MICROBIAL ISOPRENOID PRODUCTION USING A HETEROLOGOUS DXP PATHWAY

BACKGROUND OF THE INVENTION

Isoprenoids are u biquitous in nature and are made in all living organisms. They comprise a diverse family of over 40,000 compou nds and serve diverse fu nctions including maintaining cellular fluidity, electron transport, and other meta bolic functions. A vast num ber of natural and synthetic isoprenoids are useful as pharmaceuticals, cosmetics, perfumes, pigments and colorants, fungicides, antiseptics, nutraceuticals, and fine chemical intermediates.

An isoprenoid product is typically composed of repeating five carbon isopentenyl d iphosphate ("IPP") units, although irregular isoprenoids and polyterpenes have been reported. In natu re, isoprenoids are synthesized by consecutive condensations of their precursor I PP and its isomer, dimethylallyl pyrophosphate ("DMAPP"). Two pathways for these precursors are known.

Eukaryotes, with the exception of plants, use the mevalonate-dependent ("M EV") pathway to convert acetyl coenzyme A (acetyl-CoA) to IPP, which is su bsequently isomerized to DMAPP. Prokaryotes, with some exceptions, typically employ only the mevalonate-independent or deoxyxylulose-5-phosphate ("DXP") pathway to produce I PP and DMAPP. Plants use both the M EV pathway and the DXP pathway.

Traditionally, isoprenoids have been manufactured by extraction from natural sources such as plants, microbes, and animals. However, extraction yield is usually very low d ue to a num ber of profound limitations. First, most isoprenoids accumulate in nature in only small amounts. Second, the source organisms are generally not a mena ble to the large-scale cultivation that is needed to produce commercially-via ble quantities of a desired isoprenoid.

Elucidation of the M EV and DXP meta bolic pathways, as well as advances in molecular biology have made biosynthetic production of isoprenoids feasible, particularly using the M EV pathway. However, unlike the M EV pathway, the genes encoding the DXP pathway have only recently been isolated and therefore, much less work has been performed in understanding the pathway. In general, to the extent that the DXP pathway has been manipulated, it has been in E. coli which includes an endogenous DXP pathway, and thus the host cell necessarily possesses all of the required accessory proteins and cofactors for a fu nctional DXP pathway. In a few instances, the installation of a DXP pathway in a host cell (e.g., S. cerevisiae) that does not have an endogenous DXP pathway has been reported. However, only the seven enzymes that convert glyceraldehyde 3 phosphate and pyruvate to IPP and/or DMAPP were expressed and no proof was presented that the I PP or DMAPP made in the host cell was a result of the heterologous E. coli DXP enzymes (versus from the endogenously present M EV pathway in 5. cerevisiae). SUMMARY OF THE INVENTION

The present invention generally relates to methods and materials useful for isoprenoid compounds using a microbial host cell using the DXP pathway. In particular, provided herein is an iron-sulfur enzyme system (iron-sulfur enzyme plus the two accessory proteins) that can be used for enhancing DXP pathway-mediated isoprenoid production in a host cell with an endogenous DXP pathway, or for engineering a heterologous DXP pathway in a host cell lacking an endogenous DXP pathway, and using the heterologous DXP pathway for isoprenoid production. In one aspect of the invention, an iron sulfur enzyme system is provided comprising one or more nucleic acid molecules comprising one or more nucleic acid sequences encoding:

(a) a reductase that is capable of accepting electrons from an electron donor thereby converting the reductase from an oxidized state ("oxidized reductase") to a reduced state ("reduced reductase");

(b) a redox protein that is capable of accepting electrons from the reduced reductase thereby converting the redox protein from an oxidized state ("oxidized redox protein") to a reduced state ("reduced redox protein"); and

(c) an iron-sulfur enzyme that is capable of accepting electrons from the reduced redox protein thereby converting the iron-sulfur enzyme from an oxidized state ("oxidized iron-sulfur enzyme") to a reduced state ("reduced iron-sulfur enzyme") and wherein the reduced iron-sulfur enzyme is capable of converting 2C-methyl-D-erythritol-2,4-cyclodiphosphate ("cMEPP") to 1- hydroxy-2-methyl-2-(E)-butenyl-4-diphosphate ("HM BPP").

In another aspect of the invention, an iron sulfur enzyme system is provided comprising one or more nucleic acid molecules comprising one or more nucleic acid sequences encoding:

(a) a reductase that is capable of accepting electrons from an electron donor thereby converting the reductase from an oxidized state ("oxidized reductase") to a reduced state

("reduced reductase");

(b) a redox protein that is capable of accepting electrons from the reduced reductase thereby converting the redox protein from an oxidized state ("oxidized redox protein") to a reduced state ("reduced redox protein"); and

(c) an iron-sulfur enzyme that is capable of accepting electrons from the reduced redox protein thereby converting the iron-sulfur enzyme from an oxidized state ("oxidized iron-sulfur enzyme") to a reduced state ("reduced iron-sulfur enzyme") and wherein the reduced iron-sulfur enzyme is capable of converting cMEPP to isopentenyl pyrophosphate ("IPP") and dimethylallyl pyrophosphate ("DMAPP"). In another aspect of the invention, a genetically mod ified microorganism is provided comprising one or more heterologous nucleic acid molecules that encode:

(a) a first reductase that is capa ble of accepting electrons from an electron donor thereby converting the first red uctase from an oxidized state ("oxidized first reductase") to a reduced state ("reduced first reductase");

(b) a first redox protein that is capa ble of accepting electrons from the reduced first reductase thereby converting the first redox protein from an oxidized state ("oxidized first redox protein") to a reduced state ("red uced first redox protein"); and

(c) a first iron-sulfur enzyme that is capa ble of accepting electrons from the first reduced redox protein thereby converting the first iron-sulfur enzyme from an oxidized state ("oxidized first iron-sulfur enzyme") to a reduced state ("reduced first iron-sulfu r enzyme") and wherein the reduced first iron-sulfur enzyme is capa ble of converting 2C-methyl-D-erythritol-2,4- cyclodiphosphate ("cM EPP") to l-hydroxy-2-methyl-2-(E)-butenyl-4-diphosphate ("H M BPP"). In other em bodiments, the one or more heterologous nucleic acid molecules further encode: (d) a second reductase that is capa ble of accepting electrons from an electron donor thereby converting the second reductase from an oxidized state ("oxidized second reductase") to a reduced state ("reduced second reductase");

(e) a second redox protein that is capa ble of accepting electrons from the red uced second reductase thereby converting the second redox protein from an oxidized state ("oxidized second redox protein") to a reduced state ("reduced second redox protein"); and

(f) a second iron-sulfur enzyme that is capa ble of accepting electrons from the second reduced redox protein thereby converting the second iron-sulfur enzyme from an oxidized state ("oxidized second iron-sulfur enzyme") to a reduced state ("reduced second iron-sulfur enzyme") and wherein the reduced second iron-sulfur enzyme is capa ble of converting H M BPP to isopentenyl pyrophosphate ("I PP") or/and dimethylallyl pyrophosphate ("DMAPP").

In other embod iments, the first reductase and the second reductase are the same reductase. In still other embodiments, the first redox protein and the second redox protein are the same redox protein. In other embodiments, the first reductase and the second reductase are the same reductase a nd the first redox protein and the second redox protein are the same redox protein. In another aspect of the invention, a genetically mod ified microorganism is provided comprising a heterologous DXP pathway that is capa ble of converting glyceraldehydes 3-phosphate and pyruvate to IPP and DMAPP and wherein the genetically modified microorganism does not include an endogenous DXP pathway. INCORPORATION BY REFERENCE

All patents, patent applications, and pu blications mentioned in this specification are herein incorporated by reference to the same extent as though each individual pu blication or patent document was specifically and individua lly indicated to be incorporated by reference.

BRIEF DESCRIPTION OF THE DRAWINGS

Figure 1 is a schematic representation of one em bodiment of the DXP pathway as well as the additional step of converting I PP and DMAPP to FPP.

Figure 2 shows the internal cellular concentrations of DXP pathway meta bolites cM EPP and DXP in yeast strains comprising heterologous DXP pathway enzymes.

Figure 3A shows a schematic map of the pGAL expression vector with an ispG or ispH insert.

Figure 3B shows a Western plot of Escherichia coli IspG expressed as solu ble protein (lane 2) or targeted to the mitochondria (lane 3) in a yeast host cell (lane 1 = molecular weight ladder).

Figures 4A and 4B show activity profiles of IspG-fld and IspH-fld fusion proteins, respectively.

Figure 5 shows ⁵⁵ Fe incorporation in heterologously-expressed ispG and ispH proteins, normalized to the amount of protein assayed. All ⁵⁵ Fe counts^g protein were normalized to the level of ⁵⁵ Fe found in LEU1 (defined as 1.0).

Figure 6 shows iron content in representative IspG and IspH proteins purified from 5. cerevisiae. Leu l, a naturally occurring 5. cerevisiae protein with four Fe-S clusters, serves as a control. Leu l was purified using the same methods as the IspG and IspH proteins.

Figure 7 shows the intracellular concentration (μΜ ) of IspG prod uct, H DMAPP, in yeast strains expressing a full DXP pathway (Y6283), IspG a nd their redox partners only (Y6291 and Y6292), or DXP-TOP pathway only (Y4819).

Figure 8 shows the intracellular levels of DMAPP a nd FPP in yeast strains expressing a full DXP pathway, which is suppressed in Y6540 and activated in Y6533, when these strains are grown on glucose. Y6533 and Y6540 strains were grown on 2% ¹³ C-glucose synthetic medium without adenine that was supplemented with 2 g/L ¹² C-mevalonate. Y6540 + ZA and Y6533 + ZA samples represent cultures that also contained 16 μg/mL zaragozic acid.

Figures 9A through 9D show schematic maps of expression vectors used to generate host strains expressing heterologous DXP pathway enzymes.

DETAILED DESCRIPTION OF THE INVENTION

Definitions

Unless defined otherwise, all technical a nd scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. Reference is made here to a nu mber of terms that shall be defined to have the following meanings:

The term "optional" or "optionally" means that the su bsequently described feature or structure may or may not be present, or that the su bsequently described event or circumstance may or may not occu r, and that the description includes instances where a particular featu re or structure is present and instances where the feature or structure is a bsent, or instances where the event or circumstance occurs and instances where the event or circumstance does not occur.

The term "deoxyxylulose 5-phosphate pathway" or "DXP pathway" is used herein to refer to a set of enzymes that converts glyceraldehyde-3-phosphate and pyruvate to I PP and/or DMAPP.

Figure 1 illustrates one em bodiment of the DXP pathway as well as the su bsequent

interconversion of I PP and DMAPP.

The term "pyrophosphate" is used interchangea bly herein with "diphosphate".

The terms "expression vector" or "vector" refer to a nucleic acid that transd uces, transforms, or infects a host cell, thereby causing the cell to produce nucleic acids and/or proteins other than those that a re native to the cell, or to express nucleic acids and/or proteins in a manner that is not native to the cell.

The term "endogenous" refers to a su bstance or process that occurs naturally, e.g., in a non- recom binant or unmodified microorganism.

The term "nucleic acid" refers to a polymeric form of nucleotides of any length, either ribonucleotides or deoxynucleotides. Thus, this term includes, but is not limited to, single-, dou ble-, or multi-stranded DNA or RNA, genomic DNA, cDNA, DNA-RNA hybrids, or a polymer comprising purine and pyrimidine bases or other natural, chemica lly, or biochemically modified, non-natural, or derivatized nucleotide bases.

The term "operon" is used to refer to two or more contiguous nucleotide sequences that each encode a gene product such as a RNA or a protein, and the expression of which are coordinately regulated by one or more controlling elements (for exa mple, a promoter).

The term "heterologous nucleic acid" as used herein refers to a nucleic acid wherein at least one of the following is true: (a) the nucleic acid is foreign ("exogenous") to (that is, not naturally found in) a given host cell; (b) the nucleic acid comprises a nucleotide sequence that is natura lly found in (that is, is "endogenous to") a given host cell, but the nucleotide sequence is produced in an unnatural (for example, greater than expected or greater than naturally found) amount in the cell; (c) the nucleic acid comprises a nucleotide sequence that differs in sequence from a n endogenous nucleotide sequence, but the nucleotide sequence encodes the same protein (having the same or su bstantially the same amino acid sequence) and is produced in an unnatural (for example, greater than expected or greater than naturally found) amount in the cell; or (d) the nucleic acid comprises two or more nucleotide sequences that are not found in the same relationship to each other in nature (for example, two or more gene sequences are placed closer together and/or in a different order than naturally found in the host cell).

The term "recom binant host" (also referred to as a "genetically modified host cell" or "genetically modified microorganism") denotes a host cell that comprises a heterologous nucleic acid of the invention.

The term "exogenous nucleic acid" refers to a nucleic acid that is exogenously introduced into a host cell, and hence is not normally or naturally found in and/or produced by a given cell in nature.

The term "regulatory element" refers to transcriptional and translational control sequences, such as promoters, enhancers, polyadenylation signals, terminators, protein degradation signals, and the like, that provide for and/or regulate expression of a coding sequence and/or production of an encoded polypeptide in a host cell.

The term "transformation" refers to a permanent or transient genetic change induced in a cell following introduction of new nucleic acid. Genetic change ("modification") can be accomplished either by incorporation of the new DNA into the genome of the host cell, or by transient or sta ble maintenance of the new DNA as an episomal element. In eukaryotic cells, a permanent genetic change is generally achieved by introduction of the DNA into the genome of the cell. In prokaryotic cells, a permanent genetic change can be introduced into the chromosome or via extrachromosomal elements such as plasmids and expression vectors, which may contain one or more selecta ble markers to aid in their maintenance in the recombinant host cell.

The term "opera bly linked" refers to a juxtaposition wherein the components so described are in a relationship permitting them to function in their intended manner. For instance, a promoter is opera bly linked to a nucleotide sequence if the promoter affects the transcription or expression of the nucleotide sequence.

The term "host cell" and "microorganism" are used interchangea bly herein to refer to any archae, bacterial, or eukaryotic living cell into which a heterologous nucleic acid can be, or has been inserted. The term also relates to the progeny of the original cell, which may not necessarily be completely identical in morphology or in genomic or total DNA complement to the original parent, due to natural, accidental, or deliberate mutation.

The term "naturally occurring" as applied to a nucleic acid, an enzyme, a cell, or an organism, refers to a nucleic acid, enzyme, cell, or organism that is fou nd in nature. For example, a polypeptide or polynucleotide sequence that is present in an organism that can be isolated from a source in nature and that has not been intentionally modified by a human in the la boratory is naturally occurring.

The terms "isoprenoid", "isoprenoid compound", "isoprenoid product", "terpene", "terpene compound", "terpenoid", and "terpenoid compound" are used interchangea bly herein. They refer to compounds that are capa ble of being derived from I PP. Exemplary isoprenoids include, but are not limited to, monoterpenes, d iterpenes, sesquiterpenes, triterpenes, and polyterpenes. The singular forms "a," "and," and "the" include plural references unless the context clearly dictates otherwise. For example, reference to "an expression vector" includes a single expression vector as well as a plurality of expression vectors, and reference to "the host cell" includes reference to one or more host cells, and so forth. It is further noted that the claims may be drafted to exclude any optional element. As such, this statement is intended to serve as antecedent basis for use of such exclusive terminology as "solely," "only" and the like in connection with the recitation of claim elements, or use of a "negative" limitation.

Unless otherwise indicated, this invention is not limited to particular sequences, expression vectors, enzymes, host microorganisms, or processes, as such may vary in accordance with the understanding of those of ordinary skill in the arts to which this invention pertains in view of the teachings herein. Terminology used herein is for purposes of describing particu lar em bodiments only and is not intended to be limiting.

Enzymes of the DXP Pathway

The term "deoxyxylulose 5-phosphate pathway" or "DXP pathway" is used herein to refer to a set of enzymes that converts glyceraldehyde-3-phosphate and pyruvate to I PP or/and DMAPP. Figure 1 illustrates one em bodiment of the DXP pathway comprising seven enzymes as well as the su bsequent conversion of I PP and DMAPP to farnesyl pyrophosphate ("FPP"). However, the DXP pathway may comprise fewer than seven, or more than seven enzymes, so long as the starting materials are glyceraldehydes 3-phosphate and pyruvate and the products of the pathway a re IPP and DMAPP. IPP and DMAPP are referred to as universal isoprenoid intermediates because all isoprenoid compounds are derived therefrom.

The seven enzymatic reactions illustrated in Figure 1 are described in greater detail as follows: In the first step, pyruvate is condensed with D-glyceraldehyde 3-phosphate to make 1-deoxy-D- xylulose-5-phosphate ("DXP"). An enzyme known to catalyze this step is, for example, 1-deoxy-D- xylulose-5-phosphate synthase. The gene encoding l-deoxy-D-xylulose-5-phosphate synthase is referred to as dxs. Illustrative examples of nucleotide sequences for dxs encoding 1-deoxy-D- xylulose-5-phosphate synthases include, but are not limited to, (AF035440; Escherichia coli), (NC_002947, locus tag PP0527; Pseudomonas putida KT2440), (CP000026, locus tag SPA2301; Salmonella enterica subsp. enterica serovar Paratyphi A str. ATCC 9150), (NC_007493, locus tag RSP_0254; Rhodobacter sphaeroides 2.4.1), (NC_005296, locus tag RPA0952; Rhodopseudomonas palustris CGA009), (NC_004556, locus tag PD1293; Xylella fastidiosa Temeculal), (NC_003076, locus tag AT5G11380; Arabidopsis thaliana), (Y18874, Synechococcus PCC6301), (AB026631, Streptomyces sp. CL190), (AB042821, Streptomyces griseolosporeus), (AF111814, Plasmodium falciparum), (AF143812, Lycopersicon esculentum), (AJ279019, Narcissus pseudonarcissus), and (AJ291721, Nicotiana tabacum).

In the second step, DXP is converted to 2C-methyl-D-erythritol-4-phosphate ("M EP"). An enzyme known to catalyze this step is, for example, l-deoxy-D-xylulose-5-phosphate reductoisomerase. The gene encoding this enzyme is referred to as dxr or ispC. Illustrative examples of dxr or ispC nucleotide sequences encoding l-deoxy-D-xylulose-5-phosphate reductoisomerases include, but are not limited to, (AB013300; Escherichia coli), (AF148852; Arabidopsis thaliana), (NC_002947, locus tag PP1597; Pseudomonas putida KT2440), (AL939124, locus tag SC05694; Streptomyces coelicolor A3(2)), (NC_007493, locus tag RSP_2709; Rhodobacter sphaeroides 2.4.1), (NC_007492, locus tag Pfl_1107; Pseudomonas fluorescens PfO-1), (AB049187, Streptomyces griseolosporeus), (AF111813, Plasmodium falciparum), (AF116825, Mentha x piperita), (AF182287, Artemisia annua), (AF250235, Catharanthus roseus), (AF282879, Pseudomonas aeruginosa) (AJ242588, Arabidopsis thaliana), (AJ250714, Zymomonas mobilis strain ZM4), (AJ292312, Klebsiella pneumoniae), and (AJ 297566, Zea mays).

In the third step, M EP is converted to 4-diphosphocytidyl-2C-methyl-D-erythritol ("CDP-M E"). An enzyme known to catalyze this step is, for example, 4-diphosphocytidyl-2C-methyl-D-erythritol synthase. The gene encod ing this enzyme is referred to as ispD or ygbP. Illustrative examples of ispD or ygbP nucleotide sequences encod ing 4-diphosphocytidyl-2C-methyl-D-erythritol synthases include, but are not limited to, (AF230736; Escherichia coli), (NC_007493, locus_tag RSP_2835; Rhodobacter sphaeroides 2.4.1), (NC_003071, locus_tag AT2G02500; Arabidopsis thaliana), (AB037876, Arabidopsis thaliana), (AF109075, Clostridium difficile), a nd (AF230737, Ara bidopsis thaliana).

In the fourth step, CDP-M E is converted to 4-diphosphocytidyl-2C-methyl-D-erythritol-2- phosphate ("CDP-M EP"). An enzyme known to catalyze this step is, for example, 4- diphosphocytidyl-2C-methyl-D-erythritol kinase. The gene encoding this enzyme is referred to as ispE or ychB. Illustrative examples of ispE or ychB nucleotide sequences encoding 4- diphosphocytidyl-2C-methyl-D-erythritol kinases, include, but are not limited to, (AF216300; Escherichia coli), (NC_007493, locus_tag RSP_1779; Rhodobacter sphaeroides 2.4.1), (AF263101, Lycopersicon esculentum), and (AF288615, Arabidopsis thaliana). In the fifth step, CDP-M EP is converted to 2C-methyl-D-erythritol 2, 4-cyclodiphosphate

("cM EPP"). An enzyme known to catalyze this step is, for example, 2C-methyl-D-erythritol 2,4- cyclodiphosphate synthase. The gene encoding this enzyme is referred to as IspF or ygbB.

