CA2140527A1 - Methods for increasing carbon conversion efficiency in microorganisms - Google Patents

Abstract

2140527 9428154 PCTABS00034 Methods are presented for increasing the carbon conversion efficiency of a microorganism by introduction of DNA encoding enzymes of the glyoxylate cycle into the microorganism.

Description

~ ~4/2815't 21~ 0 ~ 2 7 PCT/US94/06084 METHODS FOlR IN(: REASING CARBON
CONVE~ION EFFIClENCY lN ~ClROORG~lSMS

The present invention generally relates to methods for increasing the carbon conYe~sion efficiency of microorganisms.

Baclcground Of The Invention In the presence of glucose or another carbon compound (e.g.9 a carbohydrate) capable of entry into the tsicar~oxylic acid cyc}e (hereinafter referred to as ~he "TCA cycle") as the sole source of ca3bon, bacterial cells and other microorganisms utilize the TCA cycle as a primary metabolic pathway.
The TCA cycle generates energy sources via the oxidation of acetyl groups which enter the cycle as acetyl CoA moleeules. In so doing, the TCA cycle primarily produces energy and various four-carbon intermediates, such as oxaloacetate, which serve as precursors in the synthesis of esoential cell components such as amino acids. These four-carbon intermediates result from two decarboxylation reactions in the TCA cycle, both generating carbon dioxide (CO2). The first decarboxylation reaction occurs when the six-carbon metabolit, is~itrate, is converted to the five-carbon compound, a-ketoglutarate, by the enzyme isocitra~edehydrogenase. The second carbon dioxide-producing reaction occurs in the next step of the TCA cycle, wherein ~-keto~luta~te is converted to the four-calbon , 1 20 i ntermediatej succinyl CoA.
As used herein, carbon conversion e~ficiency gen~rally re~rs to a r~tio of the amount of substrate mctabolized to the amount of biosynthetic products formed by a microorganism. The greater the loss of carbon during the process of con~rerslon from substrate :to biosynthetic products the greater the mefficiency of carbon conversion. Under conditions when the TCA cycle is a 21~05~7 predominant metabolic pathway in a microorganism, carbon conversion is limited by the amount of carbon lost through the aforementioned carbon-dioxide-evolving steps. There exists a need in the art for methods ~or increasing carbon conversion ef~lciency under conditions when the TCA cycle would normally be predominant, S i.e., when glucose or another substrate is available. The present invention provides methods for increasing carbon conversion efficiency under such conditions.

Brief Summary Of The Invention The present invention provides methods for increasing carbon conversion efficiency in a microorganism.
In a preferred embodiment of the invention, methods for inereasing carbon conversion efficiency are provided, wherein DNA encoding isocitrate Iyase, malate synthase, and isocitrate dehydrogenase kinase/phosphatase is incorporated into the microorganism which is then cultivated under appropnate conditions in the presence of a substrate rnetabolized by the TCA cycle and whereîn biosynthetic product may be isolated.
Also in a preferred ernbodiment of the invention, the DNA
encoding isocitrate lyase, malate synthase, and isocitrate dehydr~genase kinase/phosphatase compnses a portion of the glyoxylate operon.
A mic~oorganism according to the present invention may be any microorganism which normal}y utilizes the TCA cycle, but may preferably be a bacteAal cell, and most preferably an Escherichia coli cell.
In a prefelTed embodiment of the invention, the DNA encoding isocitrate Iyase, malate synthase, and Isocitrate dehydrogenase kinaselphosphatase may be incorporated in the microorganism in an appropriate vector, such as a plasmid.

`

~`

`; :

) 94~154 214 0 5 2 7 PCT/US94/06084 Also in a preferred embodiment of the invention, the method further comprises calculating the carbon conversion efficiency of a mlcroorganism.
Numerous additional aspects ar d advantages of the invention are S apparent to the skilled artisan upon consideration of ~he following detailed description which provides presen~ly preferred embodiments thereof.

Detailedi Description Of The In~ention If a two-carbon compound, such as acetate, is the sole source of carbon for the microorgar,ism, the two-carbon compound may be utilized in the TCA cycle to produce energy. However, in so doing both carbon atoms are lost as CO2, thus depleting carbon available for formation of biosynthetic products.
Microorganisms adjust to this situation by channeling some isocitrate into the glyoxylate cycle. The interrelation of the TCA cycle and the glyoxylate cycle isdepicted in Figure 1.
Carbon flow to the glyoxylate cycle is controlled by the bifunctional protein, isocit~ate dehydrogenase kinase/phospb~atase ~hereinafter referre~ to as "IDH k/p"). The glyoxylate cycle is repressed when glucose is avaiiable as a source of carbon. However, when a~etate is the sole carbon source, IDH klp reduces activity of isoci~te dehydrogenasej thus allowing the enzyme isocitrate lyase of the glyoxylate cycle to act upon isocitrate with reduced competition from isocitrate dehydrogenase.
The enzymes of the glyoxylate cycle in Escheric~ia coli are nc~ed by genes comprising the glyoxylate ~peron. ~ The known structural o~anization of the operon comprise the aceB gene which encodes mala~
synthase, the aceA gene which encodes isocitrate lyase, and the aGelY gene whichencodes IDH k/p. T~anscription of the operon in its native sequence is under thenegative regulation of the iclR gene and, to a lesser extent, thef~d~ gene which ~ , . . ... , ... .. . ..... ..... .. , .. , , ., . .. ~ , . , . , . ... ,. .. ~ . .... j wo 94/28154 2 1 4 0 5 2 7 PCT/IJS94/06084 ~

