Subscriber access provided by ORTA DOGU TEKNIK UNIVERSITESI KUTUPHANESI
Article
Identifying regulatory changes to facilitate nitrogen fixation in the non-diazotroph Synechocystis sp. PCC 6803 Thomas J. Mueller, Eric A. Welsh, Himadri B. Pakrasi, and Costas D. Maranas ACS Synth. Biol., Just Accepted Manuscript • DOI: 10.1021/acssynbio.5b00202 • Publication Date (Web): 21 Dec 2015 Downloaded from http://pubs.acs.org on December 23, 2015
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
ACS Synthetic Biology is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 33
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 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Synthetic Biology
1
Research article
2
Identifying regulatory changes to facilitate nitrogen fixation in the non-
3
diazotroph Synechocystis sp. PCC 6803
4 5
Thomas J. Mueller1
6
Eric A. Welsh2
7
Himadri B. Pakrasi3,4
8
Costas D. Maranas1*
9 10
1Department
11
Park, Pennsylvania, USA
12
2Cancer
13
Tampa, Florida, USA
14
3Department
15
University, St. Louis, Missouri, USA
16
4Department
of Chemical Engineering, Pennsylvania State University, University
Informatics Core, H Lee Moffitt Cancer Center and Research Institute,
of Energy, Environmental, and Chemical Engineering, Washington
of Biology, Washington University, St. Louis, Missouri, USA
17 18
Correspondence: Costas D. Maranas, Department of Chemical Engineering,
19
Pennsylvania State University, 112 Fenske Laboratory, University Park, PA, 16802
20
Email:
[email protected] 21
Keywords:
22
Network, Modeling, Synechocystis, Cyanothece
23
Abbreviations: TF, transcription factor
Cyanobacteria,
Computational
Biotechnology,
ACS Paragon Plus Environment
Gene
Regulatory
ACS Synthetic Biology
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 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 2 of 33
24
Abstract
25
The incorporation of biological nitrogen fixation into a non-diazotrophic
26
photosynthetic organism provides a promising solution to the increasing fixed
27
nitrogen demand, but is accompanied by a number of challenges for accommodating
28
two incompatible processes within the same organism. Here we present regulatory
29
influence networks for two cyanobacteria, Synechocystis PCC 6803 and Cyanothece
30
ATCC 51142, and evaluate them to co-opt native transcription factors that may be
31
used to control the nif gene cluster once it is transferred to Synechocystis. These
32
networks were further examined to identify candidate transcription factors for
33
other metabolic processes necessary for temporal separation of photosynthesis and
34
nitrogen
35
transcription factors native to Synechocystis, LexA and Rcp1, were identified as
36
promising candidates for the control of the nif gene cluster and other pertinent
37
metabolic processes, respectively. Lessons learned in the incorporation of nitrogen
38
fixation into a non-diazotrophic prokaryote may be leveraged to further progress
39
the incorporation of nitrogen fixation in plants.
fixation,
glycogen
catabolism
and
cyanophycin
synthesis.
Two
40 41
1 Introduction
42
Global use of nitrogen fertilizer has seen a sevenfold increase from 1960 to 1995,
43
and further increases are predicted barring substantial increases in fertilizer
44
efficiency1. Increasing levels of biological nitrogen fixation can replace a portion of
45
the required fertilizer, decreasing nitrogen demands2, 3. Some plants, such as
46
legumes, have formed symbiotic relationships with rhizobium in order to acquire
ACS Paragon Plus Environment
Page 3 of 33
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 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Synthetic Biology
47
fixed nitrogen4, whereas other plants must acquire fixed nitrogen from the
48
environment5. The inclusion of nitrogen fixation in a photosynthetic organism
49
requires strict regulation of the oxygen levels within the organism, as nitrogenase is
50
irreversibly inhibited by oxygen6. Our goal in this paper is to identify how to provide
51
the regulatory structures to control the inherently incompatible processes of
52
photosynthesis and nitrogen fixation.
53 54
Cyanobacteria are photosynthetic prokaryotes that serve as a promising test system
55
in transferring nitrogen-fixing capabilities, as both diazotrophic and non-
56
diazotrophic species exist. Some cyanobacteria, such as Synechocystis PCC 6803
57
(hereafter referred to as Synechocystis), require fixed nitrogen from the
58
environment7 whereas others, such as Cyanothece ATCC 51142 (hereafter referred
59
to as Cyanothece) and Anabaena PCC 7120, are diazotrophic8, 9. In order to
60
accommodate
61
cyanobacteria separate processes either spatially or temporally6. Organisms such as
62
Anabeana use specialized cells called heterocysts to physically separate nitrogen
63
fixation from cells performing photosynthesis9. Others, such as Cyanothece,
64
temporally separate nitrogen fixation and photosynthesis8. Cyanothece generates
65
glycogen during the day and then consumes it through a burst of respiration in the
66
early dark hours, generating both the energy necessary for nitrogen fixation as well
67
as the requisite anaerobic environment10. Therefore, on a standard 12h/12h light-
68
dark cycle, the nitrogen fixation genes are highly overexpressed during the dark
69
period under anoxic conditions8.
both
nitrogen
fixation
and
photosynthesis,
ACS Paragon Plus Environment
diazotrophic
ACS Synthetic Biology
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 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
70 71
Synechocystis and Cyanothece have been extensively studied compared to other
72
cyanobacteria. When both organisms are grown under constant light at 30°C and
73
non-nitrogen fixing conditions, the doubling times vary from approximately 24
74
hours for Synechocystis11 to 45 hours for Cyanothece12. Synechocystis was the first
75
cyanobacterium to have its genome sequenced13 and has served as a genetically
76
tractable model organism for cyanobacteria14. A number of its regulatory genes have
77
been further investigated15-17 and the structure of segments of the regulatory
78
network has been proposed18. While Synechocystis is a more common target of
79
research, recent work has compiled both transcriptomic and proteomic data for
80
Cyanothece8, 19. In addition, there have been significant efforts spent elucidating the
81
metabolism of both Cyanothece and several closely related strains11, 12, 20, 21 and the
82
regulatory network of Cyanothece22. The regulatory network presented here was
83
generated in order to facilitate comparison between the two organisms. Given the
84
close relationship in Cyanothece between nitrogen fixation and its diurnal
85
oscillations, any comparisons between different organisms focusing on nitrogen
86
fixation must trace the light-dark cycle.
87 88
Synechocystis will be used as a proof of concept for the incorporation of temporal
89
separation of photosynthesis and nitrogen fixation in a non-diazotrophic organism.
