APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Jan. 1976, p. 91-98Copyright © 1976 American Society for Microbiology

Vol. 31, No. 1Printed in U.S.A.

Growth of Enterobacter aerogenes in a Chemostat withDouble Nutrient Limitations

CHARLES L. COONEY,* DANIEL I. C. WANG, AND RICHARD I. MATELES'

Department of Nutrition and Food Science, Massachusetts Institute of Technology,Cambridge, Massachusetts 02139

Received for publication 5 June 1975

The behavior of Enterobacter aerogenes during growth in chemostats limitedby single and double nutrient restrictions was examined. On the assumptionthat different essential nutrients act to limit growth in different ways, we se-

lected pairs of nutrients likely to affect different aspects of metabolism. Resultsshow that macromolecular cell composition can be controlled by using more

than one nutrient restriction. The polysaccharide content of the cells is readilymanipulated by the ratio of carbon to nitrogen in the inlet nutrients. Also, atlow dilution rates, ratios of protein to ribonucleic acid are dependent on the ratioof phosphate to nitrogen in the input nutrients. An examination of both aceticacid and metabolite production (as measured by ultraviolet absorbance of culturefiltrates) showed that accumulation of these products was dependent on bothdilution rate and type of nutrient limitation(s). These results were examined interms of the problems of translation of batch to continuous culture processesand the use of selected nutrient limitations to control noncellular product forma-tion.

The cellular composition and biochemical ac-tivities of microorganisms are functions of thegrowth environment. Investigations (2, 9) haveshown that in continuous culture the macro-molecular composition of Enterobacter aero-genes and other organisms varies with the dilu-tion rate. The various intracellular metabolitepools also change with growth rate (1). Thesephysiological and biochemical properties aredependent on the pH, temperature, and ioniccomposition of the environment. Consequently,manipulating the environment of the cell givesthe investigator considerable control over themetabolic activities of growing and restingmicroorganisms.Many studies have been conducted to eluci-

date the relationships between growth environ-ment and cellular physiology in continuousculture (10), which provides a tool for main-taining time-invariant or steady-state growthin a constant, controllable environment. Inprevious studies using the chemostat form ofcontinuous culture (6), growth could be limitedby restricting any single essential nutrient. Byexamining possible limiting nutrients in aminimally defined medium, a hypothesis hasbeen formulated concerning the mode of limita-tion exerted by each nutrient (Table 1).

Previous work (2, 8) has established the1 Present address: Department of Applied Microbiology,

Hebrew University, Jerusalem, Israel.

degree to which one can control cell composi-tion under the influence of single nutrientlimitations. The results of these studies haveraised the question of whether it is possible togrow cells under multiple nutrient limitationsto achieve even greater control over cellularprocesses. The concept of a multiple nutrientlimitation implies that the growth and functionof a cell are regulated, or otherwise affected, bythe limited availability of more than onenutrient. It is likely that both cell compositionand the ability of cells to synthesize noncellularmetabolic products depend on the availabilityof more than one nutrient. Our hypothesizedmodes of action for various nutrient limitationssuggested that multiple nutrient limitationswere possible. To explore this question experi-mentally, we examined the growth of E. aero-genes NCTC 418 in a chemostat with variableratios of carbon to nitrogen and nitrogen tophosphate in the inlet medium. Biochemicalactivities of the culture were assessed by meas-uring macromolecular cell composition, excre-tion of metabolic products, and residual levelsof essential nutrients.

MATERIALS AND METHODS

Organism and growth media. E. aerogenes NCTC418 was obtained from D. W. Tempest and main-tained by monthly transfers on tryptone glucoseextract agar (Difco) slants. The organism was grown

91

on March 24, 2020 by guest

http://aem.asm

.org/D

ownloaded from

92 COONEY, WANG, AND MATELES

at 37 C and pH 6.5. Table 2 gives the composition ofthe growth media for carbon-limited and/or nitro-gen-limited and nitrogen-limited and/or phosphate-limited chemostats.

