SYNTHESIS OF CELLULOSE BY ACETOBACTER XYLINUM

VI. GROWTH ON CITRIC ACID-CYCLE INTERMEDIATES

ZIPPORA GROMET-ELHANAN' AND SHLOMO HESTRIN2

Department of Biological Chemistry, The Hebrew University, Jerusalem, Israel

Received for publication 20 August 1962

ABSTRACT

GROMET-ELHANAN, ZIPPoRA (The HebrewUniversity, Jerusalem, Israel) AND SHLOMOHESTRIN. Synthesis of cellulose by Acetobacterxylinum. VI. Growth on citric acid-cycle inter-mediates. J. Bacteriol. 85:284-292. 1963.-Acetobacter xylinum could be made to grow onethanol, acetate, succinate, or L-malate. Thegrowth was accompanied by formation of opaqueleathery pellicles on the surface of the growthmedium. These pellicles were identified as cellu-lose on the basis of their chemical properties,solubility behavior, and infrared absorptionspectra. Washed-cell suspensions prepared fromcultures grown on ethanol or the organic acids,in contrast to washed sugar-grown cells, wereable to transform citric-cycle intermediates intocellulose. The variations in the substrate spec-trum of cellulose synthesis between sugar-growncells and organic acids-grown cells were found tobe correlated with differences in the oxidativecapacity of the cells. The significance of the find-ings that A. xylinum could be made to grow onethanol on complex as well as synthetic media isdiscussed from the viewpoint of the wholepattern of Acetobacter classification.

Acetobacter is particularly prone to mutation,even in characters which have been used forspecies classification (Shimwell, 1956, 1957b,1959; see, however, De Ley, 1961). Of the fivebiochemical criteria established by Frateur(1950) for the classification of acetic acid bacteria,four (catalase activity, ketogenic power, produc-tion of acid from glucose, and ability to synthe-size cellulose) were shown by Shimwell to behighly mutable. On the other hand, behavior inrespect to the fifth criterion (growth on Hoyer's

I Present address: Section of Biochemistry, TheWeizmann Institute of Science, Rehovoth, Israel.

2 Deceased 2 February 1962.

synthetic medium in which ammonium andethanol serve as sole nitrogen and carbon sources)was found to be stable (Shimwell, 1957b, 1959;Shimwell and Carr, 1961).

Strains of Acetobacter, which are Hoyer-negative but positive for the other four criteriaestablished by Frateur, have been classified asAcetobacter xylinum. Previous attempts to growA. xylinum on ethanol or acetate, in either syn-thetic or complex media, have been unsuccessful(Hoyer, 1898; Beiyerinck, 1898; Tosic andWalker, 1946; Frateur, 1950; Rao and Stokes,1953; Hall et al., 1956; Shimwell, 1957a) exceptin one case, that of A. xylinum var. africansNCIB 7029 (Hall et al., 1956; Steel and Walker,1957).

In sugar-grown A. xylinum, it was shown thatcarbohydrates are oxidized by way of a pentosecycle (Gromet, Schramm, and Hestrin, 1957),and that organic acids are oxidized by way of acitric cycle which also participates in the sugaroxidation (Gromet-Elhanan, 1960a). Suspensionsof washed sugar-grown cells of this species formedcellulose from sugars and related substrates viathe pentose cycle, but formed no cellulose fromethanol or any citric-cycle intermediate(Schramm, Gromet, and Hestrin, 1957a). More-over, A. xylinum failed to form C14-cellulose evenwhen this organism was grown on C14-acetate orC14-ethanol together with glucose (Greathouse,Shirk, and Minor, 1954). However, Bourne andWeigel (1954) demonstrated synthesis of C'4-cellulose from C14-lactate by A. acetigenum(Henneberg, 1898) growing in a medium whichcontained some glucose, and Stacey (1954) men-tioned in a review that A. acetigenum also formsC14-cellulose in a medium which contains C14-acetate and glucose. The findings thus suggestthat carbon can flow in sugar-grown A. xylinumcells from the pentose cycle into the citric cyclebut that flow in the reverse direction is blocked,whereas in A. acetigenum this block either does

284

on Decem

ber 24, 2019 by guesthttp://jb.asm

.org/D

ownloaded from

SYNTHESIS OF CELLULOSE BY A. XYLINUM

not occur or an alternative pathway for cellulosesynthesis has been acquired.

It remains to be ascertained whether growthof A. xylinum in the presence of citric-cycle inter-mediates but in the absence of glucose can yieldcells which are able to transform citric-cycleintermediates into cellulose. In the present paper,it is shown that, under appropriate conditions,A. xylinum can in fact be made to grow oncitric-cycle intermediates or on ethanol or ace-tate; that such growth is accompanied by forma-tion of a cellulose pellicle; and that washed-cellsuspensions prepared from such cultures trans-form citric-cycle intermediates into cellulose. Apreliminary account of this work has been pub-lished (Gromet-Elhanan, 1960b).

MATERIALS AND METHODS

The strain of A. xylinum was the same as thatemployed in earlier investigations (Hestrin andSchramm, 1954; Schramm and Hestrin, 1954a).