90
Cyanothece is a logical source for the nitrogen fixation genes given its close
91
phylogenetic relationship to Synechocystis and the fact that it contains the largest
92
contiguous nif cluster in cyanobacteria10.
ACS Paragon Plus Environment
Page 4 of 33
Page 5 of 33
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 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Synthetic Biology
93 94
In order to identify candidate regulators for the processes necessary to incorporate
95
nitrogen fixation into Synechocystis, we generated, examined, and compared
96
regulatory networks for both Synechocystis and Cyanothece. We identified
97
transcription factors from both Synechocystis and Cyanothece that could be co-opted
98
to control these processes in Synechocystis. Two of these identified transcription
99
factors, LexA and Rcp1, are native to Synechocystis and have both expression
100
profiles and environmental responses that make them compelling candidates for
101
controlling the nif cluster and other pertinent metabolic processes, respectively.
102
Cyanobacteria have been used as a platform system for investigating CO2
103
concentrating mechanisms in plants23, 24. Therefore lessons learned in manipulating
104
regulatory structures in cyanobacteria could ultimately inform the incorporation of
105
nitrogen fixation in plants.
106 107
2 Results and Discussion
108
2.1 Comparison of peak expression timing
109
The regulatory networks for both Synechocystis PCC 6803 (Synechocystis) and
110
Cyanothece ATCC 51142 (Cyanothece) were generated by minimizing the extent of
111
experimental expression variance not explained for a fixed number of regulatory
112
influences. The number of influences for each gene cluster was chosen as the
113
minimum number of influences that would allow for over 50% of the expression
114
change to be explained. The Synechocystis network contains 45 gene clusters, 17
115
transcription factor (TF) clusters, and an average degree (i.e., TF clusters affecting a
ACS Paragon Plus Environment
ACS Synthetic Biology
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 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
116
gene cluster) of 5. The Cyanothece network contains 221 gene clusters, 16 TF
117
clusters, and has an average degree of 2.23 (Table 1). Parts of the networks are
118
shown in Figure 1. The regulatory influences of each TF cluster on each gene cluster
119
can be found in Supplemental File S1. This difference in average degree could be
120
caused in part by the differing methods used to process the original transcriptomic
121
data.
122 123
In order for a non-diazotrophic organism such as Synechocystis to temporally
124
separate nitrogen fixation and photosynthesis, a number of genes must have
125
expression profiles similar to those of their Cyanothece counterparts. The genes in
126
these two organisms that have differing expression profiles extends past those
127
involved in nitrogen fixation and its associated processes. Performing a bidirectional
128
BLAST search on the set of cycling Cyanothece genes identified 306 homolog pairs
129
between the two organisms where both genes were in the set of cycling genes. Only
130
114 of these pairs have elevated expression for both genes at the same segment of
131
the light-dark cycle. Another 77 peaked at different times during the same period
132
(e.g. a gene peaks in the early light and its homolog peaks in the late light). Figure 2
133
shows the distribution of the 115 homolog pairs that peak in different periods (i.e.
134
one homolog peaks in the dark period and the other in the light). Of the 14 homolog
135
pairs with differing peaks categorized in photosynthesis and respiration, six were
136
associated with respiratory electron flow and two with phycobilisomes25. These un-
137
matched peaks can be explained by the different temporal pattern in glycogen
138
consumption and use of phycobilisomes in the two organisms26. Given the number
ACS Paragon Plus Environment
Page 6 of 33
Page 7 of 33
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 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Synthetic Biology
139
of differences in expression across the genome, we focused only on those genes
140
directly associated with nitrogen fixation or the metabolic processes recruited
141
before or after nitrogen fixation.
142 143
2.2 Candidates for controlling nif cluster expression
144
Several different avenues are possible for imposing the necessary regulatory action
145
on genes involved in these processes. By using native transcription factors, the
146
number of genes that would need to be transferred is minimized and the
147
uncertainty in the expression profile of the transcription factor is reduced, as many
148
of its regulatory partners are already present. Candidate native TFs were identified
149
by determining the gene clusters in Synechocystis with the appropriate expression
150
profile for this process, and by investigating the TF clusters that were identified as
151
regulatory influences for a number of these gene clusters. Some of these identified
152
TFs have expression profiles similar to that of the final desired expression profile.
153
Controlling gene expression using a transcription factor with the same expression
154
profile is a concept underlying the development in gene circuits27. Gene circuits
155
developed using TFs have been designed to respond to environmental cues (i.e.,
156
oxygen level)28. The strategy outlined here, instead, aims to identify TFs native to
157
the organism that already respond to environmental conditions (e.g., low oxygen,
158
absence of light, etc.) that subsequently can be coopted to control the genes of
159
interest. Examining the Cyanothece regulatory network and the regulatory
160
influences predicted for the genes associated with the process identified additional
161
candidate regulators. Using these regulators would require the transfer of the
ACS Paragon Plus Environment
ACS Synthetic Biology
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 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
162
regulatory genes to Synechocystis but reduce the need to modify promoter regions
163
for the genes transferred from Cyanothece.
164 165
The expression profile necessary for the transferred nif genes is characterized by an
166
increase in expression during the dark and a rapid decrease in expression at the
167
beginning of illumination. The desired profile can be seen in gene clusters 4, 10, 16,
168
17, 18, 19, and 35 of Synechocystis. There are four TF clusters (10, 15, 16, and 17)
169
that are regulatory influences for at least three of these gene clusters. Each TF
170
cluster contains a single gene that has been shown to respond to environmental
171
conditions pertinent to nitrogen fixation. TF cluster 10 contains gene sll1689, which
172
codes for the group 2 sigma factor sigE. Both the transcript levels of sigE and the
173
protein levels of the associated SigE protein increase under nitrogen depletion29.
174
Studies have shown that sigE expression increases in an NtcA dependent manner29.
175
NtcA, a global nitrogen regulator in cyanobacteria30, responds to intracellular 2-
176
oxoglutarate levels and regulates a number of genes involved in nitrogen
177
assimilation31. In Cyanothece ATCC 51142 (Cyanothece), it has been shown that ntcA
178
expression is out of phase with that of the nif genes32, and has been suggested that
179
ntcA is involved primarily in nitrogen assimilation than in nitrogen fixation33.
180
Nevertheless, both ntcA and sigE do not account for the presence of oxygen, a
181
nitrogenase inhibitor, through detection of either light or intracellular oxygen levels.
