Page 1 of 31
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
1
ACS Synthetic Biology
Synthetic Quorum Sensing and Cell-Cell Communication in Gram-Positive Bacillus megaterium
2 3
Nicholas Marchanda,b and Cynthia H. Collinsa,b
4 5 6 7 8
a
Department of Chemical and Biological Engineering, Rensselaer Polytechnic Institute, 110 8th St, Troy, NY 12180, USA b
Center for Biotechnology and Interdisciplinary Studies, Rensselaer Polytechnic Institute, 110 8th St, Troy, NY 12180, USA
9 10
Corresponding Author:
11
Name: Cynthia H. Collins
12 13
Address: Center for Biotechnology and Interdisciplinary Studies, Rensselaer Polytechnic Institute, 110 8th St, Troy, NY 12180, USA
14
Phone Number: 518-276-4178
15
E-mail Address:
[email protected] 16
Fax Number: 518-276-4233
17
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
18
Abstract
19
The components of natural quorum-sensing (QS) systems can be used to engineer synthetic
20
communication systems that regulate gene expression in response to chemical signals. We have
21
used the machinery from the peptide-based agr QS system from Staphylococcus aureus to
22
engineer a synthetic QS system in Bacillus megaterium to enable autoinduction of a target gene
23
at high cell densities. Growth and gene expression from these synthetic QS cells were
24
characterized in both complex and minimal media. We also split the signal production and
25
sensing components between two strains of B. megaterium to produce “sender” and “receiver”
26
cells and characterized the resulting communication in liquid media and on semi-solid agar. The
27
system described in this work represents the first synthetic QS and cell-cell communication
28
system that has been engineered to function in a Gram-positive host, and has the potential to
29
enable the generation of dynamic gene regulatory networks in B. megaterium and other Gram-
30
positive organisms.
31 32
Keywords: Bacillus megaterium, quorum sensing, autoinduction, cell-cell communication,
33
synthetic biology
34
ACS Paragon Plus Environment
Page 2 of 31
Page 3 of 31
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
35
Traditionally, cell factories have been built by directing metabolic pathways toward the
36
production of a target of interest, often by knocking out or constitutively overexpressing select
37
genes. As the field has advanced, we have learned that these strategies can often overwhelm the
38
cell and/or divert too heavily away from natural pathways leading to decreased cell growth and
39
productivity. Additional control over gene expression levels and timing is one strategy towards
40
enhancing cell growth and increasing productivity.1 Synthetic communication systems can be
41
used to enable dynamic gene regulation, where the expression levels of target genes are
42
modulated as a function of cell density.2–4 Another strategy is to split up tasks between
43
populations of cells grown in co-culture or more complex consortia, where synthetic
44
communication pathways can be used to control and coordinate gene expression among
45
community members.2,5–8 A key challenge with respect to using synthetic communication
46
systems to build dynamic regulatory networks and microbial consortia for metabolic engineering
47
applications is the availability of regulatory components that enable communication in different
48
types of organisms.
49
Many of the existing synthetic communication pathways use the machinery from natural
50
microbial quorum-sensing (QS) systems.2,4,9 QS can be thought of as a natural autoinduction
51
mechanism, where target gene expression is turned on at high cell densities.10 Cells with natural
52
QS systems produce signaling molecules at low basal levels during growth, such that the
53
signaling molecule concentration correlates to the local cell density. At high cell densities QS
54
regulators alter gene expression to enable coordinated behaviors at the population level, such as
55
stress responses (genetic competence and uptake of complex nutrients), and protection from
56
other organisms and host defense mechanisms (biofilm formation, sporulation, and virulence
57
production).11
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
58
Synthetic autoinduction systems have proven very effective for producing high yields of
59
heterologous protein in high cell density cultures. Autoinduction systems can often express
60
higher titers of recombinant protein than more traditional inducible systems, with the added
61
advantage that there is no need to monitor cell growth and add exogenous molecules at a specific
62
cell density.12,13 These systems can be particularly useful for high-throughput screening of many
63
cultures in parallel where monitoring cell growth can be a significant burden,13,14 and also for
64
industrial applications to provide reliable induction and increased productivity.12 Some synthetic
65
autoinduction systems have been engineered to date by interrupting natural QS systems,15 and by
66
using the machinery from natural QS systems to engineer synthetic QS in host cells.16–20
67
The machinery of natural QS systems has also been split between cells to engineer
68
synthetic cell-cell communication systems.15,17,21–24 Future use of microbial consortia in
69
bioprocessing has the potential to increase process efficiency and productivity5,25 in part through
70
reducing metabolic burden on individual cells, providing modularity and compartmentalization,26
71
and constructing more dynamic systems.23 Building synthetic cell-cell communication systems
72
that function between industrially-relevant species is a key prerequisite for controlling the
73
population dynamics in microbial consortia-based bioprocesses.27
74
It is important to note that almost all of the synthetic communication systems that have
75
been developed to date are functional in and between Gram-negative organisms, and most of the
76
synthetic networks have been implemented in Escherichia coli. However, Gram-positive bacteria
77
have been used heavily in the commercial enzyme market,28 and are more efficient protein
78
secretors than their Gram-negative counterparts. It is therefore important to develop cell-cell
79
communication systems that will function in Gram-positive bacteria.
