Patent Application
20120237987
The invention encompasses a recombinant bacterium for the production of fatty acids. In particular, the invention encompasses a bacterium capable of both generating fatty acids and releasing the fatty acids into the culture medium.
Cheap renewable energy is a major goal of many countries and hardly needs a justification. Not only does renewable energy provide energy independence and security but can often provide carbon neutral or better clean energy. But not all forms of renewable energy are equal. Currently wind, solar, hydro and geothermal power are excellent supplements to the power grid, but cannot be efficiently converted into liquid biofuels needed for our cars, ships and planes. There will be a high demand for biodiesel and other liquid biofuels into the foreseeable future.
Liquid biofuels can be roughly classified by the source (feedstock) and production methods. First generation biofuels like ethanol from corn are now established industries, but the trend is to move away from using food crops as the primary feedstock for energy production. Second generation biofuels like cellulosic ethanol use biomass as the primary feedstock. The biomass is obtained from the “waste” left over after food crops have been processed, or from other energy crops like switchgrass, Miscanthus, and “energycane”. While eliminating the ethical dilemma of tapping into the world's food supply for fuel, cellulosic ethanol is still neither the most efficient way to produce biofuels nor a viable replacement for the majority of the nations fuel needs. Recycling biomass waste into fuel is a great idea, but growing energy crops just to harvest and digest them back into biofuel is not the most efficient use of energy, water, or arable land.
Third generation biofuels have typically been defined as fuel from photosynthetic organisms. The organisms are grown, harvested, and then the fuel precursors are extracted from the biomass. Photosynthetic microbes such as algae and cyanobacteria are the most efficient organisms for solar energy conversion, typically yielding lipids in the range of about 20-30% of dry weight. While photosynthetic microorganisms will theoretically out produce any plant based biofuel systems, costs associated with extraction continue to be a barrier to making these biofuels competitive with fossil fuels.
Hence, there is a need in the art for a microbe capable of producing fatty acids while decreasing processing and extraction costs.
One aspect of the present invention encompasses a recombinant bacterium. The bacterium is capable of producing fatty acids and comprises at least one modified polar cell layer.
Another aspect of the present invention encompasses a recombinant bacterium. The bacterium is capable of producing fatty acids and is capable of the inducible release of fatty acids from a cellular membrane.
Another aspect of the present invention encompasses a recombinant bacterium. The bacterium is modified to encode multiple thioesterases that specify synthesis and secretion of saturated C10 to C14 chain fatty acids that are secreted more efficiently than saturated and unsaturated C16 and C18 chain fatty acids.
Another aspect of the present invention encompasses a recombinant bacterium. The bacterium is modified to encode multiple thioesterases that specify synthesis and secretion of saturated C10 to C14 chain fatty acids that are secreted more efficiently than saturated and unsaturated C16 and C18 chain fatty acids and comprises at least one modified polar cell layer.
Another aspect of the present invention encompasses a recombinant bacterium. The bacterium is modified to encode multiple thioesterases that specify synthesis and secretion of saturated C10 to C14 chain fatty acids that are secreted more efficiently than saturated and unsaturated C16 and C18 chain fatty acids and is capable of the inducible release of fatty acids from a cellular membrane.
Yet another aspect of the present invention encompasses a method of producing fatty acids. The method comprises culturing a bacterium capable of producing fatty acids and that comprises at least one modified polar cell layer.
Still another aspect of the present invention encompasses a method of producing fatty acids. The method comprises culturing a bacterium capable of producing fatty acids and that is capable of the inducible release of fatty acids from a cellular membrane.
Still another aspect of the present invention encompasses a method of producing fatty acids. The method comprises culturing a bacterium modified to encode multiple thioesterases that specify synthesis and secretion of saturated C10 to C14 chain fatty acids that are secreted more efficiently than saturated and unsaturated C16 and C18 chain fatty acids.
Still another aspect of the present invention encompasses a method of producing fatty acids. The method comprises culturing a bacterium modified to encode multiple thioesterases that specify synthesis and secretion of saturated C10 to C14 chain fatty acids that are secreted more efficiently than saturated and unsaturated C16 and C18 chain fatty acids and comprises at least one modified polar cell layer.
Still another aspect of the present invention encompasses a method of producing fatty acids. The method comprises culturing a bacterium modified to encode multiple thioesterases that specify synthesis and secretion of saturated C10 to C14 chain fatty acids that are secreted more efficiently than saturated and unsaturated C16 and C18 chain fatty acids and is capable of the inducible release of fatty acids from a cellular membrane.
Other aspects and iterations of the invention are described more thoroughly below.
The application file contains at least one photograph executed in color. Copies of this patent application publication with color photographs will be provided by the Office upon request and payment of the necessary fee.
The present invention provides a bacterium capable of producing fatty acids. In particular, a bacterium of the invention may be used to generate fatty acids and release the fatty acids into the culture medium. Advantageously, the released fatty acids may be used as a biofuel precursor. Consequently, the invention also provides methods of producing fatty acids, and methods of harvesting the fatty acids.
A recombinant bacterium of the invention may comprise one or more alterations to increase fatty acid production, to enable fatty acid secretion, to enable fatty acid release, and to modulate fatty acid structure. These mutations are discussed in more detail below.
A bacterium of the invention is typically a cyanobacterium. In some embodiments, a bacterium belongs to the order Chroococcales. In a preferred embodiment, the bacterium is derived from the species Synechocystis. For instance, a bacterium of the invention may be derived from Synechocystis PCC sp. 6803.
A recombinant bacterium of the invention may comprise one or more alterations to increase fatty acid generation. For instance, a bacterium may comprise an alteration that enables the synthesis of an acyl-ACP thioesterase, that inhibits fatty acid degradation, that channels resources into fatty acid synthesis, that down-regulates or eliminates competing pathways, that decreases repression or feedback inhibition, and that maintains stationary phase fatty acid production. Each of these alterations are described in more detail below.
i. Acyl-ACP Thioesterase
A bacterium of the invention may comprise an alteration that enables the synthesis of at least one acyl-ACP thioesterase (hereinafter TE). Methods of altering a bacterium to synthesize a TE are known in the art. For instance, a bacterium may be altered to express a nucleic acid encoding a TE. Such a nucleic acid may be operably linked to a regulated promoter or a constitutive promoter. In certain embodiments, a bacterium may synthesize one, two, three, four or five TEs. A nucleic acid encoding a TE may be chromosomally integrated, or may be expressed on an extrachromosomal vector. Suitable vectors are known in the art. Similarly, methods of chromosomally inserting a nucleic acid are known in the art. For additional details, see the Examples.
In some embodiments, a bacterium may synthesize a TE that is restricted to the cytosol of the bacterium. For instance, in one embodiment, a bacterium of the invention may synthesize a variant of TesA that is restricted to the cytosol of the bacterium. By way of non-limiting example, a bacterium may synthesize *TesA. The expression of a nucleic acid encoding TesA may be regulated or constitutive. For instance, the nucleic acid may be operably linked to an inducible promoter. Non-limiting examples of a suitable inducible promoters may include PnrsB, PcmpA, PisiA, PsigE, PlrtA, or PsbD2. PnrsB is nickel inducible, PcmpA is inducible by CO2 limitation, PisiA is includucible under low Fe conditions, PsigE is inducible during the stationary phase, PlrtA P is dark inducible, and PsbD2 may be induced by strong light.
Alternatively, the nucleic acid encoding a TE may be operably linked to a constitutive promoter, such as PpsbA2, Pcpc, Prbc, PpetB, PpsaAB, PhspA, or PsigA.
Other TE enzymes are known in the art and may be used in the present invention. For instance, a TE from Cinnamomum camphorum, Umbellularia californica, or Cuphea hookeriana may be used. In another embodiment, a TE outlined in WO 2009/076559 may be used. In still another embodiment, a TE detailed in Table 7 below may be used.
The selection of the TE may be determined by the desired chain length of the resulting free fatty acid. For instance, see Table 7 below. In one embodiment, a TE with a preference for shorter free fatty acids may be used. For instance, a TE with a preference for C16, C14, C12, C10 or C8 fatty acids may be used.
A nucleic acid encoding a TE may be modified for high-level expression in a bacterium of the invention. As used herein, “modified” refers to an alteration of a nucleic acid sequence that results in a change in the level of transcription of a nucleic acid sequence, or that results in a change in the level of synthesis of encoded protein. For instance, in one embodiment, modify may refer to altering the start codon of a nucleic acid sequence. Generally speaking, a GTG or TTG start codon, as opposed to an ATG start codon, may decrease translation efficiency ten-fold. In another embodiment, modify may refer to altering the Shine-Dalgarno (SD) sequence of a nucleic acid sequence. The SD sequence is a ribosomal binding site (RBS) generally located 6-7 nucleotides upstream of the start codon. The SD/RBS consensus sequence is AGGAGG, and variations of the consensus sequence may alter translation efficiency. In yet another embodiment, modify may refer to altering the distance between the SD sequence and the start codon. In still another embodiment, modify may refer to altering the −35 sequence for RNA polymerase recognition. In a similar embodiment, modify may refer to altering the −10 sequence for RNA polymerase binding. In an additional embodiment, modify may refer to altering the number of nucleotides between the −35 and −10 sequences. In an alternative embodiment, modify may refer to optimizing the codons of the nucleic acid sequence to alter the level of translation of the mRNA. For instance, non-A rich codons initially after the start codon of a nucleic acid sequence may not maximize translation of the corresponding mRNA. Similarly, the codons of the nucleic acid sequence may be altered so as to mimic the codons in genes encoding highly synthesized proteins of a particular organism. In a further embodiment, modify may refer to altering the GC content of the nucleic acid sequence to change the level of translation of the corresponding mRNA. Additionally, modify may refer to alterations in the DNA sequence of a gene so that the transcribed mRNA is stabilized with a reduced rate of degradation but still able to specify a protein of the original amino acid sequence. In still another embodiment, a nucleic acid may be optimized by altering the nucleic acid such that the ability of the encoded protein to form efficient enzyme complexes is affected.
Methods of altering a nucleic acid as described above are known in the art. For more details, see the Examples.
ii. Inhibiting Fatty Acid Degradation
A recombinant bacterium of the invention may comprise an alteration that inhibits fatty acid degradation. For instance, the acyl-ACP synthetase (AAS) nucleic acid may be modified to decrease or eliminate expression of the nucleic acid. As described below the aas gene used to be referred to as the fadD gene. In one embodiment, aas is modified by replacing the aas chromosomal sequence with another sequence, such as a nucleic acid encoding a TE.
iii. Alterations that Channel Resources into Fatty Acid Synthesis
In some embodiments, a recombinant bacterium of the invention may comprise one or more alterations that channel resources into fatty acid synthesis. In certain embodiments, this may mean decreasing or eliminating expression of a nucleic acid that is not necessary for fatty acid synthesis. For instance, a bacterium may comprise a mutation that decreases or eliminates expression of a nucleic acid encoding a polyhydroxyalkanoate (PHA) synthesis enzyme. Non-limiting examples may include slr1993 and slr1994. In another embodiment, a bacterium may comprise a mutation that alters synthesis of an S-layer protein. Non-limiting examples may include mutations in sll1951, such as Δsll1951. In each of the above embodiments, the mutations should not alter the fitness of the bacterium in such a way as to reduce fatty acid synthesis.
Another way to channel resources into fatty acid synthesis is to increase the expression level of nucleic acid sequences encoding proteins in the primary free fatty acid production pathway. For instance, the expression of a nucleic acid encoding a protein involved in the generation of pyruvate may be increased. By way of non-limiting example, the expression of sll0587 or sll1275 may be increased. In another embodiment, the expression of a nucleic acid encoding a protein involved in the synthesis of acetyl-CoA from pyruvate, such as pdh or odh may be increased. In yet another embodiment, a nucleic acid sequence encoding a protein involved in the synthesis of malonyl-CoA from acetyl-CoA may be altered, such as accBCDA. Alternatively, a bacterium may be altered such that ACC may be overproduced by introducing a synthetic operon. Typically, the transcripts of the nucleic acids encoding the ACC subunits should be produced in relatively equal molar ratios. Non-limiting examples may include altering the bacterium to include the operon Pcpc accB accC Prbc accD accA. In still another embodiment, the expression of a nucleic acid sequence encoding a protein involved in the synthesis of fatty acyl-ACP may be increased. For instance, the expression of a fab nucleic acid sequence may be increased (e.g. fabD, fabF, fabG, fabZ and fabl). In an alternative embodiment, the expression of a nucleic acid encoding an acyl carrier protein (such as ssl2084) may be increased. In a further embodiment, the expression of pyk may be increased.
Yet another way to channel resourses into fatty acid synthesis is to optimize expression of a nucleic acid encoding a protein involved in fatty acid synthesis. For instance, the expression may be optimized as described in section (a) i above. Alternatively, the expression may be optimized by altering the nucleic acid sequence to increase mRNA stability. For instance, the sequence may be altered to remove stem-loop structures.
iv. Alterations to Competing Pathways
A bacterium of the invention may be altered to reduce or eliminate the expression of a nucleic acid sequence encoding a protein that competes with fatty acid synthesis for reactants. By way of non-limiting example, the expression of slr1176 for the glycogen synthesis pathway; slr0301 (pps) for the phosphoenolpyruvate pathway; slr1192 for the predicted alcohol synthesis pathway; slr0089 and slr0090 for the vitamin E pathway; sll0920 for the oxaloacetate pathway; sll0891 for the malate synthesis pathway; sll1564 for the leucine synthesis pathway, sll0401 for the citrate pathways, and slr2132, sll1299, sll0542, slr0091 and sll0090 for the acetate pathways may be reduced or eliminated. Similarly, the expression of a nucleic acid sequence encoding a cyanophycin synthetase (such as slr2002 or slr2001) can be reduced or eliminated.
Expression of the above nucleic acid sequences may be reduced by altering the promoter, SD sequence, and/or start codon, etc. as described in section (a)i. above.
v. Alterations that Decrease Repression or Feedback Inhibition
A bacterium of the invention may be altered to decrease repression of fatty acid synthesis or to decrease feedback inhibition of fatty acid synthesis. For instance, expression of a TE, as described above, may be used to decrease inhibition of ACC, FabH, and Fabl. Similarly, repression may be decreased by altering the promoters of nucleic acids encoding proteins involved in fatty acid syntheis so that they do not include the binding sequences for repressors. Further examples of methods of decreasing feedback inhibition and repression are described in the Examples.
vi. Alterations to Maintain Stationary Phase Fatty Acid Production
A recombinant bacterium of the invention may be altered so as to allow secretion of fatty acids during stationary growth phase. Generally speaking, such alterations may include supplying a bacterium with a nucleic acid sequence encoding a protein involved in fatty acid synthesis, wherein the nucleic acid is operably linked to a promoter with increased activity in the stationary phase. Non-limiting examples of such promoters are detailed in the Examples.
A bacterium of the invention may comprise alterations to enable and/or increase fatty acid secretion. In one embodiment, a polar cell layer of the bacterium may be altered so as to increase fatty acid secretion. By way of example, the peptidoglycan layer, the outer membrane layer, and/or the S layer of a bacterium may be altered to enable increased fatty acid secretion. For instance, the expression of a nucleic acid encoding an S-layer protein, such as sll1951, may be decreased or eliminated. In one embodiment, a bacterium may comprise the mutation Δsll1951.
In another embodiment, the polypeptidoglycan layer of a bacterium may be weakened to enable increased fatty acid secretion. Methods of weakening the polypeptidoglycan layer may include administering an antibiotic, such as ampicillin, to the bacterium. Care should be taken, however, to balance the ability to secrete fatty acids with the potential for cell lysis. Such a balance may be experimentally determined as detailed in the Examples.
Another method to weaken the peptidoglycan layer may comprise down-regulating the transcription efficiency of nucleic acids encoding protein involved in peptidoglycan synthesis, such as those in the mur (e.g., slr0017, slr1423, slr1656 and sll2010) and ldh (e.g., slr0528 and slr1656) families to weaken the polypeptidoglycan layer structures. In another embodiment, a nucleic acid encoding a penicillin-binding protein such as ftsl (sll1833), mrcB (slr1710) and ponA (sll0002) may be deleted or modified. These proteins are required for the assembly of the peptidoglycan.
Yet another method to interfere with peptidoglycan synthesis is by substituting a nucleic acid for a central step in an essential pathway with one from another bacterial species, such as using an exogenous asd or alr gene. Still yet another way to weaken the polypeptidoglycan layer is to introduce one or more nucleic acid sequence encoding an endolysin from a bacteriophage. Such a nucleic acid sequence may then be expressed at low levels. Endolysins are peptidoglycan-degrading enzymes that attack the covalent linkages of the peptidoglycans that maintain the integrity of the cell wall. For instance, the endolysin gp19 from Salmonella phage P22 is able to degrade the 6803 polypeptidoglycan layers, and the endolysin R from E. coli phage λ is able to compromise the 6803 polypeptidoglycan layers. These sequences may be expressed with different promoters with variant low transcription efficiencies to limit adverse growth effects.
Yet another method to increase fatty acid secretion is to express or overexpress a nucleic acid sequence encoding a transporter or porin to make channels for the lipid. Many transport and efflux proteins serve to excrete a large variety of compounds, and these can possibly be modified to be selective for fatty acids. For example, E. coli outer membrane protein FadL is a membrane-bound fatty acid transporter, which binds long chain fatty acid with a high affinity. Other suitable transport proteins may include efflux proteins and fatty acid transporter proteins (FATP). Suitable non-limiting examples may be found in Table 9.
(c) Fatty Acid Release from Cellular Membranes
Another aspect of the invention encompasses the discovery that the regulated expression of a nucleic acid encoding a protein capable of hydrolyzing the lipid membranes to free fatty acids may be used to disrupt the cells and release intracellular fatty acids. Hence, in one embodiment, the invention encompasses a cyanobacterium comprising an inducible promoter operably-linked to a nucleic acid encoding a first protein capable of hydrolyzing the lipid membranes of the bacterium and at least one endolysin protein. In another embodiment, the invention encompasses a cyanobacterium comprising a first nucleic acid, wherein the first nucleic acid comprises a first inducible promoter operably-linked to a nucleic acid encoding a first protein capable of hydrolyzing the lipid membranes of the bacterium; and a second nucleic acid, wherein the second nucleic acid comprises a second promoter operably-linked to a nucleic acid encoding at least one endolysin protein.
In certain instances, the invention encompasses a cyanobacterium comprising more than one integrated nucleic acid construct of the invention. For instance, the invention may encompass a cyanobacterium comprising a first inducible promoter operably-linked to a nucleic acid encoding a first protein capable of hydrolyzing the lipid membranes of the bacterium, a second inducible promoter operably-linked to a different nucleic acid encoding a first protein capable of hydrolyzing the lipid membranes of the bacterium, and at least two endolysin proteins. In a further embodiment, the nucleic acid sequences encoding the endolysin proteins may be operably linked to a constitutive promoter.
Methods of making a cyanobacterium of the invention are known in the art. Generally speaking, a cyanobacterium is transformed with a nucleic acid contstruct of the invention. Methods of transformation are well known in the art, and may include electroporation, natural transformation, and calcium choloride mediated transformation. Methods of screening for and verifying chromosomal integration are also known in the art.
In one embodiment, a method of making a cyanobacterium of the invention may comprise first transforming the bacterium with a vector comprising, in part, an antibiotic-resistance marker and a negative selection marker. Chromosomal integration may be selected for by selecting for antiobiotic resistance. Next, the antibiotic-resistant strain is transformed with a similar vector comprising the target genes of interest. Chromosomal integration of the target genes may be selected for by selecting for the absence of the negative marker. For instance, if the negative marker is sacB, then one would select for sucrose resistance. For more details, see Kang et al., J Bacteriol. (2002) 184(1):307-12 and Sun et al., Appl. Environ. Microbiol. (2008) 74:4241-45, hereby incorporated by reference in their entirety.
i. Nucleic Acid Constructs
The present invention encompasses a nucleic acid construct that, when introduced into a bacterium, may be used in a method for inducing the degradation of lipid membrane or the peptidoglycan layer of a bacterial cell wall. In one embodiment, the nucleic acid comprises an inducible promoter operably-linked to a nucleic acid sequence encoding a first protein capable of hydrolyzing bacterial lipid membranes into free fatty acids. In another embodiment, the nucleic acid comprises an inducible promoter operably-linked to a nucleic acid sequence encoding a first protein capable of forming a lesion in a bacterial lipid membranes. In yet another embodiment, the nucleic acid comprises a promoter operably-linked to at least one endolysin. In another embodiment, the nucleic acid comprises an inducible promoter operably-linked to both a nucleic acid sequence encoding a first protein and a nucleic acid sequence encoding at least one endolysin. In still another embodiment, the nucleic acid comprises an inducible promoter operably-linked to a nucleic acid sequence encoding a first protein and a second promoter operably-linked to a nucleic acid sequence encoding at least one endolysin. Each component of the above nucleic acid constructs is discussed in more detail below.
Methods of making a nucleic acid construct of the invention are known in the art. Additional information may be found in Sambrook et al., Molecular Cloning: A Laboratory Manual (New York: Cold Spring Harbor Laboratory Press, 1989)
ii. Promoters
A nucleic acid construct of the present invention comprises a promoter. In particular, a nucleic acid construct comprises a first inducible promoter. In some embodiments, a nucleic acid also comprises a second promoter. When a nucleic acid comprises a first and a second promoter, the promoters may read in opposite directions, or may read in the same direction.