182
Therefore, additional synthetic regulation is needed to account for such factors.
183
ACS Paragon Plus Environment
Page 8 of 33
Page 9 of 33
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 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Synthetic Biology
184
The remaining three TF clusters contain TF genes that respond to these
185
environmental conditions. TF cluster 17 contains the hik16 (slr1805) gene. While
186
hik16 is most commonly associated with the salt stress response34, it also responds
187
to oxidative stress induced by hydrogen peroxide35. The decreased expression of
188
hik16 during the day may be explained by the oxidative stress that can be caused by
189
reactive oxygen species generated by photosynthesis36. Given that it responds to
190
additional stimuli, namely salt stress, hik16 would not be an ideal native TF for use
191
in nif cluster regulation. Both TF clusters 15 and 16 contain genes (lexA and rre34
192
respectively) that respond to oxygen levels within the environment. In
193
Synechocystis, LexA binds to two different sites on the hydrogenase operon and is
194
also known to regulate the sbtA gene, which codes for a sodium-bicarbonate
195
symporter37. Both of these regulatory influences are predicted in the Synechocystis
196
network (Figure 1A). LexA is thought to function as a transcriptional activator16 and
197
is upregulated after the transition to a low O2 environment38. Similarly, rre34
198
(sll0789) was shown to be upregulated under low O2 conditions38. Given the
199
irreversible inhibition of nitrogenase by oxygen, a transcriptional activator that is
200
upregulated under low O2 conditions is a promising candidate for use in regulating
201
the nif cluster in Synechocystis. When comparing the expression profiles of the two
202
candidates in Synechocystis with that of the nif cluster in Cyanothece (Figure 3) we
203
see that lexA peaks at the same time as the nif cluster as compared to the later peak
204
in expression for rre34. This places lexA as the top candidate for regulating the nif
205
cluster in a diazotrophic Synechocystis.
206
ACS Paragon Plus Environment
ACS Synthetic Biology
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 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
207
The Cyanothece regulatory network identifies a number of transcription factors
208
involved in the regulation of genes in the nif cluster. 33 genes in the nif cluster were
209
identified as cycling and were contained within 13 different gene clusters
210
(Supplementary File S2). Many of these gene clusters have different sets of
211
regulatory influences although several TF clusters are prominent (Supplementary
212
File S3). 27 of the 33 genes have either TF cluster 4 or TF cluster 15 as a regulatory
213
influence. TF cluster 4 contains three genes, one of which is cce_1898, annotated as
214
patB. In the previous regulatory network of Cyanothece, PatB was predicted to
215
influence the expression of the nitrogenase complex22. PatB was also shown to be
216
required for diazotrophic growth, and is expressed in the heterocysts of Anabaena
217
PCC 712039. TF cluster 15 is composed of only one gene, sigE. SigE has been shown
218
to be upregulated in both Synechocystis and Anabeana PCC 7120 under nitrogen
219
depletion9,
220
influences 8 genes. The deletion of sigD in Anabeana impairs the ability of the
221
organism to establish diazotrophic growth40 (Figure 1B). Given the prediction that
222
PatB controls nif cluster expression, the introduction of patB into Synechocystis and
223
the use of LexA to control its expression is another possible approach to controlling
224
the nif cluster expression in Synechocystis (Figure 4A). By using the native nif
225
cluster regulator, this strategy circumvents the need to modify all promoter regions
226
in the nif cluster individually and also retains the native control of the nif cluster by
227
PatB.
29.
TF cluster 10, another prominent regulator, contains sigD and
228 229
2.3 Candidates for regulating other pertinent metabolic processes
ACS Paragon Plus Environment
Page 10 of 33
Page 11 of 33
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 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Synthetic Biology
230
In addition to nitrogen fixation two other processes, glycogen catabolism and
231
cyanophycin synthesis, must be active at specific times for the diazotrophic
232
Synechocystis to create the requisite anaerobic environment and to assimilate and
233
store newly fixed nitrogen. In Cyanothece the anaerobic environment and energy
234
necessary for nitrogen fixation are generated through the consumption of glycogen
235
in the early dark period8. Glycogen catabolism in Synechocystis is a metabolic
236
process that must be placed under different regulatory control. There is a pair of
237
homologs (cce_1629 and slr1367) within the two organisms that are annotated as
238
glycogen phosphorylases and identified as cycling. The current expression profile in
239
Synechocystis includes a peak in expression one hour before the onset of light, and a
240
negative log2 ratio one hour into the dark period, when its homolog peaks in
241
Cyanothece (Figure 5A). This is in contrast to the glycogen synthase genes, which
242
peak in the beginning of the light period for both organisms, and does not require
243
any modified regulation. Using the steps discussed to identify native transcription
244
factors for the nif cluster, two gene clusters, 16 and 34, were found to have the
245
desired profile; low expression during the light and peak expression in the early
246
dark. These two clusters share four regulatory influences, the most significant of
247
these is TF cluster 12, which is composed of the rcp1 (slr0474) gene. The Rcp1
248
protein is part of a two-component light sensing system with the phytochrome
249
Cph141. Garcia-Dominguez et al. showed that rcp1 is maximally expressed 15
250
minutes into the dark period, and its expression levels are almost undetectable five
251
minutes after illumination42. Given the role that glycogen catabolism plays in
252
consuming oxygen within the cell, its associated genes should be expressed prior to
ACS Paragon Plus Environment
ACS Synthetic Biology
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 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
253
those for nitrogen fixation, making rcp1 an ideal candidate for controlling glycogen
254
phosphorylase. The Cyanothece network identifies the regulatory influences for
255
glycogen phosphorylase as a hypothetical protein (cce_4141) and sigD, which is
256
involved in establishing diazotrophic growth in Anabeana40.
257 258
Cyanophycin is a dynamic nitrogen reserve composed of equimolar quantities of
259
arginine and aspartic acid26. Cyanophycin accumulation in Cyanothece follows the
260
fixation of nitrogen, being mainly generated between two and six hours after the
261
beginning of the dark period26. In contrast, cyanophycin has a minor role in
262
Synechocystis, which, instead, degrades phycobilisomes, leading to cell bleaching26.
263
Therefore, arginine, aspartic acid, and cyanophycin synthesis in Synechocystis
264
should be expressed in a manner similar to that of their counterparts in Cyanothece.