maps outside the glyoxylate operon in E. coli. Cortay et al., Embo J., 10:675-679 (1991).
Two enzymes function in the glyoxylate cycle to convert isocitrate and acetyl CoA to succinate and malate. These are isocitrate lyase, which S converts isocitrate to succinate and glyoxylate, and mala~e synthase, which catalyzes the conversion of acetyl CoA and glyoxylate to malate. The net e~ect of the glyoxylate cycle is to avoid the carbon dioxide-evolving steps of the TCAcycle, thus generating four-carbon compounds from two-carbon precursors ~or use in the formation of cellular components.
la The present invention provides methods for increasing the carbon conversion efhciency of microorganisms, ~hus resulting in numerous benefits to industrial processes utilizing microorganisms. For example, increasing carbon conversion efficiency allows the use of less substrate to manufacture a given amount of biosynthetic product as compare~ to cells not trcated according to methods of the present invention. Conversely, a given amount of metabolized substrate, such as glucose, will be converted to greater arnounts of biosynthetic products under conditions in which carbon conv~rsion efficiency is increa~ed.
Increased carbon converslon ef~lciency may also be utilized to effect increases in the production of a particular amino acid for use in industrial applications~
Microorganisms which display increased carbon conversion e~ficiency may also display increase efficiency in biomass production gene~ly.
Carbon conversion efficiency is a measure of the extent to which c~rbon from substra~es, such as glucose, are incorporated into the intennediatesand products of cellular biosynthesis. Carbon conversion efficiency may be ' " ! ; ~ ~
measured b~sed on the accumulation of any product or products of cellular biosynthesis and îts ~their~ pre~ursors or on the basis of an increase in biomass with a fixed amount of substrate or on the basis of production of a given amountof biomass utili~ing less substrate. Examples of products which may be used to measure carbon conversion efficiency include biomass generally, amino acids, :: :.

21~0S27 ~ 94/281~4 PCT/US94!06084 nucleic acids, various structural proteins, enzymes, secondary metabolites and their precursors.
Por purposes of the present application, accumulahon of L-phenylalanine and its two immediate precursors, phenylpyruvic acid and prephenicacid, are used to demonstrate increased carbon conversion efficiency using methods according to the present invention. Carbon conversion ef~lciency may be calculated by dividing the total grams of L-phenylalanine, prephenic acid, and phenylpyruvic acid by the total grams of glucose consumed during fermen~ation of the microorganism. Such a calculation provides an estimate of the extent to which carbon from metabolic precursor molecules, such as glucose, is incorporated into biosynthetic products. Enhanced carbon comrersion efficiency means that less car~on is being channeled into undesirable metabolic products.
This, in turn, means that the microorganism will more efficiently produce biosynthetic products over time.
The use of methods according to the present invention is demonstrated by the following Examples. Example 1 relates to the use of exemplary plasmids comprising glyoxylate operon genes and the incorporation of such plasmids into exemplary host cells. Example 2 provides results of enzymaticassays to confirm proper mcorporation and function of a portion of the glyoxylate genes in host cells according to the invendon. Example 3 provides culture and fermentadon condi~ions under which carbon conversion efficiency may be measured. Final~y, Example 4 relates to an exemplary means of calculating carbon conversion efficiency ac ording to the present invention.