Analytical procedures. The biuret reaction (11)was used to measure cell protein; bovine serumalbumin (Pentex) provided a standard. The totalcarbohydrate content of the cells was measured bythe anthrone reaction (3), using glucose as a stand-ard. Cellular ribonucleic acid (RNA) was extractedby the Schmidt-Tannhauser procedure as describedby Munro and Fleck (8). Absorbance of the extractwas measured at 260 nm, and the RNA was quanti-tated with an extinction coefficient of 32 liters/g percm (4). RNA from E. coli (Calbiochem) provided astandard.

Glucose in the culture supernatant was measuredwith glucose oxidase (Worthington Biochemicals).Ammonia was measured by the alkaline-phenol

TABLE 1. Nutrient limitations and their possiblemodes of action

Nutrient limita- Mode of actiontion

Carbon energy Decreases supply of carbon forbiosynthesis or energy

Nitrogen or sul- Inhibits protein synthesisfur

Phosphate Retards nucleic acid synthesisand/or energy production

Magnesium or Reduces nucleic acid synthesispotassium or cell wall and/or mem-

brane structure or permeabil-ity

TABLE 2. Growth media for E. aerogenes in carbon-,nitrogen-, and phosphate-limited chemostats

Concn (ghliter)a

Component Carbon- and/or ni- Nitrogen- and/or phos-trogen-limitedche- phate-limited chemo-

mostatb state

Glucosed 0.5-1.0 1.0(NH4)2SO4d 0.16 0.16Na2HPO4d 0.71 0.026-0.0516KH2PO4 6.1K2SO4 0.05

a Each medium was supplemented at 10 ml/literwith a trace salts solution that contained (g/liter):MgSO4, 6.0; CaC12-2H20, 0.015; FeSO4-7H20, 0.028;ZnSO4-7H20, 0.140; CUSO4-5H20, 0.025; Na2MoO4-2H20, 0.024; CaC12-6H20, 0.024; MnSO4-H20, 0.084.

b The pH in the medium reservoir was 5.2 (toprevent precipitation of salts) and was adjusted to6.5 during the continuous culture operation by theaddition of 1.0 M NaOH.

c The pH was 3;6, except during continuous cul-ture operation, when it was adjusted to 6.5.

d Exact input concentration of the limiting nutri-ent in each steady state is given in Results.

APPL. ENVIRON. MICROBIOL.

reaction (7) on an autoanalyzer (Technicon). Acetatein the supernatant was determined by the gas chro-matography procedure described by P. S. Masure-kar (MS thesis, Massachusetts Institute of Tech-nology, Cambridge, 1968). Phosphate was measuredby the procedure of Sumner (12). Ultraviolet-absorb-ing material in the culture supernatant was quan-titated by light absorption at 260 nm, and the valuesobtained were divided by the cell dry weight to nor-malize differences in cell concentrations.The results of all the analyses of the cells and

culture supernatants from steady-state culturesrepresent the average values of at least two differ-ent cell samples collected on different days, exceptwhere noted. Steady states were initially estab-lished on the basis of constant culture turbidity andlater verified by determinations of the compositionof the cell-free supernatant and cell dry weights.

Experimental apparatus. The continuous cultureapparatus used in this work was described in detailby C. L. Cooney (Ph.D. thesis, Massachusetts Insti-tute of Technology, Cambridge, 1970). The 1-litervessel with four Teflon baffles contained 400 ml ofmedium, which was agitated by a magnetic stirrer.Sterile air was sparged directly over the stirringbar at 1 liter/min. The pH control system incorpo-rated an Ingold combination pH electrode (Chem-apec, Inc.) in conjunction with a Leeds and North-rop pH controller, which actuated a pump thatadded 1.0 M NaOH on demand. Medium waspumped from a 16-liter polypropylene carboy witha variable-speed Cole-Parmer Masterflex pump,and an overflow tube maintained a constant volumein the vessel. The entire vessel was set in a waterbath controlled at 37 C.

RESULTS

Cell macromolecular composition. The mac-romolecular composition of E. aerogenes NCTC418 was measured in chemostats limited singlyby carbon, nitrogen, or phosphate and simul-taneously by nitrogen and carbon or nitrogenand phosphate (Tables 3 and 4).