A. First Inducible Promoter
In certain embodiments, a nucleic acid of the invention encompasses a first inducible promoter. Non-limiting examples of inducible promoters may include, but are not limited to, those induced by expression of an exogenous protein (e.g., T7 RNA polymerase, SP6 RNA polymerase), by the presence of a small molecule (e.g., IPTG, galactose, tetracycline, steroid hormone, abscisic acid), by absence of small molecules (e.g., CO2, iron, nitrogen), by metals or metal ions (e.g., copper, zinc, cadmium, nickel), and by environmental factors (e.g., heat, cold, stress, light, darkness), and by growth phase. In each of the above embodiments, the inducible promoter is preferably tightly regulated such that in the absence of induction, substantially no transcription is initiated through the promoter. Additionally, induction of the promoter of interest should not typically alter transcription through other promoters. Also, generally speaking, the compound or condition that induces an inducible promoter should not be naturally present in the organism or environment where expression is sought.
In one embodiment, the inducible promoter is induced by limitation of CO2 supply to the cyanobacteria culture. By way of non-limiting example, the inducible promoter may be variant sequences of the promoter sequence of Synechocystis PCC 6803 that are up-regulated under the CO2-limitation conditions, such as the cmp genes, ntp genes, ndh genes, sbt genes, chp genes, and rbc genes.
In one embodiment, the inducible promoter is induced by iron starvation or by entering the stationary growth phase. By way of non-limiting example, the inducible promo'ter may be variant sequences of the promoter sequence of the Synechocystis PCC 6803 isiA gene. In some embodiments, the inducible promoter may be variant sequences of the promoter sequence of cyanobacterial genes that are up-regulated under Fe-starvation conditions such as isiA, or when the culture enters the stationary growth phase, such as isiA, phrA, sigC, sigB, and sigH genes.
In one embodiment, the inducible promoter is induced by a metal or metal ion. By way of non-limiting example, the inducible promoter may be induced by copper, zinc, cadmium, mercury, nickel, gold, silver, cobalt, and bismuth or ions thereof. In one embodiment, the inducible promoter is induced by nickel or a nickel ion. In an exemplary embodiment, the inducible promoter is induced by a nickel ion, such as Ni2+. In another exemplary embodiment, the inducible promoter is the nickel inducible promoter from Synechocystis PCC 6803. In another embodiment, the inducible promoter may be induced by copper or a copper ion. In yet another embodiment, the inducible promoter may be induced by zinc or a zinc ion. In still another embodiment, the inducible promoter may be induced by cadmium or a cadmium ion. In yet still another embodiment, the inducible promoter may be induced by mercury or a mercury ion. In an alternative embodiment, the inducible promoter may be induced by gold or a gold ion. In another alternative embodiment, the inducible promoter may be induced by silver or a silver ion. In yet another alternative embodiment, the inducible promoter may be induced by cobalt or a cobalt ion. In still another alternative embodiment, the inducible promoter may be induced by bismuth or a bismuth ion.
In some embodiments, the promoter is induced by exposing a cell comprising the inducible promoter to a metal or metal ion. The cell may be exposed to the metal or metal ion by adding the metal to the bacterial growth media. In certain embodiments, the metal or metal ion added to the bacterial growth media may be efficiently recovered from the media. In other embodiments, the metal or metal ion remaining in the media after recovery does not substantially impede downstream processing of the media or of the bacterial gene products.
In one embodiment, the nucleic acid comprises a metal or metal ion inducible promoter operably-linked to a nucleic acid sequence encoding a first protein capable of hydrolyzing a bacterial lipid membrane. In another embodiment, the nucleic acid comprises a metal or metal ion inducible promoter operably-linked to both a nucleic acid sequence encoding a first protein and a nucleic acid sequence encoding at least one endolysin. In yet another embodiment, the nucleic acid comprises a metal or metal ion inducible promoter operably-linked to at least one endolysin. In still another embodiment, the nucleic acid comprises a metal or metal ion inducible promoter operably-linked to a nucleic acid sequence encoding a first protein and a second promoter operably-linked to a nucleic acid sequence encoding at least one endolysin.
B. Second Promoter
Certain nucleic acid constructs of the invention may comprise a second promoter. The second promoter may be an inducible promoter, or may be a constitutive promoter. If the second promoter is an inducible promoter, it may or may not be induced by the same compound or condition that induces the first inducible promoter. In one embodiment, the same compound or condition induces both the first and the second inducible promoters. In another embodiment, the first inducible promoter is induced by a different compound or condition than the second inducible promoter. Non-limiting examples of inducible promoters that may be used are detailed in section I(a)(i) above.
Constitutive promoters that may comprise the second promoter are known in the art. Non-limiting examples of constitutive promoters may include constitutive promoters from Gram-negative bacteria or a bacteriophage propogating in a Gram-negative bacterium. For instance, promoters for genes encoding highly expressed Gram-negative gene products may be used, such as the promoter for Lpp, OmpA, rRNA, and ribosomal proteins. Alternatively, regulatable promoters may be used in a strain that lacks the regulatory protein for that promoter. For instance Plac, Ptac, and Ptrc may be used as constitutive promoters in strains that lack Lad. Similarly, P22 PR and PL may be used in strains that lack the P22 C2 repressor protein, and λ PR and PL may be used in strains that lack the λ C1 repressor protein. In one embodiment, the constitutive promoter is from a bacteriophage. In another embodiment, the constitutive promoter is from a Salmonella bacteriophage. In yet another embodiment, the constitutive promoter is from a cyanophage. In some embodiments, the constitutive promoter is a Synechocystis promoter. For instance, the constitutive promoter may be the PpsbAII promoter or its variant sequences, the Prbc promoter or its variant sequences, the Pcpc promoter or its variant sequences, and the PmpB promoter or its variant sequences.
In one embodiment, a nucleic acid of the invention comprises a metal or metal ion inducible promoter operably-linked to a nucleic acid sequence encoding a first protein and a second constitutive promoter operably-linked to a nucleic acid sequence encoding at least one endolysin. In another embodiment, a nucleic acid of the invention comprises a metal or metal ion inducible promoter operably-linked to a nucleic acid sequence encoding a first protein and a second inducible promoter operably-linked to a nucleic acid sequence encoding at least one endolysin.
iii. First Protein
A nucleic acid construct of the invention also comprises a sequence encoding at least one first protein. Generally speaking, a first protein is a protein capable of degrading the lipid membranes into free fatty acid and release intracellular fatty acids. For instance, the first protein may be a lipolytic enzyme that is able to hydrolyze acylglycerols. In one embodiment, the first protein may be a lipolytic enzyme that hydrolyzes diacylglycerols, including MGDG (monogalactosyl diacylglycerol), DGDG (digalactosyl diacylglycerol), PG (phosphatidylglycerol), and SQDG (sulfoquinovosyl diacylglycerol). In another embodiment, the first protein may be a lipase that hydrolyzes triacylglycerols. In yet another embodiment, the first protein may be a lipolytic enzyme that hydrolyzes monoacylglycerols. In still yet another embodiment, the first protein may be a lipolytic enzyme from a bacterium, e.g., Staphylococcus hyicus. In a further embodiment, the first protein may be a lipolytic enzyme from a fungus, e.g., Fusarium oxysporum. In one embodiment, the first protein may be a lipolytic enzyme from an animal, e.g, guinea pig.
In other embodiments, a first protein is a protein capable of hydrolyzing a lipid membrane, such that the endolysin has access to the peptidoglycan layer of the cell wall. In these embodiments, the first protein may be a bacteriophage protein. For instance, the first protein may be a bacteriophage holin protein. In one embodiment, the first protein is a holin from a bacteriophage that infects gram-negative bacteria. In another embodiment, the first protein is a holin from a bacteriophage that infects gram-positive bacteria. In certain embodiments, the first protein is a holin from a cyanophage. In one embodiment, the first protein is a holin from a bacteriophage that infects Synechocystis. In another embodiment, the first protein may be from a bacteriophage that infects Salmonella. In still another embodiment, the first protein may be from a P22 phage. For example, the first protein may be gene 13 of the P22 phage. In yet another embodiment, the first protein may be from a λ phage. For example, the first protein may be encoded by gene S of the λ phage. In still another embodiment, the first protein may be from an E. coli phage. For instance, the first protein may be encoded by gene E of E. coli phage PhiX174. In certain embodiments, a nucleic acid of the invention may comprise at least two holins. In one embodiment, a nucleic acid may comprise a holin from P22 and a holin from λ phage. For instance, the nucleic acid may comprise gene 13 and gene S.
Additionally, a first protein may be a holin described above with at least one, or a combination of one or more, nucleic acid deletions, substitutions, additions, or insertions which result in an alteration in the corresponding amino acid sequence of the encoded holin protein, such as a homolog, ortholog, mimic or degenerative variant. For instance, a first protein may be a holin described above encoded by a nucleic acid with codons optimized for use in a particular bacterial strain, such as Synechocystis. Such a holin may be generated using recombinant techniques such as site-directed mutagenesis (Smith Annu. Rev. Genet. 19. 423 (1985)), e.g., using nucleic acid amplification techniques such as PCR (Zhao et al. Methods Enzymol. 217, 218 (1993)) to introduce deletions, insertions and point mutations. Other methods for deletion mutagenesis involve, for example, the use of either BAL 31 nuclease, which progressively shortens a double-stranded DNA fragment from both the 5′ and 3′ ends, or exonuclease III, which digests the target DNA from the 3′ end (see, e.g., Henikoff Gene 28, 351 (1984)). The extent of digestion in both cases is controlled by incubation time or the temperature of the reaction or both. Point mutations can be introduced by treatment with mutagens, such as sodium bisulfite (Botstein et al. Science 229, 1193 (1985)). Other exemplary methods for introducing point mutations involve enzymatic incorporation of nucleotide analogs or misincorporation of normal nucleotides or alpha-thionucleotide by DNA polymerases (Shortle et al. Proc. Natl. Acad. Sci. USA 79,1588 (1982)). PCR-based mutagenesis methods (or other mutagenesis methods based on nucleic acid amplification techniques), are generally preferred as they are simple and more rapid than classical techniques (Higuchi et al. Nucleic Acids Res. 16, 7351 (1988); Vallette et al. Nucleic Acids Res. 17,723 (1989)).
In addition to having a substantially similar biological function, a homolog, ortholog, mimic or degenerative variant of a holin suitable for use in the invention will also typically share substantial sequence similarity to a holin protein. In addition, suitable homologs, ortholog, mimic or degenerative variants preferably share at least 30% sequence homology with a holin protein, more preferably, 50%, and even more preferably, are greater than about 75% homologous in sequence to a holin protein. Alternatively, peptide mimics of a holin could be used that retain critical molecular recognition elements, although peptide bonds, side chain structures, chiral centers and other features of the parental active protein sequence may be replaced by chemical entities that are not native to the holin protein yet, nevertheless, confer activity.
In determining whether a polypeptide is substantially homologous to a holin polypeptide, sequence similarity may be determined by conventional algorithms, which typically allow introduction of a small number of gaps in order to achieve the best fit. In particular, “percent homology” of two polypeptides or two nucleic acid sequences is determined using the algorithm of Karlin and Altschul. Such an algorithm is incorporated into the NBLAST and XBLAST programs of Altschul, et al. (J. Mol. Biol. 215, 403 (1990)). BLAST nucleotide searches may be performed with the NBLAST program to obtain nucleotide sequences homologous to a nucleic acid molecule of the invention. Equally, BLAST protein searches may be performed with the XBLAST program to obtain amino acid sequences that are homologous to a polypeptide of the invention. To obtain gapped alignments for comparison purposes, Gapped BLAST is utilized as described in Altschul, et al. (Nucleic Acids Res. 25, 3389 (1997)). When utilizing BLAST and Gapped BLAST programs, the default parameters of the respective programs (e.g., XBLAST and NBLAST) are employed. See www.ncbi.nlm.nih.gov for more details.
In one embodiment, a nucleic acid of the invention comprises a metal or metal ion inducible promoter operably-linked to a nucleic acid sequence encoding a P22 phage holin. In another embodiment, the nucleic acid comprises a metal or metal ion inducible promoter operably-linked to both a nucleic acid sequence encoding a P22 phage holin and a nucleic acid sequence encoding at least one endolysin. In yet another embodiment, the nucleic acid comprises a metal or metal ion inducible promoter operably-linked to a nucleic acid sequence encoding a P22 phage holin and a second promoter operably-linked to a nucleic acid sequence encoding at least one endolysin.
iv. Endolysin
In some embodiments, a nucleic acid of the invention comprises at least one endolysin. In other embodiments, a nucleic acid of the invention comprises at least two endolysins. In yet another embodiment, a nucleic acid of the invention comprises at least three endolysins. In still another embodiment, a nucleic acid of the invention may comprise at least four endolysins. As used herein, “endolysin” refers to a protein capable of degrading the peptidoglycan layer of a bacterial cell wall. Generally speaking, the term endolysin encompasses proteins selected from the group consisting of lysozyme or muramidase, glucosaminidase, transglycosylase, amidase, and endopeptidase. Exemplary endolysins do not affect the cell until after the first protein creates lesions in the lipid membranes. Stated another way, the accumulation of endolysins in the cytosol of a bacterium will typically not substantially impair the growth rate of the bacterium. In another exemplary embodiment, the endolysin has a high enzymatic turnover rate. In yet another exemplary embodiment, the endolysin is from a gram positive bacterium. Because the cell walls of gram positive bacteria typically have a thicker peptidoglycan layer, an endolysin from a gram positive bacteria might be expected to have a higher enzymatic turnover rate.
In one embodiment, at least one endolysin is from a bacteriophage. In certain embodiments, suitable endolysins may be from phages detailed in section I(c)(ii) above in reference to the first protein. In another embodiment, at least one endolysin is from a Salmonella bacteriophage. In yet another embodiment, at least one endolysin is from a P22 phage. In still yet another embodiment, at least one endolysin is from a λ phage. In an alternative embodiment, at least one endolysin is gp19 from a P22 phage. In another alternative, a nucleic acid of the invention comprises gp19 and gp15 from a P22 phage. In some embodiments, at least one endolysin is R from a λ phage. In other embodiments, a nucleic acid of the invention comprises R and Rz from a λ phage. In certain embodiments, a nucleic acid of the invention comprises gp19, gp15, R, and Rz.
Additionally, an endolysin may be a protein described above with at least one, or a combination of one or more, nucleic acid deletions, substitutions, additions, or insertions which result in an alteration in the corresponding amino acid sequence of the encoded endolysin protein, such as a homolog, ortholog, mimic or degenerative variant. Such an endolysin may be generated using recombinant techniques such as those described in section I(c)(ii) above in reference to a first protein. In addition to having a substantially similar biological function, a homolog, ortholog, mimic or degenerative variant of an endolysin suitable for use in the invention will also typically share substantial sequence similarity to an endolysin protein. In addition, suitable homologs, ortholog, mimic or degenerative variants preferably share at least 30% sequence homology with an endolysin protein, more preferably, 50%, and even more preferably, are greater than about 75% homologous in sequence to an endolysin protein. Alternatively, peptide mimics of an endolysin could be used that retain critical molecular recognition elements, although peptide bonds, side chain structures, chiral centers and other features of the parental active protein sequence may be replaced by chemical entities that are not native to the endolysin protein yet, nevertheless, confer activity. Percent homology may be calculated as described in section I(c) above.
v. Additional Components
In certain embodiments, nucleic acids of the invention may further comprise additional components, such as a marker, a spacer domain, and a flanking sequence.
A. Markers
In one embodiment, a nucleic acid of the invention comprises at least one marker. Generally speaking, a marker encodes a product that the host cell cannot make, such that the cell acquires resistance to a specific compound, is able to survive under specific conditions, or is otherwise differentiable from cells that do not carry the marker. Markers may be positive or negative markers. In some embodiments, a nucleic acid of the invention may comprise both a positive marker and a negative marker. In certain embodiments, the marker may code for an antibiotic resistance factor. Suitable examples of antibiotic resistance markers may include, but are not limited to, those coding for proteins that impart resistance to kanamycin, spectromycin, streptomycin, neomycin, gentamicin (G418), ampicillin, tetracycline, and chloramphenicol. Additionally, the sacB gene may be used as a negative marker. The sacB gene is lethal in many bacteria when they are grown on sucrose media. Additionally, fluorescent proteins may be used as visually identifiable markers. Generally speaking, markers may be present during construction of the strains, but are typically removed from the final constructs. Proteins can also be marked by adding a sequence such as FLAG, HA, His tag, that can be recognized by a monoclonal antibody using immunological methods. In some embodiments, a marker may be a unique identifier of a genetically modified cyanobacterium.
B. Spacer Domain
Additionally, a nucleic acid of the invention may comprise a Shine-Dalgarno sequence, or a ribsome binding site (RBS). Generally speaking, a RBS is the nucleic acid sequence in the mRNA that binds to a 16s rRNA in the ribosome to initiate translation. For Gram-negative bacteria, the RBS is generally AGGA. The RBS may be located about 8 to about 11 bp 3′ of the start codon of the first structural gene. One skilled in the art will realize that the RBS sequence or its distance to the start codon may be altered to increase or decrease translation efficiency.
C. Flanking Sequence
Nucleic acid constructs of the invention may also comprise flanking sequences. The phrase “flanking sequence” as used herein, refers to a nucleic acid sequence homologous to a chromosomal sequence. A construct comprising a flanking sequence on either side of a construct (i.e., a left flanking sequence and a right flanking sequence) may homologously recombine with the homologous chromosome, thereby integrating the construct between the flanking sequences into the chromosome. Generally speaking, flanking sequences may be of variable length. In an exemplary embodiment, the flanking sequences may be between about 300 and about 500 bp. In another exemplary embodiment, the left flanking sequence and the right flanking sequence are substantially the same length. For more details, see the Examples.
vi. Plasmids
A nucleic acid construct of the invention may comprise a plasmid suitable for use in a bacterium. Such a plasmid may contain multiple cloning sites for ease in manipulating nucleic acid sequences. Numerous suitable plasmids are known in the art.
Non-limiting examples of first inducible promoters, first proteins, second promoters, and endolysin combinations are listed in Table A below.
A recombinant bacterium of the invention may be altered so as to modify the structure of the fatty acids produced. For instance, the chain length, the chain saturation, and the branching of the fatty acid may be modified. In certain embodiments, chain length may be altered by the choice of TE, as detailed above and in the examples. Furthermore, the expression of a TE may alter chain saturation.
A bacterium of the invention may be altered to produce branch chain fatty acids. Typically, such a bacterium will express one or more nucleic acid sequences encoding a protein involved in the production of branch chain fatty acids, such as a branched-chain amino acid aminotransferase, a branched-chain α-keto acid dehydrogenase complex, β-ketoacyl-ACP synthase III, acyl carrier protein, and β-ketoacyl-ACP synthase II. Suitable, non-limiting examples are detailed in Table 9 below.
A bacterium of the invention may also be altered so as to express a nucleic acid encoding a lipolytic enzyme. Such an enzyme may degrade membrane lipids into free fatty acids. This increases the amount of free fatty acids harvestable from a bacterium, and makes the harvest less labor intensive. Suitable lipolytic enzymes may include a galactolipase and/or a phospholipase. Examples of galactolipases and phospholipases are known in the art. In one embodiment, a lipolytic enzyme from Staphylococcus hyicus may be used. In another embodiment, a lipolytic enzyme from Fusarium oxysporum may be used. In yet another embodiment, an enzyme derived from guinea pigs (GPLRP2) may be used. In addition, a lipase encoded by the Synechocystis gene lipA (sll1969) can also be used to degrade membrane lipids.
To ensure that the synthesis of a lipolytic enzyme does not result in the premature lysis of a bacterium, the enzyme may be placed under the control of an inducible promoter. Suitable promoters may include a nickel inducible promoter and a CO2 inducible promoter. For more details, see the Examples.
The present invention also provides a recombinant bacterium comprising at least one chromosomally encoded essential nucleic acid sequence, wherein the essential nucleic acid sequence is altered so that it is not expressed, and at least one extrachromosomal vector. An “essential nucleic acid” is a native nucleic acid whose expression is necessary for cell viability. Consequently, a bacterium of the invention is non-viable if an essential nucleic acid sequence is not expressed. Therefore, the bacterium of the invention further comprises at least one extrachromosomal vector. The vector comprises a nucleic acid sequence, that when expressed, substantially functions as the essential nucleic acid. Hence, the bacterium is viable when the vector is expressed. This promotes stable maintenance of the vector in the absence of any exogenously supplied selective condition.
(a) Chromosomally Encoded Essential Nucleic Acid that is Altered so that it is not Expressed
A recombinant bacterium of the invention comprises at least one chromosomally encoded essential nucleic acid sequence, wherein the essential nucleic acid sequence is altered so that it is not expressed. As described above, an essential nucleic acid is a native nucleic acid whose expression is necessary for cell viability. In some embodiments, an individual nucleic acid sequence is not essential, but the combination of one or more sequences, together, is essential. Stated another way, if the nucleic acid sequences in an essential combination are altered, so that they are not expressed, the cell is non-viable.