21~0527 WO 94/28154 . PCTIUS94/060B4 f Escherichia cvli, strain ATCC 13281, a tyrosine auxotroph, was used as an exemplary host cell to demonstrate methods according to the present invention. The skilled artisan realizes, however, that numerous host rnicroorganisms may be used in the practice of the invention. Representative host microorganisms are reported in Kornberg et Ql., Advan. Erlzymo~., 23:401-470 (1961), incorporated by reference herein.
Plasmids pCL8 and pCI~1000 were used as the source of genes encoding glyoxylate enzymes. These plasmids contain the aceA, aceB, and aceK
genes of the glyoxylate operon in a pBR322 vector and have been described in Chung et al., J. Bacte~?ol., 170:386-392 (1988), ineorporated by reference herein.
Plasmids pCL8 and pCL1000 were incorporated into E. coli host cells by electroporation. Prior to electroporation, a single colony of ATCC 13281 E. coli cells was inoculated into a S00 ml baMed flask containing 100 mls Luria Broth (Difco, Detroit, MI). The culture was grown on a rotary incubator/shaker (New Brunswick, Edison, NJ) for approxirnately 4 hours at 32 C until it reachedmid-log phase, as determined by measuring the culture density as absor~ ce at 600 nm using a UV/vis spectrophotometer (Hewlett-Packard 8452A diode~ array spectrophotometer, Hewlett-Packard, Germany). The culture was grvwn to an optimum opdcal deosity of 0.3-0.7 abs. 1 3pon reaching the optimum density range,; cells were centrifugèd for lO minutes at 10,000 g and 4 C. The supetnatant was then decanted and the cells were washed with 50 mls cold (~ C) diseilled water and centrifuged again for 3 mînutes at 10,000 g and 4 C. The supernatant ~was then aspi~a~ so as not to disturb the~ pellet, which was ;25 ~ ~ suspended in diseilled ~water to a final volume of 500 .ul.
4~ aliquot of the ~above-described cell suspension was ansferred;to a sterile~polyvinylehloride ~PVC) tube and mixed with 2 ~1 of either .: plasmid~:pCL8 ~o~ pCLlOOO.~ ~ ~The mixture ~was then transfer~ed to a cold Bio-Rad Pulser cuvette (Bio-Rad)~wlth a 0.2~ cm gàp and electroporated with a single pulse 1' ~ 94/28154 214 0 5 2 7 PCT/IJS94/06084 using a Bio-Rad Gene Pulser (Bio-Rad) set to 2.~ Kv with 25 ~F capacitance.
The Bio-~ad pulse controller was set at 200 ohms resistance. Immediately after electroporation, 800 ~l of S.O.C., described in Hanahan, J. Mol. Biol., 166:557-580 (1983), was added to the cuvette and the cultures were allowed to incubate S for 30 minutes at 3T C. The cells were then plated on agar containing ampicillin (200 ~g/ml), ~Difco) to seleet for transformants. The plasmids contain an arnpicillin resistance marker. Additional transformation procedures are krown and available to those skille in the art and may be used in the practice of the present invention. Positive trans~ormants were re-designated NS3119 (conta~ning io pCL8) and NS3120 ~containing pCL1000). NS3120 (containing plasmid pCL1000) was deposited with the American Type Culture Collection, 12301 Parklawn Dlive, Rockville, Maryland 20852 on April 27, 1993 nd given ATCC
Accession No. 69291. In Example 2, positive transformants were tested to confirm that the glyoxylate cycle was operating.

In order to confirm that host celhl described in Example 1 were transformed with functional glyoxylate cycle genes, assays were conducted to determine the activity of glyoxylate cycle enzymes.
In preparation for the enzyme assays, cell extracts were prepared by growing cultures of transformed host cells oven~i~ht in M63 medium [Chung, et al., J. Bacteriol., 170:386-~91 ~1~88), inco~porated by reference hereinJ in a rotary/incubator shaker at 37 C and 300 rpm. A 20 ml aliquot of the culture wasansferred into a 50 ml polypropylene centrifuge tube (Corning, (~oming NY~
and centrifuged for 10 minutes at lO,O~g and 4C in a.Beckman S2-21 centn~uge (Beckman). The pellet was resuspended in 4 ml ~ 0.1 M potassium phosphate buffer, pH 7Ø The ce}ls were then disrupted using a French Pressure .
Cell (SLM Instmments, Urbana, II~ a~ 1000 psi according to the manufacturer's instruchorls. The resulting cell debns was removed by centrifuging at 1300 g for .