Illustrative examples of IspF or ygbB nucleotide sequences encoding 2C-methyl-D-erythritol 2,4- cyclodiphosphate synthases include, but are not limited to, (AF230738; Escherichia coli), (NC_007493, locus_tag RSP_6071; Rhodobacter sphaeroides 2.4.1), (NC_002947, locus_tag PP1618; Pseudomonas putida KT2440), (AB038256, Escherichia coli mecs gene), (AF250236, Catharanthus roseus (M ECS), (AF279661, Plasmodium falciparum), and (AF321531, Arabidopsis thaliana).

In the sixth step, cM EPP is converted to l-hydroxy-2-methyl-2-(E)-butenyl-4-diphosphate

("H M BPP"). An enzyme known to catalyze this step is, for example, l-hydroxy-2-methyl-2-(E)- butenyl-4-diphosphate synthase. The gene encoding this enzyme is referred to as ispG or gcpE. Illustrative examples of ispG or gcpE nucleotide sequences encoding l-hydroxy-2-methyl-2-(E)- butenyl-4-diphosphate synthases include, but are not limited to, (AY033515; Escherichia coli), (NC_002947, locus_tag PP0853; Pseudomonas putida KT2440), (NC_007493, locus_tag RSP_2982; Rhodobacter sphaeroides 2.4.1), (067496, Aquifex aeolicus), (P54482, Bacillus subtilis), (Q9pky3, Chlamydia muridarum), (Q9Z8H0, Chlamydophila pneumoniae), (084060, Chlamydia

trachomatis), (P27433, Escherichia coli), (P44667, Haemophilus influenzae), (Q9ZLL0, Helicobacter pylori J99), (033350, Mycobacterium tuberculosis), (S77159, Synechocystis sp. ), (Q9WZZ3, Thermotoga maritima), (083460, Treponema pallidum), (Q9JZ40, Neisseria meningitidis),

(Q9PPM 1, Campylobacter jejuni), (Q9RXC9, Deinococcus radiodurans), (AAG07190, Pseudomonas aeruginosa) and (Q9KTX1, Vibrio cholerae).

In the seventh step, H M BPP is converted to isopentyl pyrophosphate ("IPP") and/or its isomer, dimethylallyl diphosphate ("DMAPP"). An enzyme known to catalyze this step is, for example, isopentyl/dimethylallyl diphosphate synthase. The gene encoding this enzyme is referred to as ispH or lytB. Illustrative examples of nucleotide sequences that encode a isopentyl/dimethylallyl diphosphate synthase include, but are not limited to, (AY062212; Escherichia coli), (NC_002947, locus_tag PP0606; Pseudomonas putida KT2440), (AF027189, Acinetobacter sp. BD413),

(AF098521, Burkholderia pseudomallei), (AF291696, Streptococcus pneumoniae), (AF323927, Plasmodium falciparum), (M87645, Bacillus subtillis), (U38915, Synechocystis sp. ), and (X89371, C. jejunisp 067496)

I PP and DMAPP are enzymatically interconverted. An enzyme known to catalyze this reaction is, for example, I PP isomerase. The gene encoding this enzyme is referred to as idi. Illustrative examples of idi nucleotide sequences include, but are not limited to, (NC_000913, 3031087..3031635; Escherichia coli), and (AF082326; Haematococcus pluvialis). In certain embodiments, the genetically modified organism encodes a heterologous nucleic acid encoding an enzyme capa ble of interconverting IPP and DMAPP, which ensures that the conversion of I PP into DMAPP is not a rate-limiting step in the biosynthesis of the desired isoprenoid. IPP and DMAPP are su bsequently converted to the desired isoprenoid compound (which occurs in five carbon units) via condensation. For example, condensation of two, three, and four five-carbon units (of either I PP, DMAPP, or com binations thereof) results in geranyl pyrophosphate, farnesyl pyrophosphate a nd gera nylgeranyl pyrophosphate, respectively, which are in turn modified by a terpene synthase to a desired monoterpene, sesquiterpene, and diterpene.

Engineering the DXP Pathway

As described previously, the DXP pathway illustrated in Figure 1 includes seven reactions. Two of these seven reactions are catalyzed by iron-sulfur enzymes. The first iron-sulfur enzyme converts cM EPP to H M BPP and the second iron-sulfur enzyme converts H M BPP to IPP and/or DMAPP (see, for example, reactons 6 and 7 of Figure 1). These enzymes are so-named because they include an iron-sulfur cofactor (also called an iron-sulfur cluster) in the active site of the enzyme that is required for their respective enzymatic reactions (which typically involve a reduction) to occur. Although many heterologous proteins (including the first five enzymes of the DXP pathway illustrated in Figure 1) are expressed successfully in non-native host cells, the expression of iron- sulfur enzymes has been far less successful due to the additional complexity of having multiple proteins involved in the proper assembly of an iron-sulfur cluster and its successful incorporation into an apoprotein. In addition, once properly assembled, the iron-sulfur enzyme still requires at least two partner proteins for it to be functionally competent to carry out its designated reaction. As a result, most heterologous expression of iron-sulfur enzymes have occurred in their endogenous host cells so that the additional complexities are eliminated. In this scenario, the iron-sulfur protein expression essentially becomes simply an over-expression of an endogenous enzyme.

As described previously, the DXP pathway is the endogenous pathway for most bacteria and for plastids, which are organelles found in plants and algae. Generally, successful heterologous expression ca n be expected within the same classification. For example, expressing a non-native bacterial DXP enzyme in a bacterial host cell with an endogenous DXP pathway, or expressing a non-native plastid DXP enzyme in a plastid with an endogenous DXP pathway. Although there is more complexity in these situations than simple over-expression of an endogenous enzyme, these situations are not too far from the baseline scenario of simple over-expression in a native host. In contrast to these simpler scenarios of expressing DXP proteins in hosts having endogenous DXP pathways, the successful expression of a heterologous iron-sulfur DXP enzyme cannot be assumed in a host cell lacking an endogenous DXP pathway for the reasons previously described. As an initial matter, a threshold inquiry is whether the host cell without an endogenous DXP pathway has cellular machinery that ca n properly assem ble the iron-sulfur cluster. The present inventor's studies have esta blished that eukaryotic cells without an endogenous DXP pathway do indeed have the proper cellular machinery for assem bling an iron-sulfur cluster. Given that the cellular machinery exists, the present inventors have determined that a major consideration in determining whether a DXP iron-sulfur enzyme can be successfully expressed, and can be functionally active in the host cell's cytosol, is its solu bility.

For example, if the host cell is yeast, the following unmodified IspG proteins a re not solu ble in the cytosol and thus are not suita ble for heterologous expression in their unmodified form in yeast (and likely not in eukaryotes generally): Acidithiobacillus ferrooxidans ATCC 53993

(YP_002220034.1); Caulobacter crescentus CB15 (NP_419668.1); Desulfovibrio vulgaris RCH 1 (ZP_04791624.1); Escherichia coli (AAC75568); Gemmata obscuriglobus UQM 2246

(ZP_02736855.1); Micrococcus luteus NCTC 2665 (YP_002956780.1); P. falciparum 3 D7

(NC_004314); Pseudomonas fluorescens SBW25 (YP_002874560.1); Planctomyces maris DSM 8797 (ZP_01855505.1); Physcomitrella patens subsp. Patens (XP_001766565); Pirellula staleyi DSM 6068 (YP_003370642.1); Rhodobacter sphaeroides ATCC 17025 (YP_001168057.1); Stevia rebaudiana (ABG75916); Streptomyces sviceus ATCC 29083 (ZP_06920121.1); Streptomyces coelicolor A3(2) (N P_630839); Thalassiosira pseudonana CCM P1335 (XP_002292108.1);

Wolbachia endosymbiont of Drosophila melanogaster (NP_965936.1); and Xanthomonas campestris pv. vesicatoria str. 85-10 (YP_363560.1). And, the following u nmodified IspH proteins are not solu ble in yeast cytosol and thus not suita ble for heterologous expression in their unmodified form in yeast (and likely not in eukaryotes generally): Acidithiobacillus ferrooxidans ATCC 53993 (YP_002221250.1); Bacillus coagulans 36D1 (ZP_04432889.1); Gemmata

obscuriglobus UQM 2246 (ZP_02735110.1); Magnetococcus sp. MC-1 (YP_867320.1); Micrococcus luteus NCTC 2665 (YP_002957672.1); Ostreococcus tauri (CAL54853.1); Prevotella buccae D17 (EFC75366.1); Plasmodium falciparum 3 D7 (CAD49005); Pirellula staleyi DSM 6068

(YP_003369145.1); and Porphyra yezoensis (ACI45962.1).

In contrast, the following IspG proteins are solu ble in yeast cytosol and thus suita ble for heterologous expression: Algoriphagus sp. PRl (ZP_01720138.1); Arthrospira platensis str. Paraca (ZP_06384238.1); Chlamydomonas reinhardtii (XP_001690937.1); Brevibacillus brevis NBRC 100599 (YP_002772921.1); Catharanthus roseus (AA024774.1); Crocosphaera watsonii WH 8501 (ZP_00518798.1); Cyanobium sp. PCC 7001 (ZP_05045928.1); Eubacterium dolichum DSM 3991 (ZP_02077259.1); Nostoc punctiforme PCC 73102 (YP_001868335.1); Ostreococcus tauri (CAL55258); Prevotella buccae D17 (EFC75705); Prochlorococcus marinus subsp. pastoris str. CCM P1986 (NP_892794.1); Porphyra yezoensis red algae (ACI45961); Sorghum bicolor

(XP_002454137); Synechococcus sp. S9917 (ZP_01081635.1); Thermosynechococcus elongatus BP-1 (N P_681786); and Vitis vinifera (XP_002285130.1). The following IspH proteins are solu ble in yeast cytosol, and thus suita ble for heterologous expression: Escherichia coli (AAC73140); Fibrobacter succinogenes subsp. succinogenes S85 (YP_003250012.1); Physcomitrella patens subsp. patens (XP_001758369.1); Arabidopsis thaliana (AY168881); Sorghum bicolor

(XP_002463933.1); Stevia rebaudiana (ABB88836.2); and Thermosynechococcus elongatus BP-1 (NP_681832).

In fact, each of the iron-sulfur enzymes described a bove that were solu ble in yeast cytosol were also found to be properly folded and functional as they each were a ble to catalyze their respective reactions in the presence of artificial electron donors. In addition, if an iron-sulfur protein was found to be insolu ble, it was modified or placed in a suitable environment wherein the iron-sulfur protein could be expressed and functional. For exa mple, the yeast mitochondrion is very similar in its environment to a bacterial cytoplasm. As a consequence, when E. coli IspG, which is insolu ble in yeast cytosol, is targeted to and expressed in the yeast mitochondrion, it is properly folded and functional therein. (Suita ble targeting sequences for targeting the DXP proteins to the mitochondrion can be identified using Mitroprot, availa ble from the Institute of Human Genetics, Melmholtz Center, Munich Germany, see Claros, et al., Computational method to predict mitochondrially imported proteins and their targeting sequences. Eur. J. Biochem. (1996) 241:779-786.) Curiously, all of the previous work relating to the expression of a heterologous DXP pathway, including IspG expression in a host cell without an endogenous DXP pathway, has been done on yeast using E. coli IspG.

It is also known that many endogenous DXP proteins in plant plastids have targeting sequences to the chloroplast. Plasmodium also has endogenous DXP pathway proteins that have targeting sequences to Apicoplast. In general, the successful heterologous expression of these proteins in a non-native host can be improved by removing these targeting sequences from these proteins before heterologous expression. The task of removing these targeting sequences from these DXP pathway proteins is well within the skill of one in the arts of molecular biology or bioengineering. Programs that can be used to identify such targeting sequences for identification and removal from these DXP proteins include: TargetP, availa ble from the Center for Biological Sequence Analysis (Technical University of Denmark, DTU) (see, Emanuelsson, et al., Locating proteins in the cell using TargetP, Signal P, and related tools, Nature Protocols (2007) 2:953-971) and PATS (Prediction of Apicoplast Targeted Sequences) made availa ble by the University of Frankfurt, Germany. (See, Zuegge et al., Deciphering apicoplast targeting signals - feature extraction from nuclear-encoded precursors of Plasmodium falciparum apicoplast proteins. Gene (2001) 280:19- 26.)

In addition to solu bility in the desired compartment of the host cell (typically cytoplasm), the reactions mediated by a DXP iron-sulfur enzyme involve electron transfer and at least two accessory proteins that are required for the proper function of an iron-sulfur enzyme. The first accessory protein is a reductase that is capa ble of accepting one or more electrons from an electron donor. The second accessory protein is a redox protein that is a redox partner to the iron-sulfur enzyme. The redox partner essentially receives the one or more electrons from the reductase (which in turn received the electrons from an electron donor) and transfers the one or more electrons to the iron-sulfur cluster in the iron-sulfur enzyme. Ultimately, the electrons are transferred by the iron-sulfur enzyme to a product of the enzymatic reaction catalyzed by the iron-sulfur enzyme.

Transfer of these electrons amongst the proteins in the iron-sulfur protein and the accessory proteins includes accepta nce of one or more electrons by a first receiving protein, which converts it from its oxidized state to its reduced state, and su bsequent transfer of the same electrons by the first receiving protein to a second receiving protein, which converts the first receiving protein back to its oxidized state (from its reduced state). Redox balance is achieved within this electron transport system by regenerating the original source of the electrons (e.g., by converting NAD ⁺ or NADP ⁺ to NADH or NADPH, respectively).

Another complexity in the expression of a heterologous DXP pathway in a host cell without an endogenous DXP pathway is that the two accessory proteins that are involved in each of the two iron-sulfur enzymatic reactions have not been conclusively identified. As a consequence, almost all of the work involving the DXP pathway has been in host cells that have an endogenous DXP pathway. Because host cells with an endogenous DXP pathway necessarily include all of the cofactors and proteins needed for a functional pathway, working in such host cells eliminates the need for also expressing the two accessory proteins for each of the two iron-sulfur enzymes. However, if the DXP pathway is to be a commercially via ble option for isoprenoid production, it is almost certain that even in host cells with an endogenous DXP pathway, the respective accessory proteins need to be over-expressed so that neither of the two iron-sulfur enzymes becomes rate limiting to the DXP pathway. In order to conclusively identify redox partner(s) compatible with IspG or IspH, the present inventors used the representative eukaryotic cell yeast for screening. The ferrodoxin and flavodoxin family of proteins and their cognate reductases were tested for their respective a bility to partner with IspG and IspH. This family was selected because they are the most common redox proteins involved in electron transfer in host cells with an endogenous DXP pathway.

Unfortunately, if the desired host cell is a eukaryotic cell such as yeast, neither a ferrodoxin nor a flavodoxin is known to be present in the yeast cytoplasm.

Notwithstanding this difficulty, the inventors of the present invention have identified the necessary accessory proteins that are compatible with IspG's and IspH's such that the respective enzyme can carry out its electron transfer reactions. The simplest iron-sulfur clusters have two iron ions bridged by two sulfide ions, and are designated "[2Fe-2S]" clusters. In general, [2Fe-2S]- dependent redox partners were not generally solu ble when expressed in yeast. For example the following [2Fe-2S] ferredoxins were not solu ble in yeast cytoplasm in their unmodified forms: Arthrospira platensis (P00246); Arabidopsis thaliana (AAG40057); Arabidopsis thaliana

(NP_180320); Bacillus coagulans (EEN90783); Chlamydomonas reihardtii (XP_001692808);

Chlamydomonas reihardtii (ABC88601); Escherichia coli str. K-12 su bstr. MG1655 (NP_417057.1); Nostoc strain MAC (P00252); Nostoc strain MAC (0812211B); Peptococcus aerogenes (P00193.1); Plasmodium falciparum 3D7 (Q8IED5); Saccharomyces cerevisiae (N P_015071); Spinacia oleracea (P00221.2); Synechocystis sp. 6714 (0812212A); and Thermosynechococcus elongatus BP-1 (NP_682446.1).

In contrast, the following flavodoxins were found to be both solu ble and, when paired with a suita ble reductase, capa ble of working with IspG: Alteromonas macleodii ATCC 27126

(ZP_04714638.1); Bacillus coagulans 36D1 (ZP_04431144.1); Bacillus subtilis subsp. subtilis str. 168 Yku P (N P_389300.2); Crocosphaera watsonii WJH 8501 (ZP_00515759.1); Desulfovibrio vulgaris subsp. vulgaris DP (YP_966026.1); Escherichia coli str. K-12 su bstr. MG1655 (N P_415210); Nostoc punctiforme PCC 73102 (YP_001866434.1); Anabaena PCC 7119 (AAB20462); Wolinella succinogenes DSM 1740 (NP_906993.1); a nd Bacillus subtilis subsp. subtilis str. 168 Yku N

(NP_389298.1).

In general, components of a fully functional IspH iron-sulfur enzyme system appeared more robust because they were more readily identified than those for IspG. Certain redox partners were found to be compatible with both IspG and IspH. Illustrative examples of two such redox partners include Escherichia coli str. K-12 su bstr. MG1655 (NP_415210) and Bacillus subtilis subsp. subtilis str. 168 Yku N (N P_389298.1). However, each IspG and IspH provided the best activity if paired with a different redox partner, suggesting that the respective redox partner in the endogenous system is specific and distinct.

Because flavodoxins were identified as compatible redox partners for the two DXP iron-sulfur enzymes, the suita ble reductases included ferrodoxin, and flavodoxin NADP+ reductases ("FN Rs"). As with the other components of the iron-sulfur enzyme system, the crucial determinant of heterogeneous fu nction was solu bility of an FN R in yeast cytoplasm. In general, if an FN R is solu ble then it is likely to be compatible with flavodoxins. Illustrative examples of suita ble FN Rs include, but are not limited to, Anabaena PCC 7119 (1QUE_A); Arabidopsis thaliana (root) (NP_849734.1); Escherichia coli str. K-12 su bstr. MG1655 (NP_418359); Nostoc sp. PCC 7906 (BAG69182); Spinacia oleracea (P00455); Thermosynechococcus elongatus BP-1

(NP_682001); and lea mays (3LVB_A).

Thus, the present invention provides iron-sulfur enzyme systems comprising an iron-sulfur DXP enzyme and one or more accessory proteins that are compatible with the iron-sulfur enzyme. In one em bodiment, the present invention provides iron-sulfur enzyme systems comprising an iron- sulfur DXP enzyme and the two accessory proteins that are compatible with the iron-sulfur enzyme, so that the enzyme is a ble to catalyze its enzymatic reaction. There is no requirement that the two accessory proteins are those that are the naturally-occurring enzymatic partners for each iron-sulfu r enzyme in their endogenous host. The iron-sulfur enzyme system (iron-sulfur enzyme plus the two accessory proteins) of the present invention can be used for enhancing a DXP pathway-mediated isoprenoid prod uction in a host cell with an endogenous DXP pathway, or for engineering a heterologous DXP pathway in a host cell lacking an endogenous DXP pathway, and using the heterologous DXP pathway for isoprenoid production.

Thus, in one aspect of the invention, an iron sulfur enzyme system is provided comprising one or more nucleic acid molecules comprising one or more nucleic acid sequences encoding:

(a) a reductase that is capa ble of accepting electrons from an electron donor thereby converting the reductase from an oxidized state ("oxidized reductase") to a reduced state ("reduced reductase");

(b) a redox protein that is capa ble of accepting electrons from the reduced reductase thereby converting the redox protein from an oxidized state ("oxidized redox protein") to a reduced state ("reduced redox protein"); and

(c) an iron-sulfur enzyme that is capa ble of accepting electrons from the reduced redox protein thereby converting the iron-sulfur enzyme from an oxidized state ("oxidized iron-sulfu r enzyme") to a red uced state ("reduced iron-sulfur enzyme") and wherein the reduced iron-sulfur enzyme is capa ble of converting 2C-methyl-D-erythritol-2,4-cyclodiphosphate ("cM EPP") to 1- hydroxy-2-methyl-2-(E)-butenyl-4-diphosphate ("HM BPP").

In another aspect of the invention, an iron sulfur enzyme system is provided comprising one or more nucleic acid molecule comprising one or more nucleic acid sequences encoding:

(a) a reductase that is capa ble of accepting electrons from an electron donor thereby converting the reductase from an oxidized state ("oxidized reductase") to a reduced state ("reduced reductase");

(b) a redox protein that is capa ble of accepting electrons from the reduced reductase thereby converting the redox protein from an oxidized state ("oxidized redox protein") to a reduced state ("reduced redox protein"); and

(c) a iron-sulfur enzyme that is capa ble of accepting electrons from the reduced redox protein thereby converting the iron-sulfur enzyme from an oxidized state ("oxidized iron-sulfu r enzyme") to a red uced state ("reduced iron-sulfur enzyme") and wherein the reduced iron-sulfur enzyme is capa ble of converting cM EPP to isopentenyl pyrophosphate ("I PP") and dimethylallyl pyrophosphate ("DMAPP").