265
The genes involved in arginine and aspartic acid synthesis were identified through
266
the gene-protein-reaction relationships of the reactions that carry flux at maximum
267
biomass production for the genome-scale models iSyn731 and iCyt773 (Figure
268
5C)11. For each reaction, the elevated expression time for each associated gene was
269
compared. Several of the gene pairs shared elevated expression times, although four
270
genes in Synechocystis (sll0573, slr0585, slr1133, and sll0902) either were not
271
cycling, or had elevated expression during the beginning of the light period.
272
Conversely, their Cyanothece counterparts exhibited elevated expression during the
273
beginning of the dark period (Figure 5B). Cyanophycin synthetase catalyzes the
274
elongation reaction of cyanophycin through the incorporation of arginine and
275
aspartic acid43. The gene that codes for this enzyme in Synechocystis, slr2002, is not
ACS Paragon Plus Environment
Page 12 of 33
Page 13 of 33
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 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Synthetic Biology
276
cycling and therefore will need to be controlled by a different TF. The cyanophycin
277
synthetase gene in Cyanothece, cce_2237, peaks in expression at the beginning of the
278
dark period, with above average expression continuing through the middle of the
279
dark period. Given the similar timing of necessary expression, the native TF
280
identified as candidates for nif cluster and glycogen catabolism regulation, namely,
281
LexA and Rcp1, serve equally well as candidates for control of these genes. Two TF
282
clusters, 3 and 13, regulate the six clusters containing these genes in Cyanothece. TF
283
cluster 13 is composed of a FUR family transcription factor, and TF cluster 3
284
contains cce_1982, which is the homolog of rcp1 in Synechocystis, strengthening its
285
candidacy for regulation of cyanophycin synthesis (Figure 4B).
286 287
This work focuses on the transcriptional regulation of the nif genes and the genes
288
associated with several other key processes. Future work that considers allosteric
289
regulation will be able to provide a more comprehensive set of possible
290
modifications that allow for the incorporation of processes such as nitrogen fixation.
291
The inclusion of additional forms of regulation would capture processes such as the
292
allosteric regulation of NtcA activity through binding with 2-oxoglutarate44.
293 294
In this paper we elucidated regulatory influence networks for two cyanobacteria,
295
Synechocystis
296
transcriptomic data taken during 12h/12h light-dark cycles. These networks were
297
inspected to identify what regulation would need to be introduced or modified to
298
incorporate nitrogen fixation into the non-diazotrophic Synechocystis. One native
PCC
6803
and
Cyanothece
ATCC
ACS Paragon Plus Environment
51142,
generated
using
ACS Synthetic Biology
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 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 14 of 33
299
transcription factor, LexA, arose as a candidate for regulating nif cluster expression,
300
while PatB was identified as a promising candidate to be a regulator of nif cluster
301
genes in Cyanothece that could be transferred to Synechocystis. Using these
302
predictions together, we postulate that transferring patB to Synechocystis and
303
controlling its expression with LexA would lead to the desired nif cluster expression
304
within Synechocystis. Other processes with required expression changes dictated by
305
the temporal separation of photosynthesis and nitrogen fixation, specifically
306
glycogen catabolism and cyanophycin anabolism, were analyzed. The native Rcp1
307
transcription factor arose as an additional candidate for the regulation of these
308
processes within a newly diazotrophic Synechocystis. Given the expression profiles
309
and response to environmental conditions, LexA and Rcp1 are transcription factors
310
native to Synechocystis that should be considered for use as regulatory control of the
311
nif cluster and the associated metabolic processes, respectively.
312
3 Materials and methods
313
3.1 Cycling and homologous gene identification
314
Regulatory influence networks for both organisms were generated using time series
315
transcriptomic data over a standard 12h light-dark cycle
316
51142 (Cyanothece) data from Stockel et al.8 was processed using the IRON software
317
tool46. The log2 ratio of expression levels at each time point to the pooled samples
318
was then calculated for each gene and time point. The criteria from Kucho et al.47
319
were used to identify cycling genes in the Synechocystis PCC 6803 (Synechocystis)
320
data set45 and applied to the Cyanothece data set as well. The Synechocystis data
321
used was not normalized in the range of -1 to +1 in order to be consistent with the
ACS Paragon Plus Environment
8, 45.
The Cyanothece ATCC
Page 15 of 33
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 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Synthetic Biology
322
Cyanothece data from Stockel et al.8. The nif genes qualified as cycling using two of
323
the three criteria from Kucho et al. (i.e., appropriate period length and statistically
324
significant differences between peak and trough expression each day) but the values
325
of the error function described in Kucho et al. were above the specified cutoff. Given
326
the importance of including these genes within the network for identifying
327
regulation on the nif cluster, only the period length and statistically significant
328
differences in peak and trough expression were used to identify cycling genes. This
329
resulted in a set of 1,552 genes identified as cycling in Cyanothece as compared to
330
the 1,423 cycling genes in Synechocystis (Supplementary File S4).
331 332
To identify homolog pairs a bidirectional BLAST search between the two organisms
333
was performed using an e value cutoff of 10-2. A result was determined to be a hit if
334
the alignment length was at least 75% of the length of both the query and target
335
genes, and if the raw score was at least 25% of the self-hit score of the query gene. A
336
homolog pair was identified if both genes had only each other as a hit. A gene was
337
defined as having elevated expression if the log2 ratio at a time point was at least
338
90% of the maximum log2 ratio for that gene. To compare when peak expression
339
times for homolog pairs, elevated expression was then categorized as occurring in
340
the beginning, middle, or final third of the light or dark period.
341 342
3.2 Cluster generation
343
In order to simplify the system before inferring regulatory influences, hierarchical
344
clustering using Euclidean distances was implemented for both organisms. First the
ACS Paragon Plus Environment
ACS Synthetic Biology
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 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
345
optimal agglomeration method available in R was identified for each organism using
346
the cophenetic correlation48. For both organisms the unweighted pair group method
347
with arithmetic mean (UPGMA)49 method generated the dendrogram whose
348
cophenetic distances most closely correlated with the previously calculated
349
Euclidean distances. The SD validity index50 was then used to identify the optimal
350
height cutoff for both dendrograms, generating 45 gene clusters for Synechocystis
351
and 221 gene clusters for Cyanothece.