ACS Paragon Plus Environment
Page 4 of 31
Page 5 of 31
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
80
To this end we recently constructed a synthetic communication pathway between the
81
Gram-negative E. coli and Gram-positive Bacillus megaterium using the machinery from the agr
82
QS system of Staphylococcus aureus.29 The organism B. megaterium was chosen as a Gram-
83
positive host because of its plasmid stability, low extracellular protease activity, high protein
84
expression levels, and previous use in industry.30,31 The agr QS system of S. aureus is illustrated
85
in Figure 1A, and is responsible for cell-density-dependent control of behaviors such as virulence
86
and biofilm formation.32–34 The agr system is comprised of two promoters, P2 and P3, upstream
87
of two transcripts, RNAII and RNAIII, respectively. The RNAII transcript (or agr locus)
88
contains four genes, agrB, agrD, agrC, and agrA. The agrD gene codes for a propeptide
89
containing the autoinducing peptide (AIP) signal sequence (YSTCDFIM) between an N-terminal
90
signal peptide and a C-terminal tail. The mature, type-I AIP contains a thiolactone ring between
91
the carboxyl group on the C-terminus and the internal cysteine residue.35 Upon contact with the
92
transmembrane endopeptidase, AgrB, the C-terminal tail of AgrD is cleaved, and the thiolactone
93
ring is formed.36 The signal peptide of the AgrD intermediate is then cleaved by the S. aureus
94
type I signal peptidase (SpsB) and the mature AIP is released into the extracellular space.37 AgrC
95
and AgrA make up a two-component signal transduction system. AgrC is a transmembrane
96
histidine kinase receptor that phosphorylates the transcriptional activator AgrA in response to
97
AIP binding. Phosphorylated AgrA then activates gene expression from the P2 and P3
98
promoters.38
99
Our previously constructed communication pathway demonstrated that the “receiver”
100
components of the agr QS system, AgrC, AgrA, and the P3 promoter, function in
101
B. megaterium.29 Specifically, we showed that the B. megaterium “receiver” cells exhibited a
102
dose-dependent increases in GFP expression in the presence of supernatants containing AIP
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
103
produced by either S. aureus, which naturally produce AIP, or E. coli engineered to produce AIP.
104
Here, we have examined whether the “sender” components, AgrB and AgrD, of the agr QS
105
system can also function in B. megaterium. By combining the expression of AgrB and AgrD
106
with the AgrC and AgrA components, we have developed a synthetic QS system in
107
B. megaterium for target gene autoinduction (illustrated in Fig. 1B). Furthermore, we have
108
developed a synthetic communication pathway between B. megaterium strains by co-culturing
109
AIP-producing “sender” cells with AIP-sensing “receiver” cells (illustrated in Fig. 1C). These
110
new tools will be vital for controlling gene expression in this industrially important Gram-
111
positive host, and may lead to the expanded use of B. megaterium in laboratory and industrial
112
settings.
113
ACS Paragon Plus Environment
Page 6 of 31
Page 7 of 31
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. A. Illustration of the agr QS system of S. aureus. The AgrD propeptide is processed by the AgrB endopeptidase, and the SpsB signal peptidase, and released into the supernatant as mature AIP. The AgrC receptor and AgrA transcriptional activator make up a twocomponent signaling cascade, which control expression from the P2 and P3 promoter. B. Illustration of synthetic QS inside B. megaterium using the machinery from the agr QS system of S. aureus. B. megaterium cells express the agrBDCA genes from the xylose-inducible PxylA promoter to enable synthetic QS. The natural SipM signal peptidase functions in place of the S. aureus SpsB signal peptidase for AgrD processing. The P3 promoter controls expression of gfp reporter. C. Illustration of B. megaterium to B. megaterium synthetic communication. Sender cells (left) produce AIP through AgrB and AgrD expression. Receiver cells (right) express AgrC and AgrA to sense AIP and turn on gene expression from the P3 promoter.