In another aspect of the invention, a genetically mod ified microorganism is provided comprising one or more heterologous nucleic acid molecules comprising one or more nucleic acid sequences that encode:

(a) a first reductase that is capa ble of accepting electrons from an electron donor thereby converting the first red uctase from an oxidized state ("oxidized first reductase") to a reduced state ("reduced first reductase");

(b) a first redox protein that is capa ble of accepting electrons from the reduced first reductase thereby converting the first redox protein from an oxidized state ("oxidized first redox protein") to a reduced state ("red uced first redox protein"); and

(c) a first iron-sulfur enzyme that is capa ble of accepting electrons from the first reduced redox protein thereby converting the first iron-sulfur enzyme from an oxidized state ("oxidized first iron-sulfur enzyme") to a reduced state ("reduced first iron-sulfu r enzyme") and wherein the reduced first iron-sulfur enzyme is capa ble of converting 2C-methyl-D-erythritol-2,4- cyclodiphosphate ("cM EPP") to l-hydroxy-2-methyl-2-(E)-butenyl-4-diphosphate ("H M BPP"). In other embodiments, the one or more heterologous nucleic acid molecules further encode:

(g) a second reductase that is capa ble of accepting electrons from an electron donor thereby converting the second reductase from an oxidized state ("oxidized second reductase") to a reduced state ("reduced second reductase"); (h) a second redox protein that is capable of accepting electrons from the reduced second reductase thereby converting the second redox protein from an oxidized state ("oxidized second redox protein") to a reduced state ("reduced second redox protein"); and

(i) a second iron-sulfur enzyme that is capable of accepting electrons from the second reduced redox protein thereby converting the second iron-sulfur enzyme from an oxidized state ("oxidized second iron-sulfur enzyme") to a reduced state ("reduced second iron-sulfur enzyme") and wherein the reduced second iron-sulfur enzyme is capable of converting HMBPP to isopentenyl pyrophosphate ("IPP") or/and dimethylallyl pyrophosphate ("DMAPP").

In another aspect of the invention, a genetically modified microorganism is provided comprising one or more heterologous nucleic acid molecules wherein the nucleic acid molecules encode at least one of the following enzymes:

(a) an enzyme that converts pyruvate and glyceraldehydes 3-phosphate to 1-deoxy-D- xylulose-5-phosphate ("DXP");

(b) an enzyme that converts DXP to lC-methyl-D-erythritol-4-phosphate ("MEP");

(c) an enzyme that converts MEP to 4-diphosphocytidyl-2C-methyl-D-erythritol ("CDP-ME");

(d) an enzyme that converts CDP-ME to 4-diphosphocytidyl-2C-methyl-D-erythritol-2- phosphate ("CDP-M EP"); and

(e) an enzyme that converts CDP-MEP to 2C-methyl-D-erythritol 2,4-cyclodiphosphate ("cMEPP");

and the nucleic acid molecules further encode:

(f) a first reductase that is capable of accepting electrons from an electron donor thereby converting the first reductase from an oxidized state ("oxidized first reductase") to a reduced state ("reduced first reductase");

(g) a first redox protein that is capable of accepting electrons from the reduced first reductase thereby converting the first redox protein from an oxidized state ("oxidized first redox protein") to a reduced state ("reduced first redox protein"); and

(h) a first iron-sulfur enzyme that is capable of accepting electrons from the first reduced redox protein thereby converting the first iron-sulfur cluster enzyme from an oxidized state ("oxidized first iron-sulfur enzyme") to a reduced state ("reduced first iron-sulfur enzyme") and wherein the reduced first iron-sulfur cluster enzyme is capable of converting 2C-methyl-D- erythritol-2,4-cyclodiphosphate ("cM EPP") to l-hydroxy-2-methyl-2-(E)-butenyl-4-diphosphate ("HMBPP"); (i) a second reductase that is capable of accepting electrons from an electron donor thereby converting the second reductase from an oxidized state ("oxidized second reductase") to a reduced state ("reduced second reductase");

(j) a second redox protein that is capable of accepting electrons from the reduced second reductase thereby converting the second redox protein from an oxidized state ("oxidized second redox protein") to a reduced state ("reduced second redox protein"); and

(k) a second iron-sulfur enzyme that is capable of accepting electrons from the second reduced redox protein thereby converting the second iron-sulfur enzyme from an oxidized state ("oxidized second iron-sulfur enzyme") to a reduced state ("reduced second iron-sulfur enzyme") and wherein the reduced second iron-sulfur enzyme is capable of converting HMBPP to isopentenyl pyrophosphate ("IPP") or/and dimethylallyl pyrophosphate ("DMAPP").

For all of the above aspects, in some embodiments, the first reductase and the second reductase are the same reductase. In still other embodiments, the first redox protein and the second redox protein are the same redox protein. In other embodiments, the first reductase and the second reductase are the same reductase and the first redox protein and the second redox protein are the same redox protein.

In some embodiments, the unmodified microorganism does not comprise an endogenous DXP pathway. In other embodiments, the genetically modified microorganism is a eukaryote. In other embodiments, the genetically modified microorganism is a fungus. In other embodiments, the genetically modified microorganism is yeast. In still other embodiments, the genetically modified microorganism is S. cerevisiae.

In another aspect of the invention, the genetically modified microorganism is provided comprising a heterologous DXP pathway that is capable of making IPP or DMAPP therefrom and wherein the unmodified microorganism does not comprise an endogenous DXP pathway.

In some embodiments, first iron-sulfur enzyme is not IspG from E. coli. In other embodiments, the first iron-sulfur enzyme is IspG from a Bacillus. In other embodiments, the first iron-sulfur enzyme is IspG from B. coagulans, B. thuringiensis, or B. subtilis. In still other embodiments, the first iron-sulfur enzyme is IspG from Thermus. In yet other embodiments, the first iron-sulfur enzyme is IspG from T. thermophilus.

In some embodiments, the second iron-sulfur enzyme is IspH from E. coli. In other embodiments, the second iron-sulfur enzyme is IspH from F. succinogens. In other embodiments, the second iron-sulfur enzyme is IspH from P. patens. In other embodiments, the second iron-sulfur enzyme is IspH from 5. bicolor. In other embodiments, the second iron-sulfur enzyme is IspH from 5. rebaudiana. In other em bodiments, the second iron-sulfur enzyme is IspH from T. elongatus. In still other embodiments, the second iron-sulfur enzyme is IspH from A. thaliana.

In some cases, the activity of the iron-sulfur enzyme is improved by creating a fusion protein between it and its redox protein. As a result, in some em bodiments, the first redox protein and the first iron-sulfur enzyme form a contiguous polypeptide. In other embodiments, the first redox protein and the second iron-sulfur enzyme form a contiguous polypeptide. In still other embodiments, the second redox protein and the second iron-sulfu r enzyme form a contiguous polypepetide.

In some em bodiments, the electron donor is NADPH. In other em bodiments, the reductase is a flavodxin/ferredoxin NADP+ reductase ("FN "). In other embodiments the reductase is an FNR from E. coli. In other em bodiments, the reductase is an FNR from A. thaliana.

In some embodiments, the first redox protein is a flavodoxin. In still other em bodiments the first redox protein is a flavodoxin from B. subtilis. In yet still other em bodiments, the first redox protein is Yku N from B. subtilis.

In some embodiments, the second redox protein is a flavodoxin. In other em bodiments, the second redox protein is a flavodoxin from E. coli.

In some embodiments, the first reductase is an FN R. In other em bodiments, the second reductase is an FNR. In still other embodiments, the first and second reductase is a FN R. In yet other em bodiments, the first reductase and the second red uctase are different FN Rs.

Additional Engineering for Augmenting the DXP Pathway

Flux through the DXP pathway may also be augmented by increasing the efficiency or output of iron-sulfur cluster biogenesis, assembly and/or su bsequent insertion of these iron-sulfur clusters into the respective apoprotein. As described previously, complex, cellular machinery is involved in these processes that includes more than 25 proteins that have been identified to date.

The ISC assem bly machinery is part of this complex system. It is found in bacteria and in mitochondria and is involved in the biogenesis of all iron-sulfur proteins. Because of its importance in the biogenesis of all iron-sulfur proteins, the over-expression of any su bset of its mem bers increases the levels of iron-sulfur proteins generally, including the two DXP iron-sulfur enzymes.

In eukaroyotes, in addition to the ISC assem bly machinery, the ISC export machinery and the cytosolic CIA machinery are required. The ISC assembly, found in the mitochondrion in eukaroyotes, comprises approximately 15 proteins. Illustrative examples of proteins that comprise the ISC assem bly machinery include, but are not limited to, cysteine desulfurase Nfsl; iron bind ing protein frataxin Yfhl; redox proteins Arhl and Yahl; NADH kinase POS5; scaffold proteins Isu l and Isu2; and chaperones Ssl-ATP, Jacl and Grx5.

The mitochondrion also includes the ISC export machinery that exports key cellular components necessary for cytosolic iron-sulfur cluster assem bly a nd su bsequent insertion into the apoprotein, out of the mitochondrion. Illustrative examples of proteins that comprise the ISC export machinery include, but are not limited to, ABC transporter Atml; the sulfhydryl oxidase Ervl, and glutathione. The cytosolic CIA machinery is involved in the maturation of the iron-sulfur apoprotein. Illustrative examples of proteins that comprise the CIA machinery include, but a re not limited to, Cfd l-Nbp35; Narl, Cia l and Cia2; and Dre2. As with the ISC assembly machinery, the over-expression of any su bset of mem bers of either the ISC export machinery, the CIA machinery, or both, increases the levels of iron-sulfur proteins generally including the two DXP iron-sulfur enzymes.

General Meta bolic Engineering

Other aspects of the invention include vectors comprising the heterologous nucleic acid sequences as well as methods for using genetically modified host cells of the present invention to make isoprenoid(s).

The heterologous nucleic acid sequences of the present invention can be expressed by a single or by multiple vectors. The nucleic acid sequences can be arranged in any order in a single operon, or in separate operons that are placed in one or multiple vectors. Where desired, two or more expression vectors can be employed, each of which contains one or more heterologous sequences opera bly linked in a single operon. While the choice of single or multiple vectors and the use of single or multiple promoters may depend on the size of the heterologous sequences and the capacity of the vectors, it will largely depend on the overall yield of a given isoprenoid that the vector is a ble to provide when expressed in a selected host cell. In some instances, two- operon expression systems provide a higher yield of isoprenoid. The su bject vectors can stay replica ble episomally, or as an integral part of the host cell genome.

In certain em bodiments, the heterologous nucleic acids of the present invention are under the control of a single regulatory element. In some cases, the heterologous nucleic acid sequences are regulated by a single promoter. In other cases, the heterologous nucleic acid sequences are placed within a single operon. In still other cases, the heterologous nucleic acid sequences are placed within a single reading frame.

Where desired, the su bject nucleic acid sequences can be modified to reflect the codon preference of a selected host cell to effect a higher expression of such sequences in a host cell. For example, the su bject nucleotide sequences will in some embodiments be modified for yeast codon preference. (See, e.g., Bennetzen and Hall, J. Biol. Chem. (1982) 257(6): 3026-3031). As another non-limiting example, the nucleotide sequences will in other embodiments be modified for E. coli codon preference. (See, e.g., Gouy a nd Gautier, Nucleic Acids Res. (1982) 10(22):7055- 7074; Eyre-Walker, Mol. Biol. Evol. (1996) 13(6):864-872 (see also, Nakamura et al., Nucleic Acids Res. (2000) 28(1):292.) Codon usage ta bles for many organisms are availa ble, which can be used as a reference in designing sequences of the present invention. The use of prevalent codons of a given host microorganism generally increases the likelihood of translation, and hence the expression level of the desired sequences.

Preparation of the su bject nucleic acids can be carried out by a variety of routine recom bina nt techniques and synthetic procedures. Standard recom binant DNA and molecular cloning techniques used in the Examples of this d isclosure are well known in the art and are described by Sam brook, J., Fritsch, E. F. and Maniatis, T. Molecular Cloning: A Laboratory Manual; Cold Spring Harbor La boratory Press: Cold Spring Ha rbor, (1989) (Maniatis) and by T. J. Silhavy, M. L. Bennan, and L. W. Enquist, Experiments with Gene Fusions, Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. (1984) and by Ausu bel, F. M. et al., Current Protocols in Molecular Biology, pu b. by Greene Pu blishing Assoc. and Wiley-lnterscience (1987). Briefly, the su bject nucleic acids may be genomic DNA fragments, cDNAs, and RNAs, all of which can be synthesized, purchased commercially, extracted directly from a cell or recom binantly produced by various amplification processes including, but not limited to, PCR and rt-PCR.

Direct chemical synthesis of nucleic acids typically involves sequential addition of 3'-blocked and 5'-blocked nucleotide monomers to the terminal 5'-hydroxyl group of a growing nucleotide polymer chain, wherein each addition is effected by nucleophilic attack of the terminal 5'- hydroxyl group of the growing chain on the 3'-position of the added monomer, which is typically a phosphorus derivative, such as a phosphotriester, phosphoramidite, or the like. Such methodology is known to those of ordinary skill in the art and is described in the pertinent texts and literature (see, for exa mple, Matteuci et al. Tet. Lett. (1980) 521:719; U.S. Patent No.

4,500,707; 5,436,327; and 5,700,637).

The present invention relates in some em bodiments to an increase in the level of transcription of nucleic acids in a host organism. The level of transcription of a nucleic acid in a host

microorganism can be increased in a number of ways. For example, this can be achieved by increasing the copy num ber of the nucleotide sequence encoding the enzyme (e.g., by using a higher copy num ber expression vector comprising a nucleotide sequence encoding the enzyme, or by introducing additional copies of a nucleotide sequence encoding the enzyme into the genome of the host microorganism, for example, by recA-mediated recom bination, use of "suicide" vectors, recombination using lam bda phage recom binase, and/or insertion via a transposon or transposa ble element). In addition, it can be carried out by changing the order of the coding regions on the polycistronic m NA of an operon or breaking up an operon into individual genes, each with its own control elements, or increasing the strength of the promoter (transcription initiation or transcription control sequence) to which the enzyme coding region is opera bly linked (for example, using a consensus ara binose- or lactose-inducible promoter in an Escherichia coli host microorganism in place of a modified lactose-inducible promoter, such as the one found in pBluescript and the pBBRlMCS plasmids), or using an inducible promoter and inducing the inducible-promoter by adding a chemical to the growth medium.

The level of translation of a nucleotide sequence in a host microorganism of the invention can be increased in a num ber of ways, including, but not limited to, increasing the sta bility of the m RNA, modifying the sequence of the ribosome binding site, modifying the distance or sequence between the ribosome binding site and the start codon of the enzyme coding sequence, modifying the entire intercistronic region located upstream of, or adjacent to the 5' side of the start codon of the enzyme coding region, sta bilizing the 3'-end of the m RNA tra nscript using hairpins and specialized sequences, modifying the codon usage of enzyme, altering expression of rare codon tRNAs used in the biosynthesis of the enzyme, and/or increasing the sta bility of the enzyme, as, for example, via mutation of its coding sequence. Determination of preferred codons and rare codon tRNAs ca n be based on a sequence analysis of genes derived from the host microorganism.

The activity of a DXP pathway enzyme or prenyltransferase in a host can be altered in a num ber of ways, including, but not limited to, expressing a modified form of the enzyme that exhibits increased solu bility in the host cell, expressing an altered form of the enzyme that lacks a domain through which the activity of the enzyme is inhibited, expressing a modified form of the enzyme that has a higher Kcat or a lower Km for the su bstrate, or expressing an altered form of the enzyme that is not affected by feed-back or feed-forward regulation by another molecule in the pathway. Such variant enzymes can also be isolated through random mutagenesis of a broader specificity enzyme, and a nucleotide sequence encoding such variant enzyme can be expressed from an expression vector or from a recom binant gene integrated into the genome of a host microorganism.

The su bject vector can be constructed to yield a desired level of copy num bers of the encoded enzyme. In some em bodiments, the su bject vectors yield at least 10, between 10-20, between 20-50, between 50-100, or even higher than 100 copies of the desired enzymes. Low copy number plasmids generally provide fewer than a bout 20 plasmid copies per cell; medium copy number plasmids generally provide from a bout 20 plasmid copies per cell to a bout 50 plasmid copies per cell, or from a bout 20 plasmid copies per cell to a bout 80 plasmid copies per cell; and high copy num ber plasmids generally provide from a bout 80 plasmid copies per cell to a bout 200 plasmid copies per cell, or more.

Suita ble low copy expression vectors for Escherichia coli include, but are not limited to, pACYC184, pBeloBacll, pBR332, pBAD33, pBBRlMCS and its derivatives, pSClOl, SuperCos (cosmid), and pWE15 (cosmid). Suita ble medium copy expression vectors for Escherichia coli include, but are not limited to, pTrc99A, pBAD24, and vectors containing a ColEl origin of replication a nd its derivatives. Suita ble high copy number expression vectors for Escherichia coli include, but are not limited to, pUC, pBluescript, pGEM, and pTZ vectors. Suitable low-copy (centromeric) expression vectors for yeast include, but are not limited to, pRS415 and pRS416 (Sikorski & Hieter (1989) Genetics 122:19-27). Suita ble high-copy 2 micron expression vectors in yeast include, but are not limited to, pRS425 and pRS426 (Christainson et al. Gene (1992) 110:119-122). Alternative 2 micron expression vectors include non-selecta ble variants of the 2 micron vector (Bruschi & Ludwig Curr. Genet. (1988) 15:83-90) or intact 2 micron plasmids bearing an expression cassette (as exemplified in U.S. Patent No. 7,419,801, issued September 2, 2008) or 2 micron plasmids bearing a defective selection marker such as LEU2d (Erhanrt et al., J. Bacteriol. (1983) 156 (2): 625-635) or URA3d (Okkels, Annals of the New York Academy of Sciences (1996) 782(1): 202-207).

Regulatory elements include, for example, promoters and operators, which can also be engineered to increase the meta bolic flux of the DXP pathways by increasing the expression of one or more genes that play a significa nt role in determining the overall yield of an isoprenoid produced. A promoter is a sequence of nucleotides that initiates and controls the transcription of a nucleic acid sequence by an RNA polymerase enzyme. An operator is a sequence of nucleotides adjacent to the promoter that functions to control transcription of the desired nucleic acid sequence. The operator contains a protein-binding domain where a specific repressor protein can bind. In the a bsence of a suita ble repressor protein, transcription initiates through the promoter. In the presence of a suita ble repressor protein, the repressor protein binds to the operator and inhibits transcription from the promoter.

In some embodiments of the present invention, promoters used in expression vectors are inducible. In other embodiments, the promoters used in expression vectors are constitutive. In some embodiments, one or more nucleic acid sequences are opera bly linked to an inducible promoter, and one or more other nucleic acid sequences are opera bly linked to a constitutive promoter. Non-limiting examples of suita ble promoters for use in prokaryotic host cells include a bacteriophage T7 NA polymerase promoter; a trp promoter; a lac operon promoter; a hybrid promoter (for example, a lac/tac hybrid promoter, a tac/trc hybrid promoter, a trp/lac promoter, a T7/lac promoter; a trc promoter; a tac promoter, and the like); an ara BAD promoter; in vivo regulated promoters (such as an ssaG promoter or a related promoter; see, for example, U.S. Patent Pu blication No. 2004/0131637, pu blished July 8, 2004), a pagC promoter (Pulkkinen and Miller, J. Bacteriol. (1991) 173(l):86-93; Alpuche-Aranda et al. PNAS U.S.A. (1992) 89(21): 10079- 83), a nirB promoter (Harborne et al. Mol. Micro. (1992) 6:2805-2813), and the like (see, for example, Du nstan et al. Infect. Immun. (1999) 67:5133-5141; McKelvie et al. Vaccine (2004) 22:3243-3255; and Chatfield et al. Biotechnol. (1992) 10:888-892); a sigma70 promoter (for example, a consensus sigma70 promoter (see, GenBank Accession Nos. AX798980, AX798961, and AX798183); a stationary phase promoter, for example, a dps promoter, an spv promoter, and the like; a promoter derived from the pathogenicity island SPI-2 (see, PCT pu blication No. WO 96/17951, pu blished June 13, 1996); an actA promoter (see, Shetron-Rama et a l. (2002) Infect. Immun. 70: 1087-1096); an rpsM promoter (see, Valdivia and Falkow, Mol. Microbiol. (1996) 22:367 378); a tet promoter (see, Hillen et al. (1989) In Saenger W. and Heinemann U. (eds) Topics in Molecular and Structural Biology, Protein-Nucleic Acid Interaction. Macmillan, London, UK, Vol. 10, pp. 143-162); an SP6 promoter (see, Melton et al. Nucl. Acids Res. (1984) 12:7035- 7056); and the like.

In some embodiments of the present invention, the total activity of heterologous enzyme(s) that play a larger role in the overall yield of isoprenoid(s), relative to other enzymes in the respective pathways, may be increased by expressing the enzyme(s) from a strong promoter. Suita ble strong promoters for Escherichia coli include, but are not limited to, Trc, Tac, T5, T7, a nd PLam bda. In another embodiment of the present invention, the total activity of the one or more M EV pathway enzymes in a host is increased by expressing the enzyme from a strong promoter on a high copy number plasmid. Suita ble examples of such high copy number plasmids for Escherichia coli include, but are not limited to, Trc, Tac, T5, T7, and PLam bda promoters used with pBAD24, pBAD18, pGEM, pBluescript, pUC, and pTZ vectors.