WO 94/28154 PCT/[1594/06084 t 8 minutes in an MSE microcentrifuge (MSE, Micro Centaur, U.K.). The supernatants (lysates~ obtained were then stored on ice.
The cell lysates described above were then tested for isocitrate lyase and malate synthase activity. The isocitrate Iyase assay was based on the protocol S described in l)ixon et al., Biochem. J., 72:3P (1959) incorporated by reference herein, with 0.1 M potassium phosphate buffer substituted for the t~is buffer described in the assay ~eported in Dixon. The assay is based upon the rate of increase in optical density with increased isocitrate lyase activity as measured by the accumulation of glyoxylic acid phenylhydra~ine.
To conduct the assay, 0.10 ml of 100 mM phenylhydrazine, 0.05 ml of 2û mM cysteine, 0.05 ml of 100 mM MgCl2, and 0.05 ml of the cell lysate described above were placed in a 1 ml quartz cuvette with 0.10 m~ lûO mM
; potassium phosphate buffer in 1.10 ml distilled water at pH7.2.
The optical density~ of the above rnixture was adjusted to zero in a ` ~ ~15 Hewlett-Packard model 84S2A~ diode array spectropho~ometer equipped with a ldnetics software package in a Howlett-Packard operatillg software package A.02Ø The rea~tion whereby~ glyoxylate (an'a ultimately glyoxylic acid phenylhyd~ine) is formed from Isocit~ate was begun by addition to the cuvettè
of 0.05 ml 20 mM pot~ssium isocitrate. Absorbance was read at 3~4 nm and the 20~ ~ amount of isocitrate lyase ~was determined by an increase in optical density.
The~resulb of the isocit~ate lyase assay are shown in Table 1, wherein the specific ac~vity of UIo enzyme is~ express~d in units per miligram of protein, wherein one unit is defined as t~e amount of enzyme ca~alyzing` th~
formation of one rnicromole of glyoxylic acid phenylhydrazone per minute. i As~
~ ~ shown in that Tabl~, isocitrate lyase activity increases significantly in c 11 lysates from~transformants conta~ning the glyoxylate cycle~genes as compared to the samemeasurements~in cell lysates f~m untransf~ed~controls.
An~ assay was~also conducted to detennine malate synthase activity in transformed~and ~untransformed cell ~ly~tés.; That assay was conducted ~ _ ` i~

~ ` ~94/281~4 21~O~27 PCI/US94/0608~

according to the procedures in Dixon et al., Biochem. J., 72 :3p (1959). The malate synthase assay measures the decrease in optical density with an increase in breakdown of the thio-ester bond in acetyl COA which reacts with malate synthase to form malate in the presence of glyoxylate.
S The malate synthase assay was conducted by first mal~ng a stock solution comprising 4.0 ml 0.1 M tris, 0.50 ml 0.002 M acetyl COA and 0.1 M
MgCl2. A 0.4 ml aliquot of the stock solution was combined with 0.01 rnl of the cell lysate in a cuvette. Absorbance was rneasured in a Hewlett-Packard model 8452A UV/vis diode array spectrophotometer at 232 nm for approximately 2 minutes to ensure that no acetyl CoA deacylase, which may prevent acetyl COA
from reacting with malate synthase, was present. Upon achieving a steady baseline absorbance, 0.01 ml of 0.02 M sodium glyoxylate was added to the cuvette. The absorbance was measured every 5 seconds for a total of 200 seconds.
The results of the malate synthase assay are also presented in Table 1, wherein specific activity is measured as units per miligram of protein, wherein one unit is defined as the cleavage of one microma~e acetyl CoA per minute. As shown in that Table, the activity of malate synthase increases in transforrnantscontaining genes of the glyoxylate cycles compared to untransformed controls.
The results of both the malate synthase and isocitrate lyase assays confirm that th~ activity of enzymes of the glyoxylate eycle in transformed hostcells is greater than that in untrans~ormed host ~ells. Example 3 provides ~errnentation and growth conditions which were used to prepare cultures for car~on conversion efficiency measurements.

214d527 `
wo 94/28154 PcTIuss4l060s4 ~ ~
, Table 1 ~ ._ _ ... .. . .. ... _ --_ ~ I _ _ I
Ferrne~tation I
Construction Time (~Irs) M~late Synthase ocitrate Lyase E. coli 13281 : 8 0.066 0.009 Untransformed : : : I
. . _ . , , .
: : ~4 0. 143 0.008 _ 48 0.144 _ _0.054 :
__ . . . .. .. ~ _ ~ ,1 E.~coli 13281 8 ~ 0.731 0.164 Transformed With pCL8 :: ; `
~ _ _ . .I ~ .
;~ :: ` : : : ~ : 24 ~ 0.481 ~ 0.063 I
, . : _ : .-- . . , I
: ~ ~ ~ 48~ : ~0507 0.16 1. ~ ~ ~ ~ ~ ~ 1 : --. ~ ~ . - -: E. coli 13281 ~ 8 ; 1.615 0.428:~ 1 Transformed ~ I ~ ~ : ~ : I :
¦ With pCL1000 ~ ~ :
~ ~ 24~ ~ I.0~6 ~ ~0.198 : ~ ¦
1`~ : ~ ~a l.0l2 0 393 ~
~ .~ . , ~ ~ 1 :
~E.~coli:}3281 ~; ;~ 8 ~ ~ 0~03 : ~ 0.003 ~ ~
:Untr~nsfonned: ~; ~ ~: : ~ I
~ - ~ ::~ , _ _ , ~ ~ , ; : ~ ~ 24~ : ~ 0.098 : 0.0~ :~
~ ~ ~ I _. ~ : ~, ~ , . . . _ ~ ~ ~
, ~i D l~6 ; ; ! _0.0 ~- ~ . ~ : :: :
~ '~ ~ ~ --E~ a ~13281~ ~ 8 ~ ; ~ ; 0.626~ ;: 0.203 ~D~ W~ : ~ -- ;