Composition of culture media. The pH levelswere adjusted as necessary by addition of HClor NaOH. The media were as follows (composi-tion in per cent, w/v): (A) Substrate (glucose,fructose), 2; yeast extract (Difco), 0.5; peptone(Difco), 0.5; dipotassium phosphate, 0.1; pH7.0. For experiments at pH 5.0, the monopo-tassium salt was used. (B) Substrate (ethanol,acetate, succinate, L-malate, or citrate), 2; yeastextract (Difco), 0.4; peptone (Difco), 0.3;L-asparagine, 0.1; monopotassium phosphate,0.3; pH 5.0. (C) Substrate (succinate, L-malate,or citrate), 2; yeast extract (Difco), 0.5; peptone(Difco), 0.5; monopotassium phosphate, 0.3;pH 4.0. (D) Substrate (ethanol, acetate, orcitrate), 2; yeast extract (Difco), 1.0; peptone(Difco), 0.5; monopotassium phosphate, 0.3;pH 4.0 or 5.0, as needed. (E) As D, but withaddition of 0.1 % L-malate; pH 4.0. (F) Hoyer'ssynthetic medium (2% glucose as carbon source)prepared according to Frateur (1950). (G) As F,but with 2% ethanol as carbon source. (H) AsG, but with addition of 0.1% L-malate.

Transfers from glucose to a nonsugar mediumwere made in B. Growth in this medium wasslow except with succinate. At subsequent trans-fers from B to either C or D, a rapid growth wasachieved. In the case of acetate, the growth wasalso slow in D (Table 1), but was rapid in E,which was therefore always used for the massculture on acetate.

Effect of pH on growth. The effect of the initialpH of the growth medium on the cell yield meritscomment. Growth of A. xylinum was found to beinhibited at pH > 7.0. Elsewhere, it has beenreported that growth of A. xylinum is inhibitedat pH < 3.5 (Tosic and Walker, 1946). Accord-ingly, with acid-generating substrates, such asglucose, optimal growth was obtained when theinitial pH of the medium was held within 5.0 to7.0. With alkali-generating substrates of the typeof a salt of an organic acid, eventual rise of thepH beyond 7.0 could not be completely preventedbut was sufficiently delayed (> 60 days) whenthe initial pH was brought to 4.0. Some increaseof pH levels also occurred during growth inmedia containing fructose or ethanol as the car-bon source, but could be held within desiredlimits by setting the initial pH value at 5.0. Theextent of the variation of pH levels during growthin the presence of different substrates at selectedinitial pH values is shown in Fig. 1.

Culture conditions. Growth on substrates wasexamined in test tubes containing 5 ml of mediumheld at 30 C under static conditions. Growth wasmanifested by formation of a pellicle.

l

9-

8°/

pH 6 ,jVX

14 21Time (days)

FIG. 1. Changes of pH with time, in cultures ofAcetobacter xylinum on different carbon sources.

VOXi,. 85, 1963 285

on Decem

ber 24, 2019 by guesthttp://jb.asm

.org/D

ownloaded from

GROMET-ELHANAN AND HESTRINJ

Mass cultures were grown at 30 C in a shallowlayer of quiescent liquid contained in an alumi-num pan under loose cover. The inoculum wasprepared as follows. About 1.5 ml of culturestock in a test tube were transferred to 100 ml ofmedium incubated in a Roux flask (layer thick-ness of medium, 1.0 cm) for 3 days at 30 C. Fromthis flask, 10 ml were then transferred to 100 mlof fresh medium in a second flask which wasincubated for 2 days at 30 C and then added tothe mass medium in a ratio of 1:10. The cellswere harvested at 38 hr, but with acetate thecrop was harvested at 60 hr.

Preparation of washed-cell suspensions. Cellswere prepared as described previously (Hestrinand Schramm, 1954), but were washed twice incitrate phosphate buffer, pH 5.0 (0.032 M phos-phate; 0.16 M citrate), and twice in distilledwater. All operations were carried out in the cold.Cells were used immediately after preparation.

Test systems. To examine the effect of agita-tion, cultures were grown in 100-ml Erlenmeyerflasks containing 30 ml of medium swirled in aRose-Kershaw shaking apparatus (A. H. ThomasCo., Philadelphia, Pa.) maintained at 30 C.

Conventional Warburg techniques were usedin all manometric determinations, CO2 beingabsorbed by alkali in the central cup. The sub-strate was always put in the side arm and addedto the reaction mixture after 10 min of equilibra-tion at 30 C. The gas phase was air and the totalvolume was 2 ml. Endogenous readings repre-sented < 10% of the total readings. Values arereported after having been corrected for endoge-nous rates.