Non-limiting examples of suita ble promoters for use in eukaryotic host cells include, but are not limited to, a CMV immediate early promoter, an HSV thymidine kinase promoter, an early or late SV40 promoter, LTRs from retroviruses, and a mouse metallothionein-l promoter.

Non-limiting examples of suita ble constitutive promoters for use in prokaryotic host cells include a sigma70 promoter (for example, a consensus sigma70 promoter). Non-limiting examples of suita ble inducible promoters for use in bacterial host cells include the pL of bacteriophage λ; Plac; Ptrp; Ptac (Ptrp-lac hybrid promoter); an isopropyl-beta-D44 thiogalactopyranoside (I PTG)- inducible promoter (for example, a lacZ promoter); a tetracycline inducible promoter; an ara binose inducible promoter (for example, PBAD; see, Guzman et al. (1995) J. Bacteriol.

177:4121-4130); a xylose-inducible promoter (for example, Pxyl; see, Kim et al. (1996) Gene 181:71-76); a GAL1 promoter; a tryptophan promoter; a lac promoter; an alcohol-inducible promoter (for example, a methanol-inducible promoter or an ethanol-inducible promoter); a raffinose-inducible promoter; a heat-inducible promoter (for example, heat inducible lam bda PL promoter); a promoter controlled by a heat-sensitive repressor (for example, CI857-repressed lam bda-based expression vectors; see, Hoffmann et al. (1999) FEMS Microbiol Lett. 177(2):327- 34); and the like.

Non-limiting examples of suita ble constitutive promoters for use in yeast host cells include a n ADH1, an ADH2, a PGK, or a LEU2 promoter. Non-limiting examples of suita ble inducible promoters for use in yeast host cells include, but are not limited to, a divergent galactose- inducible promoter such as a GAL 1 or a GAL 10 promoter (West at al. (1984) Mol. Cell. Biol. 4(ll):2467-2478), or a CU P1 promoter. Where desired, the su bject vector may comprise a promoter that is stronger than a native E. Coli Lac promoter.

Non-limiting examples of operators for use in bacterial host cells include a lactose promoter operator (the Lacl repressor protein changes conformation when contacted with lactose, thereby preventing the Lacl repressor protein from binding to the operator), a tryptophan promoter operator (when complexed with tryptophan, the Trp repressor protein has a conformation that binds the operator; in the a bsence of tryptophan, the TrpR repressor protein has a conformation that does not bind to the operator), and a tac promoter operator (see, for example, deBoer et al. (1983) PNAS U.S.A. 80:21-25). The genes in the expression vector typically also encode a ribosome binding site to direct translation (that is, synthesis) of a ny encoded m RNA gene product. For suita ble ribosome binding sites for use in Escherichia coli, see Shine et al. (1975) Nature 254:34, and Steitz, in Biological Regulation and Development: Gene Expression (ed. R. F. Gold berger), vol. 1, p. 349, 1979, Plenum Pu blishing, N.Y.). Insertion of the ribosome binding site encoding nucleotide sequences such as 5'-AAAACA-3' upstream of a coding sequence facilitates efficient translation in a yeast host microorganism (Looman et al. (1993) Nuc. Ac. Res. 21:4268- 4271; Yun et. al. (1996) Mol. Microbiol. 19: 1225-1239). In some em bodiments of the invention, optimized Shine-Dalgarno sequences are inserted upstream of the coding region of a gene that is placed within a DXP module. The optimized Shine-Dalgarno sequences ensure efficient ribosome binding and translation of the coding sequence. Other regulatory elements that may be used in an expression vector include transcription enhancer elements and transcription terminators. See, for example, Bitter et al. (1987) Methods in Enzymology, 153:516-544.

An expression vector may be suita ble for use in particu lar types of host microorganisms and not others. One of ordinary skill in the art, however, can readily determine through routine experimentation whether a particular expression vector is suited for a given host microorganism. For example, the expression vector can be introduced into the host organism, which is then monitored for via bility a nd expression of any genes contained in the vector.

The expression vector may also contain one or more selecta ble marker genes that, upon expression, confer one or more phenotypic traits useful for selecting or otherwise identifying host cells that carry the expression vector. Non-limiting exa mples of suita ble selecta ble markers for eukaryotic cells include dihydrofolate reductase and neomycin resistance. Non-limiting examples of suita ble selecta ble markers for prokaryotic cells include tetracycline, ampicillin, chloramphenicol, carbenicillin, and kanamycin resistance.

For the production of isoprenoid(s) at an industrial scale, it may be impractical or too costly to use a selecta ble marker that requires the addition of an antibiotic to the fermentation media. Accordingly, some em bodiments of the present invention employ host cells that do not require the use of an antibiotic resistance-conferring selecta ble marker to ensure plasmid (expression vector) maintenance. In these em bodiments of the present invention, the expression vector contains a plasmid maintenance system such as the 60-kb IncP (RK2) plasmid, optionally together with the RK2 plasmid replication and/or segregation system, to effect plasmid retention in the a bsence of antibiotic selection (see, for example, Sia et al. (1995) 7. Bacteriol. 177:2789-97; or Pansegrau et al. (1994) J. Mol. Biol. 239:623-63). A suita ble plasmid maintenance system for this purpose is encoded by the parDE operon of RK2, which codes for a sta ble toxin and an u nsta ble antitoxin. The antitoxin can inhibit the lethal action of the toxin by direct protein-protein interaction. Cells that lose the expression vector that harbors the parDE operon are quickly deprived of the unsta ble antitoxin, resulting in cell death caused by the sta ble toxin. The RK2 plasmid replication system is encoded by the trfA gene, which codes for a DNA replication protein. The RK2 plasmid segregation system is encoded by the parCBA operon, which codes for proteins that function to resolve plasmid multimers that may arise from DNA replication.

The su bject vectors ca n be introduced into a host cell sta bly or transiently by a variety of esta blished techniques. For example, one method involves a calcium chloride treatment wherein the expression vector is introduced via a calcium precipitate. Other salts, for exa mple calcium phosphate, may also be used following a similar procedure. In addition, electroporation (that is, the application of current to increase the permea bility of cells to nucleic acids) may be used. Other transformation methods include microinjection, DEAE dextran mediated transformation, and heat shock in the presence of lithium acetate. Lipid complexes, liposomes, and dendrimers may also be employed to transfect the host microorganism.

Upon transformation, a variety of methods can be practiced to identify the host cells into which the su bject vectors have been introduced. One exemplary selection method involves su bculturing ind ividual cells to form individual colonies, followed by testing for expression of the desired gene product. Another method entails selecting transformed host cells based upon phenotypic traits conferred through the expression of selecta ble marker genes contained within the expression vector. Those of ordinary skill can identify genetically-modified host cells using these or other methods availa ble in the art.

In some embodiments, the heterlogous nucleic acid sequences can be inserted into the genome of the host organism. One method useful for the introduction of such sequences is the use of integration cassettes. Integration cassettes are typically linear, dou ble-stranded DNA fragments that can be chromosomally integrated by homologous recombination via the use of two PCR- generated fragments or one PCR-generated fragment. The integration cassette comprises a nucleic acid integration fragment that contains an expressible DNA fragment and a selecta ble marker bou nded by specific recombinase sites responsive to a site-specific recom binase, and homology arms having homology to different portions of the host cell's chromosome, (see, for example, U.S. Patent Pu blication No. 2004/0219629, pu blished November 4, 2004). Generally, the preferred length of the homology arms is a bout 10 to a bout 100 base pairs in length. From 20 to 40 base pairs of homology, the efficiency of homologous recombination increases by four orders of magnitude (Yu et al. PNAS (2000) 97:5978-5983). One method of introducing DXP modules into the host genome utilizes the λ-Red recom binase system. The λ-Red system ena bles the use of homologous recom bination as a tool for in vivo chromosomal engineering in hosts, such as E. coli, normally considered difficult to transform by homologous recom bination. The λ- Red system works in other bacteria as well (Poteete, A., supra, 2001). Use of the λ-Red recom binase system ca n be applica ble to other hosts generally used for ind ustrial production. The introd uction of various pathway sequences of the invention into a host cell can be confirmed by methods such as PCR, Southern blot or Northern blot hybridization. For example, nucleic acids can be prepared from the resultant host cells, and the specific sequences of interest can be amplified by PCR using primers specific for the sequences of interest. The amplified product is su bjected to agarose gel electrophoresis, polyacrylamide gel electrophoresis or capillary electrophoresis, followed by staining with ethidium bromide, SYBR Green solution or the like, or detection of DNA with UV light. Alternatively, nucleic acid probes specific for the sequences of interest can be employed in a hybridization reaction. The expression of a specific gene sequence can be ascertained by detecting the corresponding m RNA via reverse-transcription coupled PCR, Northern blot hybridization, or by immunoassays using antibodies reactive with the encoded gene product. Exemplary immunoassays include, but are not limited to, ELISA,

radioimmunoassays, and sandwich immunoassays.

The enzymatic activity of a given pathway enzyme can be assayed by a variety of methods known in the art. In general, the enzymatic activity can be ascertained by the formation of the product or conversion of a su bstrate of an enzymatic reaction that is under investigation. The reaction can take place in vitro or in vivo.

The yield of an isoprenoid via one or more meta bolic pathways disclosed herein can be augmented by inhibiting reactions that divert intermediates away from productive meta bolic reactions that direct the intermediates toward the formation of the desired isoprenoid product(s). Inhibition of such unproductive reactions can be achieved by reducing the expression and/or activity of enzymes involved in one or more u nproductive reactions. Such reactions include side reactions of the TCA cycle that lead to fatty acid biosynthesis, alanine biosynthesis, the aspartate superpathway, gluconeogenesis, heme biosynthesis, and/or glutamate

biosynthesis, at a level that affects the overall yield of isoprenoid production. Inhibition can be accomplished by reducing or eliminating the expression of certain genes in the target pathway, or in competing pathways, that may serve as competing sinks for energy or carbon. Where the sequence of the gene to be disrupted is known, one effective method of gene down-regulation is targeted gene disruption, where foreign DNA is inserted into a structural gene so as to disrupt transcription. This can be effected by the creation of genetic cassettes comprising the DNA to be inserted (often a genetic marker) flanked by sequence having a high degree of homology to a portion of the gene to be disrupted. Introd uction of the cassette into the host cell results in insertion of the foreign DNA into the structural gene via the native DNA replication mechanisms of the cell or by the λ-Red recombination system. (See, for example, Hamilton et al., J. Bacteriol., (1989) 171:4617-4622); Bal bas et al., Gene (1993) 136:211-213); Gueldener et a l., Nucleic Acids Res. (1996) 24:2519-2524); and Smith et al., Methods Mol. Cell. Biol. (1996) 5:270-277).

Antisense technology is another method of down-regulating genes where the sequence of the target gene is known. To accomplish this, a nucleic acid segment from the desired gene is cloned and opera bly linked to a promoter such that the anti-sense strand of RNA will be transcribed. This construct is then introduced into the host cell and the antisense strand of RNA is produced. Antisense RNA inhibits gene expression by preventing the accumu lation of m RNA that encodes the protein of interest. A person of skill in the art will know that special considerations are associated with the use of antisense technologies in order to reduce expression of particular genes. For example, the proper level of expression of antisense genes may require the use of different chimeric genes utilizing different regulatory elements known to the skilled artisan. Other less specific methodologies can also be used to down regulate undesired activity. For example, cells may be exposed to UV radiation a nd then screened for the desired phenotype. Mutagenesis with chemical agents can also be effective for generating mutants and commonly used su bstances include chemicals that affect non-replicating DNA such as HN0 ₂ and N H ₂ OH, as well as agents that affect replicating DNA such as acridine dyes, nota ble for causing frame-shift mutations. Specific methods for creating mutants using radiation or chemical agents are well documented in the art. See, for example, Thomas D. Brock in Biotechnology: A Textbook of Industrial Microbiology, Second Edition (1989) Sinauer Associates, Inc., Sunderla nd, Mass.; or Deshpande, Mukund Μ., ΑρρΙ. Biochem. Biotechnol., (1992) 36: 227).

Another non-specific method of gene disruption is the use of transposa ble elements or

"transposons." Transposons are genetic elements that insert randomly into DNA but can be later retrieved on the basis of sequence to determine where the insertion has occurred. Both in vivo and in vitro transposition methods are known. Both methods involve the use of a transposa ble element in combination with a transposase enzyme. When the transposa ble element or transposon is contacted with a nucleic acid fragment in the presence of the transposase, the transposa ble element will randomly insert into the nucleic acid fragment. The technique is useful for random mutagenesis and for gene isolation, because the disrupted gene may be identified on the basis of the sequence of the transposa ble element. Kits for in vitro transposition are commercially availa ble (see, for example, The Primer Island Transposition Kit, availa ble from Perkin Elmer Applied Biosystems, Branchburg, N.J., based upon the yeast Tyl element; The Genome Priming System, availa ble from New Engla nd Biola bs, Beverly, Mass.; based upon the bacterial transposon Tn7; and the EZ::TN Transposon Insertion System, availa ble from Epicentre Technologies, Madison, Wis., based upon the Tn5 bacterial transposa ble element). Transposon- mediated random insertion in the chromosome can be used for isolating mutants for any num ber of applications, including enhanced production of many desired products, including enzymes or other proteins, amino acids, or small organic molecules, including alcohols.

An exa mple where the reduction of a side reaction can increase the levels of a desired isoprenoid compound is the elimination or reduction of squalene synthase activity in host cells with an endogenous mevalonate pathway, such as yeast. In such systems, erg9 mutants that have a reduced a bility to convert FPP into squa lene have been shown to make more FPP-derived isoprenoid product (see e.g., Karst a nd Lacroute, Molec. Gen. Genet., 154, 269-277 (1977); U.S. Patent No. 5,589,372, issued Decem ber 31, 1996). Where the erg9 gene is blocked in yeast, such erg9 mutants may need extraneous ergosterol or other sterols added to the medium for the cells to remain via ble because yeast strains generally need ergosterol for cell mem brane fluidity. The cells normally cannot utilize this additional sterol unless grown under anaerobic conditions.

However, erg 9 mutants that take up exogenously-supplied sterols under aerobic conditions have been identified. These include those having genetic modifications in upc (uptake control mutation which allows cells to take up sterols under aerobic conditions); hem l (the H EM 1 gene encodes aminolevulinic acid synthase that is the first committed step to the heme biosynthetic pathway from FPP, and heml mutants are capa ble of taking up ergosterol under aerobic conditions following a d isruption in the ergosterol biosynthetic pathway, provided the cultures are supplemented with u nsaturated fatty acids); and overexpression of the SUT1 (sterol uptake gene, which can be used to allow for uptake of sterols under aerobic conditions; see, Bourot and Karst, Gene (1995) 165:97-102).

Host Cells

A wide variety of host cells can be used in the practice of the present invention. In one embodiment, the host cell is a genetically modified host microorganism in which nucleic acid molecules have been inserted, deleted or modified (i.e., mutated; e.g., by insertion, deletion, su bstitution, and/or inversion of nucleotides), to produce the desired isoprenoid compound or isoprenoid derivative. Illustrative examples of suita ble host cells include a ny archae, bacterial, or eukaryotic cell. Examples of an archae cell include, but are not limited to, those belonging to the genera : Aeropyrum, Archaeglobus, Halobacterium, Methanococcus, Methanobacterium, Pyrococcus, Sulfolobus, and Thermoplasma. Illustrative examples of archae strains include, but are not limited to, Aeropyrum pernix, Archaeoglobus fulgidus, Methanococcus jannaschii, Methanobacterium thermoautotrophicum, Pyrococcus abyssi, Pyrococcus horikoshii,

Thermoplasma acidophilum, and Thermoplasma volcanium.

Examples of a bacterial cell include, but are not limited to, those belonging to the genera :

Agrobacterium, Alicyclobacillus, Anabaena, Anacystis, Arthrobacter, Azobacter, Bacillus, Brevibacterium, Chromatium, Clostridium, Corynebacterium, Enterobacter, Erwinia, Escherichia, Lactobacillus, Lactococcus, Mesorhizobium, Methylobacterium, Microbacterium, Phormidium, Pseudomonas, Rhodobacter, Rhodopseudomonas, Rhodospirillum, Rhodococcus, Salmonella, Scenedesmun, Serratia, Shigella, Staphlococcus, Strepromyces, Synnecoccus, a nd Zymomonas. Illustrative examples of bacterial strains include, but are not limited to, Bacillus subtilis, Bacillus amyloliquefacines, Brevibacterium ammoniagenes, Brevibacterium immariophilum, Clostridium beigerinckii, Enterobacter sakazakii, Escherichia coli, Lactococcus lactis, Mesorhizobium loti, Pseudomonas aeruginosa, Pseudomonas mevalonii, Pseudomonas pudica, Rhodobacter capsulatus, Rhodobacter sphaeroides, Rhodospirillum rubrum, Salmonella enterica, Salmonella typhi, Salmonella typhimurium, Shigella dysenteriae, Shigella flexneri, Shigella sonnet,

Staphylococcus aureus, and the like.

In general, if a bacterial host cell is used, a non-pathogenic strain is preferred. Illustrative examples of non-pathogenic strains include, but are not limited to, Bacillus subtilis, Escherichia coli, Lactibacillus acidophilus, Lactobacillus helveticus, Pseudomonas aeruginosa, Pseudomonas mevalonii, Pseudomonas pudita, Rhodobacter sphaeroides, Rodobacter capsulatus,

Rhodospirillum rubrum, and the like.

Illustrative examples of eukaryotic strains include, but are not limited to, Aspergillus nidulans, Aspergillus niger, Aspergillus oryzae, Candida albicans, Chrysosporium lucknowense, Fusarium graminearum, Fusarium venenatum, Kluyveromyces lactis, Neurospora crassa, Pichia angusta,

Pichia finlandica, Pichia kodamae, Pichia membranaefaciens, Pichia methanolica, Pichia opuntiae, Pichia pastoris, Pichia pijperi, Pichia quercuum, Pichia salictaria, Pichia thermotolerans, Pichia trehalophila, Pichia stipitis, Streptomyces ambofaciens, Streptomyces aureofaciens, Streptomyces aureus, Saccaromyces bayanus, Saccaromyces boulardi, Saccharomyces cerevisiae, Streptomyces fungicidicus, Streptomyces griseochromogenes, Streptomyces griseus, Streptomyces lividans, Streptomyces olivogriseus, Streptomyces rameus, Streptomyces tanashiensis, Streptomyces vinaceus, Trichoderma reesei and Xanthophyllomyces dendrorhous (formerly Phaffia rhodozyma). In general, if a eukaryotic cell is used, a non-pathogenic strain is preferred. Illustrative examples of non-pathogenic strains include, but are not limited to, Fusarium graminearum, Fusarium venenatum, Pichia pastoris, Saccaromyces boulardi, and Saccaromyces cerevisiae.

Examples of eukaryotic cells include, but are not limited to, fungal cells. Examples of fungal cell include, but are not limited to, those belonging to the genera: Aspergillus, Candida,

Chrysosporium, Cryotococcus, Fusarium, Kluyveromyces, Neotyphodium, Neurospora, Penicillium, Pichia, Saccharomyces, Trichoderma and Xanthophyllomyces (formerly Phaffia). In addition, certain strains have been designated by the Food and Drug Administration as "G AS", or

Generally Regarded As Safe. These strains include: Bacillus subtilis, Lactibacillus acidophilus, Lactobacillus helveticus, and Saccharomyces cerevisiae. ISOP ENOIDS OF THE PRESENT INVENTION

The compositions and methods of the present invention can be employed to produce a wide variety of isoprenoids, including, without limitation, any C5 through C20, or higher, carbon number isoprenoid. The following describes, without limitation, exemplary isoprenoids of the invention.

C5 Compounds

The C5 isoprenoid compou nds are also known as hemiterpenes because they are derived from single isoprene unit (IPP or DMAPP). C5 compounds of the present invention generally are derived from I PP or DMAPP.

Isoprene

Isoprene has the chemical structure:

Isoprene is found in many plants and is made from I PP by isoprene synthase. Illustrative examples of suita ble nucleotide sequences encoding isoprene synthases include, but are not limited to, (AB198190; Populus alba) and (AJ294819; Polulus alba x Polulus tremula).

CIO Compou nds

The CIO isoprenoid compounds are also known as monoterpenes because they are derived from two isoprene units. CIO compounds of the invention are generally derived from geranyl pyrophosphate (GPP), which is made by the condensation of I PP with DMAPP. An enzyme known to catalyze this condensation reaction is, for example, geranyl pyrophosphate synthase. In certain embodiments, the host cells of the present invention comprise heterologous nucleic acid sequences that encode an enzyme that converts I PP and DMAPP into GPP.