For each limiting nutrient tested, the RNAcontent increased with increasing dilution rate(Fig. 1). The protein content also increasedwith the dilution rate, but to a much lesserdegree than the RNA; therefore, the ratio ofprotein to RNA decreased as the dilution rateincreased (Fig. 2). Interestingly, at a dilutionrate of 0.25/h, the ratio of protein to RNA wasnearly identical for cells grown under nitrogenor carbon limitations. However, cells in a phos-phate-limited environment exhibited a markedincrease (about 40%) in this ratio. This disparitylessened as the dilution rate increased. The dif-ference in the ratios of protein to RNA for thevarious nutrient limitations was statisticallysignificant at the low dilution rate but not atthe higher dilution rate.

Although these implications will be consid-

on March 24, 2020 by guest

http://aem.asm

.org/D

ownloaded from

DOUBLE NUTRIENT-LIMITED CHEMOSTAT 93TABLE 3. Macromolecular composition ofE.

aerogenes grown in carbon- andlor nitrogen-limitedchemostats

Limiting nutrient Dilution tein RNA Carbohy-rate (1/h) (%M) drate (%

Carbona 0.22 69 13 5.00.45 63 18 6.50.77b 72 25 8.7

Nitrogenc (am- 0.24 53 11 19.1monia) 0.50 52 17 9.7

0.75 58 21 9.5Duald (nitrogen 0.27 60 12 9.7and carbon) 0.45 60 17 7.0

0.78b 68 21 6.8

a Inlet concentrations (mg/liter): glucose, 500; am-monia, 40.

b Values were obtained for cultures at only onesteady state.

c Inlet concentrations (mg/liter): ammonia, 40;glucose 1,000.

d Inlet concentrations (mg/liter): ammonia, 40;glucose, 640.

TABLE 4. Macromolecular composition ofE.aerogenes grown in phosphate- andlor nitrogen-

limited chemostatsa

Dilu- Pro- RNA Carbohy-Limiting nutrient tion tein R%) drater%rate (1/h) (%

Nitrogenb (am- 0.25 53 9.8 25monia) 0.5 58 18 13

0.74 60 20 7.2Phosphatec 0.25 72 10 17

0.5 67 17 6.30.70 56 19 5.7

Duald (nitrogen 0.25 53 9.7 28and phos- 0.5 61 16 15phate) 0.75 71 19 9.1

a Inlet concentration (mg/liter): glucose, 1,000.b Inlet concentrations (mg/liter): ammonia, 43;

phosphate, 32.1.c Inlet concentrations (mg/liter): phosphate, 23;

ammonia, 125.d Inlet concentrations (mg/liter): ammonia 47;

phosphate, 21.

ered later, it is important to note here that thecell composition must be evaluated not only interms of absolute concentrations but also interms of pertinent concentration ratios. A com-parison of the composition of cells from glucose-and ammonia-limited cultures showed consid-erable variation in the absolute concentrationsof protein, RNA, and total carbohydrate (Table3). However, the ratio of protein to RNA in thecells remained relatively constant when thevalues were compared for cells grown at thesame dilution rate (Fig. 2). This comparison is

24.0

22.0 -

I-

z

EJ'a.

C)

44

z

4:

-J

J

U.)

20.0 _

18.0 _

16.0 _

14.0 _

12.0 _

10.0 _

8.0 _

0.2 0.3 0.4 0.5 0.6 0.7 0.8DILUTION RATE (hr'l )

FIG. 1. Cellular content of RNA in E. aerogenesas a function of the dilution rate in single and dualnutrient-limited chemostats. Symbols: O, carbonlimited; 0, phosphate limited; A, nitrogen limited;0, nitrogen and phosphate (32.1 mglliter) limited;+, nitrogen and phosphate (23 mglliter) limited.

especially important in interpreting the valuesattained from cells grown at low dilution rates,where the total carbohydrate fraction of the cellmay be inflated by accumulated material suchas glycogen (5).