A nucleic acid sequence that encodes a protein necessary for the formation of the peptidoglycan layer of the cell wall may be an essential nucleic acid. In one embodiment, an essential nucleic acid encodes a protein involved in D-alanine synthesis. For example, an essential nucleic acid may encode one or more alanine racemase proteins. In another embodiment, an essential nucleic acid may encode a protein involved in D-glutamate synthesis. In yet another embodiment, an essential nucleic acid may encode a protein involved in muramic acid synthesis. Such nucleic acid sequences are known in the art, and non-limiting examples may include asd, murA, murl, dapA, dapB, and alr. In an alternative embodiment, a nucleic acid sequence that encodes a protein whose metabolic activity is essential for growth or cell division may be an essential nucleic acid. Such nucleic acid sequences are also known in the art, and non-limiting examples may include pol nucleic acid sequences encoding DNA polymerases, rpo nucleic acid sequences encoding RNA polymerases, and rps and rpl nucleic acid sequences encoding ribosomal protein subunits. Since Synechocystis is a photoautotroph normally living in fresh water and marine environments it is capable of synthesizing all amino acids necessary for protein synthesis, all purines and pyrimidines needed for nucleic acid syntheses and all vitamins needed as co-factors for various enzymes essential for metabolic activities and survival. As such, nucleic acid sequences encoding functions for amino acid synmthesis, purine synthesis, pyrimidine synthesis and vitamin synthesis are essential nucleic acids.
A recombinant bacterium of the invention may comprise more than one chromosomally encoded essential nucleic acid sequence that is altered so that it is not expressed. For instance, a recombinant bacterium may comprise two, three, four, five, or more than five different chromosomally encoded altered essential nucleic acid sequences.
Methods of making a recombinant bacterium comprising a chromosomally encoded essential nucleic acid sequence that is altered so that it is not expressed are known in the art and detailed in the examples. Non-limiting examples of suitable alterations are detailed below.
i. Essential Nucleic Acid Encoding a Protein Involved in D-Alanine Synthesis
In one embodiment, an essential nucleic acid may encode a protein involved in D-alanine synthesis, since D-alanine is a required constituent of the peptidoglycan layer of a bacterial cell wall. Gram-positive bacteria comprise only one alanine racemase, an enzyme necessary for D-alanine synthesis. Consequently, if the essential nucleic acid sequence encoding the Gram-positive alanine racemase is altered so that it is not expressed, the bacterium is non-viable. Gram-negative bacteria, however, may comprise one more than one alanine racemases. Consequently, when the Gram-negative bacteria comprises two alanine racemases, it is the combination of both sequences that is essential, and the nucleic acid sequences encoding both alanine racemases need to be altered so that both sequences are not expressed. Suitable alterations may include deletion of two nucleic acid sequences encoding an alanine racemase. For instance, the combination of the deletions Δalr and ΔdadB will alter the essential combination such that neither racemase is expressed. Advantageously, an extrachromosomal vector need only encode one racemase to restore to the Gram-negative bacterium. In preferred embodiments, the Gram-negative bacterium is Synechocystis, and the essential nucleic acid encoding a protein involved in D-alanine synthesis is alr encoding alanine racemase.
ii. Essential Nucleic Acid Encoding a Protein Involved in Muramic Acid Synthesis
In another embodiment, an essential nucleic acid may encode a protein involved in muramic acid synthesis, as muramic acid is another required constituent of the peptidoglycan layer of the bacterial cell wall. For example, an essential nucleic acid may be murA. It is not possible to alter murA by deletion, however, because a ΔmurA mutation is lethal and can not be isolated. This is because the missing nutrient required for viability is a phosphorylated muramic acid that cannot be exogenously supplied because most, if not all, bacteria cannot internalize it. Consequently, the murA nucleic acid sequence may be altered to make expression of murA dependent on a nutrient that can be supplied during the growth of the bacterium. For instance, cyanobacteria may be altered to make expression of murA dependent on arabinose. For example, the alteration may comprise a ΔPmurA.:TT araC PBAD murA deletion-insertion mutation. During in vitro growth of the bacterium, this type of mutation makes synthesis of muramic acid dependent on the presence of arabinose in the growth medium. During growth of the bacterium in culture, arabinose may be supplied to allow for growth of the bacterium. However, the bacterium is non-viable in a natural environment where arabinose is not present. Since cyanobacteria are unable to internalize arabinose, the bacterium may be altered to enable uptake of arabinose. For instance, cyanobacteria may be altered by introducing an araE gene from E. coli that encodes an arabinose-uptake protein.
iii. Essential Protein Involved in D-Glutamate Synthesis
In yet another embodiment, an essential nucleic acid may encode a glutamate racemase, an enzyme essential for the synthesis of D-glutamic acid, which is another required constituent of the peptidoglycan layer of the bacterial cell wall. An essential nucleic acid encoding a glutamate racemase may be altered by deletion. For instance, the mutation Δmurl alters the nucleic acid sequence encoding glutamate racemase so that it is not expressed. In some instances, isolation of such mutations might require introduction of a nucleic acid sequence encoding a glutamate transporter, such as gltS, to facilitate uptake of glutamate. Such modifications are likely to facilitate introducing mutl mutations into cyanobacteria since cyanobacteria are seldom in an environment with exogenous amino acids and therefore often lack enzymes to facilitate amino acid uptake.
iv. Essential Protein Involved in DAP Synthesis
In still another embodiment, an essential nucleic acid may encode a protein involved in the synthesis of diaminopimelic acid (DAP). Various nucleic acid sequences are involved in the eventual synthesis of DAP (
v. More than One Chromosomally Encoded Essential Nucleic Acid that is Altered
In exemplary embodiments of the invention, a recombinant bacterium may comprise more than one chromosomally encoded essential nucleic acid sequence that is altered so that it is not expressed and at least one extrachromosomal vector.
For instance, in one embodiment, a recombinant bacterium may comprise a first chromosomally encoded essential nucleic acid that is altered so that the first essential nucleic acid is not expressed, a second chromosomally encoded essential nucleic acid that is altered so that the second essential nucleic acid is not expressed, a first extrachromosomal vector, the vector comprising a nucleic acid comprising a nucleic acid sequence, that when expressed, substantially functions as the first essential nucleic acid sequence, and a second extrachromosomal vector, the vector comprising a nucleic acid sequence, that when expressed, substantially functions as the second essential nucleic acid sequence.
In another embodiment, a recombinant bacterium may comprise a first chromosomally encoded essential nucleic acid that is altered so that the first essential nucleic acid is not expressed, a second chromosomally encoded essential nucleic acid that is altered so that the second essential nucleic acid is not expressed, a third chromosomally encoded essential nucleic acid that is altered so that the third essential nucleic acid is not expressed, a first extrachromosomal vector, the vector comprising a nucleic acid comprising a nucleic acid sequence, that when expressed, substantially functions as the first essential nucleic acid sequence, a second extrachromosomal vector, the vector comprising a nucleic acid sequence, that when expressed, substantially functions as the second essential nucleic acid sequence, and a third extrachromosomal vector, the vector comprising a nucleic acid sequence, that when expressed, substantially functions as the third essential nucleic acid sequence.
In yet another embodiment, a recombinant bacterium may comprise a first chromosomally encoded essential nucleic acid that is altered so that the first essential nucleic acid is not expressed, a second chromosomally encoded essential nucleic acid that is altered so that the second essential nucleic acid is not expressed, a third chromosomally encoded essential nucleic acid that is altered so that the third essential nucleic acid is not expressed, a fourth chromosomally encoded essential nucleic acid that is altered so that the fourth essential nucleic acid is not expressed, a first extrachromosomal vector, the vector comprising a nucleic acid comprising a nucleic acid sequence, that when expressed, substantially functions as the first essential nucleic acid sequence, a second extrachromosomal vector, the vector comprising a nucleic acid sequence, that when expressed, substantially functions as the second essential nucleic acid sequence, a third extrachromosomal vector, the vector comprising a nucleic acid sequence, that when expressed, substantially functions as the third essential nucleic acid sequence, and a fourth extrachromosomal vector, the vector comprising a nucleic acid sequence, that when expressed, substantially functions as the fourth essential nucleic acid sequence.
In other embodiments, a recombinant bacterium may comprise more than four chromosomally encoded essential nucleic acid sequences that are each altered so that they are not expressed, and more than four corresponding extrachromosomal vectors.
By way of non-limiting example, suitable alterations in essential nucleic acid sequences may include an alteration selected from the group consisting of ΔasdA, a ΔdapA mutation, a ΔdapB mutation, a ΔdapA mutation with a ΔdapB mutation, a Δalr mutation, a ΔPmurA::TT araC PBAD murA deletion-insertion mutation, a Δmurl mutation, a ΔaroA mutation, a ΔaroC mutation, a ΔaroD mutation, a ΔilvC mutation, and a ΔilvE mutation. For instance, a bacterium may comprise two, three, four, five, or more than five alterations in an essential nucleic acid sequence selected from the group consisting of ΔasdA, any Δdap mutation, a Δalr mutation, a ΔPmurA::TT araC PBAD murA deletion-insertion mutation, a Δmurl mutation, a ΔaroA mutation, a ΔaroC mutation, a ΔaroD mutation, a ΔilvC mutation, and a ΔilvE mutation.
A recombinant bacterium of the invention also comprises an extrachromosomal vector. The vector comprises a nucleic acid sequence that when expressed, substantially functions as the chromosomally encoded essential nucleic acid that is not expressed. Furthermore, the vector typically also comprises one or more nucleic acid sequences that encode metabolic activity capable of enhancing the productivity of the cyanobacterial strain. As used herein, “vector” refers to an autonomously replicating nucleic acid unit. The present invention may be practiced with any known type of vector, including viral, cosmid, phasmid, and plasmid vectors. The most preferred type of vector is a plasmid vector. An exemplary type of plasmid may use one or more of the small plasmid replicons normally present in a Synechocystis species as described in Section III below.
The term “extrachromosomal,” as used herein, refers to the fact that the vector is not contained within the bacterium's chromosomal DNA. The vector may comprise some sequences that are identical or similar to chromosomal sequences of the bacterium, however, the vectors used herein do not integrate with chromosomal sequences of the bacterium.
As is well known in the art, plasmids and other vectors may possess a wide array of promoters, multiple cloning sequences, transcription terminators, etc., and vectors may vary in copy number per bacterium. Selection of a vector may depend, in part, on the desired level of expression of the nucleic acid sequence substantially functioning as the essential nucleic acid or functioning to specify the metabolic activity of importance to enhance the productivity of the cyanobacterial strain.
Additionally, vector copy number may be increased by selecting for mutations that increase plasmid copy number. These mutations may occur in the bacterial chromosome but are more likely to occur in the vector.
In some embodiments, a toxin-antitoxin stability system may be used to generate the balanced-lethal system. The antitoxins may be proteins or antisense RNAs that counteract the toxins. Non-limiting examples of toxin-antitoxin gene families from cyanobacteria may be the antisense RNA-regulated toxin-antitoxin gene families hok/sok and ldr, or SOS-induced toxins such as SymE.
Although extrachromosomal vectors, such as plasmids, may be designed with unique nucleotide sequences, there is some potential for vector-vector recombination to occur that might lead to deletion of and/or alterations in one or more nucleic acid sequences encoding metabolic activity capable of enhancing the productivity of the cyanobacterial strain. Accordingly, in some embodiments, a recombinant bacterium of the invention may be deficient in one or more of the enzymes that catalyzes recombination between extrachromosomal vectors. If a bacterium comprises only a single extrachromosomal vector, then such mutations are not necessary. If two or more extrachromosomal vectors are used, however, then the recombinant bacterium may be modified so that one or more recombination enzymes known to catalyze vector-vector recombination are rendered non-functional.
In certain embodiments, the recombination enzymes do not participate in recombinations involving chromosomal nucleic acid sequences. For instance, the recombinant bacterium may comprise a ΔrecF and a ΔrecJ mutation. These mutations do not alter the attributes of the recombinant bacterium. One of skill in the art will appreciate that other recombination enzymes known to catalyze vector-vector recombination but not to participate in recombinations involving chromosomal nucleic acid sequences may be targeted for deletion or mutation in addition to RecF and RecJ.
Alternatively, the recombinant bacterium may be modified by introducing a ΔrecA mutation that prevents all recombination, whether between vectors or chromosomal nucleic acid sequences. Introducing a ΔrecA mutation may enhance sensitivity of the bacteirum to ultraviolet radiation in sunlight and render the cyanobacterial strain unsuitable for producing fatty acids or other biofuels or biofuel precursors. Alternatively, the recombinant bacterium may be modified by introducing mutations into recF and or recJ since they do not confer sensitivity to ultraviolet light.
The present invention also provides plasmids derived from one or more of the small plasmid replicons normally present in a Synechocystis species. For instance, gene functions responsible for stable maintenance of the small plasmid replicon in Synechocystis may be identified using methods known in the art. In addition, it may be determined if loss of the native Synechocystis plasmid is lethal or the plasmid is dispensable.
In some embodiments, a shuttle vector may be developed that will replicate in both E. coli and Synechocystis.
Vectors of the invention generally possess a multiple cloning site for insertion of a nucleic acid sequence that may be operably-linked to the promoter sequence and generally posses a transcription terminator (TT) sequence after a coding region. Preferably, vectors used herein do not comprise antibiotic resistance markers to select for maintenance of the vector.
Another aspect of the present invention is a method for the production of fatty acids. Generally speaking, a method of the invention comprises culturing a bacterium of the invention as detailed in section I above. Methods of culturing a cyanobacterium are known in the art and detailed in the examples. Fatty acids produced by the bacterium may be extracted from the culture media or culture biomass.
Methods of extracting the secreted fatty acids from the extracellular milieu are know in the art and detailed in the Examples. For instance, the fatty acids may be pipetted, filtered, and/or skimmed from the culture media. In addition, the culture media may be treated to extract any remaining fatty acids dissolved in the media. Such treatment is described in the Examples. Briefly, the media may be acidified and extracted with an organic solvent, such as hexane. The organic phase may then be separated and dried to give the fatty acids. In some embodiments, the media is extracted more than once with the organic solvent. For instance, the media may be extracted two, three, four or five times.
Unsecreted intracellular FFAs and lipids may also be extracted using means known in the art and detailed in the Examples.
The yield of fatty acids from a bacterium of the invention will generally be about 10, 20, 30, 40, 50, 60, 70, 80, 90, 100, 110, 120, 130, 140, 150, 160, 170, 180, 190, 200, 210, 220, 230, 240, 250, 260, 270, 280, 290, 300, 310, 320, 330, 340, 350, 360, 370, 380, 390, 400 or more than 400 times the yield of fatty acids from a wild-type bacterium. In one embodiment, the yield is greater than 100, 200, 300, 400, 500, 600, 700, 800, 900, or 1000 times the yield of a wild-type bacterium. In another embodiment, the yield is greater than about 1000, 1500, 2000, 2500, 3000, or 3500 times the yield of a wild-type bacterium.
The term “cell wall”, as used herein, refers to the peptidoglycan layer of the cell wall. Stated another way, “cell wall” as used herein refers to the rigid layer of the cell wall.
The term “operably-linked”, as used herein, means that expression of a gene is under the control of a promoter with which it is spatially connected. A promoter may be positioned 5′ (upstream) of a gene under its control. The distance between the promoter and a gene may be approximately the same as the distance between that promoter and the gene it controls in the gene from which the promoter is derived. As is known in the art, variation in this distance may be accommodated without loss of promoter function.
The term “promoter”, as used herein, may mean a synthetic or naturally-derived molecule which is capable of conferring, activating or enhancing expression of a nucleic acid in a cell. A promoter may comprise one or more specific transcriptional regulatory sequences to further enhance expression and/or to alter the spatial expression and/or temporal expression of same. A promoter may also comprise distal enhancer or repressor elements, which can be located as much as several thousand base pairs from the start site of transcription. In some embodiments, activators may bind to promoters 5′ of the −35 RNA polymerase recognition sequence, and repressors may bind 3′ to the −10 ribosome binding sequence.
The following examples are included to demonstrate preferred embodiments of the invention. It should be appreciated by those of skill in the art that the techniques disclosed in the examples that follow represent techniques discovered by the inventors to function well in the practice of the invention. Those of skill in the art should, however, in light of the present disclosure, appreciate that many changes can be made in the specific embodiments that are disclosed and still obtain a like or similar result without departing from the spirit and scope of the invention, therefore all matter set forth or shown in the accompanying drawings is to be interpreted as illustrative and not in a limiting sense.
The following examples illustrate various iterations of the invention.
6803 is an ideal organism to genetically manipulate due to its high natural transformation efficiency and high double crossover homologous recombination efficiency (Kufryk, Sachet et al. 2002). Cai and Wolk have introduced a method to counter-select cells that retain the drug markers, thus enabling construction of the desired multiple recombinants (Cai and Wolk 1990). Suicide vectors harboring a positive selection marker (e.g., Kmr) and a counter selection marker (e.g., sacB) are widely applied in a two-step gene deletion and insertion for 6803 without leaving any drug marker residuals (
Conditions for transformation and isolation of a recombinant by selection for a dominant phenotype have been well documented (Zang, Liu et al. 2007), and include DNA concentrations, cell competence, DNA uptake timing, and light dependence. However, few papers discuss the optimal conditions and efficiency of the counter selecting transformation and selection by a recessive phenotype (e.g., sucrose resistance by removal of sacB). By testing different plasmid concentrations, transformation and segregation times, and selection pressure levels, the inventors significantly improved the overall transformation efficiency for generating recombinants. A highly optimized experimental protocol for 102-105 transformants/μg DNA, which is sufficient for genetic engineering purposes, was developed. For example, after transformation, segregation without applying selective pressure is necessary to permit formation of ‘homozygous’ mutants that will phenotypically display the selected marker. The phenotypic and segregation lags for sucrose survival depend on the toxic or adverse effects of the insertion genes on the growth rate of the resulting mutant. Toxic genes or those causing slow growth rate usually require a longer segregation period. Additionally, the colonies grown from fully segregated single cells were isolated and tested to ensure that all the resulting strains are genetically homogenous. The details of optimized transformation are described further below. By using the optimized genetic manipulation methods, a series of strains with FFA secreting ability were constructed (Table 1).
Synechocystis sp. PCC
E. coli *tesA gene fused with an
In 6803, the slr1609 gene was previously annoted as encoding a long-chain acyl-CoA ligase and designated as fadD. Acyl-CoA ligase is the first key enzyme in the beta-oxidation pathway, which is the key enzyme for FFA consumption. Based on this information, we decided to delete this gene to save FFA product from being degraded. We therefore replaced slr1609 with PpsbA2 'tesA by SacB/KmR intermediate deletion/insertion in the strain SD216 described in Table 1. We developed five more generations of modifications based on strain SD216. Based on studies in yeast, deletion of genes for acyl-CoA synthetase isozymes led to efficient fatty acid secretion (Scharnewski, Pongdontri et al. 2008), and introduction of acyl-CoA synthetase mutation ameliorated a fatty acid secretion phenotype (Michinaka, Shimauchi et al. 2003). Indeed, impairment of long-chain acyl-CoA synthetase activity is expected to increase the fatty acid concentration in the cell and to enhance fatty acid secretion. In early 2010, it was reported that slr1609 is not an acyl-CoA (synthetase) ligase gene as previously thought and actually encodes an acyl-ACP synthetase, which ligates FFA and ACP together as acyl-ACP (Kaczmarzyk and Fulda 2010). The gene name is therefore aas instead of fadD and the strain genotypes in Table 1 have been corrected based on this new information. This aas catalysed reaction goes exactly in the opposite direction of FFA production catalyzed by thioesterases. The paper also showed that just by deleting slr1609, 6803 will lose the ability to recycle FFA and leaks FFA into the medium. The new published data proves that deletion of slr1906 is very important for FFA accumulation and secretion and substantiates the importance of our discovery of the benefits of deleting this gene as a starting point for constructing 6803 stains for improved fatty acid production and secretion. Barring the existence of an as yet unknown fatty acid degradation pathway, accumulation/secretion or conversion to acyl-CoA are the only two main destinations for fatty acids in 6803. Using the Kmr-sacB insertion, the full function of slr1609 in SD214 and SD216 and their derived strains (i.e., SD225, SD232, SD243, SD249 and SD277) was interrupted (
A polyhydroxyalkanoate (PHA) synthesis deletion strain SD207 was constructed by interrupting two PHA synthesis genes slr1993/1994 in 6803, which encode PHA-specific beta-ketothiolase and PHA-specific acetoacetyl-CoA reductase, respectively. PHA synthesis consumes the carbon resources from the Acetyl-CoA pool, thus inactivation of the PHA synthesis pathways will shut off the carbon flux towards the unnecessary byproducts. PHA inclusions in 6803 can be stained by Nile Red, a neutral lipid specific fluorescence dye. Flow cytometry analysis of the Nile Red stained Δslr1993/1994 strain SD207 showed disappearance of the PHA emission compared to the wild-type 6803 (
We constructed six generations of 6803 SD strains for FFA secretion (
Acetyl-CoA carboxylase (ACC) has been postulated as the rate-controlling enzyme in fatty acid biosynthesis (Davis, Solbiati et al. 2000). For the 2nd modification (SD223 and SD225), an artificial operon Pcpc accB accC Prbc accD accA was introduced into SD215 and SD216 to overproduce ACC (
In the 3rd generation strain SD232, Uc fatB1, a 12:0 acyl-ACP TE encoding gene from Umbellularia californica (Pollard, Anderson et al. 1991) was synthesized in an artificial operon *PpsbA2 Uc fatB1 and inserted to knock out sll1951 (
We initially conducted studies using the nickel-inducible FFA production and secretion strains SD215, SD219, SD223 and SD228 to compare with their constitutive counterpart stains SD216, SD220, SD225 and SD229 (Table 2). Strains with inducible synthesis had negligible FFA either intracellularly or extracellularly until after induction by addition of 7 μM Ni2+ to the culture (OD730 nm˜1.0). However, by the time cultures reached 2×109 cells/ml the amounts of intracellular and extracellular FFAs were very similar for each pair with the amount of intracellular FFAs decreasing by 35% for the constitutive strains SD216 versus SD229 and 28% for the inducible strains SD215 versus SD228. In regard to secreted FFAs, there was a 7.7-fold increase for the constitutive strains SD229 compared to SD216 and a 11.4-fold increase for the inducible strains SD228 compared to SD215.
aThe genotypes and constructions of the strains were confirmed by PCR analysis.
bFor the constitutive strains, the late-log phase FFA secreting efficiency was measured for a 24 h interval with the culture density starting at about 1.5 × 108 cells/mL and reaching about 2 × 108 cells/mL. For the inducible strains, the induced FFA secreting efficiency was measured after addition of 7 μM Ni2+ to the culture (OD730 nm ~1.0).
c2 × 109 cells in 10 mL hexane treated culture were extracted by Folch method.
dThe parent for SD216 and SD215 is wild-type 6803, the parents for the other strains are the strains on the row above the new strain.
