and van Wijk, K.J. (2014). Correlation of mRNA and protein abundance in the developing maize leaf.
The Plant journal : for cell and molecular biology 78, 424-440.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Price, J.C., Guan, S., Burlingame, A., Prusiner, S.B., and Ghaemmaghami, S. (2010). Analysis of proteome dynamics in the mouse
brain. Proc Natl Acad Sci U S A 107, 14508-14513.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Reinhold, T., Alawady, A., Grimm, B., Beran, K.C., Jahns, P., Conrath, U., Bauer, J., Reiser, J., Melzer, M., Jeblick, W., and Neuhaus,
H.E. (2007). Limitation of nocturnal import of ATP into Arabidopsis chloroplasts leads to photooxidative damage. Plant J 50, 293304.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Ren, M., Qiu, S., Venglat, P., Xiang, D., Feng, L., Selvaraj, G., and Datla, R. (2011). Target of rapamycin regulates development and
ribosomal RNA expression through kinase domain in Arabidopsis. Plant Physiol 155, 1367-1382.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Rousseeuw, P.J. (1987). Silhouettes - a Graphical Aid to the Interpretation and Validation of Cluster-Analysis. Journal of
Computational and Applied Mathematics 20, 53-65.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Rowland, E., Kim, J., Bhuiyan, N.H., and van Wijk, K.J. (2015). The Arabidopsis Chloroplast Stromal N-Terminome: Complexities of
Amino-Terminal Protein Maturation and Stability. Plant Physiol 169, 1881-1896.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Scheurwater, I., Dunnebacke, M., Eising, R., and Lambers, H. (2000). Respiratory costs and rate of protein turnover in the roots of
a fast-growing (Dactylis glomerata L.) and a slow-growing (Festuca ovina L.) grass species. J Exp Bot 51, 1089-1097.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Schmitz, G., Reinhold, T., Gobel, C., Feussner, I., Neuhaus, H.E., and Conrath, U. (2010). Limitation of nocturnal ATP import into
chloroplasts seems to affect hormonal crosstalk, prime defense, and enhance disease resistance in Arabidopsis thaliana. Mol
Plant Microbe Interact 23, 1584-1591.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Schmitz, J., Heinrichs, L., Scossa, F., Fernie, A.R., Oelze, M.L., Dietz, K.J., Rothbart, M., Grimm, B., Flugge, U.I., and Hausler, R.E.
(2014). The essential role of sugar metabolism in the acclimation response of Arabidopsis thaliana to high light intensities. J Exp
Bot 65, 1619-1636.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Schwanhausser, B., Busse, D., Li, N., Dittmar, G., Schuchhardt, J., Wolf, J., Chen, W., and Selbach, M. (2011). Global quantification
of mammalian gene expression control. Nature 473, 337-342.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Sew, Y.S., Stroher, E., Holzmann, C., Huang, S., Taylor, N.L., Jordana, X., and Millar, A.H. (2013). Multiplex micro-respiratory
measurements of Arabidopsis tissues. New Phytol 200, 922-932.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Shteynberg, D., Deutsch, E.W., Lam, H., Eng, J.K., Sun, Z., Tasman, N., Mendoza, L., Moritz, R.L., Aebersold, R., and Nesvizhskii,
A.I. (2011). iProphet: multi-level integrative analysis of shotgun proteomic data improves peptide and protein identification rates
and error estimates. Molecular & cellular proteomics : MCP 10, M111 007690.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Siddell, S.G., and Ellis, R.J. (1975). Protein synthesis in chloroplasts. Characteristics and products of protein synthesis in vitro in
etioplasts and developing chloroplasts from pea leaves. Biochem J 146, 675-685.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Smith, S.M., Fulton, D.C., Chia, T., Thorneycroft, D., Chapple, A., Dunstan, H., Hylton, C., Zeeman, S.C., and Smith, A.M. (2004).