Although the ratio of protein to RNA de-creased with increasing dilution rate, theproduct of this ratio increased (Fig. 3). Theproduct is proportional to the amount of proteinformed per unit mass of RNA in the cell perunit time and provides an estimate of the effi-ciency of protein synthesis (13, 15). The meas-urement is only approximate, however, be-cause total RNA concentration is used in thiscalculation, not just ribosomal RNA (13).

Table 3 shows the macromolecular composi-tion of cells grown under a simultaneous limita-tion of glucose and nitrogen and under a limita-tion of each individually. To achieve the doublenutrient limitation, the chemostat was firstoperated under a carbon limitation, with inputglucose and ammonia initially set at 500 mg/liter and 40 mg/liter, respectively. Efflue itammonia concentration under these conditioaswas approximately 6 mg/liter (Table 5). T.is

26.01 I I

VOL. 31, 1976

on March 24, 2020 by guest

http://aem.asm

.org/D

ownloaded from

94 COONEY, WANG, AND MATELES

9.01-

8.0 1-

7.01-

6.01-4z:

za;'I.-0

a-.

5.01F4.01

3.0

2.0

1.01p

0.2 0.3 0.4 0.5 0.6 0.7 0.8

DILUTION RATE (hr'1)FIG. 2. Ratios ofprotein to RNA in E. aerogenes

as a function of dilution rate in single and dualnutrient-limited chemostats. Symbols: O, phosphatelimited; A, nitrogen limited; O, carbon limited; 0,

nitrogen and phosphate (32.1 mglliter) limited; +,nitrogen and phosphate (23 mg/liter) limited; *, ni-trogen and carbon limited.

concentration is at least 60 times the K8 valuefor ammonia; thus the nitrogen source greatlyexceeded cell requirements. Consequently, es-

timates of the macromolecular composition ofcells taken from this system represent valuesfor bacteria grown under carbon-limited condi-tions. The glucose input was increased to 640mg/liter, whereas the ammonia input remainedconstant at 40 mg/liter. The incremental in-crease of 140 mg/liter in the input glucose con-

centration was determined by calculating theamount of carbon source that the cells wouldrequire to incorporate the remaining 6 mg ofextracellular ammonia per liter into cell mass.Glucose yield was calculated to be 0.5 g of cell/gof glucose, and nitrogen yield was assumed tobe 10 g of cell/g of ammonia. This second set ofnutrient conditions was designed to exert si-multaneous nitrogen and glucose limitations.The conclusion that the culture was nitrogen-as well as carbon-limited was based on thefollowing observations: (i) both the effluentglucose and ammonia concentrations were very

low (Table 5); (ii) a further increase in glucose

2.0

0.5__

z

z 1.0w

0a.

0.5

0.2 0.3 0.4 0.5 0.6 0.7 0.8

DILUTION RATE (hr-1)FIG. 3. Efficiency ofprotein synthesis in E. aero-

genes in single and dual nutrient-limited chemo-stats. Symbols: E, phosphate limited; E, nitrogenand phosphate (32.1 mg/liter) limited; A, nitrogenand phosphate (23 mg/liter) limited; *, nitrogen andcarbon limited; x, nitrogen limited.

TABLE 5. Effluent concentration of essentialnutrients in cultures of E. aerogenes in carbon-

andlor nitrogen-limited chemostats

Limiting nutri- Dilution Glucose Ammoniaent rate (1/h) (mg/liter) (mg/liter)

Carbona 0.22 0.3 6.00.45 0.8 6.10.77 3.0 11.1

Nitrogenb (ammo- 0.24 0.8 0.1nia) 0.50 0.1

0.75 0.2Dualc (carbon and 0.27 0.8 0.1

nitrogen) 0.45 1.0 0.10.78

a Inlet concentrations (mg/liter): glucose, 500; am-monia, 40.