24~ ~ ~ ~ ; 0~475 ~- 0~194 48~ ~ 0~584~ ~ ~0~185 .
~ ~ 941281S4 PcTrusg4/06084 Table 1 ~(: ont.) . _ _ _-- . ~
S P E CIlFIC A C TI'VIT Y
I , _ ~ _ , ,, . _ .
Fer m entation C ons~ ~ ctionTinne ~H rs) M alate Synthase Isoc;trate Lyase _ . _ -E. coli 13281 8 0.947 0.369 Trans~ornned with p C L10KND _ _ _ . .
24 1.11 0.~61 ~ _ .
48 1.073 0.257 __ __ E. coli 13281 8 O.Og9 0.CX)8 Un~ans~orrned . _ _ I 24 N.D. N.D.
_ _ _ __ _ . . . _ _ _ .
51 0.037 0.02~
. _ , .. . _ . _ _ _ . ,_ _ _ _ _ ~ _ _ E. coli 13281 8 1.119 0.37 Transform ed with p C L10(N0 . _ , ~ ~
~4 0.664 0.~34 i . __ . _ : 51 1.086 0.871 1, _ ~ , :

1 ~
.

In order to determine earbon conversion efficiency, ~ansformed cells were fermented as follows.
Fermentations were conducted using 20 L ferrnento~ ~LSL
S Biolafitte, France). The temperature, pH, agitation, and air flow were kept constant at 32~ C, 7.2, 500 rpm, and 11 L/minutes, respectively. Glucose feed, dissolved oxygen, and other parameters were varied and may be set at values determined by the skilled ar~isan.
The fermentor contained 10 L of K12 medium [Konstantinov et al., J. Fermenta~ion and Bio Engineering, 70(4):253-260 (1990), incorpo~ated by reference herein3 supplemented with 1 g/L yeast extract and was sterilized at 121 C for 30 minutes. The ~rmentor was then inoculated with 1 L of the cell culture containing either the untransformed strains or the glyoxylate cycle transformants. The back pressure of the fermentor was regulated at 10 psi and the dissolved oxygen concent~ation was initially 100% and was allowed to re~ch a value of 15% where it was maintained throughout the fennentation. A
70%(weight per volume) glucose solution was inithlly added until the fermentor contained 3.0 g/L glucose. The glucose was then continuously added into the fermentor to maintain glucose in excess during exponential growth. The final biomass in the fermentor was lintited by tyrosine (E. coli strain ATCC 13281 is a tyrosine auxotroph). The concentration of glucose was then allowed to decreaseto approximately 0.2 g/L and WRS maintained at that concentration for the rest of the fennentation. The total fermentation time was 48 hrs. The concentrations of glucose, L-phenylalanine, prephenic acid, phenylpyruvic acid, ~nd ace~ate were then measured to determine carbon conversion efflciency as taught in Example 4.
, `

, :

:" :

) 94/~8154 2 14 0 ~ 2 7 PCT/US94/û6084 Carbon conversion efficiency relates to the ratio of carbon (glucose or another carbon source) metabolized to the amount of carbon ultimately incorporated in biosynthetic precursors and end products. It is apparent to any skilled artisan that numerous biosynthetic precursors and end products may be used as a basis for the calculation of carbon conversion efficiency. Examples ofsuch compounds are amino acids, especially the aromatic amino acids, nucleotides, various enzymes, and structural proteins. By way of example, the methods of calcula~ing carbon conversion e~ficiency are exemplified herein usingL-phenylalanine and its two immediate precursors as a measure of the incorporation of caroon from the precursor, glucose. The amount of L-phenylalanine, prephenic acid, and phenylpyruvic acid and acetate which accumulated in the fermentor and the amount of glucose consumed were measured in order to calculate carbon eonversion efficiency as provid~d below.
.

A. phenylalanin~Pr~duction The accumulation of L.-phenylalanine was measured using flow-switching High Performance Liquid Chromatography. A flow-switching column allows faster Glearance of phases not containing L-PhenYlalanine through the column. High perforrnance liquid chromatography is well-known }n the art and 20the skilled artisan is aw of the application of numerous such techni~ques to determine the concentration of any biosynthetic product, such a L-phenylalanine.The chromatographic apparatus comprised two Waters {VVaters, Milford, MA3 ~del 600A solvent ~delivery systems, an Applied Biosystems (Poster (: ity, CA~
983 programmable detector, a Waters Model 710B auto sampler and model 680 25automatic 8radient controller, all used according to the manufacturer's in~tructions. A Vici (Valco Instruments, Houston, ~X) lû-part column-switching valve was used according to the manufachlrer's instruction to achieve flow :
::