During the growth of the FFA-secreting strains (SD232 for example) from a single cell to a 200 mL cell culture (8×108 cells/mL, 1.2 g of dry weight/liter), cell permeability was revealed microscopically after staining cells with the vital dye SYTOX Green (Roth, Poot et al. 1997). Our experiment showed that only 1% of the permeable cells (sorted by flow cytometry for green fluorescence) were able to form colonies on BG-11 agar plates, suggesting that they were damaged.
SD232 colonies (containing 8×105 cells) descended from a single cell and grown on BG-11 agar plates for 10 days contained 0.5% cells permeable to the dye (
FFA secretion was observed for the constitutively 'tesA expressing strains, including SD216, SD220, SD225, SD229, SD232, SD243, SD249, and SD277 (
aThe genotypes and constructions of the strains were described in Table 1.
bThe doubling times were determined in separate experiments. Briefly, daily colony forming unit (CFU) or OD730 nm measurements for exponential growing cultures (density below 108 cells/ml) were used for the calculation for the doubling times of different strains.
cThe final cell density was determined by (CFU).
d20 mL culture was extracted by hexane and analyzed by GC for FFA amount.
eThe secreted FFA per cell was calculated from the final cell density and the FFA secretion yield.
fThe intracellular unsecreted FFA were extracted Folch method, and calculated with the final cell density.
The chemical composition of the intracellular and secreted FFAs was analyzed by GC (Table 4). First, the percentage of C10:0-C14:0 FFAs increased with successive generations of strain constructions. For example, C14:0 and C12:0 increased from 16.5% (SD216 secretion) to 47.3% (SD249 secretion). Second, it was observed that the percentage of unsaturated fatty acids decreased in the FFA-overproducing strains. Third, we observed that the percentage of middle chain FFAs was higher in the secreted extracts than the intracellular unsecreted extracts (for example, the C12:0 and C14:0 FFA weight % for SD232, SD243 and SD249 in Table 4), suggesting that the middle chain FFAs are easier to secrete and especially when fully saturated.
aThe number before the colon in the fatty acid name refers to the number of carbons, and the number after the colon refers to the number of double bonds.
bThe fatty acid percentages for SD100 membrane lipids were obtained from Wada's report (Wada and Murata 1990) as a baseline for the SD100 fatty acid profile. The fatty acid percentages for the other samples were based on the free fatty acids. Secreted samples (Secr) mean secreted FFAs in the culture medium extracted by hexane without disrupting the cells. Cell samples represent the unsecreted FFAs remaining inside the cells extracted from sedimented cells by the Folch method.
ct, trace amount (less than 4%) (Wada and Murata 1990).
dND, not detected.
Cyanobacterial cells often have multiple surface layers including extracellular polysaccaride capsules (Panoff, Priem et al. 1988), surface layers composed of regularly arrayed proteins (Karlsson, Vaara et al. 1983), outer membranes containing lipopolysaccharide (LPS) and numerous outer membrane proteins with diverse functions, rigid cell wall peptidoglycan layers (Hoiczyk and Hansel 2000) and a cytoplasmic membrane (see
Since peptidoglycan layers (Hoiczyk and Hansel 2000) are also polar, imparing their structural integrity should also facilitate the secretion of FFAs. We used the controllable 'tesA strains SD125 and SD228 (S-layer deficient) to evaluate the FFA secretion efficiency after partial inhibition of peptidoglycan layer synthesis by addition of ampicillin (Amp). After induction of 'tesA, the FFA secretion efficiency increased with Amp doses lower than needed to cause cell lysis (Table 5). However, too much damage to the cell walls also killed the cells. Based on these observations, we further genetically weakened peptidoglycan layers by deleting a gene encoding a peptidoglycan assembly protein (PBP2) in SD249, and this also increased FFA secretion (Table 3).
a200 mL cultures (about 1.5 × 108 cells/mL) were induced by adding 7.0 μM Ni2+ to the medium and treated with ampicillin of various concentrations.
We found both disadvantages and advantages of FFA-overproduction to the growth of SD strains. The disadvantage was the fragility of SD cells with CO2 aeration at low cell density, which caused a long lag phase for FFA secreting SD cultures (e.g., SD232). The permeability to the vital dye SYTOX Green indicated that the cytoplasmic membranes of some cells were damaged when CO2 aeration started. In addition, elimination of the surface-layer protein and accumulation of intracellular FFAs contributed to cell fragility. However, the high percentage of damaged cells (
FFA-overproduction strains exhibited less cell damage than wild-type cells at stationary phase (
We showed that two factors significantly increased the FFA secretion quality and quantity. One factor is increasing the cytosol TE activities. Our data have shown that introducing extra TEs into 6803 did increase the FFA secretion efficiency (Table 3, from SD216 to SD249), and shorten the FFA chain length (Table 4, from SD216 to SD249). However, the product chain length of plant TEs in 6803 did not totally match their substrate preference in plants or E. coli. For example, Ch FatB2 produced substantive C12:0 in SD243 (Table 4), while it accumulated C10:0 and C8:0 in Cuphea hookeriana plants and in recombinant E. coli (Dehesh, Jones et al. 1996). Extra TEs also increased the fraction of fully saturated FFAs (Table 4). We hypothesize that the 6803 fatty acid dehydrogenases are located in the di-glycerolipid membranes. When the FFAs were released from acyl-ACP, they would not be incorporated into membrane lipids for further desaturation. This phenomenon is beneficial for biofuel production, since unsaturated carbon chains result in a lower octane rating, and they are less stable and could potentially compromise storage.
Another important factor is weakening cell envelopes to facilitate FFA export. As Hamilton and colleagues have shown, long chain fatty acids perform fast free diffusion by ‘flip-flop’ in the phospholipid bilayer (Hamilton 2007). As membranes are not substantial barriers to FFA secretion, we focused on compromising the layers made of polar molecules, i.e., surface-layer proteins and peptidoglycan layers. Quite possibly as detailed in Example 17, alteration or elimination of some LPS components of the outer membrane will also facilitate FFA secretion. Removal of FFAs from the cyanobacterial cell interior will increase intracellular FFA production, as removing the final products from a reaction system into a metabolic sink will push the equilibrium toward products (Fell 1996) as well as relieve the consequences of feedback inhibition.
In SD277, insertion of the codon-optimized tesA137 gene driven by the artificial promoter Ptrc not only increased FFA secretion from 14.3×10−11 to 20.0×10−11 mg/cell, but also restored growth rate (Table 3). The secreted FFA accounts for about 13% of the biomass (the dry weigh of an SD100 cell is about 150×10−11 mg). We are continuing to investigate a diversity of means for further genetic improvements in our strains to give increased production and secretion of fatty acids. For example, we will enhance the primary FFA pathway genes, and eliminate or down regulate the competing pathway genes (
Bacterial strains, media and growth conditions. All SD strains are derived from Synechocystis sp. PCC 6803. SD strains were grown at 30° C. in BG-11 medium (Rippka et al. 1979) under continuous illumination (140 μmol photons m−2 s−1) and bubbled with 1% CO2-enriched air. The details for growing an SD culture from a colony descended from a single cell are described below. For plating and transformant selection, 50 μg/mL kanamycin or 4.5% (wt/vol) sucrose is added to 1.5% agar plates (wt/vol) and plates were grown under continuous illumination (50 μmol photons m−2 s−1). All of our strains are maintained as concentrated cultures in BG-11 medium with 20% glycerol and stored at −80° C.
Growth and Cell Damage Measurement.
Bacterial growth in liquid culture was monitored spectrophotometrically or by flow cytometry and/or by plating. The relationship between 6803 culture optical density and cell density is used for conversion (
Synthetic Molecular Procedures.
Methods for DNA manipulation are standard (Sambrook et al. 1989). The primers for constructions and genotype verifications are listed in Table 14. DNA sequences were analyzed in the DNA Sequence Laboratory at Arizona State University. Gene segments are synthesized at Genscript (Piscataway, N.J., USA). The nucleic acid sequences of heterogenous genes were redesigned by codon optimization based on the codon frequencies of highly expressed 6803 genes (Table 15) Also the stem-loop hairpins in the predicted mRNA secondary structure were removed to smooth the transcription and to stabilize mRNA by prolonging its half-life (Hamilton 2007).
Gene Deletion and Introducing New Genes into 6803.
Multiple gene modifications are applied into SD strains by using a sacB-KmR cassette (Liu and Curtiss, 2009). Detailed methods are described below.
FFA Separation and Measurement.
The FFAs in the medium are quantitatively separated from the culture medium by hexane, which is unable to release FFAs and other lipids from intact SD100 cells. Twenty mL of culture is acidified by 0.4 mL H3PO4 (1 M) containing 0.4 g NaCl, and extracted with 10 mL hexane. For the unsecreted intracellular FFAs and lipids, the cells are extracted by the Folch method (Folch et al. 1957) for total lipids. The FFA samples were analyzed by GC (Lalman and Bagley, 2004) (
To measure the FFA secretion of the constitutively producing strains, the accumulation of FFAs were measured for late-log phase cultures with a density of about 109 cells/mL. Briefly, during the continuous cultivation of a 50 mL culture, aeration was switched from air to 1% CO2 enriched air when culture density reached about 1.5×108 cells/mL. After the cell density reached about 109 cells/mL with 1% CO2 aeration (2-3 days later), a 20 ml sample was extracted by hexane. To measure the FFA secretion efficiencies of the nickel inducible strains, the secretion efficiencies in one day were calculated from the difference of the FFA secretion values between the before induction and 24 h after induction. In the experiment to evaluate the effect of weakening cell walls on FFA secretion, five subcultures (about 1.5×108 cells/mL) of 200 mL were induced by adding 7.0 μM Ni2+ to the medium and treated with 0, 1, 3, 9, and 25 μg/mL ampicillin.
Statistical Analysis.
Most data were expressed as means±standard deviation. The means were evaluated with one-way Analysis of Variance (ANOVA) for multiple comparisons among groups. Student's T-test was used for pair-wise comparisons. P<0.05 was considered statistically significant.
Growth of an SD Culture Started from a Single Cell Descended Colony.
As described above, cultures of extensively modified SD strains suffered a long lag phase when started at low cell density, probably because a significant percentage of the cells was damaged. A principle mechanism for culturing extensively modified SD strains is therefore not to start cultures below a density of 107 cells/mL, since low cell densities will create a long lag phase prior to exponential growth. We used the following procedure to grow a 6803 culture from a colony descended from a single cell. A single SD colony is picked by a sterilized needle and used to inoculate 1 mL BG-11 medium buffered by 10 mM N-[tris(hydroxymethyl)methyl]-2-aminoethanesulfonic acid (TES) NaOH (pH 8.2) in a glass test tube. The tube is incubated with illumination and intermittent shaking for 2-4 days. These starter cultures can be scaled up by ‘1 into 10’ inoculations after achieving an OD730 nm of 0.6 (108 cells/mL) by adding 10 mL BG-11 medium culture grown in a 50-mL flask with 50 rpm rotation. We added 90 mL BG-11 medium to the 10 mL culture, and grew it in a 250-mL flask with 100 mL/min aeration with air and without shaking. Then we added 900 mL BG-11 medium to the 100 mL culture, and grew it in a 2-L flask with 300 mL/min air sparged with an air stone. Once a 1 L culture achieves OD730 nm˜0.6, aeration is switched from air to 1% CO2-enriched air. This protocol uses TES buffer and air aeration to keep the pH around 8 at the beginning inoculation stages to minimize the lag phase. However, TES buffer reduces FFA production, so TES buffer is not used after the cell density achieves 107 cells/mL. Free fatty acid (FFA)-producing strains need a sufficient CO2 supply and a pH above 8 to maximize FFA-secretion yields. When the 6803 cell density achieves 108 cells/mL, the culture is able to maintain its pH above 8 and can be supplied with CO2-enriched air.
Transformation Procedures for 6803.
We optimized the current genetic modification techniques for 6803 gene deletion, insertion and substitution (Liu and Curtiss, 2009). Suicide vectors harboring a positive selection marker (e.g., KmR) and a counter selection marker (e.g., sacB) are widely applied in a two-step gene deletion and insertion for 6803 without leaving any drug marker residuals (
i. Transformation of Suicide Vectors Containing sacB-KmR Cassette.
About 106 SD cells in 10 μl BG-11 medium are mixed with 400 ng suicide vector DNA containing the sacB-KmR cassette and incubated for 5 h. Then the mixtures were plated onto a filter membrane (Whatman PC MB 90mM 0.4 μm) layered on a BG-11 agar plate. After segregation on the BG-11 plate for about 24 h, the membrane carrying the cyanobacteria was transferred onto a BG-11 plate containing 50 μg/mL of kanamycin. Generally, the colonies appear 4-5 days later. Individual colonies are restreaked with a sterile loop onto a kanamycin BG-11 agar plate and a 4.5% sucrose BG-11 agar plate. Those patches growing on a kanamycin BG-11 agar plate and not growing on a 4.5% sucrose BG-11 agar plate have the correct insertions with the sacB-KmR cassette.
ii. Transformation with Markerless Constructs.
To replace the sacB-KmR selective marker with target gene segments, about 106 sacB-KmR cells in 10 μl BG-11 medium are mixed with 400 ng suicide vector DNA containing the target genes and incubated for 5 h. The mixtures are inoculated into 2 mL buffered BG-11 medium and grown for 3-4 days. 1 mL inoculation is plated onto a 4.5% sucrose-containing BG-11 agar plate. Generally, the colonies appear 5-8 days later. Individual colonies are restreaked onto a kanamycin BG-11 agar plates and a 4.5% sucrose BG-11 agar plates. The patches growing on sucrose plates and not growing on kanamycin plates are positive candidates for further evaluation by PCR.
iii. Confirmation of Replacement.
Cells from a colony are resuspended in 2 μl water in a 200-μl PCR tube. The cell suspension is frozen at −80° C. for 2 min, and then thawed in a 60° C. water bath. This freeze-thaw cycle needs to be performed three times. 1 μl frozen-thawed cell suspension is used as the PCR template for a 30 μl PCR reaction including the primers specific for the inserted gene segments or the deleted region.
Genetic Stability Tests.
When a foreign gene is introduced into 6803, it may cause an adverse effect on growth of the culture and be subjected to gene loss or modification, since any cell losing the genetic alteration will likely have a higher growth rate to eventually take over the population. The genetic stability of foreign genes in 6803 is therefore tested by growing a culture of the strain with periodic dilution and subculturing for at least two months. After this time, the cells from the culture are plated onto BG-11 agar plates to obtain single isolated colonies. One hundred single colonies are picked and tested for all genetic attributes and confirmed for the presence of the foreign gene by PCR as described above. The percentage of positive colonies in the culture reflects the genetic stability of the foreign gene. Genes found to be unstable can be modified to eliminate synthesis of non-functional hydrophobic domains that often are responsible for poor growth due to association with and impairment of cytoplasmic membrane function.
GC for the FFA Produced by the SD Strains.
GC was performed to determine the FFA amount in the hexane extracts (Lalman and Bagley 2004). After 6000 g×10 min centrifugation, 5 ml hexane was taken out from the upper organic layer, filled in a glass tube (13×100 mm, Fisherbrand), and dried on a nitrogen evaporator (N-EVAP111, Organomation Associates Inc.). The dried samples were then re-dissolved by a known volume of hexane and analyzed by gas chromatography (Shimadzu GC 2010) equipped with a Supelco Nukol capillary column (30 m×0.53 mm×0.50 μm) and flame ionization detector (FID).
For the unsecreted intracellular FFAs, the cells are collected by centrifugation, and extracted by the Folch method (Folch, Lees et al. 1957) for total lipids. The intracellular unsecreted FFA were extracted from the cell pellet after hexane extraction, and calculated based on the final cell density.
GC operating conditions were as follows: split ratio 1:5; inject volume 1 μL; helium carrier gas with constant flow rate 30 ml/min; H2 40 ml/min, Air 400 ml/min, make up gas (helium) 5 ml/min; injector and detector temperature 250° C.; and oven temperature started at 150° C. and increased at a rate of 10° C./min to 220° C. and held for 10 min. Each FFA compound was identified by comparing its retention time with that of standard (Sigma, St. Louis, Mo.). Compound concentrations in samples were quantified based on the area under the chromatogram peak in comparison with the standards.
Preliminary results have shown that 6803 is able to overproduce and secrete FFA after genetic modification, and the FFA chain length can be adjusted for the production of biodiesel. On the basis of the FFA secreting 6803 strain, the whole fatty acid synthesis pathway may be genetically optimized to increase the production yields of FFA. The optimization for higher yields includes channeling carbon flow to fatty acid synthesis, attenuating or eliminating the competing pathways, and decreasing repression and feedback inhibition. The structure of secreted fatty acids may also be modified to match the requirement for high quality biofuels.
The 6803 fatty acid synthesis pathways are different from those in heterotrophic bacteria. Starting with photosynthesis, the cyanobacterial carbon source for fatty acid biosynthesis comes from the Calvin-Benson-Bassham cycle rather than from glycolysis cycles (
As shown in
To overexpress the primary pathway genes, the coding sequences of these genes may be optimized by the synthetic procedures described in the Materials and Methods below. The codon optimization is necessary even for some of the native 6803 genes, because by removing the possible stem-loop mRNA structures, the mRNAs may be stabilized, and the transcription and translation may be enhanced, thus increasing the enzyme synthesis efficiency. Then the optimized enzyme encoding open reading frame may be operably linked to a strong promoter or tandem promoter active in different growth phases or under different environmental conditions or in response to different stresses or responsive to different activators or repressors (see Materials and Methods below) resulting in a constitutive or regulatable overexpression cassette. By the methods described in Materials and Methods below, the primary pathway gene overexpression cassettes may be inserted into the 6803 chromosome or contained on a plasmid vector.
Once an overexpression strain is constructed, the genetic stability of the introduced overexpression cassette(s) may be tested by the methods described Materials and Methods above and below. If the overexpressed cassette is genetically stable in 6803, the FFA production efficiency of this strain may be measured by the methods described in the Materials and Methods below, and compared with its parent strain to evaluate the success of the modification(s).
Overexpression of a gene within a pathway does not necessarily result in maximum protein levels. To optimize the whole pathway, the amount of a pathway enzyme should match the other enzymes in this pathway and the overall material-product flow. For example, the four subunits in the ACC complex should have correct stoichiometry to keep the appropriate relative equal molar ratio for maximum synthesis efficiency (Davis, Solbiati et al. 2000). The amount of the ACC proteins also needs to agree with the substrate acetyl-CoA supply and the product malonyl-CoA demand. As was stated before, straight over production of ACCs without presence of exogenous thioesterases in 6803 will not lead to increase of lipid or fatty acid contents, but cause significant adverse effects on the normal growth of 6803. For this reason, the optimal transcriptional and translational levels of the FFA primary pathway genes will be tested to match the maximum FFA-overproduction carbon flow in 6803.
As discussed above, in the FFA-secreting 6803 strains, the FFA synthetic pathway starts at the Calvin-Benson-Bassham cycle and ends at TEs (
By using optimized genetic approaches (Materials and Methods below), we will knock out or knock down the competing pathway genes by using the Kmr-sacB double-crossover mediated gene interruption. The competing genes that may be attenuated include slr1176 for glycogen synthesis; sll0920 for the oxaloacetate pathway; sll0891 for malate synthesis; sll1564 for leucine synthesis, sll0401 for the citrate pathways, slr2132, sll1299, sll0542, slr0091 and sll0090 for the acetate pathways. Since there is reason to believe that these synthetic functions cannot be totally deleted, constructions to reduce gene expression levels (attenuation) may be made by altering the promoter, SD sequence, and/or start codon, etc. and suicide vector methods may be used to make these alterations. Cyanophycin (a non-ribosomally produced amino acid polymer composed of an aspartic acid backbone and arginine side groups) synthetase genes (encoded by slr2002 and slr2001) may be deleted in 6803 to limit the competing carbon flow.