" Inlet concentrations (mg/liter): ammonia, 40;glucose, 1,000.

c Inlet concentrations (mg/liter): ammonia, 40;glucose, 640.

did not proportionately increase cell mass; (iii)total cell carbohydrate at D = 0.27/h washigher than that found under glucose-limitingconditions alone (Table 3) (these cells accumu-late glycogen when starved for nitrogen in the

+

APPL. ENVIRON. MICROBIOL.

on March 24, 2020 by guest

http://aem.asm

.org/D

ownloaded from

DOUBLE NUTRIENT-LIMITED CHEMOSTAT

presence of excess carbon source [5]); and (iv)the cells excreted about the same amount ofacetate as those in a strictly carbon-limitedculture (Fig. 4).

After this series of steady states, the glucoseinput was increased to 1,000 mg/liter to createa nitrogen-limited chemostat. The result was

an increase in carbohydrate at low dilutionrates (Table 3) and a further increase inacetate excretion (Fig. 4). When cellular pro-tein and RNA were compared at a constantdilution rate for increasing concentration ratiosof glucose to ammonia in the inflowing me-dium, the absolute concentration of both cellcomponents generally tended to decrease. Thisphenomenon was most noticeable at low growthrates. However, the ratio of protein to RNAwas relatively constant (Fig. 2) when com-

pared at a given dilution rate.The procedure used to examine the influence

of phosphate on the composition of cells from anitrogen-limited chemostat (Table 4) was simi-lar to the one used to study cells from carbon-limited and/or nitrogen-limited systems. Table4 lists values for cells from an ammonia-limited chemostat that have a low, but not

0.4-

E 0.3

-.I.

z 0.20

0.2

0

IL

0..I

0.2 0.3 0.4 0.5 0.6 0.7 0.8

DILUTION RATE ( hr-1)FIG. 4. Acetic acid excretion by E. aerogenes

grown in single and dual nutrient-limited chemo-stats. Symbols: *, nitrogen limited; +, nitrogenand phosphate (23 mglliter) limited; 0, nitrogenand phosphate (32.1 mglliter) limited; A, phosphatelimited; O, nitrogen and carbon limited; O, carbonlimited.

limiting, concentration of phosphate. In thiscase, the effluent phosphate concentrationvaried between 8.4 and 11.1 mg/liter, depend-ing on the dilution rate (Table 6). Theseeffluent phosphate concentrations are at least10 times the K8 value. The ammonia-limitedcultures (cell compositions presented in Tables 3and 4) were grown under slightly different con-ditions. In the first case, the inlet phosphatewas greater than 3,000 mg/liter, and in thesecond it was 32.1 mg/liter. Despite this differ-ence in medium compositions, the values arenot strikingly different.The inlet phosphate concentration was low-

ered to 21 mg/liter, whereas the inlet ammoniaconcentration remained constant (Table 4).Macromolecular compositions were similar forcells from ammonia-limited cultures with ex-cessive or restricted phosphate. A slightly in-creased ratio of protein to RNA was noted atdilution rates of 0.5 and 0.75/h for cells fromcultures limited by ammonia and restricted byphosphate compared with values for cells fromcultures limited by ammonia alone.When the ammonia inlet concentration was

increased to permit phosphate limitation only,the ratio of protein to RNA in the cells in-creased dramatically at D = 0.25/h. At higherdilution rates, the ratio returned to valuessimilar to those observed under ammonia limi-tation.The evidence for a dual limitation of ammo-

nia and phosphate includes the following: (i)the effluent concentrations of both nutrientswere at limiting values (Table 6); and (ii) the

TABLE 6. Effluent concentrations of essentialnutrients in cultures ofE. aerogenes in nitrogen-

and/or phosphate-limited chemostatsa

Dilu- Ammo- Phos- Glu-Limiting nutrient tion nia phate cose

rate (1/h) (mg/Iiter) (ite) (imteg)

Nitrogenb (am- 0.25 0.1 9.6 0.8monia) 0.50 0.1 8.4 0.8

0.74 0.2 11.1Phosphatec 0.26 67 0.6 1.4

0.52 79 1.2 1.50.70 90 0.6 9.9

Duald (nitrogen 0.25 0.1 0.54 0.6and phos- 0.50 0.2 0.6 0.8phate) 0.75 0.2 0.33a Inlet concentration (mg/liter): glucose, 1,000.b Inlet concentrations (mg/liter): ammonia, 43;

phosphate 32.1.c Inlet concentrations (mg/liter): phosphate, 23;

ammonia, 125.d Inlet concentrations (mg/liter): ammonia, 47;

phosphate, 21.