Illustrative examples of nucleotide sequences for geranyl pyrophosphate synthase include but are not limited to: (AF513111; Abies grandis), (AF513112; Abies grandis), (AF513113; Abies grandis), (AY534686; Antirrhinu m majus), (AY534687; Antirrhinum majus), (Y17376; Ara bidopsis thaliana), (AE016877, Locus AP11092; Bacillus cereus; ATCC 14579), (AJ243739; Citrus sinensis), (AY534745; Clarkia breweri), (AY953508; Ips pini), (DQ286930; Lycopersicon esculentum), (AF182828;

Mentha x piperita), (AF182827; Mentha x piperita), (M PI249453; Mentha x piperita), (PZE431697, Locus CAD24425; Paracoccus zeaxanthinifaciens), (AY866498; Picrorhiza kurrooa), (AY351862; Vitis vinifera), and (AF203881, Locus AAF12843; Zymomonas mobilis).

GPP is then su bsequently converted to a variety of CIO compounds. Illustrative examples of CIO compounds include but are not limited : Carene

Carene has the chemical structure:

Careneis found in the resin of many trees, particularly pine trees. Carene is made from GPP from carene synthase. Illustrative examples of suita ble nucleotide sequences encoding carene synthases include, but are not limited to, (AF461460, REGION 43..1926; Picea abies) and (AF527416, REGION : 78..1871; Salvia stenophylla).

Geraniol

Geraniol (also known as rhodn

Geraniol is the main component of oil-of-rose and palmarosa oil. It also occurs in geranium, lemon, and citronella. Geraniol is made from GPP by geraniol synthase. Illustrative examples of suita ble nucleotide sequences encoding geraniol synthases include, but are not limited to, (AJ457070; Cinnamomum tenuipilum), (AY362553; Ocimum basilicum), (DQ234300; Perilla frutescens strain 1864), (DQ234299; Perilla citriodora strain 1861), (DQ234298; Perilla citriodora strain 4935), and (DQ088667; Perilla citriodora).

Linalool

Linalool has the chemical struct

Linalool is found in many flowers and spice plants, such as coriander seeds. Linalool is made from GPP by linalool synthase. Illustrative examples of a suita ble nucleotide sequence encoding linalool synthases include, but are not limited to, (AF497485; Arabidopsis thaliana), (AC002294, Locus AAB71482; Arabidopsis thaliana), (AY059757; Arabidopsis thaliana), (N M_104793;

Arabidopsis thaliana), (AF154124; Artemisia annua), (AF067603; Clarkia breweri), (AF067602; Clarkia concinna), (AF067601; Clarkia breweri), (U58314; Clarkia breweri), (AY840091;

Lycopersicon esculentum), (DQ263741; Lavandula angustifolia), (AY083653; Mentha citrate), (AY693647; Ocimum basilicum), (XM_463918; Oryza sativa), (AP004078, Locus BAD07605; Oryza sativa), (XM_463918, Locus XP_463918; Oryza sativa), (AY917193; Perilla citriodora), (AF271259; Perilla frutescens), (AY473623; Picea abies), (DQ195274; Picea sitchensis), and (AF444798; Perilla frutescens var. crispa cultivar No. 79).

Limonene

Limonene has the chemical structure:

Limonene is found in the rind of citrus fruits and peppermint. Limonene is made from GPP by limonene synthase. Illustrative examples of suita ble nucleotide sequences encoding limonene synthases include, but are not limited to, (+)-limonene synthases (AF514287, REGION : 47..1867; Citrus limon) and (AY055214, REGION : 48..1889; Agastache rugosa) and (-)-limonene synthases (DQ195275, REGION: 1..1905; Picea sitchensis), (AF006193, REGION: 73..1986; Abies grandis), and (M HC4SLSP, REGION : 29..1828; Mentha spicata).

Myrcene

Myrcene has the chemical struct

Myrcene is found in the essential oil in many plants including bay, verbena, and myrcia from which it gets its name. Myrcene is made from GPP by ocimene synthase. Illustrative examples of suita ble nucleotide sequences encoding ocimene synthases include, but are not limited to: (U87908; Abies grandis), (AY195609; Antirrhinum majus), (AY195608; Antirrhinum ma jus), (NM_127982; Arabidopsis thaliana TPS10), (NM_113485; Arabidopsis thaliana ATTPS-CI N), (NM_113483; Arabidopsis thaliana ATTPS-CI N), (AF271259; Perilla frutescens), (AY473626; Picea abies), (AF369919; Picea abies), and (AJ304839; Quercus ilex).

Ocimene

a- and -Ocimene have the chemical structur and , respectively, a- and β-Ocimene are found in a variety of plants and fruits, including Ocimum basilicum a nd are made from GPP by ocimene synthase. Illustrative examples of su ita ble nucleotide sequences encoding an ocimene synthase include, but are not limited to, (AY195607; Antirrhinum majus), (AY195609; Antirrhinum majus), (AY195608; Antirrhinum majus), (AK221024; Arabidopsis thaliana), (NM_113485; Arabidopsis thaliana ATTPS-CIN), (NM_113483; Arabidopsis thaliana ATTPS-CIN), (NM_117775; Arabidopsis thaliana ATTPS03), (NM_001036574; Arabidopsis thaliana ATTPS03), (NM_127982; Arabidopsis thaliana TPS10), (AB110642; Citrus unshiu CitMTSL4), and (AY575970; Lotus corniculatus var. japonicus).

g-Pinene

a-Pinene has the chemical structure:

a-Pinene is found in pine trees and eucalyptus. a-Pinene is made from GPP by α-pinene synthase. Illustrative examples of suitable nucleotide sequences encoding α-pinene synthases include, but are not limited to, (+) α-pinene synthase (AF543530, REGION: 1..1887; Pinus taeda), (-)a-pinene synthase (AF543527, REGION: 32..1921; Pinus taeda), and (+)/(-)a-pinene synthase (AGU87909, REGION: 6111892; Abies grandis).

3-Pinene

β-Pinene has the chemical structure

β-Pinene is found in pine trees, rosemary, parsley, dill, basil, and rose. β-Pinene is made from GPP by β-pinene synthase. Illustrative examples of suitable nucleotide sequences encoding β- pinene synthase include, but are not limited to, (-) β-pinene synthases (AF276072, REGION: 1..1749; Artemisia annua) and (AF514288, REGION: 26..1834; Citrus limon).

Sabinene

Sabinene has the chemical structure:

Sabinene is found in black pepper, carrot seed, sage, and tea trees. Sabinene is made from GPP by sabinene synthase. An illustrative example of a suitable nucleotide sequence encoding a sabinene synthased includes, but is not limited to, AF051901, REGION: 26..1798 from Salvia officinalis. y-Terpinene

γ-Terpinene has the chemical structure:

γ-Terpinene is a constituent of essential oils from citrus fruits. Biochemically, γ-terpinene is made from GPP by a γ-terpinene synthase. Illustrative examples of suitable nucleotide sequences encoding γ-terpinene synthases include, but are not limited to, (AF514286, REGION: 30..1832 from Citrus limon) and (AB110640, REGION 1..1803 from Citrus unshiu).

Terpinolene

Terpinolene has the chemical structure:

Terpinolene is found in black currant, cypress, guava, lychee, papaya, pine, and tea. Terpinolene is made from GPP by terpinolene synthase. Illustrative examples of suitable nucleotide sequences encoding terpinolene synthases include, but are not limited to, (AY693650 from Oscimum basilicum) and (AY906866, REGION: 10..1887 from Pseudotsuga menziesii).

C15 Compounds

The C15 isoprenoid compounds are also known as sesquiterpenes because they are derived from three isoprene units. C15 compounds of the present invention generally derive from farnesyl pyrophosphate (FPP), which is made by the condensation of two molecules of IPP with one molecule of DMAPP. An enzyme known to catalyze this condensation reaction is, for example, farnesyl pyrophosphate synthase. In certain embodiments, the host cells of the present invention comprise a heterologous nucleic acid sequence that encodes an enzyme that converts IPP and DMAPP into FPP. Illustrative examples of nucleotide sequences for farnesyl pyrophosphate synthase and the ispA gene which encodes it are described above.

Alternatively, FPP can also be made by adding IPP to GPP. Illustrative examples of nucleotide sequences encoding an enzyme that catalyzes this reaction include, but are not limited to:

(AE000657, Locus AAC06913; Aquifex aeolicus VF5), (NM_202836; Arabidopsis thaliana), (D84432, Locus BAA12575; Bacillus subtilis), (U12678, Locus AAC28894; Bradyrhizobium japonicum USDA 110), (BACFDPS; Geobacillus stearothermophilus), (NC_002940, Locus

N P_873754; Haemophilus ducreyi 35000HP), (L42023, Locus AAC23087; Haemophilus influenzae Rd KW20), (J05262; Homo sapiens), (YP_395294; Lactobacillus sakei subsp. sakei 23K),

(NC_005823, Locus YP_000273; Leptospira interrogans serovar Copenhagen/ str. Fiocruz Ll-130), (AB003187; Micrococcus luteus), (NC_002946, Locus YP_208768; Neisseria gonorrhoeae FA 1090), (U00090, Locus AAB91752; Rhizobium sp. NGR234), (J05091; Saccharomyces cerevisae),

(CP000031, Locus AAV93568; Silicibacter pomeroyi DSS-3), (AE008481, Locus AAK99890;

Streptococcus pneumoniae R6), a nd (NC_004556, Locus NP 779706; Xylella fastidiosa Temecula l). FPP is then su bsequently converted to a variety of C15 compou nds. Illustrative exa mples of C15 compounds include but are not limited to:

Amorphadiene

Amorphadiene has the chemical structu

Amorphadiene is a precursor to artemisinin, which is made by Artemisia anna. Amorphadiene is made from FPP by amorphadiene synthase. An illustrative exa mple of a su ita ble nucleotide sequence encod ing amorphadiene synthase is SEQ I D NO. 37 of U.S. Patent No. 7,192,751, issued March 20, 2007.

q-Farnesene

a-Farnesene has the chemical structure:

a-Farnesene is found in various biological sources including, but not limited to, the Dufour's gland in ants and in the coating of apple and pear peels. α-Farnesene is made from FPP by a-farnesene synthase. Illustrative examples of suita ble nucleotide sequences encoding α-Farnesene synthases include, but are not limited to, DQ309034 from Pyrus communis cultivar d'Anjou (pear; gene name AFS1) and AY182241 from Malus domestica (apple; gene AFS1). (see, Pechouus et al., Planta (2004) 219(l):84-94). 3-Farnesene

β-Farnesene has the chemical structure:

β-Farnesene is found in various biological sources including, but not limited to, aphids and essential oils, such as peppermint. In some plants, such as wild potato, β-farnesene is synthesized as a natural insect repellent. β-Farnesene is made from FPP by β-farnesene synthase. Illustrative examples of suita ble nucleotide sequences encoding β-farnesene synthase include, but are not limited to, GenBank accession num ber AF024615 from Mentha x piperita

(peppermint; gene Tspa ll), and AY835398 from Artemisia annua, (see, Picaud et al.,

Phytochemistry (2005) 66(9):961-967).

Farnesol

Farnesol has the chemical structure

Farnesol is found in various biological sources including insects and essential oils such as from cintronella, neroli, cyclamen, lemon grass, tu berose, and rose. Farnesol is made from FPP by a hydroxylase such as farnesol synthase. Illustrative examples of suita ble nucleotide sequences encoding farnesol synthases include, but are not limited to, GenBank accession number

AF529266 from Zea mays and YD 481C from Saccharomyces cerevisiae (gene Pho8). (see, Song L, Applied Biochemistry and Biotechnology, (2006) 128:149-158).

Nerolidol

Nerolidol has the chemical structure:

Nerolidol, also known as peruviol, is found in various biological sources as essential oils, such as from neroli, ginger, jasmine, lavender, tea tree, a nd lemon grass. Nerolidol is made from FPP by a hydroxylase such as nerolidol synthase. An illustrative exa mple of a suita ble nucleotide sequence encoding a nerolidol synthase includes, but is not limited to, AF529266 from Zea mays (maize; gene tpsl). Patchoulol

Patchoulol has the chemical structure:

Patchou lol, a lso known as patchouli alcohol, is a constituent of the essential oil of Pogostemon patchouli. Patchoulol is made from FPP by patchoulol synthase. An illustrative example of a suita ble nucleotide sequence encoding a patchoulol synthase includes, but is not limited to, AY508730 REGION : 1..1659 from Pogostemon cablin.

Valenecene

Valencene has the chemical structure:

Valenecene is one of the main chemical components of the smell and flavor of oranges and is found in orange peels. Valencene is made from FPP by nootkatone synthase. Illustrative examples of a suita ble nucleotide sequence encoding a nootkatone synthase includes, but are not limited to, AF441124 REGION : 1..1647 from Citrus sinensis and AY917195 REGION : 1..1653 from Perilla frutescens.

C20 Compou nds

The C20 isoprenoid compounds are known as diterpenes because they are derived from four isoprene u nits. The C20 compounds of the invention are generally derived from geranylgeraniol pyrophosphate (GGPP), which is made by the condensation of three molecules of I PP with one molecule of DMAPP. In certain embodiments, the host cells of the present invention comprise a heterologous nucleic acid sequence that encodes an enzyme that converts I PP and DMAPP into GGPP. An enzyme known to catalyze this step is, for example, geranylgeranyl pyrophosphate synthase.

Illustrative examples of nucleotide sequences encoding geranylgeranyl pyrophosphate synthases include, but are not limited to: (ATHGERPYRS; Arabidopsis thaliana), (BT005328; Arabidopsis thaliana), (NM_119845; Arabidopsis thaliana), (NZ_AAJ M01000380, Locus ZP_00743052; Bacillus thuringiensis serovar israelensis, ATCC 35646 sql563), (CRGGPPS; Catharanthus roseus), (NZ_AABF02000074, Locus ZP_00144509; Fusobacterium nucleatum subsp. vincentii, ATCC 49256), (GFGGPPSGN; Gibberella fujikuroi), (AY371321; Ginkgo biloba), (AB055496; Hevea brasiliensis), (AB017971; Homo sapiens), (MCI276129; Mucor circinelloides f. lusitanicus), (AB016044; Mus musculus), (AABX01000298, Locus NCU01427; Neurospora crassa), (NCU20940; Neurospora crassa), (NZ_AAKL01000008, Locus ZP_00943566; Ralstonia solanacearum UW551), (AB118238; Rattus norvegicus), (SCU31632; Saccharomyces cerevisiae), (AB016095;

Synechococcus elongates), (SAGGPS; Sinapis alba), (SSOGDS; Sulfolobus acidocaldarius), (NC_007759, Locus YP_461832; Syntrophus aciditrophicus SB), and (NC_006840, Locus

YP_204095; Vibrio fischeri ES114).

Alternatively, GGPP can also be made by adding IPP to FPP. Illustrative examples of nucleotide sequences encoding an enzyme capable of this reaction include but are not limited to:

(NM_112315; Arabidopsis thaliana), (E WC TE; Pantoea agglomerans), (D90087, Locus BAA14124; Pantoea ananatis), (X52291, Locus CAA36538; Rhodobacter capsulatus), (AF195122, Locus AAF24294; Rhodobacter sphaeroides), and (NC_004350, Locus NP_721015; Streptococcus mutans UA159).

GGPP may subsequently be converted to a variety of C20 isoprenoids. Illustrative examples of

C20 compounds include but are not limited to:

Geranylgeraniol

Geranylgeran

Geranylgeraniol is a constituent of wood oil from Cedrela toona and of linseed oil.

Geranylgeraniol can be made by e.g., adding to the expression constructs of the present invention a phosphatase gene, after the gene for GGPP synthase.

Abietadiene

Abietadiene encompasses the following isomers:

and is found in trees such as Abies grandis. Abietadiene is made by abietadiene synthase.

Illustrative examples of suitable nucleotide sequences encoding an abietadiene synthase i but are not limited to: (U50768; Abies grandis) and (AY473621; Picea abies). C20+ Compounds

C20+ compounds are also within the scope of the present invention. Illustrative examples of such compounds include sesterterpenes (C25 compounds made from five isoprene units), triterpenes (C30 compounds made from six isoprene units), and tetraterpenes (C40 compounds made from eight isoprene units). These compounds are made using methods described herein, while substituting or adding nucleotide sequences for the appropriate synthase(s).

Although the invention has been described in conjunction with specific embodiments thereof, the foregoing description is intended to illustrate and not limit the scope of the invention. Other aspects, advantages, and modifications within the scope of the invention will be apparent to those skilled in the art to which the invention pertains.

EXAMPLES

Unless otherwise indicated, the practice of the present invention can employ conventional techniques known to those skilled in the arts of the biosynthetic industry and the like. To the extent such techniques are not described fully herein, one can find ample reference to them in the scientific literature.

In the following examples, efforts have been made to ensure accuracy with respect to numbers used (for example, amounts, temperatures, etc.), but variations and deviations can be accommodated, and in the event a clerical error in the numbers reported herein exists, one of ordinary skill in the arts to which this invention pertains can deduce the correct amount in view of the remaining disclosure herein. Unless otherwise indicated, temperature is reported in degrees Celsius, and pressure is at or near atmospheric pressure at sea level. All reagents, unless otherwise indicated, were obtained commercially. The following examples are intended for illustrative purposes only and do not limit in any way the scope of the present invention.

Example 1

This example describes methods for making DNA constructs useful in the generation of host cells comprising heterologous DXP pathway enzymes.

The DNA constructs were generated from PC -amplified DNA fragments. Table 1 lists the DNA fragments, and the templates and primers used to generate them by PCR amplification. PCR amplifications were done using the PHUSION™ DNA polymerase (New England Biolabs, Ipswich, MA) per the manufacturer's suggested protocol. The PCR reactions were resolved by gel electrophoresis, the DNA fragments were gel-purified using the gel extraction kit from Omega Bio-tek, Inc., (Norcross, GA), and the purified DNA fragments were treated with T4 polynucleotide kinase (PNK) (New England Biolabs, Ipswich, MA) per the manufacturer's suggested protocol. The PNK was heat-inactivated at 65°C for 20 minutes, and the samples were stored at -20°C. Table 1 - PCR Amplified DNA Fragments

DNA Fragment Primers Template

GW392 (SEQ ID NO: 1)

ERG13 US Y002 genomic DNA

GW393 (SEQ ID NO: 2)

GW394 (SEQ ID NO: 3)

ERG13 DS Y002 genomic DNA

GW395 (SEQ ID NO: 4)

2604 0 link-fwd (SEQ ID NO: 5)

HO US Y002 genomic DNA

2604 2 link-rev (SEQ ID NO: 6)

2605 0 link-fwd (SEQ ID NO: 7)

HO DS Y002 genomic DNA

2605 2 link-rev (SEQ ID NO: 8)

R3519-fwd (SEQ ID NO: 9)

ADE2 US Y002 genomic DNA

R3519-rev (SEQ ID NO: 10)

GW522 (SEQ ID NO: 11)

ADE2 DS Y002 genomic DNA

GW523 (SEQ ID NO: 12)

R2862-fwd (SEQ ID NO: 13)

TRP1 ½ Y002 genomic DNA

R2862-rev (SEQ ID NO: 14)

GW526 (SEQ ID NO: 15)

TRP2 2/2-1 Y002 genomic DNA

GW525 (SEQ ID NO: 16)

GW524 (SEQ ID NO: 60)

TRP2 2/2-2 Y002 genomic DNA

GW525 (SEQ ID NO: 16)

JWL7 (SEQ ID NO: 17)

GAL80 US Y002 genomic DNA

JWL8 (SEQ ID NO: 18)

R3209-fwd (SEQ ID NO: 19)

GAL80 DS Y002 genomic DNA

R3209-rev (SEQ ID NO: 20)

R4389-fwd (SEQ ID NO: 21)

ADH5 US Y002 genomic DNA

R4389-rev (SEQ ID NO: 22)

R3202-fwd (SEQ ID NO: 23)

ADH5 DS Y002 genomic DNA

R3202-rev (SEQ ID NO: 24)

M187-fwd (SEQ ID NO: 25)

LEU2 US Y002 genomic DNA

M187-rev (SEQ ID NO: 26)

M3818-F (SEQ ID NO: 27)

LEU2 DS Y002 genomic DNA

M3818-R (SEQ ID NO: 28)

GW396 (SEQ ID NO: 29)

LEU2 Y002 genomic DNA

GW397 (SEQ ID NO: 30)

M2850-fwd (SEQ ID NO: 31)

HIS3 Y002 genomic DNA

M2850-rev (SEQ ID NO: 32)

M2865 5 link-fwd (SEQ ID NO: 33)

URA3 ½ Y002 genomic DNA

M2865 9 link-rev (SEQ ID NO: 34)

2077 4 link-fwd (SEQ ID NO: 35)

URA3 2/2 Y002 genomic DNA

2077 9 link-rev (SEQ ID NO: 36)

R2246-fwd (SEQ ID NO: 37)

PGALl/10-1 Y002 genomic DNA

R2246-rev (SEQ ID NO: 38)

M2820-PTDH3-fwd (SEQ ID NO: 39)

PTDH3-1 Y002 genomic DNA

M2820-PTDH3-rev (SEQ ID NO: 40)

M2823-PTDH3-fwd (SEQ ID NO: 41)