We identified more than 40 dispensible genes in the chrosome of Synechocystis 6803 (Table 12). These genes are all deletable in 6803 without significant adverse effects on the cell growth. Some deletions increased FFA production or secretion. See example 14 for details. The deficient strain(s) may be tested for growth rate and FFA production. By deleting a competing pathway, the FFA production may not be significantly increased. However, if the deficient mutant grows well and has no decrease in FFA production compared with its parent strain, the mutation causing this loss of function may be kept in the 6803 chromosome and serve as an integration platform to insert gene cassettes specifying a beneficial new activity. In this case, the intermediate Kmr-sacB insertion may be replaced by other positive FFA-production gene cassettes.
Eliminating repression of expression of genes is important in channeling resources toward maximal synthesis and secretion of FFAs. If it occurs, the repression may likely be eliminated by the adopted experimental design. For instance, substituting promoters with high activities may be used to maximize gene expression. In so doing wild-type promoters of genes may be replaced with improved promoters. Thus potential for binding of repressors may be eliminated as a problem. In this regard, using active constitutive promoters from other gram-negative bacteria such as E. coli and S. enterica will invariably eliminate the potential for repression of transcription. Contending with feedback inhibition is another matter. To accomplish this, the following strategy is proposed based on the predicted ability of a fusion protein subject to feedback inhibition to lead to an inability to synthesize the peptidoglycan layer of the cell wall resulting in cell death, a powerful selection force. Diaminopimelic acid (DAP), is a unique essential constituent of the peptidoglycan layer of the cell wall. Strains with deletion mutations of the asdA gene are unable to synthesize threonine, methionine and DAP (which by decarboxylation is converted to lysine). Such mutants in the absence of DAP undergo DAP-less death and lyse. A plasmid specifying synthesis of the Asd enzyme from Streptococcus mutans could complement and allow the growth of ΔasdA mutants of either E. coli or Salmonella (Nakayama et al. 1988). Interestingly, if a ΔasdA Salmonella culture containing a plasmid with the S. mutans asdA gene was vigorously shaken, cells lysed. This was because the S. mutans enzyme was unable to form a functional enzyme complex with the Salmonella aspartate kinases that produced the unstable β-aspartyl phosphate which is the substrate for the Asd enzyme. This original observation is serving as an additional means to interfer with peptidoglycan synthesis and thus facilitate secretion of FFAs (see Example 10). To develop the selective enrichment for feedback-resistant mutants may require several steps. First the 6803 asdA gene may be deleted as is also described in Example 10. The efficiency of complementation of this defect may then be determined by introducing plasmids (Materials and Methods below) encoding the 6803, Salmonella and S. mutans asdA genes and determining the complementation efficiency dependant on means of culture growth. Assuming that complementation and survival for slow growth and lethality and lysis for vigorous growth may be observed in some of these complementing pairs, the appropriate asdA gene may be chosen, and a fusion with a 6803 gene whose product is subject to feedback inhibition by intracellular FFAs may be made. For example, the accB gene may be fused with the selected asdA gene on a plasmid vector and the plasmid introduced into a ΔasdA 6803 strain maximally producing FFA precursors but lacking *TesA or other FFA export attributes. It is expected that feedback inhibition will decrease the activity of the fused Asd enzyme and lead to some degree of cell lysis. Feedback inhibition resistant mutants should, however, survive. Minor adjustments may be empirically determined to optimize enzymes with varying degrees of feedback inhibition in WT 6803.
The 6803 FFA products are useful as biofuel precursors. The chain structure of the FFA products can be modified to contain different branch points, level of saturation, and carbon chain length, thus making these products desirable starting materials for the biofuel application. Preliminary results have shown that the saturation level of FFA has been increased from the native 6803 fatty acid profiles, which was postulated by the fact that the 6803 desaturases are located in the cytoplasmic membranes. The chemical composition of 6803 secreted FFAs also showed that introducing middle chain preferring thioesterases enabled 6803 to produce shorter chain FFAs. The quality of 6803 secreted FFAs for biofuel production may be further improved by increasing the amount of shorter chain FFA production (this is discussed in Materials and Methods below) and by engineering 6803 to produce, or to overproduce branched chain fatty acids (brFA).
As described in Materials and Methods below, exogenous genes for the three steps of brFA synthesis may be introduced into 6803 to engineer 6803 to produce brFAs. The first step in forming brFAs is the production of the corresponding α-keto acids by a branched-chain amino acid aminotransferase (IlvE; EC 2.6.1.42). The ilvE gene (slr0032) may be overexpressed in 6803 by the methods described in Materials and Methods below. The ilvE candidates from other bacteria, which are listed in Table 6 may also be introduced. However, in Salmonella there is a series of balanced-lethal and balanced-attenuation plasmid vector-host systems and one of these is dependent on use of IlvE+ plasmid vectors. The aminotransferase reaction encoded by the ilvE gene should be the rate limiting step in brFA biosynthesis in 6803.
The second step, the oxidative decarboxylation of the α-keto acids to the corresponding branched-chain acyl-CoA, is catalyzed by a branched-chain α-keto acid dehydrogenase complex (bkd; EC 1.2.4.4) (Denoya, Fedechko et al. 1995). This complex consists of E1α/β (decarboxylase), E2 (dihydrolipoyltransacylase) and E3 (dihydrolipoyl dehydrogenase) subunits. The bkd genes may also be introduced into 6803. The bkd candidates are listed in Table 6.
Because 6803 does not naturally make brFA, heterogeneous components of fatty acid synthesis machinery with specificity for brFAs need to be introduced into 6803 in the final step. For example, the initiation of brFA biosynthesis utilizes β-ketoacyl-ACP synthase 111 (FabH; EC 2.3.1.41) with specificity for branched chain acyl CoAs (Li et al 2005). Additionally, other components of fatty acid synthesis machineries with specificity for brFAs need to be introduced into 6803, such as acyl carrier protein (ACP) and β-ketoacyl-ACP synthase II (FabF; EC 2.3.1.41). The brFA synthesis machinery gene candidates we will overexpress in 6803 are listed in Table 6.
E. coli
Lactococcus lactis
Streptomyces coelicolor
Strptomyces avermitilis
Psedomonas putida
Bacillus subtilis
By the methods described in Materials and Methods below, the brFA synthesis genes from Streptomyces coelicolor may be introduced and overexpressed in the chromosome of one of the FFA-secreting strains (e.g., SD 232). The FFA chain structure of the resulting strain may be analyzed. If brFAs can be produced and secreted by 6803, that will justify the hypothesis of improving fuel quality by branching the FFA chain structure. Depending on the quality and yields of the brFAs produced by the modified 6803, other brFA synthesis genes from other organisms and their optimal expression levels may be tested by a 6803 plasmid expression system which will be discussed in Materials and Methods below.
In addition to directly enhancing FAS, multiple means may be used to increase uncoupling FFA from FAS, even when the cells enter the stationary growth phase, and to facilitate the secretion of FFAs. The optimization means include the following three points.
Duplication of Enzyme Activities from Multiple Organisms Using Chromosomal and Plasmid-Specified Gene Activities
As proven by preliminary results, introducing and overproducing exogenous Acyl-ACP TE in 6803 will change the metabolic destination of a portion of fatty acids. Some fatty acids may be decoupled from the FAS and membrane lipid synthesis resulting in FFAs. TEs are the essential enzymes in our FFA overproducing 6803 strains. Therefore, multiple TEs may be overproduced to improve the FFA overproduction and determine their optimal expression levels. Our most recent construction of SD277 completely validates this approach.
In bacteria and plants, each FAS reaction is catalyzed by a discrete, monofunctional enzyme and the growing acyl chain is bound to ACP. This FAS version has been termed type II (FASII). Thus plant and bacterial FAS systems resemble each other in machinery and the plant fatty acid synthase resides in the chloroplast (Ohlrogge, Kuhn et al. 1979), which is considered to be of prokaryotic (cyanobacterial) origin. Therefore, the common ancestry of plant and bacterial FAS results in their structural and functional similarities (Ohlrogge 1982). Thus the plant FAS enzymes are usually functional in cyanobacteria, which also has been demonstrated by our results. For this reason, a variety of additional plant TEs may be tested in 6803 to achieve maximum FFA production yields with optimal chain lengths.
The known plant TEs can be divided into two main classes, based on their specificity for acyl-ACPs of different chain lengths and degrees of unsaturation. The “FatA” type of plant TE has preferential activities on oleoyl-ACP (C18:1). The “FatB” type of plant TE has preferential activity on saturated acyl-ACPs with different chain length preferences. The TE candidates are listed in Table 7. Besides the plant TEs, various TEs from other bacteria such as S. enterica may also be used.
E. coli
Desulfovibrio vulgaris
Alkaliphilus
oremlandii OhlLAs
Anaeromyxobacter
dehalogenans 2CP-1
Streptococcus
mutans UA159
Cinnamonum
camphorum
Arabidopsis thaliana
Arabidopsis thaliana
Bradyrhiizobium
japonicum
Cuphea hookeriana
Each of the genes listed in Table 7 may be cloned into pSD504 and into pSD505 (Materials and Methods below) to investigate introduction of one or more combinations of two genes introduced into a suitable strain of 6803 and determine the effect on level and chain length of synthesized FFAs and on the composition of secreted FFAs. pSD504 confers resistance to gentamycin and cloned genes are induced for expression by addition of arabinose (see Table 13). pSD505 confers resistance to streptomycin and spectinomycin and cloned genes are expressed under the control of Ptrc, which is inducible with IPTG. Thus both plasmids may be selected and maintained independently and their cloned genes regulated for expression independently. The two plasmids share sequence homology and can undergo recombination at a low frequency but this will not alter the structure and regulation of inserted genes. A more likely possibility would be incompatibility in maintenance of the two plasmids. However, since 6803 has some 6 to 8 chromosome copies it is expected that the number of the RSF1010 derived plasmids will be sufficient to force co-maintenance by presence of the selective antibiotics to which the plasmids confer resistance.
The results may enable us to determine which combination of tes and fat genes will generate the most productive strain. Of course, the choice of genes may also be influenced by the specific chain length desired to be secreted by a given 6803 strain. The gene may then be optimized for high-level expression (transcription and translation) and the optimal site within the chromosome or in one of the indispensible large plasmids to maximize stability and level of expression may be investigated. In most cases, the insertion site may be chosen to inactivate a gene or genes encoding a pathway that competes for carbon flow that may be channelled to synthesize and secrete FFAs.
One aim of our research is to enhance secreting FFA from 6803 cells throughout their life cycle, including in stationary growth phase. The objective is that more of the energy derived from photosynthetic CO2 fixation can be used for fatty acid production and less applied toward growth of non-lipid biomass. For this purpose, most of the FFA-overproducing genes need to be driven by promoters that initiate high-level protein synthesis even in the stationary growth phrase.
Three methods to identify and redesign the 6803 stationary phase promoters are proposed. (1) According to the 6803 mRNA microarray data (Foster, Singh et al. 2007), the fluctuation in expression of many but presumably not all genes during various growth stages is known. Some genes maintain their expression levels and (of interest to the research) some increase transcription levels when cells enter stationary growth phase. Table 8 shows the top 10 stationary phase up-regulated genes, most of which are not up-regulated in the ΔsigB mutant. Bioinformatics tools may be used to analyze the promoters for these genes that are not decreased in the stationary growth phase to identify the consensus sequence for the stationary promoters. (2) The 5′ upstream regions of genes that contain consensus sequences recognized by the group 2 sigma factors (especially SigB and SigC), which might be important in facilitating transcription in stationary phase, may be searched and compared. (3) Four promoter trap vectors (Materials and Methods below) were designed with different reporters to search for promoters with maximum transcription efficiencies in stationary phase. Libraries of short DNA sequences in pSD502 may therefore be made such that the promoters would drive expression of the gfp gene such that we could use flow cytometry and cell sorting to recover cells with plasmid clones giving maximum GFP activity in stationary phase.
Promoters identified in these studies may be used to fuse to genes for any step in the FFA production and secretion pathway that would benefit from improved gene expression in stationary phase. It is also anticipated that hybrid as well as sequential promoters may be designed to maximize expression levels.
It is has been proven that increasing synthesis of fatty acids will automatically lead to increased secretion of fatty acids. However, results have shown that the FFA-secreting 6803 strains overproduce FFAs with a portion retained inside the cells to inhibit cell functions (by feedback inhibition) or introduce membrane leakage. Our results also showed that if the polypeptidoglycan layers of fatty acid secreting cyanobacteria are compromised by ampicillin, there was significant leakage/export of FFAs outside the cells. Removal of the S-layer also facilitated FFA secretion with an overall increased FFA yield. This implies that facilitating the removal of intracellular FFA releases the product inhibition on the FFA producing equilibrium. Therefore, it is important to further facilitate fatty acid secretion into the culture medium through the cyanobacterial cell envelope without causing damage that would lead to leakage of diverse intracellular contents.
The cyanobacterial cell envelope is composed of four layers (Lounatmaa, Vaara et al. 1980); the external surface layers, such as S-layers and carbohydrate structures (Karlsson, Vaara et al. 1983), outer membrane, peptidoglycan layer (Hoiczyk and Hansel 2000), and cytoplasmic membrane (
To weaken the peptidoglycan layer, which actually is composed of several layers, three solutions are proposed. The first solution is to reduce peptidoglycan synthesis in 6803 by down-regulating the efficiency of peptidoglycan synthesis genes, such as those in the mur (essential peptidoglycan synthesis and ligation genes, e.g., slr0017, slr1423, slr1656 and sll2010) and ldh (involved in peptidoglycan precursor UDP-N-acetylmuramyl-pentapeptide synthesis, e.g., slr0528 and slr1656) families to weaken the polypeptidoglycan layer structures. In addition, some of the penicillin-binding protein genes, such as ftsl (sll1833), mrcB (slr1710) and ponA (sll0002) which are required in assembly of the peptidoglycan, may be deleted or modified to favor the release of fatty acids. In this regard, we have already deleted the slr1710 gene encoding PBP2 and included that mutation in SD248 and desendants (Table 1). We also deleted PBP3 in SD263, and PBP4 in SD264. Actually, we deleted combinations of all three PBP genes but were unable to incorporate more than one mutation into 6803 strains to retain good growth properties. For example, we deleted PBP2 and PBP3 in SD255, but this strain grew very slowly with a doubling time of 22 h. We also deleted PBP2 and PBP4 in SD259, and this strain also grew very slowly with a doubling time above 20 h.
Another way to interfere with peptidoglycan synthesis is by substituting a gene for a central step in an essential pathway for synthesis of an essential cell wall component with one from another bacterial species, such as using a foreign asd or alr gene. The asdA gene from Streptococcus mutans (Cardineau and Curtiss 1987) may be expressed in E. coli to enable synthesis of DAP (diaminopimelic acid) for incorporation into peptidoglycan. However, due to the fact that a foreign enzyme does not generate a stable poly-enzyme complex to efficiently catalyze the entire biosynthetic pathway to synthesize DAP rapid growth with vigorous aeration destabilizes the peptidoglycan layer in E. coli. Addition of pSD506 plasmid derivatives encoding asdA genes from diverse bacterial species into a FFA producing and secreting ΔasdA 6803 mutant may be compared to evaluate whether FFA secretion is increased. In this regard, we have cloned the 6803 alr (for alanine racemase to synthesize D-alanine required for peptidoglycan synthesis) and asdA genes and expressed them in E. coli and S. Typhimurium strains with deletion alr and dadX (E. coli) or dadB (Salmonella) (Lobockam M. et al 1994, Lilley, P. E. et al 1993, and Wasserman, S. A. et al 1983) and asdA mutations to enable synthesis of peptidoglycan and growth. E. coli and Salmonella have two genes encoding alanine rasemases whereas 6802 only has a single alr gene. We have also generated His-tagged Alr, AsdA and Murl (glutamate racemase) (Doublet et al., 1993) proteins to enable production of rabbit antibodies to enable immunological detection. Lastly, we have made suicide vectors to enable allele replacement mutagenesis to generate 6803 Δalr (Δslr0827), ΔasdA (Δslr0549) and Δmurl (Δslr1746) mutants (
The third solution to weaken the peptidoglycan layer may be to introduce endolysin genes from bacteriophages into 6803 and to express such genes at a low level. Endolysins are peptidoglycan-degrading enzymes that attack the covalent linkages of the peptidoglycans that maintain the integrity of the cell wall (Loessner 2005). It has been demonstrated that the endolysin gp19 from Salmonella phage P22 is able to degrade the 6803 polypeptidoglycan layers, and the endolysin R from E. coli phage λ is able to compromise the 6803 polypeptidoglycan layers. Thus, the 19 or R endolysin genes in 6803 may be expressed at very low levels to weaken the peptidoglycan to facilitate fatty acid secretion. Different promoters with variant low transcription efficiencies of the endolysin genes may be attempted to limit adverse growth effects. We can also insert infrequently used codons into the genes encoding these endolysins to reduce levels of endolysin synthesis.
Besides weakening the cell envelopes to facilitate lipid secretion, transporter or porin genes may also be expressed or overexpressed to make channels for the lipid. Most bacterial fatty acid translocation studies were focused on fatty acid uptake, that is, the transmembrane movement of fatty acids from the outside to the inside of the cells (Black and DiRusso 2003). However, a few studies showed evidence that, in some bacteria, intracellularly synthesized lipids were transported outside the cell by transporter proteins. For example, the cell envelop of Mycobacterium tuberculosis includes a thick layer of lipids on the outer part of the cell, which protects the tubercle bacillus from the host's immune system. Mycolic acids, the major constituents of this protective layer, are the long chain fatty acids found in the bacteria of the Mycobacterium genus. The precursor of mycolic acids, trehalose monomycolate (TMM), is synthesized inside the cell, and then transported outside the cell (into the outer membrane) by the ATP-Binding Cassette (ABC) transporter (Rv1272c Rv1273c). Use of these genes in 6803 may be examined to determine whether they enhance FFA secretion.
Many transport and efflux proteins serve to excrete a large variety of compounds, and these may possibly be modified to be selective for fatty acids. For example, E. coli outer membrane protein FadL is a membrane-bound fatty acid transporter, which binds long chain fatty acid with a high affinity (Dirusso and Black 2004). Other suitable transport proteins include, efflux proteins (Thompson, Lobo et al. 2009), and fatty acid transporter proteins (FATP) (Hirsch, Stahl et al. 1998). These proteins or their derivatives may be introduced into the fatty acid producing 6803 to determine whether they facilitate lipid secretion. Some transporter proteins are listed in Table 9.
Escherichia coli K-
Escherichia coli K-
Escherichia coli K-
Pseudomonas
syringae
Mycobacterium
tuberculosis
Rhodocuccus
erythropolis
Rhodococcus jostii
Pseudomonas
stutzeri A1501
Pseudomonas
putida KT2440
Arabidopsis
thaliana
Yersinia pestis
Yersinia pestis
Pestoides
Candidatus
amoebophila
Shigella sonnei
Thermosynechococcus
elongates BP-1
Traditional downstream recovery of microbial lipids, requires physical cell lysis followed by chemical solvent extraction and accounts for 70-80% of the total cost of biofuel production (Molina Grima, Belarbi et al. 2003). We propose a more cost-effective way to harvest lipids from cyanobacterial biomass for biofuel production that precludes the need for mechanically disrupting the cells. To this end, we developed a Green Recovery system in which lipolytic enzymes degrade the membrane lipids into free fatty acids (FFA) with the collapse of cells. The Green Recovery system controls the synthesis of lipolytic enzymes using CO2-limitation inducible promoters, which induce expression of the lipolytic genes upon cessation of CO2 aeration. This same system described below could be adapted for Green Recovery of lipids and other biofuel precursors from a diversity of photosynthetic microorganisms not limited to cyanobacteria and including micro algae, unicellular algae, diatoms and purple-sulfur bacteria and even from non-photosynthestic bacteria such as Escherichia coli and members of the Enterobacteriaceae and Pseudomonaceae. In these cases, one could use oxygen or nitrogen regulated expression of genes encoding lipases instead of using CO2 regulated gene expression.
The lipolytic enzymes (EC 3.1.1) including galactolipase and phospholipase B (Svendsen 2000) hydrolyze the carboxylic ester bonds to release the fatty acids from diacylglycerols. Galactolipase (EC 3.1.1.26) catalyzes the hydrolysis of galactolipids by removing one or two fatty acids (Helmsing 1969). Phospholipase B is an enzyme with a combination of both Phospholipase A1 (EC 3.1.1.32) and Phospholipase A2 (EC 3.1.1.4) activities, which can cleave acyl chains from both the sn-1 and sn-2 positions of a phospholipid (Kohler, Brenot et al. 2006). For the purpose of fatty acid recovery from membrane lipids, we tested the performance of three lipolytic enzymes (from a bacterium, a fungus, and an herbivorous animal, respectively) in cyanobacterium Synechocystis sp. PCC 6803. The lipase from Staphylococcus hyicus (Shl) was selected because it has a very broad substrate specificity ranging from triacylglycerol lipids of various chain lengths to phospholipids and lysophospholipids (Rosenstein and Gotz 2000). The second candidate was a modified fungal phospholipase from Fusarium oxysporum (Fol) that exhibited galactolipase activity as well as increased phospholipase activity (Rapp 1995). Third, the guinea pig lipase (Gpl, also called GPLRP2, guinea-pig pancreatic lipase-related protein 2) from the digestive juice of guinea pig (Andersson, Carriere et al. 1996) was also tested. Gpl shows the highest galactolipase activity known to date, and plays a dual role in the digestion of galactolipids and phospholipids, the most abundant lipids occurring in plant thylakoid membranes (Andersson, Carriere et al. 1996).