VOL. 31, 1976 95

on March 24, 2020 by guest

http://aem.asm

.org/D

ownloaded from

96 COONEY, WANG, AND MATELES

patterns of acetate excretion at D = 0.25 and0.5/h fell between the values for a singleammonia or phosphate limitation.Formation of products under different nu-

trient limitations. (i) Essential nutrients. Ef-fluent concentrations of several essential nutri-ents, including the limiting nutrient(s), weremeasured at each steady state. Table 5 givesresults for assays of culture supernatantsunder carbon and/or nitrogen limitations, andTable 6 gives results for those under nitrogenand/or phosphate limitations. The quantity ofeffluent glucose is not shown for those samplesin which there was interference with the glui-cose oxidase procedure. When interference oc-curred, samples tested for glucose gave apurple-gray product instead of the normalbrown-orange of the oxidized chromogen (3,3'-dimethyoxybenzidine). The degree of interfer-ence was a function of both dilution rate andlimiting substrate, and the reaction generallyoccurred at D = 0.75/h for all nitrogen-limitedconditions.The effluent glucose concentration under

carbon limitation (Table 5) fits the Monodgrowth kinetic model quite well and yields aK8 value for glucose of 1.0 mg/liter. For allgrowth conditions, effluent glucose approacheda very low value near its Ks concentration. Theeffluent ammonia concentration under nitro-gen limitation was so low that it was difficult tomeasure. The alkaline-phenol procedure usedin conjunction with a Technicon autoanalyzerhas a lower sensitivity limit of about 0.1 gg/ml.

Unlike glucose, neither ammonia nor phos-phate were completely removed from the me-dium when they were not the primary or sec-ondary limiting nutrient.

(ii) Catabolite excretion. The productionof catabolic intermediates from glucose indi-cated inefficient utilization of the carbon-energy source. Acetic acid excretion variedmarkedly with changes in the nutrient compo-sition of the media during growth (Fig. 4). At agiven dilution rate acetic acid excretion wasminimal in carbon-limited cultures. Undernitrogen limitation, the amount of acetic acidexcreted was proportional to the amount ofexcess glucose available to the cell (comparethe slopes of the curves for nitrogen-limitedcultures with those for nitrogen- and carbon-limited cultures [Fig. 4]). Furthermore, theinlet concentration of phosphate affected ace-tate production (Fig. 4). Thus, the availabilityof at least three essential nutrients (glucose,ammonia, and phosphate) affected the produc-tion of acetic acid.

(iii) Metabolite excretion. The third class ofcompounds examined in the extracellular en-

APPL. ENVIRON. MICROBIOL.

vironment was metabolites that did not appearto be direct products of glucose catabolism.These materials were characterized by theirlight absorption at 260 nm; the amounts ex-creted into the medium varied over a fivefoldrange under the growth conditions investigatedbut were greatest in nitrogen-limited cultures(Fig. 5). Although the relative positions changewith dilution rate, excretion generally de-creased when the quantity of carbon or phos-phate was lowered to make it a second limit-ing nutrient (Fig. 5). A double limitation ofnitrogen and phosphate suppressed excretionmore than a limitation of phosphate alone, anda carbon shortage produced the greatest sup-pression.Much of the excreted material may have

been nucleotides and/or nucleosides, judgingfrom the positive orcinol reaction obtained.Since orcinol reacts not only with ribose, butalso to a limited extent with hexoses, and pos-sibly other interfering materials, the presenceof nucleic acid components was not unequivo-cally established by this reaction. Anion-ex-

-4.0cc~~~~~~~~~-

Cd //

otX& Iw~~~~~

0

IL0_

0.2 0.3 0.4 0.5 0.6 0.7 0.8DILUTION RATE (hr'1)