352 353
The transcription factor (TF) clusters were generated in a similar manner. Genes
354
with a regulatory function were first identified using literature sources. Of the 146
355
regulatory genes identified by Singh et al in Synechocystis51, 76 are present in the set
356
of cycling genes of Saha et al.45. The cycling Cyanothece gene set contained 72 genes
357
that were either annotated as having a regulatory function by Stockel et al.8 or were
358
identified as regulators by McDermott et al.22. Dendrograms for these gene sets
359
were then generated using the same UPGMA agglomeration method. The SD validity
360
index was used to identify the optimal height cutoff for Cyanothece, and these 72
361
genes were clustered into 16 different TF clusters. The identified optimal height
362
cutoff for the Synechocystis transcription factors generated only six TF clusters. The
363
subsequently generated network did not adequately capture the expression of the
364
gene clusters, as the percentage of experimental expression change explained by the
365
network (see Network Formulation) ranged from 8.1-56.4% at maximum regulatory
366
influences. A lower sampled height cutoff yielded 17 TF clusters, and was chosen to
367
give a comparable number of TF clusters for the two networks. The members of
ACS Paragon Plus Environment
Page 16 of 33
Page 17 of 33
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 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Synthetic Biology
368
each cluster for both organisms are listed in Supplemental File S5. All clustering and
369
cluster analysis was performed using the R language along with the cluster and
370
clusterCrit packages52-54
371 372
3.3 Network elucidation using optimization formulation
373
The network of regulatory influences for each organism was generated by
374
iteratively determining for each gene cluster the maximum amount of experimental
375
expression change that can be accounted for a given number of regulatory
376
influences. J, K, and T denote the total number of gene clusters, TF clusters, and time
377
points, respectively. The experimental expression level of gene cluster j at time
378
point t is represented as Xjt, and the experimental expression level of TF cluster k at
379
time point t is denoted as TFkt. Ckj is the interaction coefficient that describes the
380
influence of TF cluster k on gene cluster j.
381 382
Gene expression was assumed to be a linear additive contribution of the set of
383
influencing TFs. This can be postulated for each gene cluster j and time point t as
384
shown in equation 1.
385
= ∀ ∈ & ∀ ∈
(1)
386
Backward finite difference was used to approximate the rate of expression term (i.e.,
387
left-hand side) of equation 1 in equation 3 in the mixed integer linear program
388
formulation shown below. Solving this formulation identifies the interaction
ACS Paragon Plus Environment
ACS Synthetic Biology
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 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
389
coefficients (Ckj) that minimize the amount of change in gene cluster expression not
390
explained by the regulatory influences. The formulation was solved iteratively for
391
every gene cluster and a range of maximum regulatory influences.
E + (2) D D D s. t. D (3)D ∀ ∈ − − ' = − ∀ ∈ {1, … , − 1} D ≤ ≤ ./ ∀1 ∈ 2 (4) −./ D D / ≤ max regulatory influences (5)D C 392
The slack variables and represent the positive and negative deviations from
393
experimental measurements. Equations 4 and 5 use the binary variables, ykj, to
394
restrict the number of regulatory influences per gene cluster. If a transcription
395
factor in TF cluster k is in gene cluster j then a regulatory interaction will be
396
imposed for that TF cluster gene cluster pair by fixing ykj to one for that pair.
397 398
In order to determine the number of regulatory influences per cluster, the
399
percentage of experimental expression change accounted for by the network was
400
calculated for the range from one to eleven regulatory influences for both
401
organisms. For a given number of regulatory influences this percentage of explained
402
expression variance is defined as the difference between the error (i.e. the sum of
403
the slack variables) with zero regulatory influences and the given number of
404
regulatory influences, divided by the error with zero influences. The number of
405
regulatory influences was chosen for each cluster as the minimum number of
ACS Paragon Plus Environment
Page 18 of 33
Page 19 of 33
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 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Synthetic Biology
406
influences that brought the percentage of explained expression change above 50%.
407
The percentage of explained expression change for each cluster at every number of
408
regulatory influences can be found in Supplemental File S6.
409
Author Contributions
410
TJM created and analyzed regulatory networks. EAW processed Cyanothece data.
411
TJM and CDM wrote the paper. All authors read and approved the final manuscript.
412
Acknowledgements
413
We thank Fuzhong Zhang and Tae Seok Moon for discussions during the writing of
414
the manuscript. Supported by funding from the National Science Foundation, grant
415
MCB-1331194.
416
Conflict of Interest
417
The authors declare that they have no conflict of interest.
ACS Paragon Plus Environment
ACS Synthetic Biology
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 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
418
Tables
419
Table 1: Statistics for both regulatory networks
Page 20 of 33
Synechocystis
Cyanothece
Genes
1,423
1,552
Gene clusters
45
221
TFs
76
72
TF clusters
17
16
Average degree
5
2.23
Average % explained
54.25%
60.70%
expression change 420 421
ACS Paragon Plus Environment
Page 21 of 33
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 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
422
ACS Synthetic Biology
Figure Captions
423 424
Figure 1: Parts of the regulatory networks for A) Synechocystis and B) Cyanothece.
425
Edges are colored based on whether the predicted interaction is activating (green)
426
or inhibiting (red). A subset of the gene clusters containing nif genes is shown
427
above, all regulatory influences on the nif genes are listed in Supplemental File S2.
428
Dashed lines indicate edges with associated literature support discussed in the
429
Results section.
430 431
Figure 2: Distribution of homolog pairs in which one peaks in the light period and
432
one peaks in the dark between the different subsystems. Subsystem annotations
433
from Saha et al. were used45. *“DNA replication and other processes” refers to the
434
subsystem “DNA replication, restriction, modification, recombination, and repair”.
435 436
Figure 3: Comparison of candidate native transcription factor expression profiles in
437
Synechocystis (lexA: black line, rre34: cyan line) with those of the nif cluster in
438
Cyanothece (dashed lines). The dark periods of the light-dark cycle (hours 0-12 and
439
24-36) are represented with gray shading.
440 441
Figure 4: All proposed regulatory changes. Genes native to Synechocystis are in
442
green; those from Cyanothece are in orange. Proposed regulatory changes are
443
represented as dashed lines. A) Proposed strategy for control of the nif cluster in
444
Synechocystis using lexA and patB. B) Strategies for control of glycogen catabolism
ACS Paragon Plus Environment
ACS Synthetic Biology
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 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
445
(horizontal stripes) and cyanophycin synthesis (vertical stripes) genes in
446
Synechocystis using rcp1.