PTDH3-2 Y002 genomic DNA

M2823-PTDH3-rev (SEQ ID NO: 42)

M3819 Prom-F (SEQ ID NO: 43)

PTDH3-3 Y002 genomic DNA

M3819 Prom-R (SEQ ID NO: 44)

M2821-PTEF2-fwd (SEQ ID NO: 45)

PTEF2-1 Y002 genomic DNA

M2821-PTEF2-rev (SEQ ID NO: 46)

M2824-PTEF2-fwd (SEQ ID NO: 47)

PTEF2-2 Y002 genomic DNA

M2824-PTEF2-rev (SEQ ID NO: 48)

R3419-prom-fwd (SEQ ID NO: 65)

PTEF2-3 Y002 genomic DNA

R3419-prom-rev (SEQ ID NO: 66) GW527 (SEQ ID NO: 67)

PTEF2-4 Y002 genomic DNA

R5593-prom-rev (SEQ ID NO: 68)

GW527 (SEQ ID NO: 67)

PTEF2-5 Y002 genomic DNA

GW546 (SEQ ID NO: 87)

GW550 (SEQ ID NO: 86)

PTEF2-6 Y002 genomic DNA

GW546 (SEQ ID NO: 87)

M2822-PTDH2-fwd (SEQ ID NO: 49)

PTDH2 Y002 genomic DNA

M2822-PTDH2-rev (SEQ ID NO: 50)

M2820-dxs-l-fwd (SEQ ID NO: 51)

Ec_dxs-TSHM2 SEQ ID NO: 97 ^a)

M2820-dxs-2-rev (SEQ ID NO: 52)

M2821-dxr-fwd (SEQ ID NO: 83)

Ec_dxr-TAD01-1 SEQ ID NO: 98 ^a)

M2821-dxr-rev (SEQ ID NO: 82)

GW577 (SEQ ID NO: 93)

Ec_dxr-TAD01-2 SEQ ID NO: 99 ^a)

GW578 (SEQ ID NO: 94)

M2822-ispD-fwd (SEQ ID NO: 53)

EcjspD-TYSBl SEQ ID NO: 99 ^a)

M2822-ispD-rev (SEQ ID NO: 54)

GW563 (SEQ ID NO: 55)

hisG SEQ ID NO: 109

GW564 (SEQ ID NO: 56)

M2823-ispE-fwd (SEQ ID NO: 57)

Ec_ispE-TH0M2 SEQ ID NO: 100 ^a)

M2823-ispE-rev (SEQ ID NO: 58)

M2824-ispF-fwd (SEQ ID NO: 85)

Ec_ispF-TADE17 SEQ ID NO: 101 ^a)

M2824-ispF-rev (SEQ ID NO: 84)

GW565 (SEQ ID NO: 136)

Ec_ispF-TH0M2 SEQ ID NO: 101 ^a)

GW566 (SEQ ID NO: 139)

M3819 Gene-F (SEQ ID NO: 61)

Ec_xyl B-TPUP2 SEQ ID NO: 102 ^a)

M3819 Gene-R (SEQ ID NO: 62)

GW556 (SEQ ID NO: 63)

PEN02 Y002 genomic DNA

GW574 (SEQ ID NO: 64)

GW573 (SEQ ID NO: 69)

BsyN_fld-TSUI3 SEQ ID NO: 104 ^a)

GW544 (SEQ ID NO: 70)

R3419-ORF-fwd (SEQ ID NO: 71)

At_RFNR2trunc-T0LEl-l SEQ ID NO: 103 ^a)

R3419-ORF-rev (SEQ ID NO: 72)

R6121-fwd (SEQ ID NO: 95)

At_RFNR2trunc-TOLEl-2 SEQ ID NO: 103 ^a)

R6121-rev (SEQ ID NO: 96)

R5593-ORF-fwd (SEQ ID NO: 73)

Ec_ispH-TTHSl SEQ ID NO: 106 ^a)

GW528 (SEQ ID NO: 74)

GW545 (SEQ ID NO: 75)

STREP-Ec_ispH-TTHSl SEQ ID NO: 201 ^a)

GW528 (SEQ ID NO: 74)

GW545 (SEQ ID NO: 75)

STR E P-Te_is p H -TO LA 1 SEQ ID NO: 105

GW529 (SEQ ID NO: 76)

STREP-At_RFNR2trunc- GW545 (SEQ ID NO: 75)

SEQ ID NO: 107

TOLE1 GW551 (SEQ ID NO: 77)

GW594 (SEQ ID NO: 78)

STREP-Ec_fldA-Ec_ispH SEQ ID NO: 108

GW595 (SEQ ID NO: 79)

GW596 (SEQ ID NO: 80)

STREP-Bt_ispG-BsyN_fld SEQ ID NO: 59

GW597 (SEQ ID NO: 81)

a) The synthetically generated DNA fragments were provided by Integrated DNA Technologies,

Inc., (Coraville, IA).

The DNA fragments were su bsequently assem bled into a first set of "stitches" as outlined in Ta ble 2, using methods described in U.S. Patent No. 8, 110,360. Additional DNA fragments from Ta ble 1 and stitches from Ta ble 2 were assem bled into a second set of stitches as outlined in Ta ble 3. For assembly reactions, 100 fmole of each DNA fragment or stitch were placed together in one tu be. The samples were split into three, 30 μί reactions; water, buffer, d NTPs, and DNA polymerase were added to each reaction mixture, and a first round of PC -amplification was initiated.

Samples were placed on ice, 0.5 μΜ of each terminal primer were added to the reaction mixtures, and a second rou nd of PCR amplification was performed. The three PCR reaction 5 mixtures were com bined in one tu be, the reaction mixtures were resolved by gel electrophoresis, and the PCR products were gel purified.

Table 3 - Terminal Primers Used for Assembly of Second Set of Stitches

Stitch DNA Fragments (see Table 1) or Stitches (see Table 2) Terminal Primer 1 Terminal Primer 2

M2822-ispD-fwd GW566 i3850 Ec_ispD-TYSBl, hisG, Ec_ispF-THOM2

(SEQ ID NO: 53) (SEQ ID NO: 139) i3830 ERG13 US, LEU2, ERG13 DS, linearized pAM2041 ^a)

HO US, PTDH3-Ec_dxs-TSHM2, PTEF2-Ec_dxr-TAD01, PTDH2-

S2075

Ec_ispD-TYSBl, URA3 1/2, linearized pAM2041 ^a)

HO DS, PTDH3-Ec_ispE-THOM2, PTEF2-Ec_ispF-TADE17, URA3

S2076

2/2, linearized pAM2041 ^a)

LEU2 US, HIS3, PTDH3-Ec_xyl B-TPUP2, PTEF2-Ec_dxr-TAD01,

Ϊ2145

LEU2 DS, linearized pAM2041 ^a)

ADE2 US, PEN02-BsyN_fld-TSUI3, PTEF2-At_RFNR2trunc- si 1469

TOLE1, TRP1 1/2, linearized pAM2041 ^a)

ADE2 US, PEN02-BsyN_fld-TSUI3, PTEF2-STREP-

S11472

At_RFNR2trunc-TOLEl, TRP1 1/2, linearized pAM2041 ^a)

ADE2 DS, PTEF2 ispH-TTHSl, TRP1 2/2, linearized pAM2041 terminal end of terminal end of

S11475

linearized linearized

S11478 ADE2 DS, TRP1 2/2-2, linearized pAM2041 ^a) pAM2041 pAM2041

ADE2 DS, PTEF2-STREP-Ec_ispH-TTHSl, TRP1 2/2-1, linearized

S11479

pAM2041 ^a)

ADE2 DS, PTEF2-STREP-Te_ispH-TOLAl, TRP1 2/2-1, linearized

si 1480

pAM2041 ^a)

GAL80 US, STREP-Bt_ispG-BsyN_fld, pGALl/10-1,

S11629

At_RFNR2trunc-TOLEl-2, TRP1 1/2, linearized pAM2041 ^a)

ADH5 US, STREP-Bt_ispG-BsyN_fld, pGALl/10-1,

S11632

At_RFNR2trunc-TOLEl-2, TRP1 1/2, linearized pAM2041 ^a)

GAL80 DS, STREP-Ec_fldA-Ec_ispH, pGALl/10-1, Ec_dxr-TAD01-

S11634

2, TRP1 2/2, linearized pAM2041 ^a)

ADH5 DS, STREP-Ec_fldA-Ec_ispH, pGALl/10-1, Ec_dxr-TAD01-

S11635

2, TRP1 2/2, linearized pAM2041 ^a)

The first round of PCR amplification was performed as follows: one cycle of denature at 98°C for 2 minutes; 5 cycles of denature at 98°C for 30 seconds and anneal/extend at 72°C for 30 seconds per kilobase PCR product. The second round of PCR amplification was performed as follows: one cycle of denature at 98°C for 2 minutes; 35 rounds of denature at 98°C for 12 seconds and anneal/extend at 72°C for 20-25 seconds per kilobase PCR product; one cycle of final extend at 72°C for 7 minutes; and a final hold at 4°C. When the annealing temperature was not 72°C (i.e., when it was either 54°C or 65°C), in the first round of PCR amplification a 1 minute annealing step followed by a 30 seconds per kilobase PCR product extension step at 72°C was used, and for the second round of PCR amplification a 15 second annealing step followed by a 20 seconds per kilobase PCR product extension step at 72°C was used.

a) Plasmid pAM2041 (SEQ ID NO: 92) was digested to completion using Schl restriction endonuclease (Fementas Life Sciences, Glen Burnie, Maryland), and the purified, linearized plasmid was included in the PCR reaction.

Example 2

This example describes methods for ma king yeast strains comprising heterologous DXP enzymes. The yeast strains listed in Ta ble 4 were generated by introducing the indicated genetic

5 modifications into the indicated parental strains. To these ends, exponentially-growing parental

strain cells were transformed with the indicated DNA fragment(s), stitch(es), or 2μ plasmid, and host cell transformants were grown on the indicated selective agar. Genomic integrations in selected clones were confirmed by colony PCR using PCR primers homologous to the 5' and 3' junctions of the chromosomal integration sites.

all heterologous sequences being codon-optimized for

expression in S. cerev.

replace LEU2 coding sequence with S. cerev. HIS3 coding

sequence, E. coli dxr coding sequence with S. cerev. TEF2 CSM-H + 2 promoter and ADOlterminator, E. coli xylB coding sequence g/L

Y4819 Y4161 Ϊ2145 ^b) - with S. cerev. TDH3 promoter and PUP2terminator, all mevalolacton heterologous sequences being codon-optimized for e ^d) expression in S. cerev.

replace URA3 coding sequence at deleted HO locus with hisG CSM + 5 FOA

Y6237 Y4819 i3850 - coding sequence + U replace ADE2 coding sequence with S. cerev. TRP1 coding

sequence, Bacillus subtilis flavodoxin coding sequence with S.

cerev. EN02 promoter and SUB terminator, Arabidopsis CSM-W + 2 thaliana FNR truncated coding sequence with S. cerev. TEF2 sll469 ^{c) f)} g/L

Y6276 Y6237 - promoter and OLE1 terminator, E. coli ispH coding sequence sll475 ^{c) f)} mevalolacton with 5. cerev. TEF2 promoter and THSl terminator, all e ^d) heterologous sequences being codon-optimized for

expression in S. cerev.

replace ADE2 coding sequence with S. cerev. TRP1 coding

sequence, Bacillus subtilis flavodoxin coding sequence with S. CSM-W + 2 cerev. EN02 promoter and SUI3 terminator, Strepll-tagged sll472 ^{c) f)} g/L

Y6280 Y6237 - Arabidopsis thaliana FNR truncated coding sequence with 5. sll478 ^{c) f)} mevalolacton cerev. TEF2 promoter and OLE1 terminator, all heterologous e ^d) sequences being codon-optimized for expression in S. cerev.

PGAL1- STREP- express Strepll-tagged Thermus thermophilus ispG with 5.

Tt_ispG

Y6292 Y6280 cerev. GAL1 promoter and ILV5 terminator, all heterologous - CSM-U

(SEQ ID

sequences being codon-optimized for expression in 5. cerev.

NO: 89;

FIG. 9B)

PGAL1- STREP- express Strepll-tagged Bacillus thuringiensis ispG with S. cerev.

BtjspG

Y6291 Y6280 GAL1 promoter and ILV5 terminator, all heterologous - CSM-U

(SEQ ID

sequences being codon-optimized for expression in S. cerev.

NO: 90;

FIG. 9C)

express Strepll-tagged Bacillus thuringiensis ispG with S. cerev.

FBA1 promoter and ILV5 terminator, all heterologous

sequences being codon-optimized for expression in 5. cerev.

PFBA1- replace ADE2 coding sequence with S. cerev. TRP1 coding STREP- sequence, Bacillus subtilis flavodoxin coding sequence with S. sll472 ^{c) f)} BtjspG

Y6281 Y6237 CSM-U cerev. EN02 promoter and SUI3 terminator, Strepll-tagged si 1479 ^{c) fl} (SEQ ID Arabidopsis thaliana FNR truncated coding sequence with S. NO: 91; cerev. TEF2 promoter and OLEl terminator, and Stepll-tagged FIG. 9D)

E. coli ispH with S. cerev. TEF2 promoter and THSl terminator,

all heterologous sequences being codon-optimized for

expression in S. cerev.

PGAL1- STREP- express Strepll-tagged Bacillus thuringiensis ispG with S. cerev.

BtjspG

Y6283 Y6276 GAL1 promoter and ILV5terminator, all heterologous - CSM-U

(SEQ ID

sequences being codon-optimized for expression in S. cerev.

NO: 90;

FIG. 9C)

replace ADE2 coding sequence with S. cerev. TRP1 coding sll472 ^{c, fl} CSM-W + 2

Y6279 Y6237 - sequence, Bacillus subtilis flavodoxin coding sequence with S. sll480 ^{c) f)} g/L cerev. EN02 promoter and SUB terminator, Strepll-tagged mevalolacton Arabidopsis thaliana FNR truncated coding sequence with 5. e ^d) cerev. TEF2 promoter and OLE1 terminator, and Stepll-tagged

Thermosynechococcus elongates ispH with5. cerev. TEF2

promoter and OLA1 terminator, all heterologous sequences

being codon-optimized for expression in S. cerev.

PGAL1- STREP- express Strepll-tagged Bacillus thuringiensis ispG with S. cerev.

BtjspG

Y6289 Y6279 GAL1 promoter and ILV5 terminator, all heterologous - CSM-U

(SEQ ID

sequences being codon-optimized for expression in S. cerev.

NO: 90;

FIG. 9C)

CSM = 6.7 g/L Yeast Nitrogen Base (YNB; BD Diagnostics, Franklin Lakes, NJ), 2 g/L Synthetic Complete Mix (CSM; Bufferad, Lake Bluff, IL), 2% glucose, 1% methionine, 1% leucine, 0.2% adenine sulfate, 0.2% uracil, 1% lysine, 1% trypophan, 1% histidine; CSM-L = CSM comprising all amino acids except leucine; CSM-U = CSM comprising all amino acids except uracil; CSM-H = CSM comprising all amino acids except histidine; CSM-W = CSM comprising all amino acids except tryptophan; CSM + 5 FOA + U = CSM comprising all amino acids and 5- fluoroorotic acid.

a) CEN.PK2 background MAT A; ura3-52; trpl-289; Ieu2-3,112; his3Al; MAL2-8C; SUC2 (van Dijken et al. Enzyme Microb. Technol. 2000, 26:706-714)

b) The construct was digested to completion using Pmel restriction endonuclease (New England Biolabs, Inc., Ipswich, MA) to liberate the insert from the pAM2041 vector, before the isolated insert was transformed into the parent strain.

c) The constructs were digested to completion using Pmel restriction endonuclease (New England Biolabs, Inc., Ipswich, MA) to liberate the inserts from the pAM2041 vectors, and the purified inserts were then co-transformed into the parent strain where they recombined with each other and with the host cell genome by host cell mediated homologous recombination. d) Simga-Aldrich, St. Louis, MO.

e) See Example 6.

f) BJ5459, ATCC # 208284

Example 3

This example describes methods for the detection and quantitation of DXP pathway metabolites by Mass Spectrometry.

5 Engineered yeast cells were rapidly quenched in -80°C 8:1:1 methanokwatenglycerol at a ratio of

1:4 cell broth:quenching solution, and pelleted by centrifugation. After disposing of quench, pellets were extracted by sonication, vortexing with beads, resting on wet ice, and centrifugation in two fractions of 7:3 methanol:20 mM ammonium acetate, pH 10 extraction buffer spiked with farnesyl-thiolodiphosphate (TFPP; Echelon Biosciences, Salt Lake City, UT) as an internal standard.

10 The extracts were combined, diluted in methanol, centrifuged to pellet salts, and transferred to

LC vials for injection. Analyte separation was achieved via a gradient from acetonitrile to aqueous 20 mM ammonium acetate, pH 10, on a LUNA™ NH2 column (Phenomenex, Torrance, CA) using high pressure liquid chromatography (Shimadzu, Kyoto, Japan) (see Table 5 for gradient

conditions). Mass spectral data was acquired via scheduled multiple reaction monitoring (MRM)

15 on an API 5000 instrument (Applied Biosystems, Carlsbad, CA). Negative ionization mode was

used to monitor two MRM transitions each for DXP, cMEPP, HDMAPP, DMAPP/IPP, and FPP, with retention time and fragmentation confirmed with purchased standards (Sigma, St Louis, MO; and Echelon Biosciences, Salt Lake City, UT) (see Table 6). Experiments involving labeled isotope feeding also included MRM transitions for the aforementioned analytes plus 5 atomic mass units, and plus 15 atomic mass units for FPP. Data was analyzed using ANALYST™ (AB SCIEX, Foster City, CA), and normalized to cell number by measuring the optical density of cultures at 600nm (OD ₆₀ o). Calibration curves using TFPP and standards were used for absolute quantitation.

Table 6 - List of Masses

Scheduled

Declustering Collision Collision Cell

Ql Mass Q3 Mass Retention Analyte, Transition #

Potential Energy Exit Potential Time

212.982 78.8 7.5 DXP_1 -55 -42 -21

212.982 96.8 7.5 DXP_2 -55 -20 -13

212.982 138.9 7.5 DXP_3 -55 -18 -19

217.982 78.8 7.5 DXP_l,+5 AM U -55 -42 -21

217.982 96.8 7.5 DXP_2,+5 AM U -55 -20 -13

217.982 138.9 7.5 DXP_3,+5 AM U -55 -18 -19

276.71 79 7.5 cMEPP_l -100 -38 -9

276.71 158.8 7.5 cMEPP_2 -100 -32 -23

281.71 79 7.5 cMEPP_l,+5 AM U -100 -38 -9

281.71 158.8 7.5 cMEPP_2,+5 AM U -100 -32 -23

260.916 78.9 9.3 HDMAPP_1 -75 -52 -9

260.916 62.9 9.3 HDMAPP_2 -75 -130 -13

260.916 158.9 9.3 HDMAPP_3 -75 -22 -21

265.916 78.9 9.3 HDMAPP_l,+5 AM U -75 -52 -9

265.916 62.9 9.3 HDMAPP_2,+5 AM U -75 -130 -13

265.916 158.9 9.3 HDMAPP_3,+5 AM U -75 -22 -21

245.021 78.9 9.3 DMAPP_1 -35 -24 -9

245.021 158.9 9.3 DMAPP_2 -35 -24 -9

250.021 78.9 9.3 DMAPP_l,+5 AM U -35 -24 -9

250.021 158.9 9.3 DMAPP_2,+5 AM U -35 -24 -9

381.065 78.9 9 FPP_1 -95 -38 -11

381.065 63 9 FPP_2 -95 -130 -11

386.065 78.9 9 FPP_l,+5 AM U -95 -38 -11

386.065 63 9 FPP_2,+5 AM U -95 -130 -11

391.065 78.9 9 FPP_1,+10 AMU -95 -38 -11

391.065 63 9 FPP_2,+10 AMU -95 -130 -11

396.065 78.9 9 FPP_1,+15 AMU -95 -38 -11

396.065 63 9 FPP_2,+15 AMU -95 -130 -11

397.055 158.9 9.4 TFPP -65 -26 -21

Example 4

This example describes methods for detection and quantitation of DXP pathway proteins in yeast by Mass Spectrometry.