FIG. 5. Excretion of ultraviolet-absorbing (ab-sorbance at 260 nm [A260) material by E. aerogenesgrown in single and dual nutrient-limited chemo-stats. Symbols: 0, nitrogen limited; x, nitrogen andphosphate (23 mg/liter) limited; 0, nitrogen andphosphate (32.1 mglliter) limited; A, phosphatelimited; 0, nitrogen and carbon limited; O, carbonlimited.

on March 24, 2020 by guest

http://aem.asm

.org/D

ownloaded from

DOUBLE NUTRIENT-LIMITED CHEMOSTAT

change chromatography of culture superna-tant fluids showed four to five major com-ponents in the medium from phosphate- orammonia-limited cultures, but these did notcorrespond with any of the common nucleo-tides (C. L. Cooney, Ph.D. thesis). Thus, theidentity of the extracellular, ultraviolet-ab-sorbing products remains unknown. Interest-ingly, ion-exchange chromatograms showedthat growth of the bacteria under nitrogenand phosphate limitations led to excretion ofdifferent products.

DISCUSSIONWhen carbon or nitrogen alone limited

growth, the ratio of protein to RNA was afunction of only the dilution rate (and not ofthe ratio of input carbon to nitrogen); the ratioof protein to RNA decreased as the dilutionrate increased. This decrease was primarilythe result of an increased level of RNAwhile the protein concentration remainedrelatively constant. This behavior patternmay be explained by the need of the cells tosynthesize protein faster as their growth rateincreased.When phosphate was the growth-limiting nu-

trient, the ratio of protein to RNA was higherat low dilution rates compared with the ratioresulting from carbon- or nitrogen-limitedgrowth. The increased ratio of protein to RNAwas interpreted here as being consistent withthe hypothesis that a phosphate limitation re-stricts nucleic acid synthesis. Under a phos-phate limitation the cell would have a minimalquantity of nucleic acid, which is predominatelyRNA.When a double limitation of carbon and

nitrogen was imposed, the macromolecularcomposition fell between the limits establishedby single nutrient limitations. Consequently,it became possible to design a medium thatwould produce cells with a range of composi-tions between the limits established by thesingle nutrient limitations. The major con-straint appeared to be the ratio of protein toRNA, which was determined by the dilutionrate under carbon and nitrogen limitations.However, the limits of variation in compositionwere somewhat more flexible when cells werecultured with a phosphate limitation, becauseat low growth rates the ratio of protein toRNA was not only a function of the dilutionrates, but also of the nature of the growthlimitation.One major variable influencing cell composi-

tion is the concentration of storage materials,such as polysaccharide, lipid, or poly-,f-hy-

droxybutyric acid. When these materials accu-mulate, they dilute the relative proportions ofother cell components. Such carbon-energy re-serves are not uncommon in cells grown undernitrogen-limited conditions (5). Although Tem-pest and Dicks (14) found that E. aerogenesdid not accumulate carbon reserve materialunder phosphate-limited conditions, our find-ings indicate that intracellular carbohydratesaccumulated to comprise 15 to 20% of the drycell weight, but only at dilution rates lessthan 0.50/h. This discrepancy might be ex-plained by a change in the nature of thegrowth-limiting nutrient at low growth rates.Most of the cellular phosphate was in the RNAfraction, and the RNA level was directly pro-portional to the growth rate. At low growth ordilution rates, the demand of the cell for phos-phate decreased, and a given amount of phos-phate was able to support a larger cell popula-tion. Therefore, in the cultures incubated at adilution rate of 0.25/h, the cells might havebeen nitrogen-limited and/or phosphate-lim-ited, even though at higher dilution rates onthe same medium they were phosphate-lim-ited.