WO 941:t8154 PCTIUS94106084 reversal and a Fisions Multichrome Data Collection System (Danvers, MA) was used to analyze the data.
A 10 ~11 aliquot of sample obtained from fermentation broth containing E. coli transformed with glyoxylate cycle genes or from broth S containing untransformed E. COI! cells was injected onto a Supelco (Bellefonte, PA) LC-1~ DB column (97.5 mm x 4.~ mm) at a flow rate of 1.5 ml/minutes Detection was at 214 nm and retention time was 4.1 minutes. Extemal standards of 0.1, 0.5, and 1.0 mg/ml were diluted with the HPLC mobile phase, and a standard curve was developed according to procedures krown in the art. The 10 mobile phase for the HPLC runs and for the standards comprised 5 % acetonitrile, 95% 0.005 M pentane sulfonic acid (sodium salt) àt pH 2.5.
The results of HPLC analysis of L-phenylalanine concentraeion in fennentadon broths are presented in Table 2.

B. Phen~,rlpvruvic Acid Prod_rtinn Phenylpyruvic acid is the immediate precursor of phenylalanine in the phenylalanine biosynthetic pathway. Phenylpyruvic acid concentraeion provides, in part, a measur~ of earbon conversion because, at any point in time,not all the carbon from glucose which has been transferred ~ the biosynthesis ofcellular components will be in the form of L-phenylalanine. Some of the carbon 20 will be in the form of the immediate precursors of phenylalanine. Thus, measurement of those precursors in addition to phenylalanine provides a more accurate measure of carbon conversion than would phenylalanine measurements ~on~
Phenylpyruvic acid concentration was measured in samples taken 25 from ~ermentation broths containing either kansformed or untransformed E. coli cells as described above by reverse-phase High Performance I~iquid :
Chromatography at a detection wavelength of 214 nm, 0.2 AUFS (Absorbance Units Full Scale~ using a Whatman RAC II Partisil S ODS-3 10 cm column :

`

) 94l~8154 2 lL 4 0 5 ~ 7 PCT/US94106084 (Whatman, Hillsboro, OR) according to the manu~acturer's instructions. An isocratic method was used, with a mobile phase comprising 10 mls of 1% 0.5 M
pentane sulfonic acid solution, 950 ml water, and 40 ml acetonitlile, pH.2.5. A
volume of 10 ~l was injected onto the column at a flow rate of 2 ml/minutes and S the run time was 10 minutes Phenylpyruvic aeid standards were also mn at concentrations of 0.0~, 0.08, and 0.4 mg/ml.
The results are shown in Table 2, wherein phenylpymvic acid concentration is shown as grams/liter for the~ ious samples, including untransformed controls.

C. ephenic Acid Production Prephenic acid is the immediate precursor of phenylpyruvic acid in the pathway leading to synthesis of phenylalanine. The concentration of prephenic acid was determined by High Performance Liquid Chromatography at a detection wavelength of 220 nm, 0.2 AUFS, using a Supelco LC10 column ~Supelco). An isocratic method was used, with the mobile phase comprising 10% acetonitnle in 0.1 M ~ammonium phosphate buffer, pH 6.7 withO.005 M tetrabutylammonium phosphate. A 10 ful volume of each sample was injected onto the column at a flow rate of 1.5 ml/minutes. Equivalent-size standards were also run at concentrations of 0.02, 0.15, 0.2 mg/ml. ;~
The results ~e shown in Table 2, wherein prephenic acid is labelled PA.

D. Iucose ~oneentration ~ ~
The amount of glucose concentr~tion of the ~eed stoclc was measured using High Perfo~rnance Liquid Chromatograph equipped with a refractive index detector and a column heater. A 400 mg sample was obtained from the glucose ~eing ~d mto the fermentor and diluted to 100 ml. A Bior~d HPX-878 column equilibrat~d to 85 C was used with a mobile phase compIising .

.

WO ~4/Z8154 21 ~ O 5 2 7 PCT/US94/06084 Milli-Q water (Millipore, Bedford, MA). The flow rate was 0.6 ml/minutes and the run time was 20 minutes. A 20 ~I volume of sample was injected on the column at tirne zero. The column was equilibrated for 30 minutes until a stable base line was reached. External standards were also run using Glucose obtained S from Sigma Chemical ~G-5500, Sigma St. Louis, MO).
The results are presented in Table 2 and are used in the calculation of carbon conversion efficiency below.

E. Acetate oduction During the fermentation process glucose may be broken down to }O form aeetate. While such glucose will be detected as being consumed, it should not properly be incorporated into calculations of carbon efficiency according tothe presently-claimed methods. The reason for ~his is that any glucose used ~o produce acetate is not available to directly produce biosynthetic products.
There~ore, acetate production should be calculated in order to form a basis for subtracting glucose which is dive~ted from pathways generating biosynthetic products and intermediates as noted above.
Ion chromatography was us~d to detect acetate in fermentation broths according to the pr~ure set for~h in the Dionex (Sunnvale, CA) Ion Chromatography Cookbook, Issue I, p. II-16. A 50 ~il aliquot of sample was injected on an HPIcE-ASl column (Dionex) with a flow rate of 0.8 ml/min. The eluant was 1.0 mM octane-sulfonic acid in 2% 2-propanol. The regeneran~ was 5 mM tetrabutyl ammonillm hydroxide. Standards were also run at concentrations of 0.02, 0.1, 0.2 mg/ml.
Results are included in Table 2 and used in the exemplified 2~ calculations below.