447 448
Figure 5: Metabolic processes requiring modified regulation in a diazotrophic
449
Synechocystis. A) Expression profiles of rcp1 (solid black line) and glycogen
450
phosphorylase (solid brown line) in Synechocystis, and glycogen phosphorylase
451
(dashed brown line) in Cyanothece. B) Expression profiles of rcp1 (solid black line)
452
and cycling cyanophycin synthesis genes (solid lines, colors corresponding to C) in
453
Synechocystis, and the expression profiles of the cycling cyanophycin synthesis
454
genes in Cyanothece (dashed lines). Sll0573 and slr2002 are not shown, as they were
455
not identified as cycling. The dark periods of the light-dark cycle (hours 0-12 and
456
24-36) are represented with gray shading. C) Reactions and genes identified by the
457
iSyn731 and iCyt773 genome-scale models as being involved in cyanophycin
458
synthesis. Reactions with associated genes whose elevated expression times differ
459
between the two organisms are color-coded.
460
ACS Paragon Plus Environment
Page 22 of 33
Page 23 of 33
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 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Synthetic Biology
461
Supporting Information
462
S1 (.xlsx): Regulatory influences for each gene cluster for both networks
463
S2 (.docx): Annotation, gene cluster, and regulatory influences for the cycling genes
464
within the nif cluster.
465
S3 (.docx): Venn Diagram depicting those transcription factors that are influences
466
for nif genes and have cycling homologs in Synechocystis.
467
S4 (.xlsx): Transcriptomic log2 ratios of cycling Cyanothece genes and cycling
468
Synechocystis genes
469
S5 (.xlsx): Members of each gene and TF cluster for Synechocystis and Cyanothece
470
S6 (.xlsx): Percentage of explained expression change for each gene cluster for 1-10
471
TF influences
472 473
References
474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493
[1] Tilman, D., Cassman, K. G., Matson, P. A., Naylor, R., and Polasky, S. (2002) Agricultural sustainability and intensive production practices, Nature 418, 671-677. [2] Peoples, M. B., Herridge, D. F., and Ladha, J. K. (1995) Biological NitrogenFixation - an Efficient Source of Nitrogen for Sustainable Agricultural Production, Plant Soil 174, 3-28. [3] Dobereiner, J. (1997) Biological nitrogen fixation in the tropics: Social and economic contributions, Soil Biol Biochem 29, 771-774. [4] Gage, D. J. (2004) Infection and invasion of roots by symbiotic, nitrogen-fixing rhizobia during nodulation of temperate legumes, Microbiol Mol Biol Rev 68, 280-300. [5] Kraiser, T., Gras, D. E., Gutierrez, A. G., Gonzalez, B., and Gutierrez, R. A. (2011) A holistic view of nitrogen acquisition in plants, J Exp Bot 62, 1455-1466. [6] Berman-Frank, I., Lundgren, P., and Falkowski, P. (2003) Nitrogen fixation and photosynthetic oxygen evolution in cyanobacteria, Res Microbiol 154, 157164. [7] Herrero, A., Muro-Pastor, A. M., and Flores, E. (2001) Nitrogen control in cyanobacteria, J Bacteriol 183, 411-425. [8] Stockel, J., Welsh, E. A., Liberton, M., Kunnvakkam, R., Aurora, R., and Pakrasi, H. B. (2008) Global transcriptomic analysis of Cyanothece 51142 reveals robust
ACS Paragon Plus Environment
ACS Synthetic Biology
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 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538
diurnal oscillation of central metabolic processes, P Natl Acad Sci USA 105, 6156-6161. [9] Kumar, K., Mella-Herrera, R. A., and Golden, J. W. (2010) Cyanobacterial Heterocysts, Csh Perspect Biol 2. [10] Welsh, E. A., Liberton, M., Stoeckel, J., Loh, T., Elvitigala, T., Wang, C., Wollam, A., Fulton, R. S., Clifton, S. W., Jacobs, J. M., Aurora, R., Ghosh, B. K., Sherman, L. A., Smith, R. D., Wilson, R. K., and Pakrasi, H. B. (2008) The genome of Cyanothece 51142, a unicellular diazotrophic cyanobacterium important in the marine nitrogen cycle, P Natl Acad Sci USA 105, 15094-15099. [11] Saha, R., Verseput, A. T., Berla, B. M., Mueller, T. J., Pakrasi, H. B., and Maranas, C. D. (2012) Reconstruction and Comparison of the Metabolic Potential of Cyanobacteria Cyanothece sp ATCC 51142 and Synechocystis sp PCC 6803, Plos One 7. [12] Feng, X. Y., Bandyopadhyay, A., Berla, B., Page, L., Wu, B., Pakrasi, H. B., and Tang, Y. J. (2010) Mixotrophic and photoheterotrophic metabolism in Cyanothece sp ATCC 51142 under continuous light, Microbiol-Sgm 156, 25662574. [13] Kaneko, T., Sato, S., Kotani, H., Tanaka, A., Asamizu, E., Nakamura, Y., Miyajima, N., Hirosawa, M., Sugiura, M., Sasamoto, S., Kimura, T., Hosouchi, T., Matsuno, A., Muraki, A., Nakazaki, N., Naruo, K., Okumura, S., Shimpo, S., Takeuchi, C., Wada, T., Watanabe, A., Yamada, M., Yasuda, M., and Tabata, S. (1996) Sequence analysis of the genome of the unicellular cyanobacterium Synechocystis sp. strain PCC6803. II. Sequence determination of the entire genome and assignment of potential protein-coding regions, DNA Res 3, 109136. [14] Mitschke, J., Georg, J., Scholz, I., Sharma, C. M., Dienst, D., Bantscheff, J., Voss, B., Steglich, C., Wilde, A., Vogel, J., and Hess, W. R. (2011) An experimentally anchored map of transcriptional start sites in the model cyanobacterium Synechocystis sp. PCC6803, Proc Natl Acad Sci U S A 108, 2124-2129. [15] Garcia-Dominguez, M., Reyes, J. C., and Florencio, F. J. (2000) NtcA represses transcription of gifA and gifB, genes that encode inhibitors of glutamine synthetase type I from Synechocystis sp PCC 6803, Mol Microbiol 35, 11921201. [16] Gutekunst, K., Phunpruch, S., Schwarz, C., Schuchardt, S., Schulz-Friedrich, R., and Appel, J. (2005) LexA regulates the bidirectional hydrogenase in the cyanobacterium Synechocystis sp PCC 6803 as a transcription activator, Mol Microbiol 58, 810-823. [17] Li, H., and Sherman, L. A. (2000) A redox-responsive regulator of photosynthesis gene expression in the cyanobacterium Synechocystis sp strain PCC 6803, J Bacteriol 182, 4268-4277. [18] Lemeille, S., Latifi, A., and Geiselmann, J. (2005) Inferring the connectivity of a regulatory network from mRNA quantification in Synechocystis PCC6803, Nucleic Acids Res 33, 3381-3389. [19] Aryal, U. K., Stockel, J., Krovvidi, R. K., Gritsenko, M. A., Monroe, M. E., Moore, R. J., Koppenaal, D. W., Smith, R. D., Pakrasi, H. B., and Jacobs, J. M. (2011)
ACS Paragon Plus Environment