We also developed a nickel inducible lysis system for 6803, which is able to induce phage lysis genes to break down cell walls by adding nickel as an inducer (Liu and Curtiss 2009). This system shows the feasibility of controlling lethal genes by inducible promoters. However, nickel is toxic to the environment and also adds cost. Therefore, we developed a 6803 CO2 limitation response mechanism (McGinn, Price et al. 2003) for an inducible transcription system regulated by CO2 rather than nickel. It has been reported that aeration of illuminated 6803 cells with CO2-free air for 30 min depleted the culture CO2 concentration to near zero levels. Under these conditions, transcripts for the three inducible inorganic carbon uptake systems, ndhF3, sbtA, and cmpA, showed near-maximal abundance after 15 min under CO2 limitation (McGinn, Price et al. 2003). By utilizing the promoter sequences for these genes to control the lipase genes, Green Recovery of lipids can be initiated by CO2 limitation resulting from stopping aeration of the 6803 culture. Aeration to the photobioreactor is necessary and easy to regulate, thus limiting the CO2 supply is an economical and environmentally friendly method to initiate lipid hydrolysis.
We constructed a series of 6803 strains for testing and optimizing Green Recovery (
Synechocystis sp. PCC
Cuphea hookeriana) and E. coli
Salmonella phage P22 lysis
aGenetic information: lipA (sll1969), a putative lipase gene in 6803 chromosome; sacB, sacB gene, which is lethal for cyanobacteria in the presence of sucrose; KmR, kanamycin resistance gene; slr1609, the 6803 fatty acid activation gene, encoding an acyl-ACP synthetase (Kaczmarzyk and Fulda 2010)); PpsbA2, the promoter of 6803 psbA2 gene, a 6803 strong promoter (Agrawal, Kato et al. 2001); ′tesA, an E. coli thioesterase gene without the export signal sequence (Cho and Cronan 1995); slr1993 and slr1994, two poly-3-hydroxybutyrate (PHB) synthesis genes; Pcpc, the promoter of 6803 cpc operon (Imashimizu, Fujiwara et al. 2003); Prbc, the promoter of 6803 rbc operon (Onizuka, Akiyama et al. 2003); accB, accC, cD, and accA, four genes coding for 6803 acetyl-CoA carboxylase (ACC) subunits (Davis, Solbiati et al. 2000); sll1951, encoding a hemolysin-like protein, which is a 6803 cell surface layer protein (Sakiyama, Ueno et al. 2006); *PpsbA2, an improved promoter from PpsbA2; Uc fatB1, a TE gene from Umbellularia californica (Pollard, Anderson et al. 1991); Ch fatB2, a TE gene from Cuphea hookeriana (Dehesh, Jones et al. 1996); Pcmp, the promoter of 6803 cmp operon; fol, a synthesized gene based on the amino acid sequence of the fungal phospholipase from Fusarium oxysporum (Fol); shl, a synthesized gene based on he amino acid sequence of the lipase from Staphylococcus hyicus (Shl); gpl, a synthesized gene based on the amino acid sequence of guinea pig pancreatic lipase-related protein 2 (GPLRP2); Psbt, the promoter of 6803 sbtA gene; 13 19 15, Salmonella phage P22 lysis cassette.
Limiting the CO2 supply to the Green Recovery strain cultures (density >108 cells/ml) resulted in culture discoloration (
aDetailed genetic information is described in Table 10.
bMembrane damage was detected by Sytox staining. The damage rates (%/day) were estimated from starting time point to the 50% permeable cell time point, when over 99.9% cells were dead at last. If the permeable cell percentage did not drop to 50%, the damage rates were estimated over the entire experimental period.
cRecovered FFAs were extracted by 10 ml hexane from 16 ml cultures after four days of CO2 limitation.
dThe data for 1/16 dilutions were similar to the data for ¼ dilutions.
eThe secreted FFAs before CO2 limitation.
fTotal FFAs after CO2 limitation, including secreted FFAs and released FFAs.
gCalculated released FFAs for the combination strains by subtracting the total amount from the secreted amount.
Wild-type 6803 also showed membrane damage at high cell density under CO2 limitation conditions (
We tested several optimal conditions for Green Recovery, such as cell density of the cultures (
Gas Chromatography (GC) (Lalman and Bagley 2004) showed that a significant amount of FFAs were released by lipolytic degradation of the membrane lipids (
As detailed in Examples 1 through 9, we constructed 6803 strains that secrete FFAs into the culture medium by introducing the acyl-acyl carrier protein (acyl-ACP) thioesterase genes into 6803 along with other genetic modifications. The FFA-secretion strains that harbor the Green Recovery system (SD239, SD254 and SD262 in Table 10) are still able to release membrane lipids as FFAs at rates faster than non-FFA-secretion strains following CO2 limitation. GC analysis showed that the profile of the FFAs overproduced by acyl-ACP thioesterases by FFA-secretion strains was different from the profile of the membrane-released lipids. The FFAs recovered from overproducing strains are highly saturated and rich in C12:0 and C14:0, whereas FFAs obtained via the Green Recovery system contained substantial amounts of unsaturated fatty acids and only a small portion of C12:0 and C14:0, which is the same composition observed in membrane lipids (
The Green Recovery system was designed for production of scalable and cost-effective renewable biofuels in photobioreators. Productive photobioreators require aeration systems to supply the photosynthetic microorganisms with CO2. Lipid recovery from biomass by limiting CO2 supply is clearly an efficient and effective method. The system we describe here does not require traditional biomass processes (Molina Grima, Belarbi et al. 2003), such as cell harvesting, dewatering, cell disruption, solvent extraction or inducer molecules (Liu and Curtiss 2009), thus considerably reducing the cost of lipid recovery. Since continuous agitation is not required for Green Recovery (
Green Recovery exhibits other advantages when combined with the previously described cyanobacterial FFA secretion system (Examples 1 to 9). The FFA secretion system avoids the energy intensive biomass processes such as concentration and extraction by directly recovering the secreted FFA from the culture medium. However, the FFA secretion system still requires substantial biomass to achieve cost-effective FFA production, which means a significant amount of fixed carbon has to be converted and stored as lipid membranes. It is expected that the Green Recovery system will recover the membrane lipids in the potential spent cyanobacterial biomass generated by the FFA secretion system, and also will cause cell lysis and release of the unsecreted intracellular FFAs. To our surprise, incorporation of the Green Recovery system into the FFA secretion strains resulted in an increased damage rate upon CO2 limitation (Table 11, SD239 and SD262). We postulate that secretion of FFAs through the cytoplamstic membranes creates some lesions in the membranes, which facilitate the contact of lipolytic enzymes to the acyl glycerol ester bonds. These findings demonstrate the practical combination of the FFA-secretion system and Green Recovery in photobioreactors, where the old cultures or the spent biomass can be utilized for extra FFA yield. We believe that cyanobacterial biofuel will be instrumental in developing a carbon neutral source of sustainable fuels.
Developing an inexpensive means to achieve effective biological containment for a system designed for very large culture volumes with very high populations is a daunting task. The reason is that even an improbable occurrence like multiple mutations occurring in the same cell starts to increase in probability as the number of cells becomes large. The features of such a system may also depend on whether the eventual photobioreactor system is open or closed. Conceptually, initial focus may be limited on obtaining a better understanding of the enzymology for synthesis of precursors and assembly of the peptidoglycan layer. Some of this will derive from efforts described above in Example 10 to slightly impair synthesis of DAP and thus weaken the peptidoglycan to facilitate FFA secretion. Success in these endeavors may contribute some biological containment attributes. It is also likely that the extensive genetic engineering of 6803 may preclude its ability to survive in most environments encountered outside the photobioreactor. Because of this possibility, such studies may be conducted during the course of constructions. Listed below, are the types of efforts that may be used to achieve biological containment.
Strains as constructed may be evaluated for growth attributes in diverse environments as a function of temperature, light intensity, inoculation density and in the presence of competing bacteria of diverse genera. For example, we have found that FFA secreting strains such as SD243 can be grown in media with 0.8 M NaCl with no reduction in production and secretion levels of FFAs and have furthermore found that these salt concentrations very much inhibit growth of many heterotrophs that can contaminate cultures of FFA-producing cyanobacteria. Thus growth of strains in 0.8 M NaCl at pH 10 in the precense of sunlight that delivers UV that kills heterotrophs more than photosynthetic cyanobacteria should very much reduce presence of contaminating heterotrophs that might consume some of the secreted FFA produced. Nevertheless, we are examining other means to out-compete contaminants as well as to engineer resistance to predation by bacteriophages and protozoans such as amoebas.
Development of a Two Strain System with Interdependence on Synthesis and Excretion of One Cell Wall Constituent and the Inability to Synthesize Another
Two 6803 strains with identical genotypes relative to all aspects of FFA biosynthesis and secretion may be obtained. One strain may be designed so that it is unable to synthesize D-alanine (due to a Δalr mutation to eliminate alanine racemase) but be able to secrete either DAP or D-glutamate. The other strain may be unable to synthesize either DAP (due to a ΔasdA mutation) or D-glutamate (due to a Δmurl mutation) but be able to secrete D-alanine. Using this design neither strain could live alone nor even survive together at low cell densities. A potential concern for the success of this system is the possible presence of low levels of D-alanine, DAP or D-glutamate in lakes, streams, soil, etc. due to the death of other bacteria. We will thus measure the concentrations of these amino acids in environmental samples to assess this potential problem. However, as described in Example 10 we have cloned the 6803 alr, asdA and murl genes and expressed them in suitable E. coli mutants, have generated His-tagged Alr, AsdA and Murl proteins to generate antibodies, and have constructed suicide vectors to enable generation of 6803 derivatives with delation mutations in these genes. Other means to generate a system of mutual interdependency to confer biological containment are contemplated.
It may be explored whether a quorum sensing system (not so far described in 6803 or other cyanobacteria) can be introduced into 6803 so that a critical cell function (such as DNA synthesis or cell division) is dependent on maintaining a high cell density.
We have made extensive use of bioinformatic searches to identify heterologous genes encoding functions or activities that would be beneficial to express in our biofuel production strains. Similarly, we have used bioinformatic searches using amino acid and nucleotide sequence information of genes with known beneficial functions from other organisms to search the cyanobase data set to identify putative 6803 genes that can be evaluated to determine whether they encode a protein of the desired function. For example, we have used this approach to identify potential cell surface proteins that might constitute a surface-layer protein or be necessary for export or anchoring of such surface proteins.
In this regard, we conducted a global search for S-layer homology (SLH) domains by screening 64 known SLH genes (Beverage et al., 1997; Sara and Sleytr, 2000; McCarren et al., 2005; Kawai et al., 1998; Smit et al., 1992) across all genes in PCC 6803. The Venn diagram in
Another use of bioinformatic searches is to enable searches of the cyanobase data set to identify genes that are likely dispensible and which can therefore be deleted to establish a potential placement for insertion of heterologous genes to encode some desired function. Such knock-outs of dipensible genes/functions are detailed in Example 14.
In the study of Synechocystis genetic transformation, we identified some deletable genes in the 6803 chromosome. These genes can be deleted by the Kmr/sacB intermediated double crossover recombination without significant adverse effects on the growth of 6803 cells. Some of these deletable genes encode enzymes that would divert energy and cellular resourses away from production, for example, the slr1993 and slr1994 genes for PHB synthesis. Some of these deletable genes encode proteins that are components of cell walls, for example, the sll1951 gene for the surface layer protein. Some of these deletable genes encode enzymes that would direct FFA to other metabolic pathway, for example, the slr1609 for acyl-ACP synthetase. Some of these deletable genes encode proteins with unknown functions, and deletion of them can save the carbon source and energy for synthesizing them. These dispensible genes in 6803 and their functions are described in Table 12.
All strains were derived from Synechocystis sp. PCC 6803 and the SD100 strain used as the parent of all other strains was maintained at −80° C. in BG-11 medium supplemented with 20% glycerol. Defined deletion mutations with and without specific insertions are described in the Examples above and below (see Tables 1 and 10). 6803 wild-type and mutant strains were generally grown at 30° C. in modified BG-11 medium (Rippka, Derulles et al. 1979) under continuous illumination (140 mmol photons m−2 s−1). Bacterial growth in liquid culture was monitored spectrophotometrically or by flow cytometry and/or by plating. For growth on plates, 10 mM TES-NaOH (pH 8.2) and 0.3% (w/v) sodium thiosulfate were added to BG-11 medium, and the medium is also solidified by addition of 1.5% agar for plating and colony isolation. For transformant selection, 50 μg/ml kanamycin or 4.5% (w/v) sucrose are supplemented in the BG-11 agar plates.
Cell density, CO2 supply and pH are very important to the growth of 6803, and these factors interact with each other. A typical guideline for SD culture is not to start cultures below a cell density of 107 cells/ml, since low cell densities will create a long lag phase prior to exponential growth. We used the following procedure to grow a 6803 culture from a colony descended from a single cell. A single SD colony is picked by a sterilized needle and used to inoculate 1 ml modified BG-11 medium buffered by 10 mM TES-NaOH (pH 8.2) in a glass test tube. The tube is incubated with illumination and intermittent shaking for 2-4 days. These starter cultures can be scaled up after the OD730 nm reaches 0.6 (108 cells/ml) by inoculating the 1 ml culture into 10 ml buffered BG-11 medium. The culture is grown in 50 ml flasks with 50 rpm rotation. 100 ml buffered BG-11 medium cultures are grown in 250 flasks with 100 ml/min aeration with air, and 1 L modified BG-11 medium cultures are grown without TES buffer with 300 ml/min air sparged with an air stone. Once the 1 L culture achieves OD730 nm˜0.6, aeration is switched from air to CO2-enriched air. This protocol uses TES buffer and air aeration to keep the pH around 8 at the beginning inoculation stage to minimize the lag phase. FFA-producing strains need a sufficient CO2 supply and a pH above 8 to maximize FFA secretion yields. When the 6803 cell density achieves 108 cells/ml, the culture is able to maintain their pH above 8, and can be supplied with CO2-enriched air.
Methods for DNA isolation, restriction enzyme digestion, DNA cloning and use of PCR for construction and verification of vectors are standard (Sambrook, Fritsch et al. 1989). The E. coli K-12 strains χ6097 and χ6212 were used for initial cloning. DNA sequence analysis was performed in the DNA Sequence Laboratory in the School of Life Sciences. All oligonucleotide and/or gene segment syntheses were done commercially (Genscript USA, Piscataway, N.J.). To increase the transcriptional and translational efficiency of heterogenous genes in 6803, as well as their genetic stabilities, the nucleic acid sequences of gene segments are redesigned by codon optimization based on the codon frequencies of highly expressed 6803 genes. Also the stem-loop hairpins in the predicted mRNA secondary structure are removed to smooth the transcription and to stabilize mRNA by prolonging its half-life. This may involve site-directed mutagenesis to “destroy” RNase E cleavage sites (Smolke 2000; Liou 2001). In some genes, the second codon was replaced by AAA to increase protein translation efficiency (Stenstrom, Jin et al. 2001). HA-tagged or FLAG-tagged proteins are used to obtain anti-protein rabbit antisera for western blot analyses and for quantitating protein synthesis levels in 6803 strains. Stabilization of plasmid constructs may be evaluated by DNA sequencing, and by ability to complement various cyanobacterial mutant strains and synthesize specific proteins as determined by western blot analyses. Stability of genetic modifications may be evaluated by growth under non-selective conditions for at least 50 generations of growth.
Gene Deletion and Introducing New Genes into 6803 by Using Kmr-sacB Cassette
Multiple gene modifications were included into the cyanobacteria strains. Suicide vectors with a Kmr-sacB cassette and sequences flanking the gene targeted for deletion are inserted with high efficiency into the desired chromosome site. The same vector having a second gene of interest may be used to replace the Kmr-sacB cassette with a desired sequence, thus substituting an undesired gene with a gene of interest in just two sequential Kmr-sacB transformation steps. The same strategies may be used to delete native promoters and replace them with other constitutive or regulatable or improved promoters. The important feature is to be able to select for kanamycin resistance to make the initial Kmr-sacB insertion/interruption and selection for sucrose resistance to introduce the final modification to eliminate the drug-resistance marker. One can thus make a succession of sequential changes in a strain genotype, resulting in a markerless strain with as many modifications as desired.
6803 is transformable at high efficiency and integrates DNA by homologous double crossover recombination for gene deletion, insertion and modification. General conditions for transformation of 6803 were optimized (Kufryk, Sachet et al. 2002) and in previous studies procedures for efficient segregation, clonal selection, and genetic identification were optimized. These procedures continue to be improved.
i. Transformation of Suicide Vectors Containing Kmr/sacB Cassette.
About 106 SD cells in 10 μl BG-11 medium are mixed with 400 ng suicide vector DNA containing the Kmr/sacB cassette and incubated for 5 h. Then the mixtures were plated onto a filter membrane (Whatman PC MB 90mM 0.4 μm) layered on a BG-11 agar plate. After segregation on the BG-11 plate for about 24 h, the membrane carrying the cyanobacteria was transferred onto a BG-11 plate containing 50 μg/ml of kanamycin. Generally, the colonies appear 4-5 days later. Individual colonies are picked into a small volume of BG-11 medium and restreaked with a sterile loop onto a kanamycin BG-11 agar plate and a 4.5% sucrose BG-11 agar plate. Those patches growing on the kanamycin BG-11 agar plate and not growing on the 4.5% sucrose BG-11 agar plate are the desired correct insertions with the Kmr/sacB cassette.
ii. Transformation with Markerless Constructs.
To replace the Kmr/sacB selective marker with target gene segments, about 106 Kmr/sacB cells in 10 μl BG-11 medium are mixed with 400 ng suicide vector DNA containing the target genes and incubated for 5 h. The mixtures are inoculated into 2 ml buffered BG-11 medium and grown for 3-4 days. 1 ml is plated onto a BG-11 agar plate containing 4.5% sucrose. Generally, the colonies appear 5-8 days later. Individual colonies are picked into a small volume of BG-11 medium and restreaked onto kanamycin BG-11 agar plates and 4.5% sucrose BG-11 agar plates. The patches growing on sucrose plates and not growing on kanamycin plates are positive candidates for further PCR identification.
iii. Confirmation of Replacement.
Cells from a colony (decended from a sigle cell) are resuspended in 2 μl water in a 200-μl PCR tube. The cell suspension is frozen at −80° C. for 2 min, and then thawed in a 60° C. water bath. This freeze-thaw cycle needs to be performed three times. 1 μl frozen-thawed cell suspension is used as the PCR template for a 30 μl PCR system including the primers specific for the inserted gene segments or the deleted region. The products of the various PCR reactions are separated on gels that are stained with ethidium bromide. If the PCR products of the expected sizes are observed with an absence of fragments that are unexpected, the correction of the construction can be inferred. In some cases, enzyme assays can be performed to demonstate the absence or insertion of a gene. In still other cases, other phenotypes associated with the genetic alterations can be observed or tested for.
iv. Stocking of Strains.
The cells with correct genotype are suspended from plates, transferred into glycerol-BG-11 medium (20% glycerol, v/v), distributed into at least four tubes and frozen at −80° C. and stored in two separate freezers on separate power supply with backup generators.
When a foreign gene is introduced into 6803, it may cause adverse effect on the growth and be subjected to gene loss or modification, since any cell losing the genetic alteration will have a higher growth rate to eventually take over the population. The genetic stability of foreign genes in 6803 is therefore tested by growing a culture of the strain with periodic dilution and subculturing for at least two months. After this time, the cells from the culture are plated onto BG-11 agar plates to obtain single isolated colonies. One hundred single colonies are picked and tested for all genetic attributes and confirmed for the presence of the foreign gene by PCR as described above. The percentage of positive colonies in the culture reflects the genetic stability of the foreign gene. For example, the presence ratio of E. coli *tesA gene driven by PpsbAII (in SD216) or PnrsB (SD215) is 100% for two months, thus we can say the E. coli *tesA gene is genetically stable in 6803. Genes found to be unstable can be modified to eliminate non-functional hydrophobic domains that often are responsible for poor growth due to association with and impairment of cytoplasmic membrane function.
As the FFA-secreting strain grows, FFA will be secreted into the culture medium and form insoluble FFA deposits on top of the culture media. The precipitated FFAs can be directly separated out and recovered by pipetting, filtration and/or skimming. However, some FFAs stay in the medium as dissolved acid anions because the pH of the culture is above 8. We thus developed a quantitative method for separation of the secreted FFAs from the culture medium. One hundred ml of culture is acidified by 2 ml H3PO4 (1M) containing 1 g NaCl, and extracted with 100 ml hexane. After 30 min shaking, the mixture is centrifuged, and the organic phase is separated by a separation funnel and dried by a vacuum. The chemical composition of FFAs is analyzed by GC-MS. Briefly, 1 ml 3M methanolic HCl (Supelco, St. Louis, Mo.) was added to the sample, which is heated at 85° C. for 2.5 h. After cooling to room temperature, 0.5 ml water and 1 ml of hexane are added and well mixed. The sample hexane layer with transesterification product FAME (fatty acid methyl ester) is collected and the remaining aqueous phase is twice extracted by 1 ml hexane. In total, 3 ml hexane is collected and mixed for GC analysis. Over 99% of FAMEs can be recovered after three hexane extractions.