.

:

2 i 4 o ~ 2 7 r ~ 94/;!81~4 - 17 - PCT/US94/06084 r e. ~ _ _ _ =_ .~ _ . o~
_ __ _ _ __ ~ c~l ~
___ _ _ _ _ . .
~ ~Q ~
~ ~i cr~
__. __ _ , , __ ~_ ~: ~ .
o' _ c~ ` ~ ~o ~, v~

~a ~ _ _ _ , __ ~:~ ~ ~ o ~ o . ~o ~ o ~ ~
E~ , _ _ _ . _ _ . , . ~ o ~ c~ 8 ~ o o o o o o _ o o __ _ . _ . ._ , ~ 8 ~ ~ 8 o~ ~ o ~
~ o ~ r_ o ~, ~ ~ ~, ---~ ~ _ _ _ __~

~ oo ~i q W ~ ~ C~ C~
~3 :
_ ~_ ~ _ __ _ __. _ _ ,_ _ _ _ _ ~ _ _ _ __ _ ~ ~ a 3 ~ ~L
::: :

: :

WO 94/~15'l 21 4 0 5 2 7 18 - PCT/US94/06084 ~
I ~i ~ T 1~ 1 ~
I ~- ~
. .............. . .
~ ~ o ~
~ I ~ r ~
~ Z ~+ O ~ O ~ O ~ ~ O ;

V ~ ~ ~ o ~ ~ d ~. O .
...
. ~ o o t~ o ~ ~ 8 ..
. t~ o ~ x _= ~ x o 1~ ~ 1 I ~ ~
oo ~ ~ ~o ~ ~ eo .

-o~ _ __ __ _ _ _, 1 ~ O ~i . _ , _ ~ _ . _ _ __ _ ~ ' L S
~'.

21~27 94/2815~ 19 PCT/US94/06084 --- = _ = _ ~ ~

_ - - _ _____ _ _ C~`~ o~ C~ .

:- _ _ .. _ . . , ~ ~ .
~ ~ ~ ~ ~ g C~ ~ $ U~ o 11~ L~
cli~ _ o v~ 8 r _ _ o l o E~~ ~i ~ g ~i o , g o ~
_ .__ _ _ ~ . __ _ 8 ~ ~ 8 ~t ~o ~S ~, ~o o ~ oo o ~ W

- 3~ ____ _ . _ _ rl ~ x C~l ~n , ~ ~
. _ _ ______ _ '1 ' ; 1 ~ I ~ ~ : l O ` .

~ l~
:: :
:

21~0527 I`

Fo Calculation Of Carboll ~onYersion Efficiencv Carbon conversion efficiency of transformed and untransformed cells was calculated as the total combined grams of L-phenylalanine, phenylpyruvic acid, and prephenic acid produced at 48 hrs. of fe~nentation S divided by the total grams of glucose consumed dunng that fenTentation. Ifacetate accumulates duIing the ~ermentation, carbon conversion efficiency may becalculated by subtracting 180.16 g (I mole) of glucose ~or every 121 g (2 moles~of acetate produced and then calculating as indicated above. The carbon conversion efficiencies for various batches of transform~ and untransformed hostcells are shown in Table 2. As is evident in that Table, earbon conversion efficiency significantly increased in cells which had been transformed`with glyoxylate cycle genes.
The results ob~ained in one fermen~ation run are used herein to demonstrate the calculation s)f carbon conversion efficiency according to the invention. Specifically, the results obtained in fermentation run 4 in Table 2, wherein E. coli cells had been transformed with plasmid pCL1000 as described above. As noted in Table 2, the total phenyl~anine ~6.11 g) prephenic acid (15.55 g), and ph~nylpyruvic acid (1.81) was 23.47 g/L in a fermentation volume of 13.11 L. Additionally, 2.86 g/L acetate were produced and 2,844.18 grams of glucose wcre consumed during fermentation.
Tbe above da~a are used to calculate carbon conversion ef~lciency, wherein the grams of glucose which were used to produce acetate in the fermenta~on are calculated as follows:

h~olo~ C~C~I~O .Dlod. durln~ r~c tlor~ os ~C~C~ or~t~ L~

(mole9 ~c~tae~ prod~ ol~ ) G~ms of gl~ose used to prad. flceeate = 2 .