Engineered yeast cells were grown to yield sufficient biomass (OD ₆₀ o x culture volumes (mL) > 5). Pellets were lysed in the presence of 6 M urea, 500 mM triethylamine bicarbonate pH 8.5, with glass beads on a mini-beadbeater (Biospec, Bartlesville, OK). Total protein measurements were taken using a QUBIT™ Fluorometer (Invitrogen, Carlsbad, CA), and 20-30 μg total protein was reduced by incubating with 6 mM tris(2-carboxyethyl)phosphine at 37°C for 30 min, alkylated by incubating with 11 mM iodoacetamide at room temperature for 30 min in the dark, and digested with tosyl phenylalanyl chloromethyl ketone (TPCK)-treated trypsin at a trypsin:total protein ratio of at least 1:20 at 37°C, overnight. The digestion reaction (approximately 50 μί total volume) was then quenched with 450 μί 0.1% formic acid in water, and analyzed on a QT AP™ 4000 Mass Spectrometer (AB SCIEX, Foster City, CA) (see Table 7 for gradient conditions). Table 7 - LC Gradient

Time (min) % A % B

0 98 2

13.5 60 40

14.5 15 85

17.5 98 2

A = 0.1% formic acid in water

B = 0.1% formic acid in acetonitrile

For LC-MS/MS acquisition method design, sequence information for proteins of interest were obtained and fed into a FASTA file. The FASTA file was fed into Skyline software (MacCoss Lab, University of Washington, WA), and an in-silico digestion was performed by the software, yielding tryptic fragments amenable to detection by LC-MRM. The resulting MRM transitions were then transferred to an acquisition method in ANALYST™ (AB SCIEX, Foster City, CA). The acquisition method was an MRM-IDA (Multiple Reaction Monitoring - Information Dependent Acquisition), and suitable MRM's triggered MS/MS that were then used to confirm the identity of the desired tryptic fragments belonging to the proteins of interest.

The acquired MS/MS spectra were searched in Mascot against the Swissprot database, and spectra with p-values of less than 0.05 were accepted as positive IDs. The retention time of the positively-identified MS/MS spectra was matched to the retention time of the MRM traces of concurrent y-ions corresponding to the matched tryptic peptide. The identified tryptic peptide signature could then be used to determine the presence of a specific protein, or even to perform relative quantitation of proteins.

Example 5

This example describes the optimization of expression of the first five enzymes in the DXP pathway (DXP-TOP pathway).

Six independent isolates of Y6080 were grown up in CSM comprising 2 g/L mevalolactone to an OD ₆₀ o of between 2 and 3, and analyzed for metabolites of the DXP-TOP pathway as described in Example 3.

As shown in FIG. 2, all Y6080 isolates produced cMEPP at an internal cellular concentration of between 300 to 400 μΜ, suggesting that the DXP-TOP pathway was active in these cells.

However, significant levels of DXP were also detected, suggesting that DXR enzyme activity was limiting.

To increase carbon flux through the first five DXP enzymes to cMEPP, strain Y6081 was generated, which comprises an additional copy of the Escherichia coli dxr coding sequence, in addition to a copy of the Escherichia coli gene xylB, which encodes an enzyme that can convert deoxy-xylulose into DXP. Six independent isolates of Y6081 were analyzed for meta bolites of the DXP-TOP pathway as described in Example 3. As shown in FIG. 2, all but one Y6081 isolate (#36) showed an increase in cM EPP concentration. Isolate lineage #27 of Y6080 and Y6081 was chosen for all future strain construction work.

Example 6

This example describes the selection of yeast strains that can grow in the a bsence of an endogenous M EV pathway. Because these strains a re mevalonate auxotrophs, they can only survive if provided mevalonate in their growth medium. As a consequence, a heterologous DXP pathway can be engineered in these strains and tested to determine if it can rescue these strains by making I PP and DMAPP through the DXP pathway when placed in medium without mevalonate. In addition, inactivation of YPKl allows yeast strains to be more efficient in their use of mevalonate.

Three positive transforma nts of Y5821 were streaked on CSM agar plates lacking leucine and comprising 5 g/L mevalolactone or 0 g/L mevalolactone. Plates were incu bated at 30°C for 7 - 10 days. No growth was observed on CSM agar plates containing 0 g/L mevalolactone, confirming that Y5821 is a mevalonate auxotroph. Cells were scraped from the CSM agar plates comprising 5 g/L mevalolactone, washed in sterile water, resuspended to a concentration of 5 x 10 ⁷ cells in 1 m L of sterile water, and plated (5 x 10 ^s cells per plate) onto ten CSM agar plates comprising 2 g/L mevalolactone, a mevalolactone concentration that supports colony growth by Y5821 cells only after 14 days. Three single colonies visible after 5 days of growth at 30°C were picked, and were confirmed to grow faster than their Y5821 parent in liquid CSM lacking leucine (see Ta ble 8). The three strains were also plated on CSM agar plates comprising 0 g/L mevalonolactone to confirm that they were still mevalonate auxotrophs.

Y5822, Y5823, and Y5824 were su bjected to another round of screening for faster growth on CSM agar plates comprising 0.267 g/L mevalolactone. In this second round of screening, strain Y5825 was derived from Y5824. Y5825 was confirmed to grow even faster than Y5824 in liquid CSM lacking leucine.

Additional Y5821 cells were scraped from CSM agar plates comprising 5 g/L mevalolactone, washed in sterile water, resuspended to a concentration of 5 x 10 ^s cells in 1 mL of sterile water, and plated onto ten CSM agar plates comprising 0 g/L mevalolactone (5 x 10 ⁷ cells per plate). After three weeks of growth at 30°C, no colonies were visible, indicating the frequency of spontaneous suppression of the ergl3 deletion is less than 1 in 5 x 10 ^s cells.

Strains Y5822, Y5823, Y5824, and Y5825 were su bmitted to whole genome sequencing and compared to a reference CEN PK parent genome. As shown in Ta ble 9, each mutant strain comprised only a single mutation, except Y5825, which had two mutations. The mutations fell into two genes: YPK1 and P M 10/YJL107C. YPK1 is annotated in the Saccharomyces Genome Data base (yeastgenome.org, Feb 21, 2011) as a "serine/threonine protein kinase that phosphorylates and downregulates flippase activator Fpklp; mutations affect receptor-mediated endocytosis and sphingolipid-med iated and cell integrity signaling pathways." M utations found in YPK1 in Y5822, Y5824, and Y5823 are presuma bly inactive or negatively affect YPK1, indicating that loss of function of this protein allows 5. cerevisiae to grow on a lower concentration of mevalolactone. YJL107C is directly upstream from PRM 10 and the coding sequences overlap at the a nnotated stop (TGA) and start (ATG) codons for the two genes. In the yeast

Zygosaccharomyces rouxii, these two sequences are annotated as a single gene, and

Saccharomyces cerevisiae transcript data indicates that PRM 10/YJ L107C have identical transcriptional profiles (Xu Z. et al. Nature 2009, 457: 1033-1037). The data therefore support the conclusion that PRM 10/YJL107C is a single transcriptional unit. The entire PRM 10/YJL107C protein contains a Major Facilitator Superfamily (M FS) domain, which is a domain found in a large family of transporters.

Example 7

This example describes methods for expressing and purifying solu ble DXP pathway enzymes in yeast cells.

A pGALI expression vector was generated for use in purifications of heterologous DXP proteins in yeast cells. To this end, the DNA fragments shown in Ta ble 9 were PCR-amplified as described in Example 1.

The DNA fragments were assem bled into stitches as outlined in Ta ble 10 a nd described in

Example 1.

The final stitch, P S02-PGALl-Strepl l-ispG -PRS39-pAM09, was transformed into chemically competent Escherchia coli cells. Positive transformants were selected on Luria Bertani (LB) agar containing ca rbenicillin. Selected clones were grown in LB containing 100 μg/m L carbenicillin, the plasmid was extracted by standard techniques, and the isolated plasmid was sequenced to verify the DNA was assembled correctly. A sequence-verified plasmid was digested with the restriction enzyme Pmel (New England BioLa bs, Inc., Ipswich MA) to release the PRS02-PGALl-Strepl l-ispG- PRS39 insert from the vector backbone. Vector pAM70 was digested with Xma l a nd Sacl-HF restriction endonucleases (New England BioLa bs, Inc., Ipswich MA). 100 ng of purified linearized pAM70 was mixed with 150 ng of purified PRS02-PGALl-Strepl l-ispG-PRS39 insert, and the mixture was used to transform a yeast strain in which the final vector pAM2130 (FIG. 3A) was created in vivo via host cell mediated gap-repair homologous recombination. The vector was recovered using the ZYMOP EP Yeast Plasmid Miniprep II kit (Zymo Research, Irvine, CA), and sequence verified. pAM2130 was digested using Pacll and Aatll restriction endonucleases (New England Biolabs, Ipswich, MA) to remove the ispG coding sequence, and the vector backbone was purified using a gel extraction kit (Qiagen, Valencia, CA), yielding the pGALl expression vector. All proteins analyzed by in vitro biochemistry in this invention were cloned into the pGALl expression vector using gap repair into yeast to create an in-frame fusion protein containing the Strepll tag at the N terminus of the protein. The Strepll tag could then be used to identify the expressed proteins via Western blot, and to purify the expressed proteins using a STREP- TACTIN™ resin. Cloning into the pGALl expression vector was accomplished by PCR-amplifying a sequence encoding a protein of interest using a primer comprising, in addition to gene-specific sequences, the 5'-terminal sequence GGGAGCTGAAACCGCGGTCCCAAATTCC, and another primer comprising, in addition to gene-specific sequences, the 5'-terminal sequence

AACCCTCACTAAAGGGAACAAAAGCTGG, and by co-transforming yeast host cells with the purified PCR product and the linear pGALl expression vector. The terminal sequences of the PCR product served to direct the sequence encoding the protein of interest to be recombined into the co- transformed pGALl expression vector by the native recombination machinery in Saccharomyces cerevisiae (see, for example, the cloning of fusion proteins in Example 11).

For protein expression, strain Y4743 (Macbeth and Bass. Large-scale overexpression and purification of ADARs from Saccharomyces cerevisiae for biophysical and biochemical studies. Methods Enzymol. 2007, 424:319-31) was used. This strain is deficient in the expression of several proteases and has an extra copy of the GAL4 gene of Saccharomyces cerevisiae under regulatory control of the promoter of the GAL1 gene of Saccharomyces cerevisiae, modifications that are expected to increase heterologous protein stability and expression, respectively.

For biomass build and induction, a 50 mL culture of Y4743 cells containing the plasmid of interest were grown overnight in modified CSM (6.7 g/L YNB, 10 mL/L succinic acid, 6 g/L NaOH pellets, 1.92 g/L CSM, 0.002% adenine sulfate) containing all amino acids with the exception of uracil and containing 2% glucose. This culture was diluted to an OD ₆₀ o of 0.2 in 50 mL fresh modified CSM lacking uracil and containing 2% glucose (to maintain selection for the plasmid), and allowed to grow for 24 hours. The complete culture was then harvested by centrifugation, the supernatant discarded, and the culture pellet resuspended to a total volume of 50 mL in modified CSM lacking uracil and containing 2.7% galactose as the only carbon source. This switch to galactose was the induction phase, and lasted a minimum of 16 hours and a maximum of 24 hours. Cell pellets were harvested, washed in lx phosphate buffered saline (PBS), and stored at -80 °C until needed. When necessary, cell pellets were taken into the anaerobic cham ber to seal the pellet in a layer of gas free of oxygen.

For protein purification, the Strepl l tagged proteins were purified using ST EP-TACTI N™ sepharose resin or the STREP-TACTI N™ spin columns (IBA Gm bH, Goettingen, Germany) according to the manufacturer's suggested protocol with the exception that avidin was added at a final concentration of 40 μg/mL to prevent native biotinylated proteins in the cell free extract from bind ing the STREP-TACTIN™ resin.

To determine the fraction of solu ble Escherichia coli ispG protein produced using the pGALl expression vector, Y4743 cells comprising pAM2130 were lysed by either the protein denatu ring extraction method or the solu ble protein extraction method. For the solu ble protein extraction method, cell pellets were resuspended in a non-denaturing buffer (100 m M Tris-HCI pH 8.0, 150 mM NaCI, and 1 complete mini protease inhibitor cocktail ta blet (Roche Applied Science, Indianapolis, I N) per 10 m L of buffer), 0.5 mm glass beads were added, and the cells were lysed using a M P bead beater. Tu bes were centrifuged, total protein concentration in the su pernatant was determined by Bradford analysis, a nd 2.5 μg of total protein was separated by SDS PAGE and su bjected to Western blot using an anti-ispG primary antibody. For the protein denaturing extraction method, cell pellets were resuspended in a denaturing buffer (20 m M Tris pH 7.4, 8M urea, 2%CHAPS, 1%DTT, and 1 complete mini protease inhibitor cocktail ta blet per 10 m L of buffer), 0.5 mm glass beads were added, and the cells were lysed using a M P bead beater. Tu bes were centrifuged, total protein concentration in the su pernatant was determined by Bradford analysis, and 3.5 μg of total protein was separated by SDS PAGE and su bjected to Western blot using an anti-ispG primary antibody.

As shown in FIG. 3 B, the signal obtained with the denaturing protein extraction method (lane 3) was significantly stronger than the signal obtained using the solu ble protein extraction method (lane 2), suggesting that the large majority of Escherichia coli IspG was insolu ble when expressed in the yeast cytosol.

To investigate whether a lack of solu ble IspG expression in Y4743 cells comprising pAM2130 was related to an Fe-S incorporation issue, a mitochondria l-targeted Escherichia coli ispG expression construct was created. To this end, the DNA fragments shown in Ta ble 11 were PCR-amplified as described in Example 1. Ta ble 11 - PC Amplified DNA Fragments

DNA Fragment Primers Template

GW114 (SEQ I D NO: 120)

PGALl/10-2 Y002 genomic DNA

GW115 (SEQ I D NO: 121)

GW116 (SEQ I D NO: 123)

MTS ^a) SEQ ID NO: 122 ^b)

GW150 (SEQ I D NO: 124)

GW400 (SEQ I D NO: 125)

PGAL1-MTS-2 PGALl/lO-MTS-1 (see Table 13)

GW401 (SEQ I D NO: 126)

a) MTS = mitochondrial targeting signal of the Su9 gene of Neurospora crassa (Westermann and Neupert, Mitochondria-targeted green fluorescent proteins: convenient tools for the study of organelle biogenesis in S. cerevisiae. Yeast 2000,16, 1421-1427), codon-optimized for expression in Saccharomyces cerevisiae.

b) The synthetically-generated DNA fragment was provided by Integrated DNA Technologies, Inc.

(Coraville, IA).

The DNA fragments were assem bled into stitches as outlined in Ta ble 13 a nd as described in Example 1. The final PGAL1-MTS-2 PCR fragment had 500 bp of homology to the PGAL1 promoter at its 5'- end and 30 bp of homology to the Strepll tag on its 3'-end. It was mixed with linearized pAM2130, and the mixture was used to transform Y4743 cells, in which the two DNA fragments were ligated via host cell mediated gap-repair homologous recombination to yield a plasmid. The plasmid was recovered using the ZYMOPREP™ Yeast Plasmid Miniprep I I kit (Zymo Research, Irvine, CA), and sequence verified. The plasmid was digested using Agel restriction endonuclease to remove the ispG coding sequence, and the vector backbone was purified using a QU IAGEN™ gel extraction kit, yielding the pGALl-MTS expression vector. The mitochondrial-targeted IspG protein (EcJspG-MTS) was found to be mostly solu ble, readily purified, and active.

Example 8

This example describes the in vitro identification of active IspG enzymes in Saccharomyces cerevisiae.

Using the pGALl expression vector, various IspG homologs were expressed in Y4743, extracted from the host cells via the solu ble protein extraction method, and analyzed by Western blot as described in Example 7.

As shown in Ta ble 13, 29 of the IspG homologs tested proved to be solu ble in the host cell cytosol. The activity of the 29 solu ble IspG homologs was then determined using artificial electron donors. For in vitro enzyme activity assays with artificial electron donors, the 50 μί reaction contained degassed 100 mM Tris buffer pH8.0, cM EPP at 320 μΜ, 5 mM dithionite, 1 m M Methyl viologen, and 5 μί of freshly purified IspG protein. The total volume was brought up to 50 μί using degassed, deionized water. Samples were incu bated in the anaerobic cham ber at am bient temperature. At the desired time points, 10 μί of the reaction was added to 490 μί of methanol with 1 μg/mL of internal standard (TFPP). Samples were then kept at -80°C until they were analyzed by LC-M M as described in Example 3. The activity of the IspG enzymes was measured by the rate of H DMAPP formation. Relative activity was determined by the relative amount of H DMAPP formation observed with equiva lent a mounts of IspG protein over the same amount of time.

Pb_G

Prevotella buccae D17 EFC75705 ++++ +

(SEQ ID NO: 170)

Pc_G

Paulinella chromatophora YP_002049460.1 ++++ -

(SEQ ID NO: 171)

Pm_G Prochlorococcus marinus subsp.

NP_892794.1 ++++ +

(SEQ ID NO: 172) pastor is str. CC MP 1986

Py_G

Porphyra yezoensis red algae ACI45961 ++++ +

(SEQ ID NO: 173)

Sb_G

Sorghum bicolor XP_002454137 ++ ++

(SEQ ID NO: 174)

Sm_G Selaginella moellendorffii XP_002986683.1 ++ -

Syn_G

Synechococcus sp. RS9917 ZP_01081635.1 ++++ ++

(SEQ ID NO: 175)

Te_G Thermosynechococcus elongatus

NP_681786 ++++ ++

(SEQ ID NO: 176) BP-1

Tt_G

Thermus thermophilus HB8 YP_143571.1 ++++ +++++

(SEQ ID NO: 177)

Vv_G

Vitis vinifera XP_002285130.1 ++++ ++

(SEQ ID NO: 178)

All IspG homologues tested were codon-optimized for expression in Saccharomyces cerevisiae, and devoid of any organelle targeting sequences.

- = no activity detected; ++++, +++, ++, + = most soluble or active, highly soluble or active, medium soluble or active, poorly soluble or active, respectively.

Example 9

This example describes the in vitro identification of active IspH enzymes in Saccharomyces cerevisiae.

Using the pGALl expression vector, various IspHs were expressed in Y4743, extracted from the host cells via the solu ble protein extraction method, and analyzed by Western blot as described in Example 7. Table 14 illustrates seven exemplary IspHs that were solu ble in the host cell cytosol.

The activity of the exemplary IspH homologs was then determined using artificial electron donors as described for IspG except that in place of cMEPP at 320 μΜ, H DMAPP at 380 μΜ was used in the 50 μϋ^^ϊοη, and the activities of the IspH enzymes were measured by the rate of

DMAPP/I PP formation.

IspH enzymes were generally much more robust than IspG enzymes. In general, if the enzymes were solu ble in the host cell's cytosol, they were more likely to be fully functional, displaying good to excellent enzymatic activity. Illustratives exa mples of suita ble IspH's include but are not limited to IspH enzymes from: Escherichia coli, Fibrobacter succinogens, Physcometrella patens, Sorghum bicolor, Stevia rebaudiana, Thermosynecoccus elongatus, and Arabidopsis thaliana. Table 14 - Solubilities and Activities of IspH Homologs in Yeast Host Cells

Activity with

Solubility in

Protein Accession Artificial

IspH Homolog Species Origin Host Cell

Number Electron

Cytosol

Donors

At-H

Arabidopsis thaliana AY168881 ++++ ++++

(SEQ ID NO: 179)

Ec-H

Escherichia coli AAC73140 ++ ++++

(SEQ ID NO: 180)

Fs-H Fibrobacter succinogenes subsp.

YP_003250012.1 ++ ++

(SEQ ID NO: 181) succinogenes S85

Pp-H

Physcomitrella patens subsp. patens (1) XP_001758369.1 ++ ++++

(SEQ ID NO: 182)

Sb-H

Sorghum bicolor XP_002463933.1 ++++ ++++

(SEQ ID NO: 183)

Sr-H

Stevia rebaudiana ABB88836.2 ++++ +++

(SEQ ID NO: 184)

Te-H

Thermosynechococcus elongatus BP-1 NP_681832 ++++ ++++

(SEQ ID NO: 185)

All IspH homologs tested were codon-optimized for expression in Saccharomyces cerevisiae, and devoid of any organelle targeting sequences.

NA = not available (e.g., not expressed in soluble form in yeast); NT = not tested; ++++, +++, ++, + = most soluble or active, highly soluble or active, medium soluble or active, poorly soluble or active, respectively.

Example 10

This example describes the in vitro identification of redox partners for DXP pathway enzymes IspG a nd IspH in Saccharomyces cerevisiae.

Using the pGALl expression vector, the 14 flavodoxin and 10 flavodoxin/ferredoxin NADP ⁺ - reductases (FN ) homologs listed in Tables 14 and 15, respectively, were expressed in Y4743, extracted from the host cells via the solu ble protein extraction method, a nd analyzed by Western blotting as described in Example 7.

Ta ble 15 shows the results of exemplary flavodoxins that were solu ble in the cytosol and their respective degree of compatibility with IspG. As stated previously, IspH's are generally more robust than IspG's and suita ble redox partners and reductases that provide a fully functional IspH were more readily identified.

Table 15 - Solubilities and Activities as Redox Partner for IspG Enzymes of Flavodoxin Homologs in Yeast Host Cells

Redox

Protein Solubility

Flavodoxin Partner

Species Origin Accession in Host Cell

Homolog Activity with

Number Cytosol

IspG

AM_fld

Alteromonas macleodii ATCC 27126 ZP_04714638.1 ++++ +

(SEQ ID NO: 127)

Anabaena_fld

Anabaena PCC 7119 AAB20462 ++++ ++ (SEQ ID NO: 128)

Bc_fld

Bacillus coagulans 36D1 ZP_04431144.1 ++++ +

(SEQ ID NO: 129)

BsyN_fld Bacillus subtilis subsp. subtilis str.

