Synthetic Quorum Sensing and Cell–Cell Communication in Gram

Jul 23, 2015 - peptide-based agr QS system from Staphylococcus aureus to ... communication system that has been engineered to function in a Gram-posit...
0 downloads 3 Views 1MB Size
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