Determination of FAMEs was carried out using gas chromatography (Shimadzu GC 2010) equipped with a Supelco SP2380 capillary column (30m×0.25 mm×0.20 μm) and a flame ionization detector (FID) was used to quantify FAMEs. Operating conditions were as follows: split ratio 1:10; inject volume 1 μl; helium carrier gas with constant linear velocity 20 cm/s; H2 40 ml/min, air 400 ml/min, make up 30 ml/min; injector and detector temperature 240° C.; oven temperature started at 140° C. for 1 min and increased at rate of 4° C./min to 220° C. and held for 5 min. Supelco 37 Component FAME Mix standard (Supelco, St. Louis, Mo.) was used to make a calibration curve for each FAME compound. The peaks from samples were identified by comparing retention times of unknown compounds with those of standard compounds and were also confirmed by GC-MS. Unknown compounds in samples were quantified based on their specific areas.
For the unsecreted intracellular FFAs, the cells are collected by centrifugation, and extracted by the Folch method (Folch, Lees et al. 1957) for total lipids. GC needs to be performed to analyze the FFAs profile from the total cell lipid extraction, which includes significant quantities of phospho-, galacto-, and sulfo-diglycerides as described above.
A series of plasmids may be constructed to facilitate research on the genetic manipulation of 6803. As listed in Table 13 and diagrammed in
Transposon vectors with Tn5 and Tn10 and derivatives may also be made with capabilities for operon and protein fusions or for inducible expression of genes downstream from the inserted transposons (Wolk, Cai et al. 1991; Milcamps, Ragatz et al. 1998; Bhaya, Takahashi et al. 2001).
aSm, streptomycin; Sp, spectinomycin.
We can use the ΔasdA 6803 mutant described in Example 10 to establish a balanced-lethal plasmid-host system that ensures maintenance of the plasmid vector by placing the wild-type asdA gene on a plasmid derivative of pSD505 (Table 13). We would thus replace the gene encoding streptomycin and spectinomycin resistance with the wild-type asdA gene. This plasmid has the strong prompter Ptrc and if we deleted the lacI gene we could engineer this plasmid by having Ptrc control over the expression of the accABCD genes to lead to an over production of manonyl Co-A. This technology should further increase FFA production and secretion.
As part of efforts towards biofuel production and recovery from cyanobacterial biomass, the current genetic modification techniques for gene deletion, insertion and substitution were optimized. Conditions for transformation and isolation of a recombinant by dominant phenotype have been well documented. However, if a recombinant is selected by a revertant phenotype (e.g., sucrose resistance by removal of sacB), the apparent overall transformation efficiency of Synechocystis 6803 is usually low because of low segregation efficiency. Few papers discuss the optimal conditions and efficiency of the counter selecting transformation. By testing different plasmid concentrations, transformation and segregation times, and selection pressure levels, the overall transformation efficiency was thus significantly improved. A highly optimized experimental protocol for 102-105 transformants/μg DNA has been developed, which is sufficient for genetic engineering purposes. On the basis of the optimized transformation technique, a series of gene expression systems for Synechocystis 6803 were developed, including both constitutive and inducible expression strategies. Various Synechocystis 6803 native promoters with different transcriptional levels were cloned, e.g., PcpcBA, PpsbAII, PpsaAB, Prbc, and PsigA, so that the target genes can be transcribed with the desired frequency. Genetic modifications of the native promoters such as removing or mutating the negative regulatory elements to stabilize their transcription levels in the dark or under stress conditions were also performed. For the inducible expression strategy, Synechocystis 6803 promoters responsive to different inducing factors were cloned, such as PnrsB, which is activated by addition of 7-50 μM nickel ion (Liu and Curtiss 2009), PisiA, which is activated by iron deficiency or when the cells grow into stationary-growth-phase (Singh and Sherman 2006), and PcmpABCD and PsbtA, which are activated by CO2 limitation in the culture media (McGinn, Price et al. 2003).
Using the above techniques, a lysis induction system in Synechocystis 6803 was designed and constructed to facilitate extracting lipids for the production of biodiesel. Several bacteriophage-derived lysis genes were integrated into the Synechocystis 6803 genome and placed downstream of a nickel-inducible signal transduction system (nrsRS-PnrsB). Three strategies were applied to utilize the phage lysis genes. Strategy 1 used the lysozymes from P22 and λ, respectively, to test the lysing abilities of lysozymes from different bacteriophages. Significant autolysis has been induced in the Synechocystis cells with this system by addition of NiSO4. Strategy 2 was designed to over express the endolysin genes (P22 19 15) under a constitutive promoter PpsbAII, while restricting the control of the expression of the holin gene (P22 13). As a result, before induced expression of the holin gene, endolysins accumulate in the cytosol. Once the holin gene is expressed, the holins synthesized would produce holes in the cytoplasmic membrane from within and allow the accumulated endolysins to gain access to the cell wall, resulting in rapid destruction of the murein. Strategy 3 incorporated the lysis genes from λ with P22 lysis genes. A faster lysis rate resulted since different lysozymes attacked different bonds in the cell envelope.
The complete genomes for E. coli K-12, S. Typhimurium LT-2, and Synechocystis 6803 were downloaded from NCBI GenBank and gene information was extracted using Perl scripts. The COG database (Tatusov et. Al. 2000) was also downloaded for use in categorizing genes. All of the features were stored in a MySQL database. For gene identification, BLASTP implemented in NCBI blast-2.2.18 was used to identify possible homologous genes of E. coli or Salmonella in the Synechocystis 6803 genome with a threshold e-value less than 1.0−4 and identity greater than 35%. Three methods were used to define genes of synthesis and assembly of cell wall components: (1) Based on a set of genes in E. coli K-12 and Salmonella LT-2, the homologous genes in Synechocystis 6803 were determined using BLASTP. (2) A set of genes based on the functional descriptions of the COG database were determined, and then the homologous genes in Synechocystis 6803 were found using BLASTP (3) Functional annotation of Synechocystis 6803 was used to search for genes directly. Using these methods, over 300 putative genes potentially involved in the synthesis and assembly of the Synechocystis 6803 cell wall and cell-surface macromolecular components were identified, and 25 of these genes were arranged into 7 putative operons. These are diagramed in
There is no comprehensive study of LPS or EPS in Synechocystis 6803, in contrast to the depth and detail published about photosynthesis by this organism. Therefore, genes and operons in LPS or EPS synthesis may be systematically identified and characterized, using a combination of bioinformatic analysis and experimental methods. For the previously characterized genes in pili and surface-protein layer synthesis, knock-outs may be made to assess their role in biofilm formation. Recent work has elucidated synthesis and regulation of pili-mediated motility, but this work has not been assessed in the context of biofilms.
The LPS and specifically the O-antigen component is normally covered by the surface-layer protein, and therefore does not directly mediate attachment and adhesion of WT cells in biofilms. LPS mutants in S-layer minus strains may likely have modified biofilm characteristics due to their different surface biochemistry and therefore different adhesion characteristics. We have found this to be so. These differences may be important if the Synechocystis strain used for industrial-scale production is a S-layer minus variant. Bioinformatic assessment of Synechocystis 6803 (see Examples 17 and 20) indicates that the O-antigen synthesis pathways (rfb operons) of previously characterized gram-negatives such as E. coli (Paton and Paton 1999) are at least partially conserved in Synechocystis 6803. A putative operon slr0982, slr0983, slr0984, slr0985 (rfbB, rfbF, rfbG, rfbC), has been identified and many additional rfb homologs are found individually or in pairs in the genome. In total, rfbABCDEFG, M, P, U and W each occur at least once, for a total of at least twenty rfb homologs. Synechocystis 6803 LPS may be characterized via LPS gels to identify components of the LPS. These components may be used to focus a bioinformatic search for genes related to LPS synthesis in Synechocystis 6803. These genes may be knocked out individually and in combination in both WT and S-layer minus backgrounds, and mutants may again be assessed using LPS gels. LPS mutations that result in an S-layer shedding phenotype may help identify which O-antigen moieties are required to anchor the S-layer.
Lipid A is the lipid component of LPS and in most bactreria is essential for viability. We postulate that blocking lipid A synthesis in 6803 can enhance fatty acid production. Each 6803 cell has ≈106 lipid A molecules as a structural component of LPS (Raetz et al., 2009), and each lipid A has about 4 FA molecules, so blocking lipid A synthesis should enhance FFA production. Removal of lipid A may also enhance outer membrane permeability, facilitating secretion of FFA. Surprizingly, deletion of the IpxA (sll0379), IpxD (sll0776) and IpxB (sll0015) genes essential for lipid A synthesis is not lethal in 6803. We have thus constructed strains with both individual deletions and combination deletion mutations to examine for increase in FFA production and potential enhanced secretion. We are also contemplating deleting many other genes associated with lipid A synthesis or modification such as IpxC and msbA since they are also likely to be dispensible and their deletion will reduce unnecessary energy use and facilitate FFA production. Other likely genes identified by bioinformatic searches that could possibly be deleted are the sll1276, sll1725, sll1149, sll1180 and sll0615 genes.
Characterization of operons for synthesis and activation of pili required for motility has been published in detail (Yoshihara, Geng et al. 2001; Bhaya 2004; Bhaya, Nakasugi et al. 2006); however, the role of pili and motility in attachment and adhesion to surfaces is unknown. Synechocystis 6803 is covered with a dense layer of pili, including thick Type IV pili required for motility, and thinner pili with no known function. Mutations in previously characterized genes may allow us to perform experiments and to understand the role of pili in biofilm formation. Specifically, comparison of mutants that are unpiliated (pilC, slr0162), have paralyzed pili (pilH sll0415), have pili lacking tip adhesins (pilO, sll1276), and lacking Type IV pili (pilN, sll1275) may be compared using the crystal violet assay to identify the role of pili in attachment and adhesion of cells, and development of biofilms.
Two putative EPS export proteins were identified and may be inactivated to elucidate their role in biofilm formation. BLASTP of E. coli W3110 wza, a gene coding for an OM EPS export protein (Dong, Beis et al. 2006), shows 28% identity with sll1581 (with e-value of 10−16). sll1581 is annotated as gumB (a gene coding for an EPS export protein) in Synechocystis 6803 from Cyanobase. This protein was identified in the OM proteome of Synechocystis 6803, which would be consistent with its predicted role as a porin for EPS export (Huang, Hedman et al. 2004). wzc, a gene coding for the EPS membrane translocator in E. coli W3110, has 21% identity (with e-value of 10−16) with sll0923. Strains of E. coli, C. crescentus, and X. campestris with deletions in wza homologs are deficient in biofilm formation (Smith, Hinz et al. 2003; Dong, Beis et al. 2006).
BLASTP of E. coli EPS synthesis genes such as wzxE (flippase) and wzyE (polymerase) against the Synechocystis 6803 proteome showed no homologs present. A possible operon or operons for polysaccharide synthesis in Synechocystis 6803 were identified bioinformatically by searching for E. coli EPS homologs. slr0527, slr0528, slr0529, slr0530, slr0531, slr0533, slr0534, and slr0535 were found to be a putative operon, since all are in the same orientation in the genome and encode proteins that have homology to proteins for sugar synthesis and transport. However, deletion of polysaccharide synthesis genes frequently have pleiotropic phenotypes because polysaccharides are used for multiple cellular functions, in addition to EPS.
To characterize EPS in Synechocystis 6803, EPS export protein homologs sll1581 and sll0923 may be knocked out, and these mutants screened along with WT using a lectin library. Since individual lectins bind to a specific sugar moiety, each lectin may indicate whether or not its specific sugar ligand is found in Synechocystis 6803 EPS. Cells treated with each of the fluorescein-conjugated lectins may be examined using fluorescence microscopy to identify which lectins bind to WT and mutant strains. Identification of polysaccharides thus found in Synechocystis 6803 EPS may be used to focus our bioinformatic search for genes related to EPS synthesis. Knockouts of putative EPS synthesis genes may again be characterized by fluorescein-conjugated lectins, as well as by LPS gels, to determine whether the deleted gene contributes to either EPS, LPS, or both. If Synechocystis 6803 strains do bind at least one lectin, then EPS gels may also be used to characterize Synechocystis 6803 EPS.
Insertional knockouts and deletions described above, as well as those isolated from screens of phage-resistant or mutagenesis libraries, may be screened using the crystal violet assay in order to identify structural features important for biofilm formation. Those mutants showing atypical biofilms may be further characterized to elucidate the specific function of each feature in biofilm formation, and whether they are required for initial attachment, adhesion, biofilm maturation and structure, and/or biofilm dispersal. In addition, strains may be compared under different physiological conditions, to see if the deleted structure plays a role in biofilms that is modulated by heterotrophic growth, stress response, light response, or different growth stages, for example. These studies may identify genetic modifications resulting in a strain of Synechocystis that cannot form biofilms, which may be assessed for its suitability for large-scale biodiesel production, in collaboration with the rest of the ASU biodiesel research team. Specifically, the biofilm-minus strain of Synechocystis 6803 may be assessed for ease of lysis using our nickel-induced promoter system, for ease of lipid extraction from lysed cells, for susceptibility to phage and other predators identified by the microbial ecology group of the biodiesel team, for growth rate and general robustness of the strain in photobioreactor growth conditions, and for the amount and quality of lipid per gram of biomass.
In addition to screening mutants for atypical biofilm formation, the crystal violet assay may also be used in preliminary assessment of biofilms. Specifically, the crystal violet assay may be used to visualize where biofilms form relative to the meniscus of media, to determine whether there is differential attachment for glass compared to other surface materials found in our photobioreactors used for biodiesel production, to quantify the amount of biofilm relative to the optical density and growth phase of the culture, and assess the effects of DNase, proteinase, and other additives on biofilm formation. Also, it may be used in competitive binding assays, to identify which specific EPS moieties (as identified by binding to fluorescent lectins in section Materials and Methods below) contribute to surface adhesion. Cells pretreated with lectins may show less binding or no binding in this assay, compared to untreated cells. Mutants lacking EPS may be characterized using the motility assay to determine whether EPS is required for motility, as has been suggested by the appearance of a corona of translucent extracellular material preceding phototactic cells (Burriesci and Bhaya 2008).
Additional characterization of WT and mutant strains of interest may be performed using a biofilm reactor. This reactor uses larger culture volumes than the crystal violet assay (1000 ml vs 3 ml), and also allows better control of growth conditions, including an option for chemostatic growth, and ability to create both illuminated and dark conditions in the same culture. Biofilm samples may be assessed for stages of development and maturation under various conditions. For example, WT strains cultivated in a reactor with illuminated and unilluminated areas may only form biofilms in the illuminated areas. By allowing these biofilms to develop, and then switching the conditions so that the biofilms are in the dark may allow the assessment of how biofilms respond to a change in light conditions. This experiment may also be conducted with WT and phototaxis mutants that are growing heterotrophically (addition of glucose and DCMU, which inhibits photosynthesis).
Reports of biofilm formation by cyanobacteria are limited to mixed-species biofilms related to environmental microbiology, such as epilithic cyanobacteria that degrade stone Mayan artifacts (Scheerer, Ortega-Morales et al. 2009), hot spring biofilms (Boomer, Noll et al. 2009), or symbionts of marine corals or crops important to agriculture (Arboleda and Reichardt 2009; Zheng, Bergman et al. 2009). There are no studies published characterizing single-species biofilm formation by a cyanobacterium. Thus, detailed studies to characterize biofilm formation by axenic cultures of WT Synechocystis 6803 may be undertaken using confocal laser scanning microscopy of biofilm reactor samples. Time-course experiments may be used to describe the stages of Synechocystis 6803 biofilm formation, from initial attachment and adhesion of cells, to formation of microcolonies, monolayers, biofilm maturation and development of biofilm architechture such as mushroom morphology, and biofilm dispersal. Biofilms may be studied using growth curves, and also under chemostatic conditions. Live-dead staining may be used to identify stratification of biofilm architecture by layers of live cells and dead cells. Mutants of interest identified in screens using the crystal violet assay may be used to clarify the role of specific structures in each stage of biofilm formation, by comparing confocal micrographs of biofilms formed by mutant monocultures with those of WT.
Preliminary work by the microbial ecology group of the biodiesel team has already identified non-Synechocystis species within the biodiesel photobioreactors. These non-Synechocystis species include cyanobacterial predators such as phage and protozoa, as well as heterotrophic and autotrophic bacteria whose roles in the health of the Synechocystis 6803 reactor culture have yet to be established. Studies may be undertaken to assess the role of these invaders on biofilm formation by Synechocystis 6803 in photobioreactor cultures.
We propose to use biofuel strains that have been genetically engineered for inducible aggregation and/or flocculation of the biofuel strain with the following benefits:
We propose to use environmental or physiological signals such as CO2 limitation or entry into stationary phase in order to induce expression of aggregation/flocculation signals. Specific mechanisms for aggregation/flocculation include but are not limited to modulating the expression of surface protein adhesins or exopolysaccharides.
Proteins that have high binding affinity can act as adhesins and cause cell-cell binding and aggregaton (Miller and Falkow 1988, Isberg and Falkow 1985). Inducible synthesis of homo-dimers such as YadA-YadA (Y pseudotuberclosis) or heterodimers such as invasin-β1-integrin (Y pseudotuberculosis) will display these adhesins on the cell surface, causing aggregation. The Km (binding affinity) of invasin with β1 integrin is the one of the strongest protein-protein interactions known (Leong and Isberg 1990).
In addition to transgenic proteins, the expression of native surface structures of the biofuel strain can also be induced using environmental signals to cause auto-flocculaton or attachment to surfaces. For example, preliminary data indicate that the exopolysaccharide of one biofuel strain (Synechocystis PCC 6803) contributes to biofilm formation in WT cells; over-expression of native exopolysaccharide synthesis genes may increase the aggregation/biofilm formation phenotype. Similarly, inducible expression of exopolysaccharide genes from Caulobacter or other organisms in other biofuel strains may be used to increase the degree and rate of flocculation for biomass harvest. The exopolysaccharide synthesized by Caulobacter has been shown to be strongest adhesive known (Tsang, Li et al 2006). Additional modification to the biofuel strain may include inducible down-regulation of O-antigen biosynthesis genes. Furthermore, our preliminary data show that deleting O-antigen genes slr0977 and slr1610 cause a flocculation phenotype in Synechocystis PCC 6803.
Inducible flocculation or aggregation (cell-cell binding) and biofilm formation (attachment of cells to the surface of reactors or other abiological surfaces) can be used to create microbial mats at the surface of open ponds to reduce evaporation when induced in buoyant biofuel strains. In strains that are not naturally buoyant, synthesis of gas-filled vacuoles may be genetically engineered by expressing vacuole synthesis genes, which include but are not limited to gvpA and gvpC (Hayes et. al. 1988).
Biofilm formation by strains secreting fatty acids or other biofuels may be induced. Growing the biofuel strain as a biofilm instead of a suspended culture negates the need to induce aggregation as a separate step in the biofuel production process. Biofilm photobioreactors have been shown to have higher performance and productivity in both bioenergy and bioremediation applications (Tian, Liao et al 2010; Syed and Henshaw 2003). Additionally, biofilm photobioreactors do not have reduced performance due to biofouling, as is the case in traditional suspended-biomass photobioreactors.
Although natural transformation for suicide vectors is very efficient in Synechocystis (Zang, Liu et al. 2007), the observation that double crossover recombination with transformed suicide vectors is far more prevalent than single crossover plasmid insertion (Porter 1986) suggests that single crossover insertion might be lethal. This might occur if plasmids need to be linearized for efficient DNA uptake. This seems logical since the presence of Type IV pili on Synechocystis 6803 cells seems to be required for transformation and competence to take up DNA (Yoshihara, Geng et al. 2001). In fact, plasmid transformation into Synechocystis 6803 is at an exceedingly low frequency (˜10−8) unless some sequence homologous to a Synechocystis sequence exists (Mermet-Bouvier, Cassier-Chauvat et al. 1993). Therefore, transformation efficiency may be compared using plasmid RSF1010 (Bagdasarian, Lurz et al. 1981) monomer versus dimer molecules stabilized in a recA E. coli mutant. Synechocystis likely has restriction enzymes and we will investigate whether administration of a five to ten minute pulse of 50° C. overcomes this barrier (Middleton and Mojica-a 1971; Mojica and Middleton 1971; Robeson, Goldschmidt et al. 1980). If so, this treatment could also be used to enhance conjugational transfer from E. coli to Synechocystis.
Construction of an E. coli-Synechocystis 6803 Shuttle Vector.
One of the small dispensable Synechocystis plasmids (e.g. pCB2.4, pCC5.2 or pCA2.4) may be cloned into enteric plasmids with p15A ori and pSC101 ori (which may be tested for inability to replicate or be maintained in Synechocystis 6803), and identify, using transformation of monomer or dimer plasmids, a shuttle vector with the desired properties. Several different regulatable promoters and multiple cloning sites followed by a transcription terminator sequence may then be introduced. This shuttle vector may be useful in complementation studies and also in gene regulation studies. As more is learned, a set of plasmids that replicate with different copy numbers in Synechocystis 6803 may be generated.