NP_389298.1 ++++ ++++ (SEQ ID NO: 130) 168 YkuN

BsyP_fld Bacillus subtilis subsp. subtilis str.

NP_389300.2 ++++ +

(SEQ ID NO: 131) 168 YkuP

Cw_fld

Crocosphaera watsonii WH 8501 ZP_00515759.1 ++++ +

(SEQ ID NO: 132)

Dv_fld Desulfovibrio vulgaris subsp. vulgaris

YP_966026.1 ++++ +

(SEQ ID NO: 133) DP4

Ec_fld Escherichia coli str. K-12 substr.

NP_415210 ++++ +

(SEQ ID NO: 134) MG1655

Np_fld

Nostoc punctiforme PCC 73102 YP_001866434.1 ++++ +

(SEQ ID NO: 135)

Rs_fld Rhodobacter sphaeroides ATCC

YP_001169375.1 ++++ -

(SEQ ID NO: 137) 17025

Syn_fld

Synechococcus sp. CC9605 YP_382866.2 ++++ -

(SEQ ID NO: 138)

Vh_fld

Vibrio harveyi 1DA3 ZP_06175070.1 ++++ -

(SEQ ID NO: 140)

Ws_fld

Wolinella succinogenes DSM 1740 NP_906993.1 ++++ ++

(SEQ ID NO: 141)

All flavodoxin homologs tested were codon-optimized for expression in Saccharomyces cerevisiae, and devoid of any organelle targeting sequences.

- = no activity detected; ++++, +++, ++, + = most soluble or active, highly soluble or active, medium soluble or active, poorly soluble or active, respectively.

Table 16 - Solubilities and Activities as Redox Partner for IspG and IspH Enzymes of FNR Homologs

in Yeast Host Cells

Redox Partner Redox Partner

Protein Solubility in

Activity with Activity with

FNR Homolog Species Origin Accession Host Cell

IspGs and IspHs and Number Cytosol

Flavodoxins Flavodoxins

Anabaena_FNR

Anabaena PCC 7119 1QUE_A ++++ ++++ NT

(SEQ ID NO: 143)

At_RFNR2 Arabidopsis thaliana

NP_849734.1 ++++ ++++ ++++

(SEQ ID NO: 145) (root)

Ec_FNR Escherichia coli str.

NP_418359 ++++ ++ +++

(SEQ ID NO: 146) K-12 substr. MG1655

Nostoc_FNR

Nostoc sp. PCC 7906 BAG69182 ++++ ++++ ++++

(SEQ ID NO: 147)

So_FNR

Spinacia oleracea P00455 +++ ++ NT

(SEQ ID NO: 150)

Te_FNR Thermosynechococcus

NP_682001 ++++ ++++ ++++

(SEQ ID NO: 151) elongatus BP-1

Zm_FNR

Zea mays 3LVB_A ++++ ++++ ++++

(SEQ ID NO: 152)

All FNR homologs tested were codon-optimized for expression in Saccharomyces cerevisiae, and devoid of any organelle targeting sequences.

NA = not available (e.g., not expressed in soluble form in yeast); NT = not tested; ++++, +++, ++, + = most soluble or active, highly soluble or active, medium soluble or active, poorly soluble or active, respectively.

Assays with flavodoxin homologs and FNR homologs were performed essentially as described a bove for artificial electron donors, except the 50 μί reaction contained degassed 100 mM Tris buffer pH8.0, cM EPP at 320 μΜ (for ispG) or H DMAPP at 380 μΜ (for ispH), 5 μΐ of the freshly purified FN R, 5 μί of the freshly purified flavodoxin, and 5 μί of freshly purified IspG or IspH protein.

The four FN R homologs proved to be of equal activity, such that a ny of them could be used in com bination with IspG or IspH and a flavodoxin without a change in overall enzyme activity.

Certain redox partners were found to be compatible with both IspG a nd IspH. Illustrative examples of two such redox partners include Escherichia coli str. K-12 su bstr. MG1655

(NP_415210) and Bacillus subtilis subsp. subtilis str. 168 Yku N (NP_389298.1). However, each IspG a nd IspH provided the best activity if paired with a different redox partner, suggesting that the respective redox partner in the endogenous system is specific and distinct.

The most compatible flavodoxin/lspG pair that was identified from the a bove screen was the flavodoxin of Bacillus subtilis Yku N with IspG from Bacillus coagulans, Bacillus thuringiensis, Bacillus subtilis, and Thermus thermophilus.

The most compatible flavodoxin/lspH pair that was identified was Escherichia coli flavodoxin with IspH from 5. bicolor. The Bacillus subtilis flavodoxin Yku N also worked well with IspH but was not nearly as active as the Escherichia coli flavodoxin. Not Furnished Upon Filing

Example 11

This example describes the generation of protein fusions of IspG or IspH with their flavodoxin redox partner proteins.

Fusion proteins of the IspG and IspH enzymes with their preferred flavodoxin redox partner were created to increase the local effective concentration of flavodoxin in vivo. Fusion proteins have the additional advantages of ensuring one-to-one stoichiometries of interacting enzymes in vivo, and of providing a single template for enzyme engineering to enhance enzyme interactions and catalytic properties.

N- and C-terminal fusions were made between Bc_G, Bt_G, and Tt_G with BsyN_fld, and between Ec_H and Ec_fld. To this end, the DNA fragments shown in Ta ble 17 were PC -amplified as described in Example 1.

The DNA fragments were assem bled into stitches as outlined in Ta ble 18 a nd as described in Example 1.

Finally, Y4743 cells were co-transformed with the purified stitches and linear pGALl expression vector, and the fusion proteins were purified as described.

The activities of the fusion proteins were assayed using the in vitro enzyme activity assay with the physiological electron donor as described a bove except only one source of IspG or IspH and flavodoxin redox partner was needed. The percent prod uction of HTMAPP or DMAPP/I PP after 16 hours was compared to product production by solu ble Bt_G with BsyN_fld (the best IspG flavodoxin pair identified in Example 10) or Ec_H with Ec_fld (the best IspH flavodoxin pair identified in Example 10), respectively.

As shown in FIGs. 4A and 4B, Bc_G-BsyN_fld, Bt_G-BsyN_fld, and Tt_G-BsyN_fld, and Ec_fld-Ec_H fusion proteins were active and converted a significant amount of CM EPP into H DMAPP or H DMAPP to DMAPP/I PP, respectively.

Example 12

This example describes methods for measuring the amount of Fe incorporation in DXP pathway enzymes expressed in yeast.

This method was adapted from a reported procedure (Pierik AG, Netz DJ, Lill . Nat Protoc 2009, 4:753-766). Yeast cells were grown in 50 m L of modified CSM comprising 2% glucose u ntil the OD ₆₀ o was greater than 2. At that point, cells were spun down, resuspended in 50 m L of low iron modified CSM comprising 2.7% galactose, and grown overnight. The next morning, the cell cultures were spun down, cell pellets were resuspended in 87 mL of low iron CSM comprising

2.7% galactose, and the resuspended cell culture was split into four different flasks: three 250 mL disposa ble flasks received 25 mL of cell culture each. 10 μθ of ⁵⁵ Fe ( ⁵⁵ FeCI ₃ (PerkinElmer, Boston MA) added to 250 μί of 0.1 M ascorbate solution per flask) was added to each flask, and one 125 mL flask received 12 mL of culture and u nla belled FeCI ₃ to a final concentration of 300 μΜ (25 μί of 150 m M FeCI ₃ was added to 125 μΐ of 0.1 M ascorbate solution, and 150 μΐ of the FeCI ₃ in ascorbate was added to 12 m L of cell culture). All cultures were then grown for 20-24 hours prior to cell harvesting and processing.

For processing, the cells were pelleted, washed in 10 m L of citrate buffer (50 m M citrate, 1 m M EDTA, pH 7.0), pelleted, washed in 1 m L of H EPES buffer (20 m M H EPES/KOH pH 7.4), transferred to a 2 mL centrifuge tu be, and pelleted. The HEPES buffer was carefully removed with a pipette and the wet pellet was weighed. 500 μΐ of cold Buffer W (100 m M Tirs pH 8.0, 150 m M NaCI) with complete EDTA-free protease inhibitor (Roche Applied Science, Indianapolis, IN) was added per 0.3 g of wet pellet weight. 700 μί of resuspended cells (equivalent to 0.3 g of pellet wet weight) was added to 2.0 m L O-ring tu bes containing 500 μί of 0.5 mm diameter glass beads and the cells were lysed by bead-beating in a FASTPREP™ instrument (M P Biomedicals, Solon, OH); cells were beat for 40 sec at 6.0 m/s, iced for 5 min, and beat for another 40 sec at 6.0 m/s then iced for 5 min. All su bsequent procedures were done on ice or at 4°C. Tu bes were centrifuged at max speed for 5 min, and all supernatant was carefully removed and transferred to a pre-chilled 1.5 m L centrifuge tu be. Sa mples were centrifuged at max speed again for 5 min to remove cellular debris, and the clarified supernatant was carefully transferred to a fresh 1.5 m L centrifuge tu be. 15 μg of avidin was added to each sample, gently mixed, and allowed to sit on ice for 30 min; avidin was added to prevent native biotinylated proteins in the cell-free extract from bind ing the streptactin resin. Clarified supernatant with avidin was then loaded onto a ST EP-TACTIN™ spin colu mn (IBA Gm bH) according to the manufacturer's suggested protocol. The concentration of protein in the elution was determined using a QUBIT™ reader (Invitrogen, Carlsbad, CA) according to the manufacturer's instructions. All remaining protein elution was added into a glass scintillation vial containing 10 m L of scintillation fluid and the radioactivity was detected on a Beckman Coulter LS 6500 scintillation counter. Total counts of radioactivity were normalized to the amount of protein in the sample.

The unla belled cell culture was prepped in the method described a bove, and the eluted protein was run out on a polyacrylamide gel to verify that the protein was expressed, solu ble, and pure. The endogenous yeast gene LEUl is a cytosolic Fe-S containing protein, and this protein was also cloned into the 2 micron PGALl-Strepll N-terminal tag plasmid as a control. In every experiment Y4763 containing LEUl was cultured, la beled, a nd extracted as a positive control and to normalize ⁵⁵ Fe la beling among experiments. Y4763 with the PGALl-Strepl l N-terminal tagged plasmid containing no gene insert was run as a negative control to control for background ⁵⁵ Fe signal. No yeast protein was ever detected on a protein gel using this negative control.

As shown in FIG. 5, heterologously-expressed ispG and ispH proteins incorporated la beled ⁵⁵ Fe, suggesting that the endogenous cytosolic Fe-S machinery in 5. cerevisiae is capa ble of incorporating the necessary metallic cofactors needed for catalysis in these proteins. Note that the ispG protein from Escherichia co// ^' was shown to be insolu ble in 5. cerevisiae, and therefore no signal was detected for ⁵⁵ Fe as no protein was eluted from the Strepl l purification column.

Example 13

This example describes methods for the quantitation of complexed iron in the purified IspG and IspH enzymes.

The amount of iron present in purified IspG and IspH protein samples was quantitated using the method reported previously (Fish, W. Rapid Colormetric Micromethod for the Quantitation of Complexed Iron in Biological Samples, Methods in Enzymology, 1988, 158:357-364). Protein concentration was determined using the QUBIT™ Fluorometer (P/N Q32860, Invitrogen, Chicago, I L) and the QUANT-IT™ Protein Assay Kit (P/N Q33212, Invitrogen, Chicago, IL) per the manufacturer's instructions. Glassware used for reagent preparation was de-mineralized by acid treatment, and then rinsed in Hyclone ultra pure water, WFI grade (P/N 16750-108, VWR, Pittsburgh, PA). All reagents were made using Hyclone ultra pure water, WFI grade (P/N 16750- 108, VW , Pittsburgh, PA). The standard for Fe calibration was prepared as follows: a 0.041% solution of Iron (II) ethylenediammonium sulfate tetrahydrate solution was serially diluted in 0.01 N HCI to a range of 0.1875 μg/mL-6 μg/mL. Purified protein samples were diluted, if necessary, in Protein Diluent (100 m M Tris-HCI, pH 8.0, 150 mM NaCI, 2.5 m M Biotin). A blank sample to measure reagent background was prepared at the same time using 600 μί of Protein Diluent in place of the purified protein. To release the bound iron, 300 μίοί Reagent A (0.6 N HCL, 2.25% (w/v) KMn0 ₄ ) was added to 600 μί of purified protein or blank solution in an Eppendorf tu be. 250 μί of Reagent A was also added to 600 μί of each of the standard solutions. Samples were then incu bated for two hours at 60°C. After incu bation, 0.06 m L of Reagent B (6.5 m M Ferrozine (P/N 82950, Sigma-Aldrich, Inc., Atla nta, CA), 13.1 m M neocuprione (P/N 72080, Sigma-Aldrich, Inc., Atlanta, GA), 2 M ascorbic acid, 5 M a mmonium acetate) was added to each sample.

Samples were then incu bated at room temperature (22-25°C) for 30 minutes while protected from light. Absorbance was read at 562 nM (A ₅₆₂ ) on a GEN ESYS™ 10 spectrophotometer (Thermo Scientific, Asheville, NC) using disposa ble polystyrene cuvettes. A calibration curve was esta blished and the iron concentrations in the protein samples were calculated from their A ₅₆₂ based on the calibration curve.

As shown in FIG. 6, the heterologously-expressed IspG and IspH proteins contained

approximately 1 mol Fe per mol of protein. The amount of Fe contained in IspG and IspH proteins was lower than what is expected for a protein with four Fe-S clusters. However, the fact that the native yeast [4Fe-4S] enzyme Leu l, which also contained approximately 1 mol Fe per mol of protein, contained the same amount of Fe as isolated IspG a nd IspH, suggests that the low level of Fe-S incorporation is due to the limited capacity of the yeast Fe-S cluster assem bly machinery. Example 14

This example describes the detection of IspG meta bolites from 5. cerevisiae strains bearing heterologous DXP pathways.

In the reported in vitro experiments, IspG homologues exhibited lower catalytic activity than IspH homologues even though both enzymes contain a [4Fe-4S] cluster. The results point to the IspG- catalyzed conversion of cM EPP to H DMAPP as a likely bottleneck in the DXP pathway in yeast host cells.

To determine whether the in vivo activity of the IspG homologues was indeed reduced, strains Y4819, Y6283, Y6291, and Y6292 were induced for IspG expression, and then analyzed for intracellular H DMAPP. To this end, the strains were inoculated into 3 mL of CSM comprising 2% (w/v) glucose and lacking uracil, and grown overnight at 30°C, with shaking. Fresh 50 mL of CSM comprising 2% glucose and lacking uracil was inoculated with 3 m L of the overnight cultures, and the resultant cultures were allowed to grow at 30°C until they reached an OD ₆₀ o of between 1 and 4. The cultures were spun down at 5,000xg for 10 min to separate cells from the supernatant, the cells were transferred to 50 mL of CSM comprising 4% (w/v) galactose and lacking uracil, and the cultures were incubated for an additional 18-22 hours at 30°C, with shaking. Metabolite extraction and analysis of HDMAPP were carried out as described in Example 3.

As shown in FIG. 7, no HDMAPP accumulation was detected in the control strain, Y4819, which only contains the DXP-TOP pathway. A small amount of HDMAPP was detected in Y6283, which harbors a complete DXP pathway. Much higher levels of HDMAPP (approximately 10 μΜ) were observed in strains Y6291 and Y6292, which lack an IspH gene but express B. subtilis flavodoxin YkuN along with Bt-G and Tt-G, respectively.

To further verify the chemical identity of an HDMAPP signal observed in the strains depicted in FIG. 7, strains Y6291 and Y6289 were grown on fully ¹³ C-labeled sugar as a carbon source, the incorporation of which shifts the HDMAPP metabolite accumulated in these strains by 5 atomic mass units (AMU) from 260.916 to 265.916 in the LC/MS analysis. Strains Y4819, Y6291, and Y6289 were inoculated into 3 mL of CSM comprising 2% (w/v) ¹³ C-glucose and lacking uracil, and grown overnight at 30°C with shaking. A fresh 10 mL aliquot of CSM comprising 2% ¹³ C-glucose and lacking uracil was inoculated with 1 mL of the overnight cultures, and the resultant cultures were allowed to grow at 30°C until they reached an OD ₆₀ o between 1 and 4. A 2 mL aliquot of CSM comprising 4% (w/v) ¹³ C-galactose and lacking uracil was then inoculated with the above cultures to a final OD ₆₀ o of 0.03 and allowed to grow for an additional 48 hours at 30°C, with shaking. Metabolite extraction and analysis of HDMAPP were carried out as previously described. No detectable ¹³ C-labeled HDMAPP signal was observed in Y4819. ¹³ C-labeled HDMAPP signal was also missing in Y6289 and Y6291 when they were cultured in ¹³ C-labeled glucose (IspG expression was not induced). In contrast, ¹³ C-labeled HDMAPP accumulated in Y6289 and Y6291 when ¹³ C- labeled galactose was used as the carbon source to induce IspG and YkuN expression.

Example 15

This example describes the detection of FPP production via a heterologous DXP pathway in 5. cerevisiae.

To determine whether the heterologous DXP pathway is functional, yeast strains harboring a full DXP pathway and a deleted E G13 coding sequence were grown on ¹³ C-labeled media supplemented with ¹² C-mevalonate. Under the cultivation conditions, unlabeled IPP/DMAPP metabolites were only made from the supplemented ¹² C-mevalonate via the mevalonate pathway. A functional DXP pathway, on the other hand, would produce only ¹³ C-labeled

DMAPP/IPP. FPP, which is made by condensing 1 molecule of DMAPP and 2 molecules of IPP, may therefore exist in either an unla beled form (mass = 381.065), partially la beled forms (mass + 5 AM U = 386.065, and mass + 10 AM U = 391.065), or a fully la beled form (mass + 15 AM U = 397.055). Detection of any la beled form of FPP (i.e. mass + 5 AM U, mass + 10 AM U, or mass + 15 AM U) would demonstrate that the heterologous DXP pathway was capa ble of providing flux to isoprenoids. The ratio of ¹³ C-la beled FPP to unla beled FPP was determined by the level of DXP flux in yeast relative to the native mevalonate pathway flux (mevalonate concentrations in the growth medium).

Strains Y6540 and Y6533 were inoculated into 3 mL of CSM comprising 2% (w/v) ¹³ C-glucose a nd lacking adenine, and grown overnight at 30°C with shaking. Fresh 10 mL of CSM comprising 2% ¹³ C-glucose a nd lacking adenine was inoculated with 1 mL of the overnight cultures, and the resultant cultures were allowed to grow at 30°C until they reached an OD ₆₀ o of between 1 and 4. The cultures were then equally split into two flasks, and grown in the presence or a bsence of zaragozic acid (a squalene synthase inhibitor). A final concentration of 16 μg/m L of zaragozic acid was added to one flask while the same volume of water (6 μί) was added to the other. Four of the resultant cultures were then allowed to grow for an additional 18-22 hours at 30°C, with shaking. Meta bolite extraction and analysis of DMAPP and FPP were carried out as described in Example 3.

As shown in FIG. 8, the intracellular DMAPP/I PP meta bolite concentrations were extremely low in these strains. No ¹³ C-la beled DMAPP/I PP was detected in either strain. In cultures without zaragozic acid, the level of ¹² C-la beled FPP was very low, and no ¹³ C-la beled FPP was detected. Partially la beled FPP (mass + 5 AM U or 386.065) was detected in Y6533 only when zaragozic acid was added to cause FPP to accumulate by inhibiting the FPP-consuming enzyme squalene synthase. In contrast, Y6540, in which the expression of IspG, IspH, and redox proteins is only induced by the addition of galactose, did not accumulate any ¹³ C-la beled FPP even in the presence of zaragozic acid. It was estimated that the FPP + 5 AM U signal was at approximately 1% of the total FPP, indicating that the DXP flux is a bout 100-fold less active than the native mevalonate pathway. Since la beled FPP of mass +10 AM U and +15 AM U is made via

condensation of 2 or 3 ¹³ C-la beled DMAPP/I PP u nits, respectively, these la beled meta bolites are expected to be below the limit of LC/MS detection, which explains why only the FPP +5 form of the ¹³ C-la beled FPP was detected.

The foregoing examples of the present invention have been presented for pu rposes of illustration and description. Furthermore, these examples are not intended to limit the invention to the form disclosed herein. Consequently, variations and modifications commensurate with the teachings of the description of the invention, and the skill or knowledge of the relevant art, are within the scope of the present invention. The specific em bodiments described in the examples provided herein are intended to further explain the best mode known for practicing the invention and to ena ble others skilled in the art to utilize the invention in such, or other, em bodiments and with various modifications required by the particular applications or uses of the present invention. It is intended that the appended claims be construed to include alternative embodiments to the extent permitted by the prior art.