Page 24 of 33
Page 25 of 33
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 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584
ACS Synthetic Biology
Dynamic proteomic profiling of a unicellular cyanobacterium Cyanothece ATCC51142 across light-dark diurnal cycles, Bmc Syst Biol 5. [20] Mueller, T. J., Berla, B. M., Pakrasi, H. B., and Maranas, C. D. (2013) Rapid construction of metabolic models for a family of Cyanobacteria using a multiple source annotation workflow, Bmc Syst Biol 7. [21] Bandyopadhyay, A., Elvitigala, T., Welsh, E., Stockel, J., Liberton, M., Min, H. T., Sherman, L. A., and Pakrasi, H. B. (2011) Novel Metabolic Attributes of the Genus Cyanothece, Comprising a Group of Unicellular Nitrogen-Fixing Cyanobacteria, Mbio 2. [22] McDermott, J. E., Oehmen, C. S., McCue, L. A., Hill, E., Choi, D. M., Stockel, J., Liberton, M., Pakrasi, H. B., and Sherman, L. A. (2011) A model of cyclic transcriptomic behavior in the cyanobacterium Cyanothece sp ATCC 51142, Mol Biosyst 7, 2407-2418. [23] Price, G. D., Badger, M. R., Woodger, F. J., and Long, B. M. (2008) Advances in understanding the cyanobacterial CO2-concentrating-mechanism (CCM): functional components, Ci transporters, diversity, genetic regulation and prospects for engineering into plants, J Exp Bot 59, 1441-1461. [24] Lieman-Hurwitz, J., Rachmilevitch, S., Mittler, R., Marcus, Y., and Kaplan, A. (2003) Enhanced photosynthesis and growth of transgenic plants that express ictB, a gene involved in HCO3- accumulation in cyanobacteria, Plant Biotechnol J 1, 43-50. [25] Vermaas, W. F. J. (2001) Photosynthesis and Respiration in Cyanobacteria, In eLS, John Wiley & Sons, Ltd. [26] Li, H., Sherman, D. M., Bao, S. L., and Sherman, L. A. (2001) Pattern of cyanophycin accumulation in nitrogen-fixing and non-nitrogen-fixing cyanobacteria, Arch Microbiol 176, 9-18. [27] Brophy, J. A. N., and Voigt, C. A. (2014) Principles of genetic circuit design, Nat Methods 11, 508-520. [28] Immethun, C. M., Ng, K. M., DeLorenzo, D. M., Waldron-Feinstein, B., Lee, Y. C., and Moon, T. S. (2015) Oxygen-responsive Genetic Circuits Constructed in Synechocystis sp. PCC 6803, Biotechnol Bioeng. [29] Osanai, T., Ikeuchi, M., and Tanaka, K. (2008) Group 2 sigma factors in cyanobacteria, Physiol Plant 133, 490-506. [30] Vega-Palas, M. A., Flores, E., and Herrero, A. (1992) NtcA, a global nitrogen regulator from the cyanobacterium Synechococcus that belongs to the Crp family of bacterial regulators, Mol Microbiol 6, 1853-1859. [31] Liu, D., and Yang, C. (2014) The nitrogen-regulated response regulator NrrA controls cyanophycin synthesis and glycogen catabolism in the cyanobacterium Synechocystis sp. PCC 6803, J Biol Chem 289, 2055-2071. [32] Gaudana, S. B., Krishnakumar, S., Alagesan, S., Digmurti, M. G., Viswanathan, G. A., Chetty, M., and Wangikar, P. P. (2013) Rhythmic and sustained oscillations in metabolism and gene expression of Cyanothece sp ATCC 51142 under constant light, Front Microbiol 4. [33] Bradley, R. L., and Reddy, K. J. (1997) Cloning, sequencing, and regulation of the global nitrogen regulator gene ntcA in the unicellular diazotrophic cyanobacterium Cyanothece sp. strain BH68K, J Bacteriol 179, 4407-4410.
ACS Paragon Plus Environment
ACS Synthetic Biology
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 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629
[34] Marin, K., Suzuki, L., Yamaguchi, K., Ribbeck, K., Yamamoto, H., Kanesaki, Y., Hagemann, M., and Murata, N. (2003) Identification of histidine kinases that act as sensors in the perception of salt stress in Synechocystis sp PCC 6803, P Natl Acad Sci USA 100, 9061-9066. [35] Kanesaki, Y., Yamamoto, H., Paithoonrangsarid, K., Shoumskaya, M., Suzuki, I., Hayashi, H., and Murata, N. (2007) Histidine kinases play important roles in the perception and signal transduction of hydrogen peroxide in the cyanobacterium, Synechocystis sp PCC 6803, Plant J 49, 313-324. [36] Niyogi, K. K. (1999) Photoprotection revisited: Genetic and molecular approaches, Annu Rev Plant Phys 50, 333-359. [37] Oliveira, P., and Lindblad, P. (2009) Transcriptional regulation of the cyanobacterial bidirectional Hox-hydrogenase, Dalton T, 9990-9996. [38] Summerfield, T. C., Nagarajan, S., and Sherman, L. A. (2011) Gene expression under low-oxygen conditions in the cyanobacterium Synechocystis sp. PCC 6803 demonstrates Hik31-dependent and -independent responses, Microbiol-Sgm 157, 301-312. [39] Jones, K. M., Buikema, W. J., and Haselkorn, R. (2003) Heterocyst-specific expression of patB, a gene required for nitrogen fixation in Anabaena sp strain PCC 7120, J Bacteriol 185, 2306-2314. [40] Khudyakov, I. Y., and Golden, J. W. (2001) Identification and inactivation of three group 2 sigma factor genes in Anabaena sp strain PCC 7120, J Bacteriol 183, 6667-6675. [41] Yeh, K. C., Wu, S. H., Murphy, J. T., and Lagarias, J. C. (1997) A cyanobacterial phytochrome two-component light sensory system, Science 277, 1505-1508. [42] Garcia-Dominguez, M., Muro-Pastor, M. I., Reyes, J. C., and Florencio, F. J. (2000) Light-dependent regulation of cyanobacterial phytochrome expression, J Bacteriol 182, 38-44. [43] Berg, H., Ziegler, K., Piotukh, K., Baier, K., Lockau, W., and Volkmer-Engert, R. (2000) Biosynthesis of the cyanobacterial reserve polymer multi-L-arginylpoly-L-aspartic acid (cyanophycin) - Mechanism of the cyanophycin synthetase reaction studied with synthetic primers, Eur J Biochem 267, 55615570. [44] Flores, E., and Herrero, A. (2005) Nitrogen assimilation and nitrogen control in cyanobacteria, Biochem Soc T 33, 164-167. [45] Saha, R., Liu, D., Hoynes-O'Connor, A., Liberton, M., Yu, J., Bhattacharyya, M., Balassy, A., Zhang, F., Moon, T., Maranas, C. D., and Pakrasi, H. B. (2015) Diurnal Processes in the Cyanobacterium Synechocystis sp. PCC 6803: Insights from Transcriptomic, Fluxomic and Physiological Analyses. [46] Welsh, E. A., Eschrich, S. A., Berglund, A. E., and Fenstermacher, D. A. (2013) Iterative rank-order normalization of gene expression microarray data, BMC Bioinformatics 14, 153. [47] Kucho, K., Okamoto, K., Tsuchiya, Y., Nomura, S., Nango, M., Kanehisa, M., and Ishiura, M. (2005) Global analysis of circadian expression in the cyanobacterium Synechocystis sp strain PCC 6803, J Bacteriol 187, 21902199.