Suicide plasmid donor strains originally described (Miller and Mekalanos 1988) were modified to possess ΔasdA, Δalr and ΔdadB mutations that impose requirements for diaminopimelic acid (DAP) and D-alanine, two essential unique ingredients of the rigid peptidoglycan layer of the cell wall. Thus a donor-recipient conjugating mixture may be moved to medium devoid of DAP and D-alanine and the donor population undergoes cell wall-less death. Since E. coli strains grow much more rapidly than Synechocystis strains, inclusion of dnaA (Ts), dnaB (Ts) and dnaE (Ts) alleles into the E. coli donor may be investigated to arrest cell growth but without diminishing conjugational gene transfer. These E. coli donor strains have an integrated broad host range IncP conjugative plasmid that should facilitate gene transfer to Synechocystis 6803 recipients (Marraccini, Bulteau et al. 1993). However, there exists a collection of other Inc group plasmids (with drug-resistance markers) that can be easily evaluated for conjugational transfer to Synechocystis 6803 should the IncP system not be optimal. Although the potential for conjugational gene transfer of a shuttle vector described above may be examined in liquid medium, it is likely that mating and transfer may be at higher frequencies if recipient and donor cells are impinged on a filter that can be aseptically moved from a permissive agar (at various temperatures) to a non-permissive agar (no DAP and D-alanine). Glucose and LB broth may be added to supplement the BG-11 medium during mating since conjugational transfer in both donor and recipient requires active energy metabolism (Curtiss, Charamella et al. 1968). Contingent on success, conditions may be systematically varied to optimize the frequency of plasmid transfer. Ultimately the possibility of a conjugational system between Synechocystis 6803 mutant strains may be investigated.
The procedure may involve taking a collection of Synechocystis strains, treating cells with mitomicin C or UV and after a suitable period for induction of putative lysogens, add CHCl3 and continue aeration for another hour. Cells may then be sedimented and the putative lysates stored over CHCl3 and then tested for presence of phages causing lysis of different Synechocystis strains imbedded in soft agar overlays. Phage may be propagated on sensitive strains either in liquid medium or by confluent plate lysis and the lysates titered. Lysates may be treated with DNase to eliminate contaminating cyanobacterial DNA. The DNase may then be inactivated or removed and DNA harvested from lysates to screen for the presence of rRNA-encoding DNA sequences whose presence would predict the occurrence of generalized transduction. If multiple successes are achieved, the phage system that gives the highest titers of lysates and the highest frequency of transduction of drug-resistance markers may be determined. In addition, host-range mutants that will specifically propagate on and transduce Synechocystis 6803 may also have to be selected.
Screening of Mutant Libraries and Strains with Modified Surface Molecules for Resistance to Lytic Phage.
Bacteriophages often use surface molecules such as pili, outer membrane proteins, LPS or S-layers to attach to and infect cells (Boyd and Brussow 2002). Thus the library of 6803 mutants with altered pili, EPS, LPS and S-layer features may be screened for resistance to phages isolated from the environment or from screens (Materials and Methods below). To facilitate such screening, virulent non-lysogenizing mutants of temperate phages (Materials and Methods below) may be isolated and recovered as described in Materials and Methods below. Phage-resistant mutants of WT Synechocystis 6803 may also be isolated and screened for defects in biofilm formation (as described in Materials and Methods below). These phage may also be used to determine whether sensitivity and infection of Synechocystis 6803 and its mutants varies depending on stresses and/or growth conditions that favor or inhibit biofilm formation.
All strains may be derived from the Synechocystis 6803. Defined deletion mutations with and without specific insertions are described in the Examples. Synechocystis 6803 wild-type and mutant strains may be grown at 30° C. in modified BG-11 medium buffered with 10 mM TES-NaOH (pH 8.2) with a supplement of 1.5 g/l NaNO3 (Rippka, Derulles et al. 1979) and bubbled with a continuous stream of filtered air under continuous illumination (140 mmol photons m−2 s−1). For growth on plates, 1.5% (w/v) agar and 0.3% (w/v) sodium thiosulfate may be added to BG-11. BG-11 medium is also solidified by addition of 1.5% agar for plating and mutant selection. Antibiotics may be used at the following concentrations: chloramphenicol (Cm) (100 μg/ml), kanamycin (Km) (50 μg/ml), streptomycin (Sm) (50 μg/ml) gentamicin (Gen) (50 μg/ml), and Zeocin (Zeo) (100 μg/ml). Bacterial growth will be monitored spectrophotometrically, by flow cytometry and/or by plating.
Methods for DNA isolation, restriction enzyme digestion, DNA cloning and use of PCR for construction and verification of vectors are standard (Sambrook, Fritsch et al. 1989). The E. coli K-12 strains χ6097 and χ6212 may be used for initial cloning. DNA sequence analysis may be performed by the DNA Laboratory in the School of Life Sciences, ASU. All oligonucleotide and/or gene segment syntheses may be done commercially. Site-directed mutagenesis was used to optimize codons for translational efficiency in Synechocystis 6803. Plasmid constructs may be evaluated by DNA sequencing, ability to complement various cyanobacterial mutant strains and for ability to specify synthesis of proteins using gel electrophoresis and western blot analyses. His-tagged or GST-tagged proteins may be produced and used for antibody production to perform western blot analyses.
Multiple gene modifications may be included in the strains and strain attributes may be evaluated after every step in strain construction. If needed −35 and −10 RNA polymerase recognition and binding sites, SD sequences, and start codons may be altered to modulate up or down expression of genes for regulatory proteins or those to sustain cell integrity. Presence of fimbrial adhesins may be assayed using agglutination of appropriate cells and in the presence and absence of sugars as a function of growth conditions, and by transmission electron microscopy (TEM) using negative staining with phosphotungstic acid (Qadri, Hossain et al. 1988).
ColE1 replicons cannot be maintained in Synechocystis 6803. Therefore, pUC vectors may be constructed with Tn5, Tn10, Tn5-with LacI synthesis for ability to generate Ptrc fusions (inducible by IPTG) and Tn5 reporter fusions with LacZ, Lux, mcherry or Green Fluorescent Protein. The use of fusaric acid selection (Bogosian, Bilyeu et al. 1993) for loss of Tn10 insertions may be investigated as a means to inactivate genes by deletion. Other methods are as described above.
An enrichment assay to find suppressor mutants for strains defective in adhesion is useful to elucidate the signal transduction pathways regulating EPS, which is likely to be coordinated with phototactic, photosynthetic, and other pathways. For example, a knockout of the wzc homolog (sll0923) or other EPS-export response regulators may be repeatedly subcultured in test tubes. Test tubes may be vortexed vigorously to remove unattached cells, and it would be expected that any attached cells would have acquired suppressor mutations that are in the same pathway as the original wzc mutation. These suppressors may be enriched by adding new media and incubating. Mapping of the suppressor mutations may require either plasmids that replicate independently in Synechocystis 6803, or use of a transducing phage, in order to locate the suppressor mutation by screening complementation libraries, or mapping via co-transduction frequency of marker lysate libraries (Pierce, O'Donnol et al. 2006).
A crystal violet assay may be adapted from the study of biofilm formation in Caulobacter crescentus to use for studying Synechocystis 6803 biofilms. 12-well plates may be inoculated with log-phase liquid cultures. Glass coverslips may be inserted into the wells as a substrate for biofilm attachment and maturation. For each time point, coverslips may be removed and unattached cells rinsed off with a strong jet of water. Coverslips may be stained in 1% crystal violet solution. Excess stain is removed by repeated rinsing until no purple is observed on paper used to blot the coverslips. Biofilms may be assessed qualitatively and quantitatively. Biofilm quantitation may be performed by eluting the stain from the biofilm by immersing the coverslip in 3 ml of DMSO, and measuring the crystal violet absorbance of the eluant at OD600 nm. This assay can be adapted to 96-well plates, by using microplate readers to screen mutant libraries for adhesion defects to identify important genetic pathways for biofilm formation.
5-10 μl of log-phase culture may be streaked onto BG-11 plates with 0.4% agar, 15 mM glucose. Unidirectional light at 40 mmol/m2 sec may be used for phototaxis, spreading of colonies under uniform light shows general motility; colonies that remain the same size are non-motile (Bhaya, Watanabe et al. 1999).
Commercially available libraries of lectins may be used to screen Synechocystis 6803 EPS for specific sugars. For example, the lectin wheat-germ agglutinin has been found to bind specifically to polymers of N-acetyl-glucosamine. This lectin can be conjugated to FITC (fluorescein isothiocyanate) for fluorescence microscopy imaging of EPS localization in Synechocystis 6803 cultures, and for imaging gels of EPS samples (see EPS gel assay below).
This assay may be used to detect EPS with HRP (horse radish peroxidase)-conjugated WGA (wheat germ agglutinin) lectin (Hitchcock and Brown 1983). Solubilize cell pellet in lysis buffer with or without proteinase K. Run samples on PAGE gel without SDS, using a 4% acrylamide stacking layer and 10% acrylamide separating layer. Blot onto nitrocellulose (include stacking layer in blot). Block with 3% BSA and rinse with TBST. Incubate blot in 1:150,000 dilution of HRP-conjugated WGA lectin. Incubate with Dura chemiluminescent substrate from Pierce Protein Research Products.
Samples from large-volume and benchtop-volume photobioreactors may be used to characterize the microbial ecology of mixed-species biofilms. Cell pellets may be pre-treated with lysozyme and then genomic DNA extracted using the Qiagen DNeasy kit for gram-positives. Universal primers, and also Synechocystis 6803-specific primers, may be used to amplify 16S RNA encoding DNA sequence, and T-RFLP analysis used to determine the relative prevalence of 16S RNA encoding DNA from each bacterial species in these samples.
The LPS (lipopolysaccharide) may be analyzed on polyacrylamide gels and visualized with silver stain solution (Tsai and Frasch 1982). The cell pellet may be resuspended in 100 ml of 2× dissociation buffer with 5% β-mercaptoethanol and boiled for 10 min. The sample is then centrifuged at 13,000 rpm for 15 min. Supernatant is diluted 1:10 (50 ml) in 2×SDS (sodium dodecyl sulfate) loading buffer and 50 mg/ml Proteinase K. Incubate at room temperature for 1 h. Run samples on PAGE gel without SDS, using a 4% acrylamide stacking layer and 12% acrylamide separating layer. Run gel in SDS running buffer at 100 volts for 1 h. Resolve LPS signature on the gel by shaking the gel for 4 h in a fixative solution. Add 1.4 g of periodic acid to 200 ml of fixative and shake for 10 min. Rinse with 15 ohm ddH20 3×. Add silver nitrate mixture for 10 min and rinse 3×. Add developer and shake for 5 min or until bands appear. Rinse once with ddH2O before adding Stop Solution (a 1% acetic acid solution that stops the development).
After filtration through 0.22 mm polycarbonate filter (Whatman), the filtrates suspected of containing cyanophages are inoculated with a variety of Synechocystis cultures from our strain collection including Synechocystis 6803. After amplification on the host strain, chloroform is added to the inoculum and the mixture is agitated for 60 min and then centrifuged (7000×g, 4° C., 15 min). The supernatant is mixed again with chloroform (4 ml/ml) and stored at 20° C. Cyanophages are then titered on the propogating host using soft agar overlays on BG-11 plates. Plaque formation is used to enumerate phages (Wilson, Joint et al. 1993).
The collection of Synechocystis strains may be screened for presence of temperate phages that propagate on Synechocystis 6803 (see Examples above). Isolated phages and especially phages isolated from lysogens are likely to be temperate and have turbid plaques. Virulent mutants forming clear plaques can readily be isolated from temperate phages by making almost confluent lysis plates using the soft agar overlay method screening for areas of clear lysis (Levine and Curtiss 1961). Screening of 20 such plates may identify between one and ten clear areas of lysis, which may be picked by using a sterile needle, plaque-purified on the Synechocystis host and then amplified by propagation in Synechocystis liquid cultures to create lysates. We may also determine whether it is possible to select host-range mutants that will infect Synechocystis 6803 or its biofilm-defective mutants.
Synechocystis 6803 subjected to chemical or other mutagens may be screened for mutants resistant to various phages or their host-range mutants. Virulent cyanophage mutants are used to avoid problems with generating phage-resistant lysogens. Resistant mutants can be selected after prolonged cultivation of strains in the presence of phage or my spreading high titers of phage on BG-11 plates followed by plating the Synechocystis 6803 culture. Mutants may be picked by sterile needle and pure cultures obtained by streaking for isolated colonies on new plates. Mutants may be characterized for loss or alteration of surface structures. TEM may also be used to help identify structures to which given phage strains attach.
Mojica, A. T. and R. B. Middleton (1971). “Fertility of Salmonella typhimurium crosses with Escherichia coli.” J Bacteriol 108(3): 1161-1167.
Suicide vectors listed in Table 16 were constructed using the techniques previously described under the section “Materials and methods for examples 1-8” (see sections i-iii) to enable deletion of the essential 6803 asd, alr, dapA, dapB, and murl genes.
The two suicide plasmids needed to delete the 6803 alr gene are shown in
Synechocystis 6803 strains with alr deletion
A series of plasmids based on the IncQ conjugative plasmid RSF1010 may be constructed to facilitate research on the genetic manipulation of E. coli and cyanobacteria. The 5.7 kb region of RSF1010 containing three rep genes, A, B, and C, is necessary for its replication in 6803(Scherzinger, Bagdasarian et al. 1984; Marraccini, Bulteau et al. 1993). RSF1010 was modified to construct a family of promoter fusion vectors with various reporter genes. For example, the promoter search vectorp W575 (
All of the above-described vectors derived from RSF1010 have the original aadA gene encoding resistance to the antibiotics streptomycin and spectinomycin or the substituted aph gene specifying resistance to kanamycin. Since use of antibiotics to maintain plasmid vectors would be expensive on a commercial scale and be environmentally unacceptable, genes encoding an essential gene may be inserted to complement a deletion of an essential gene in cyanobacteria. In the first step toward making such balanced-lethal vectors, the essential gene may be first inserted and the antibiotic resistance gene may be retained, affording two means of selecting for plasmid vector transfer during strain construction and in maintaining the plasmid after construction. These constructions are illustrated in the next Example.
The 6803 asd and alr genes were initially cloned in the low copy E. coli plasmid pWSK29 (pSC101 ori) to yield pv566 and pv567 and their inheritance was selected for by their ability to complement asd and alr mutations in E. coli strain χ6097 (E. coli K-12 F−, araΔ[pro-lac] rpsLφ80d lacZ ΔM15ΔasdA4 Δ[zhf-2::Tn10]) and S. Typhimurium strain χ8901 (S. Typhimurium UK-1 Δalr-3 ΔdadB4), respectively. Since the strains grow, it is evident that the 6808 asd+ and alr+ homologs can functionally complement the asd and alr gene deletions in E. coli and/or Salmonella. Plasmids containing other essential genes such as the 6803 murl+ gene may also be constructed to evaluate their use as selective markers on shuttle vectors. The 6808 asd+ and alr+ genes have also been inserted into RSF1010 to yield pΨ569 (
The means to transfer any RSF1010 derived plasmid (for example pΨ569 and pΨ570) can be done using the triparental mating method (Elhai, J. & Wolk, C. P. Conjugal transfer of DNA to cyanobacteria. Methods Enzymol 1988.167, 747-754). The three parents are (1) E. coli strains HB101 harboring helper plasmid pRL528 (methylases) and the target plasmid (our Asd+ and Alr+ expression vectors, for example), (2) E. coli bearing the conjugal plasmid pRL443, and the target 6803 recipient cells that lack either the asd or alr gene. The three strains are mixed and allowed to mate on a nitrocellulose membrane on solid BG-11 medium for 12 h under 40 mmol m−2 g−1 white light at 30° C. Exconjugants are selected by transferring the filter to solid BG-11 medium supplemented with either DAP plus methionine and threonine or D-Alanine, respectively. Then the filter will be transferred to solid BG-11 medium without supplements to select only the complemented exconjugants.
The Δalr 6803 mutant SD546 (Table 17) was used to establish a balanced-lethal plasmid-host system that ensures maintenance of the plasmid vector by placing the wild-type alr gene in a RSF1010 derivative pΨ568 (
To test for the reciprocal function of E. coli and Salmonella wild-type essential genes to complement deletion of Synechocystis essential genes, the Salmonella wild-type alanine racemasealr and dadB genes have also been inserted into pΨ568 to yield pΨ591 (
As the last step in generating a balanced-lethal vector without antibiotic-resistance genes, the aadA gene encoding streptomycin and spectinomycin resistance was replaced with the wild-type 6803 alr gene to yield pΨ642 (
An essential nucleic acid such as murA may encode a protein involved in muramic acid synthesis, as muramic acid is another required constituent of the peptidoglycan layer of the bacterial cell wall. It is not possible to alter murA by deletion, however, because a ΔmurA mutation is lethal and cannot be isolated. This is because the missing nutrient required for viability is a phosphorylated muramic acid that cannot be exogenously supplied because most, if not all, bacteria cannot internalize it. Consequently, the murA nucleic acid sequence may be altered to make expression of murA dependent on a nutrient (e.g., arabinose) that can be supplied during the growth of the bacterium. In this case, since cyanobacteria are often unable to take up arabinose, the strain may first be genetically modified to enable uptake of arabinose by introducing an araE gene from E. coli that encodes an arabinose-uptake protein. This may be accomplished by cloning the E. coli araE gene and inserting it into a shuttle vector to enable insertion in place of a dispensible Synechocystis gene. Such genes have been identified in preceding Examples and generally encode products that would compete for energy or biosynthetic ability that would decrease ability of strains to produce desired biofuels or biofuel precursors. Such a dispensable/competing gene could be the sll1951 gene encoding an S-Layer protein. Using such an araE+ strain enables the introduction of a ΔPmurA::TT araC PBADmurA deletion-insertion mutation chromosomal mutation (
Cyanobacterial strains with the ΔPmurA:TT araC PBADmurA deletion-insertion mutation may be used to develop an additional balanced-lethal vector-host system. This may be accomplished by cloning the wild-type murA gene and inserting it into a shuttle vector plasmid (
The RSF1010 plasmid possesses the aadA gene conferring resistance to both streptomycin and spectinomycin. Inactivation of both antibiotics is by adenylylation and both antibiotics block protein synthesis by binding to two different ribosomal proteins encoded by two separate chromosomal genes. It is also likely that both antibiotics enter bacterial cells by different pathways. Many years ago, many investigators investigated whether levels of antibiotic resistance in bacteria could be increased in R plasmid containing strains. Results were generally positive and several mechanisms for achieving this were discovered. One mechanism resulted in increased plasmid copy numbers (Macrina et al. 1974. J. Bacteriol. 120:1387). The aadA gene in RSF1010 confers resistance to about 30 μg/ml of either streptomycin or spectinomycin and to slightly lower concentrations when added simultaneously to the BG-11 growth medium. Selection for increased resistance to both spectinomycin and streptomycin may selectively enrich for strains synthesizing increased amounts of the AadA enzyme and this may be by mutations increasing aadA gene expression by improvements in promoters, ribosome binding sites or by optimization of codons to improve translation or stabilize mRNA also to increase translation efficiency. These changes may likely be small and incremental whereas mutations that increased plasmid copy number may be expected to have a more significant increase in dual antibiotic resistance. By optimizing the concentrations of the two antibiotics, It may be possible to enrich for mutations to increase plasmid copy number. It should be emphasized that mutations in two chromosomal genes to confer increased resistance to both antibiotics would essentially never occur since the mutation rates to each type of resistance are exceedingly rare being 10−10/bacterium/generation and the mutation rate to both antibiotics would be the product of the individual mutation rates to resistance (10−20/bacterium/generation). Since the Synechocystis strain in these examples is ampicillin sensitive, the bla gene encoding β-lactamase may also be added on balanced-lethal or shuttle vectors to enable use of ampicillin to select for high-level ampicillin resistance, which would also be expected to result in plasmid mutants with increased copy numbers. The two selection methods may also be combined to provide even greater selective pressure to yield the desired high copy number plasmids. Upon isolating plasmids with increased copy numbers, their use in various applications of the balanced-lethal vector technologies described in these examples may be investigated.
Genes and operons like the accBCDA, fatB1, fatB2, and rbcLXS may be inserted into multi-copy plasmids stabilized by any one of the balanced-lethal systems to amplify the overall productivity of strains to synthesize and secrete free fatty acids or other biofuels or biofuel precursors. The plasmid pΨ627 containing accBCDA along with the balanced-lethal system has been constructed and is shown in
It should also be evident that one may investigate use of additional plasmids to establish balanced-lethal systems for the various uses described that are either broad host range being able to replicate in either E. coli or Synechocystis or only in Synechocystis in which case a fusion replicon (Hansen et al. 1981. Infect. Immun. 31:1034) able to replicate in both bacterial hosts may be generated. In this regard, Synechocystis 6803 has three small plasmids that may be exploited in these ways to establish additional balanced-lethal or shuttle vector systems.
TGGCTCCCTGACCAATTTTTCGG
This application is a Continuation in Part of PCT application PCT/US2010/054494, filed Oct. 28, 2010, which claims the priority of U.S. provisional application No. 61/255,729, filed Oct. 28, 2009, U.S. provisional application No. 61/307,185, filed Feb. 23, 2010, and U.S. provisional application No. 61/382,156, filed Sep. 13, 2010, each of which is hereby incorporated by reference in its entirety.
| Number | Date | Country | |
|---|---|---|---|
| 61255729 | Oct 2009 | US | |
| 61307185 | Feb 2010 | US | |
| 61382156 | Sep 2010 | US |
| Number | Date | Country | |
|---|---|---|---|
| Parent | PCT/US2010/054494 | Oct 2010 | US |
| Child | 13459774 | US |