Diurnal changes in the transcriptome encoding enzymes of starch metabolism provide evidence for both transcriptional and
posttranscriptional regulation of starch metabolism in Arabidopsis leaves. Plant physiology 136, 2687-2699.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Suzuki, K., Nakanishi, H., Bower, J., Yoder, D.W., Osteryoung, K.W., and Miyagishima, S.Y. (2009). Plastid chaperonin proteins
Cpn60 alpha and Cpn60 beta are required for plastid division in Arabidopsis thaliana. BMC Plant Biol 9, 38.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Takahashi, S., and Badger, M.R. (2011). Photoprotection in plants: a new light on photosystem II damage. Trends Plant Sci 16, 5360.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Tanz, S.K., Castleden, I., Hooper, C.M., Vacher, M., Small, I., and Millar, H.A. (2013). SUBA3: a database for integrating
experimentation and prediction to define the SUBcellular location of proteins in Arabidopsis. Nucleic Acids Res 41, D1185-1191.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Umena, Y., Kawakami, K., Shen, J.R., and Kamiya, N. (2011). Crystal structure of oxygen-evolving photosystem II at a resolution of
1.9 A. Nature 473, 55-60.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
van Wijk, K.J. (2015). Protein maturation and proteolysis in plant plastids, mitochondria, and peroxisomes. Annu Rev Plant Biol 66,
75-111.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Vierstra, R.D. (2009). The ubiquitin-26S proteasome system at the nexus of plant biology. Nat Rev Mol Cell Biol 10, 385-397.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Vogtle, F.N., Wortelkamp, S., Zahedi, R.P., Becker, D., Leidhold, C., Gevaert, K., Kellermann, J., Voos, W., Sickmann, A., Pfanner, N.,
and Meisinger, C. (2009). Global Analysis of the Mitochondrial N-Proteome Identifies a Processing Peptidase Critical for Protein
Stability. Cell 139, 428-439.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Walsh, C.T. (1984). Suicide substrates, mechanism-based enzyme inactivators: recent developments. Annu Rev Biochem 53, 493535.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Wang, D., and Portis, A.R., Jr. (2007). A novel nucleus-encoded chloroplast protein, PIFI, is involved in NAD(P)H dehydrogenase
complex-mediated chlororespiratory electron transport in Arabidopsis. Plant Physiol 144, 1742-1752.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Wang, M., Weiss, M., Simonovic, M., Haertinger, G., Schrimpf, S.P., Hengartner, M.O., and von Mering, C. (2012). PaxDb, a database
of protein abundance averages across all three domains of life. Mol Cell Proteomics 11, 492-500.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Waters, M.T., Nelson, D.C., Scaffidi, A., Flematti, G.R., Sun, Y.K., Dixon, K.W., and Smith, S.M. (2012). Specialisation within the
DWARF14 protein family confers distinct responses to karrikins and strigolactones in Arabidopsis. Development 139, 1285-1295.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Wei, X., Su, X., Cao, P., Liu, X., Chang, W., Li, M., Zhang, X., and Liu, Z. (2016). Structure of spinach photosystem II-LHCII
supercomplex at 3.2 A resolution. Nature 534, 69-74.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Woodson, J.D., Joens, M.S., Sinson, A.B., Gilkerson, J., Salome, P.A., Weigel, D., Fitzpatrick, J.A., and Chory, J. (2015). Ubiquitin
facilitates a quality-control pathway that removes damaged chloroplasts. Science 350, 450-454.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Yang, X.Y., Chen, W.P., Rendahl, A.K., Hegeman, A.D., Gray, W.M., and Cohen, J.D. (2010). Measuring the turnover rates of
Arabidopsis proteins using deuterium oxide: an auxin signaling case study. Plant J 63, 680-695.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Yip, C.Y., Harbour, M.E., Jayawardena, K., Fearnley, I.M., and Sazanov, L.A. (2011). Evolution of respiratory complex I:
"supernumerary" subunits are present in the alpha-proteobacterial enzyme. J Biol Chem 286, 5023-5033.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Zaltsman, A., Feder, A., and Adam, Z. (2005). Developmental and light effects on the accumulation of FtsH protease in Arabidopsis
chloroplasts--implications for thylakoid formation and photosystem II maintenance. Plant J 42, 609-617.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Zhang, H., and Cramer, W.A. (2004). Purification and crystallization of the cytochrome b6f complex in oxygenic photosynthesis.
Methods Mol Biol 274, 67-78.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Zhang, H., Deery, M.J., Gannon, L., Powers, S.J., Lilley, K.S., and Theodoulou, F.L. (2015). Quantitative proteomics analysis of the
Arg/N-end rule pathway of targeted degradation in Arabidopsis roots. Proteomics 15, 2447-2457.
Pubmed: Author and Title
CrossRef: Author and Title
Google Scholar: Author Only Title Only Author and Title
Protein Degradation Rate in Arabidopsis thaliana Leaf Growth and Development
Lei Li, Clark J Nelson, Josua Trösch, Ian Castleden, Shaobai Huang and A. Harvey Millar
Plant Cell; originally published online January 30, 2017;
DOI 10.1105/tpc.16.00768
This information is current as of June 13, 2017
Supplemental Data
/content/suppl/2017/02/23/tpc.16.00768.DC2.html
/content/suppl/2017/01/30/tpc.16.00768.DC1.html
Permissions
https://www.copyright.com/ccc/openurl.do?sid=pd_hw1532298X&issn=1532298X&WT.mc_id=pd_hw1532298X
eTOCs
Sign up for eTOCs at:
http://www.plantcell.org/cgi/alerts/ctmain
CiteTrack Alerts
Sign up for CiteTrack Alerts at:
http://www.plantcell.org/cgi/alerts/ctmain
Subscription Information
Subscription Information for The Plant Cell and Plant Physiology is available at:
http://www.aspb.org/publications/subscriptions.cfm
© American Society of Plant Biologists
ADVANCING THE SCIENCE OF PLANT BIOLOGY