~ 2140527` ^
(. .`~ 94/28l54 PCT/US94/06084 Carbon conversion efficiency may then be calculated as follows using the above data: .

cg = I t,q~r ~ph~ ~ gjL~PPA) ~ ~/I,(PA) ] x f~rm~ntat~on volum~ ~r-) ~ x 100 ~ ~glucose con6urned) - g~gluco6e used toprod. acet~te) To calculate carbon conversion eff`iciency from the foregoing exemplary data, 18û.16g ~1 mole~ of glucose is subtracted for every 121 g (2 S moles) acetate produced to give the total amount of glucose actually consumed.
Doing so reveals that 63.11 g of glucose were used to produce acetate, leavin~g 2,781.07 g of glucose consumed. Thus, carbon efficiency expressed as a percentage is 11.04, the total grams of phenylalanine, phenylpyruvic acidiand prephenic acid divided~ by the total glucose consumed.

11 04 = (6.11 + 15.55 + 1.81) x 13.11 X 100 - 2844 . 18 - 63 . 11 Host cells which had b~n transi~nned with genes encoding enzymes of the glyoxylate cycle utiliz~d carbon in a significantly more efficient ~:` manner as compared to untransformed controls. This effect is likçly due to the fact that transformed cells utilized :the glyoxylate cycle to bypass the CO2-evolving , steps of the TCA cycle, thus malang avaîlable greater amounts of carbon for IS incorporation into biosynthetic products. T he greater: efficiency of carbon utilization further results ~in increases in the synthesis of ~hose products.
While the present invention~ has~ been characterized in terms of a preferred embodiment thereof, it is readily apparent to the skilled artisan thatm~reased c rbon conversion efficiency may aiso~be obtained through use of host 20~ cells and plasmids~other than those presently:disclosed.;~: For example9 a host cell:
or plasmid com~ising~ ~he phe~ gene reported~ in~ lJ.S. Pat~nt No. 5,120? 837 isexpect~ed to produce~gT~ter quantities of L-phenylalanine when such a host cell:or plasmid is~used in me~hods according to the present inven~ion than would host cells of plasmids without the pheA gene. In addition, it is apparent to the skil1ed artisan that the use of speci~lc promoters and other constituents of transcription and translation, such as, e.g., a tac promoter [Russell, et al., Gene, 20:231-243 (1982)], may improve upon the carbon conversion efficiency, biosynthetic prodùctyield, or biomass resulting from practice of the presently-claimed invention.
Mutations in the glyoxylate genes may also be used to obtain increases in carboncom~ersion efficiency. Por example, a muta~ed aceK gene, which possesses primarily Icinase activity, may be used in methods according to the invention.
The foregoing parameters are known in the art and may be varied according to the use to which the presently-claimed methods are put.
While the present invention has been described in tenns of its preferred embodiments, the skilled artisan ~ealizes that numerous modifications may be made. For example, there are numerous microorganisms which may serve as host cells of the invention and numerous means of incorporating glyoxylate cycle genes into a host cell. Thus, the present invention should onlybe limited by the scope of the appended claims.
d~

!

:~

Claims (8)

? 94/28154 PCT/US94/06084 We claim:

1. A method for increasing the carbon conversion efficiency of a microorganism, comprising the steps of:
(a) incorporating in a microorganism, said microorganism having a functional glyoxylate operon, DNA encoding isocitrate lyase, malate synthase, and isocitrate dehydrogenase kinase/phosphatase;
(b) fermenting said microorganism in a suitable fermentation medium containing a carbon compound capable of entry into a tricarboxylic acid cycle; and (c) selecting one or more biosynthetic product(s) from said fermentation medium.

2. The method according to claim 1, wherein the DNA encoding isocitrate lyase, malate synthase, and isocitrate dehydrogenase kinase/phosphatase comprises a portion of the glyoxylate operon.

3. The method according to claim 1, wherein said microorganism is a bacterial cell.

4. The method according to claim 3, wherein said bacterial cell is an Escherichia coli cell.

5. The method of claim 1, wherein said incorporating step comprises transformation of said microorganism with a vector comprising said DNA encoding isocitrate lyase, malate synthase, and isocitrate dehydrogenase kinase/phosphatase.

? 94/28154 PCT/US94/06084

6. The method according to claim 1, wherein said DNA encoding isocitrate dehydrogenase kinase/phosphatase is mutated to encode an isocitrate dehydrogenase kinase/phosphatase which possesses primarily kinase activity.

7. A method for producing an amino acid comprising the steps of:
(a) incorporating in a microorganism, said microorganisms having a functional glyoxylate operon, DNA encoding isocitrate lyase, malate synthase, and isocitrate dehydrogenase;
(b) fermenting said microorganism in a suitable fermentation medium containing a carbon compound capable of entry into a tricarboxylic acid cycle; and (c) isolating the amino acid from said fermentation medium.

8. The method according to claim 7, wherein said amino acid is selected from the group consisting of phenylalanine, tyrosine, and tryptophan.