ACS Paragon Plus Environment
Page 26 of 33
Page 27 of 33
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 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647
ACS Synthetic Biology
[48] Halkidi, M., Batistakis, Y., and Vazirgiannis, M. (2001) On clustering validation techniques, J Intell Inf Syst 17, 107-145. [49] Sneath, P. H. A., and Sokal, R. R. (1973) Numerical taxonomy; the principles and practice of numerical classification, W. H. Freeman, San Francisco,. [50] Halkidi, M., Vazirgiannis, M., and Batistakis, Y. (2000) Quality Scheme Assessment in the Clustering Process, In Principles of Data Mining and Knowledge Discovery (Zighed, D., Komorowski, J., and Żytkow, J., Eds.), pp 265-276, Springer Berlin Heidelberg. [51] Singh, A. K., Elvitigala, T., Cameron, J. C., Ghosh, B. K., Bhattacharyya-Pakrasi, M., and Pakrasi, H. B. (2010) Integrative analysis of large scale expression profiles reveals core transcriptional response and coordination between multiple cellular processes in a cyanobacterium, Bmc Syst Biol 4. [52] R Core Team. (2014) R: A Language and Environment for Statistical Computing, R Foundation for Statistical Computing, Vienna Austria. [53] Martin Maechler, Peter Rousseeuw, Anja Struyf, Mia Hubert, and Hornik, K. (2015) cluster: Cluster Analysis Basics and Extensions. [54] Desgraupes, B. (2014) clusterCrit: Clustering Indices.
648
ACS Paragon Plus Environment
ACS Synthetic Biology
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 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
649
For Table of Contents Use Only
650
Title: “Identifying regulatory changes to facilitate nitrogen fixation in the non-
651
diazotroph Synechocystis sp. PCC 6803”
652
Authors: Thomas J. Mueller, Eric A. Welsh, Himadri B. Pakrasi, and Costas D.
653
Maranas
654 655
Table of Contents Figure:
656
ACS Paragon Plus Environment
Page 28 of 33
Page 29 of 33
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 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Synthetic Biology
Figure 1: Parts of the regulatory networks for A) Synechocystis and B) Cyanothece. Edges are colored based on whether the predicted interaction is activating (green) or inhibiting (red). A subset of the gene clusters containing nif genes is shown above, all regulatory influences on the nif genes are listed in Supplemental File S2. Dashed lines indicate edges with associated literature support discussed in the Results section. 157x139mm (300 x 300 DPI)
ACS Paragon Plus Environment
ACS Synthetic Biology
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 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Figure 2: Distribution of homolog pairs in which one peaks in the light period and one peaks in the dark between the different subsystems. Subsystem annotations from Saha et al. were used45. *“DNA replication and other processes” refers to the subsystem “DNA replication, restriction, modification, recombination, and repair”. 84x155mm (300 x 300 DPI)
ACS Paragon Plus Environment
Page 30 of 33
Page 31 of 33
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 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Synthetic Biology
Figure 3: Comparison of candidate native transcription factor expression profiles in Synechocystis (lexA: black line, rre34: cyan line)) with those of the nif cluster in Cyanothece (dashed lines). The dark periods of the light-dark cycle (hours 0-12 and 24-36) are represented with gray shading. 84x63mm (300 x 300 DPI)
ACS Paragon Plus Environment
ACS Synthetic Biology
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 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Figure 4: All proposed regulatory changes. Genes native to Synechocystis are in green; those from Cyanothece are in orange. Proposed regulatory changes are represented as dashed lines. A) Proposed strategy for control of the nif cluster in Synechocystis using lexA and patB. B) Strategies for control of glycogen catabolism (horizontal stripes) and cyanophycin synthesis (vertical stripes) genes in Synechocystis using rcp1. 177x109mm (300 x 300 DPI)
ACS Paragon Plus Environment
Page 32 of 33
Page 33 of 33
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 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Synthetic Biology
Figure 5: Metabolic processes requiring modified regulation in a diazotrophic Synechocystis. A) Expression profiles of rcp1 (solid black line) and glycogen phosphorylase (solid brown line) in Synechocystis, and glycogen phosphorylase (dashed brown line) in Cyanothece. B) Expression profiles of rcp1 (solid black line) and cycling cyanophycin synthesis genes (solid lines, colors corresponding to C) in Synechocystis, and the expression profiles of the cycling cyanophycin synthesis genes in Cyanothece (dashed lines). Sll0573 and slr2002 are not shown, as they were not identified as cycling. The dark periods of the light-dark cycle (hours 0-12 and 24-36) are represented with gray shading. C) Reactions and genes identified by the iSyn731 and iCyt773 genome-scale models as being involved in cyanophycin synthesis. Reactions with associated genes whose elevated expression times differ between the two organisms are color-coded. 177x110mm (300 x 300 DPI)
ACS Paragon Plus Environment