In addition to differences in cell compositionunder double-nutrient limitations, componentsof the extracellular environment varied withthe type of nutrient limitation. When glucosewas the sole limiting nutrient, the cells re-sponded according to the Monod model; yetwhen other nutrients limited growth, glucosewas still maintained in low (i.e., near K) con-centrations, even though it was supplied in con-siderable excess. Then, under ammonia and/orphosphate limitation(s), the cells were stillreadily able to assimilate glucose and to formcell carbohydrate, catabolic products of glucose(such as acetic acid), and a variety of ultra-violet-absorbing excretion products. Therefore,it may be possible to choose a limiting nutrientthat restricts cell growth yet permits the car-bon source to be shunted through pathways foroversynthesis of desired products.By knowing how cultural conditions affect

the kinetics of growth and product formation,one may be able to design conditions formaintaining, and possibly enhancing, productformation. In this respect, the imposition of adual nutrient restriction on cell metabolismmay be valuable in achieving a tighter controlof useful metabolic pathways.

LITERATURE CITED

1. Brown, C. M., and S. 0. Stanley. 1972. Environment-mediated changes in the cellular content of the'pool" constituents and their associated changes incell physiology. J. Appl. Chem. Biotechnol. 22:363-389.

97VOL. 31, 1976

on March 24, 2020 by guest

http://aem.asm

.org/D

ownloaded from

98 COONEY, WANG, AND MATELES

2. Dean, A. C., S. J. Pirt, and D. W. Tempest (ed.). 1972.Environmental control of cell synthesis and function.Academic Press Inc., New York.

3. Fales, F. W. 1971. The assimilation and degradationof carbohydrates by yeast cells. J. Biol. Chem.193:113-124.

4. Fleck, A., and H. N. Munro. 1962. The precision ofultraviolet absorption measurements in the Schmidt-Tannhauser procedure for nucleic acid estimation.Biochim. Biophys. Acta 55:571-583.

5. Herbert, D. 1961. A theoretical analysis of continuousculture systems, p. 21-53. In Continuous cultureof microorganisms. Monograph No. 12. Society ofChemical Industries, London.

6. Herbert, D., R. Elsworth, and R. C. Telling. 1956.The continuous culture of bacteria; a theoreticaland experimental study. J. Gen. Microbiol. 14:601-622.

7. Logsdon, E. E. 1960. A method for the determinationof ammonia in biological materials on the autoana-lyzer. Ann. N.Y. Acad. Sci. 87:801-807.

8. Munro, H. N., and A. Fleck. 1966. Recent develop-ments in the measurement of nucleic acids in biologi-cal materials. Analyst 91:78-88.

9. Postgate, J. R., and J. R. Hunter. 1962. The survivalof starved bacteria. J. Gen. Microbiol. 29:233-263.

10. Powell, E. 0. 1967. The growth rate of microorganisms

APPL. ENVIRON. MICROBIOL.

as a function of substrate concentration, p. 34-55.In E. 0. Powell, C. G. T. Evans, R. E. Strange, andD. W. Tempest (ed.), Microbial physiology and con-tinuous culture. Her Majesty's Stationery Office,London.

11. Stickland, L. H. 1951. The determination of smallquantities of bacteria by means of the biuret reaction.J. Gen. Microbiol. 5:698-703.

12. Sumner, J. B. 1944. A method for the colorimetricdetermination of phosphorus. Science 160:413-414.

13. Sykes, J., and T. W. Young. 1968. Studies on theribosomes and ribonucleic acids of Aerobacter aero-genes grown at different rates in carbon-limited con-tinuous culture. Biochim. Biophys. Acta 169:103-116.

14. Tempest, D. W., and J. W. Dicks. 1967. Interrela-tionships between potassium, magnesium, phospho-rus, and ribonucleic acid in the growth of Aerobacteraerogenes in a chemostat, p. 140-154. In E. 0. Powell,C. G. T. Evans, R. E. Strange, and D. W. Tempest,(ed.), Microbial physiology and continuous culture.Her Majesty's Stationery Office, London.

15. Young, T. W., and J. Sykes. 1968. Studies on theribosomes and ribonucleic acids of Aerobacter aero-genes grown at different rates in magnesium-limitedcontinuous culture. Biochim. Biophys. Acta 169:117-128.

on March 24, 2020 by guest

http://aem.asm

.org/D

ownloaded from