114 115
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
Results and Discussion
117
QS Peptide Production in B. megaterium
118
In previous work, we constructed B. megaterium cells that can sense and respond to an
119
extracellular AIP QS peptide using machinery from the agr QS system of S. aureus.29 These
120
B. megaterium “receiver” cells express the AgrC receptor and AgrA transcriptional activator
121
from a xylose inducible promoter, and respond to extracellular AIP through activation of the P3
122
promoter in front of a reporter gfp gene. To develop new synthetic communication pathways in
123
B. megaterium, we tested the functionality of the two remaining components of the agr QS
124
system: AgrB and AgrD. As shown in Figure 1A, the AgrD pre-protein is processed first by the
125
AgrB endopeptidase and then by a type-I signal peptidase (SpsB in S. aureus) to produce and
126
secrete the AIP QS peptide. It has been previously shown that expression of the agrBD genes in
127
E. coli is sufficient for AIP production,29,39 suggesting that the type-I signal peptidase of E. coli
128
(LepB) can replace the function of SpsB. We therefore hypothesized that the type-I signal
129
peptidase of B. megaterium (SipM) could similarly replace the action of SpsB (Fig. 1B).
130
To determine if B. megaterium cells are capable of producing mature AIP when agrBD is
131
expressed, the pXylA-agrBD-I plasmid was first constructed. Similar to our previous construct
132
pXylA-agrCA-I,29 the type-I agrBD genes of S. aureus RN4220 were cloned downstream of the
133
xylose-inducible PxylA promoter of pT7-RNAP (MoBiTec), replacing the gene encoding T7
134
RNAP. This new plasmid was transformed into B. megaterium MS941 cells containing the pP3-
135
GFP plasmid,29 which contains the reporter gfp gene downstream of the S. aureus P3 promoter.
136
B. megaterium control cells lacking any plasmids, and the new B. megaterium AIP
137
“sender” cells containing the pXylA-agrBD-I and pP3-GFP plasmids were cultured in LB in
138
shake flasks with and without the xylose inducer molecule (1% w/v). Following 12 h of growth
ACS Paragon Plus Environment
Page 8 of 31
Page 9 of 31
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
supernatant samples were isolated, concentrated 10-fold through lyophilization, and diluted 10-
140
fold into growing cultures of B. megaterium AIP receiver cells (MS941 with pXylA-agrCA-I and
141
pP3-GFP; used here as a reporter for AIP). Fluorescence from the receiver cells following 24 h
142
of growth can be seen in Figure 2A. Receiver cells grown with supernatants from induced
143
cultures of B. megaterium expressing agrBD showed fluorescence values > 5-fold above the
144
background (ANOVA, P < 0.01), demonstrating that the AgrB and AgrD components are
145
functional in B. megaterium, and their expression leads to secretion of the AIP QS peptide.
146
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
Page 10 of 31
Figure 2. AIP production from B. megaterium. A. Supernatants were isolated from cultures of B. megaterium control cells lacking any plasmids and B. megaterium AIP sender cells. Induced cultures were supplemented with 1% xylose at the beginning of the culture. Supernatants were concentrated 10-fold through lyophilization, then diluted 10-fold into cultures of AIP receiver cells. Fluorescence from receiver cells after 24 h of growth is shown. Data was evaluated using ANOVA; double asterisk indicates P 7-fold above control
200
cell fluorescence after 28 h of growth (ANOVA, P < 0.01). Uninduced QS cells appeared to
201
fluoresce slightly earlier than induced QS cells, though this can likely be attributed to the shorter
202
lag phase (Figure 3C). Interestingly, no difference was observed between the fluorescence from
203
induced vs. uninduced synthetic QS cells grown in minimal media, and the fluorescence levels
204
were ~2.5-fold lower than that of induced cells grown in LB. It is possible that subtle differences
205
in cell metabolism when grown in LB vs. minimal media have an effect on the xylose regulatory
206
pathway. Previous work has shown that uptake of xylose into the cell, as well as xylose
207
metabolism by the cell are regulated by simple sugars present in the culture.43 We hypothesized
208
that adding an amino acid source to the media could improve cell growth and AIP production. As
209
can be seen by comparing Figure 3C to 3E, the addition of 1% casamino acids to the minimal
210
media led to a decrease in lag time. The addition of casamino acids also led to an increase in
211
synthetic QS cell fluorescence from ~ 7-fold (Fig. 3D) to > 25-fold (Fig. 3F; ANOVA, P < 0.01)
212
above the control cell fluorescence. Additionally, a small but significant 1.3-fold increase in
213
fluorescence was seen between induced and uninduced synthetic QS cells (paired T-test, P