Cellulose synthesis was tested in a system(10 ml) consisting of 30 mg (dry wt) of cells and10 mg of substrate in potassium phosphate buffer(pH 6.5; 0.025 M phosphate) contained instoppered 125-ml Erlenmeyer flasks shaken for5 hr at 100 oscillations/min in a water bath at30 C. Except as otherwise stated, the gas phasewas 02. Synthesis of cellulose was stopped byflooding the mixture with ether. The celluloseformed and the enmeshed cells were then sedi-mented by centrifugation, and the precipitatewas repeatedly washed. The cellulose initiallypresent (never exceeding 0.2 mg) was deductedfrom the total amount found after incubation ofcells with substrate.

Determination of infrared absorption spectra.

The pellicles produced under static growth con-ditions in 100 ml of medium in a Roux flask wereused for optical analysis. In one case, an analysiswas also made on a pellicle produced by washedcells in a system (20 ml) consisting of 27 mg(dry wt) of ethanol-grown cells and 200 mg ofglucose in potassium phosphate buffer (pH 6.5;0.025 M phosphate) contained in a petri dish andincubated at 30 C for 20 hr. The pellicles werewashed in water. Proteins were removed fromthe pellicles by treatment with 4% NaOH at30 C for 3 days under N2, and then for 10 min at100 C in air. Transparent pellicles obtained bythese means were then washed twice in diluteacetic acid and twice in water and finally lyophil-ized (4 hr) while spread out on petri dishes, toafford thin films which were analyzed with theprocedure of Forziati and Rowen (1951) in adouble-beam infrared recording spectrophotome-ter (Baird Ltd.), kindly placed at our disposal bythe Department of Organic Chemistry, HebrewUniversity.

Analytical methods. Cellulose was determinedessentially by the method of Schramm andHestrin (1954b), except that the tedious neutrali-zation stel) was circumvented by the use of aphenol-sulphuric acid reagent (Dubois et al.,1956) to measure the glucose released by acetoly-sis-hydrolysis from cellulose. Values for celluloseobtained with this modification and by theoriginal method of hexose determination using acopper reagent were identical. Reducing powerwas estimated with Somogyi's copper reagent(Nelson, 1944), fructose with resorcinol (Roe,Epstein, and Goldstein, 1949), and glucose (asaldose) by a microadaptation of an iodometricmethod (Macleod and Robison, 1929).The radioactivity of C'4 was measured in a

thin-window Geiger-Mifller tube. Substrate solu-tions were put on thin lens paper discs togetherwith 0.5 ml of 0.2% agar, dried in planchets, andcounted. Cellulose samples were washed, freedfrom protein, then washed again and homogenizedin a small amount of water, collected on Milliporefilter discs, dried in planchets, and counted.Counts were corrected for self-absorption.

Substrates. Radioactive fructose T-C14 was ob-tained from the Radiochemical Centre,Amersham, England. Acetate 1-C'4 was fromResearch Specialties Co., Berkeley, Calif.

286 J. BACTERIOL.

on Decem

ber 24, 2019 by guesthttp://jb.asm

.org/D

ownloaded from

SYNTHESIS OF CELLULOSE BY A. XYLINUM

RESULTS

Growth of A. xylinum on ethanol and some or-

ganic acids. Conditions enabling rapid pellicleformation by A. xylinum growing on differentsubstrates are summarized in Table 1. It can beseen that A. xylinum could be made to grow on

ethanol, acetate, succinate, and L-malate. Understatic conditions, growth was always indicatedby formation of a pellicle on the surface of theliquid medium. The medium beneath the pellicleremained transparent with no diffuse growth.Transfer from static to agitated conditionscaused appearance of diffuse turbid growth ac-

companied by small cellulose bodies. But, as ex-

pected from the work of Schramm and Hestrin(1954a), no cellulose was formed after a fewtransfers under agitation, growth being indicatedonly by turbidity. Transfer from such turbidcultures to test tubes kept under static conditionsgave no pellicles, only diffuse growth with a

precipitate at the bottom.Growth on ethanol was obtained at the first

transfer from a glucose-grown culture. At a

second transfer on ethanol, a pellicle was formedin only one of four test tubes, but thereafter all

further transfers in ethanol from this positivetube gave pellicle formation. In this case, theability to form cellulose from ethanol has obvi-ously been acquired, and the results cannot beattributed to traces of glucose in the medium (seeSteel and Walker, 1957).Growth on acetate did not appear at the first

transfer from either glucose- or ethanol-growncultures. After many unsuccessful attempts togrow A. xylinum on acetate, a pellicle appearedin one test tube. All subsequent transfers fromthis tube to acetate medium gave pellicle for-mation.With succinate and L-malate as carbon sources,

pellicle formation was observed at the first trans-fer from either glucose- or ethanol-grown cultures.With citrate, on the other hand, only a very fineand incoherent membrane appeared on mediumB, C, or D at pH values in the range of 3.5 to 7.0after 7 days of incubation; it showed no furtherincrease in mass with time.

Growth of A. xylinum on Hoyer's medium. Theability to grow on ethanol and acetate could bedue neither to simultaneous mass adaptation nor

to selection of a pre-existing variant, since itappeared only in one of many cultures examined,

TABLE 1. Production of cellulose pellicleby Acetobacter xylinum growing on

different carbon sources

Culture medium

CarbonsourceDesig-Timea

Carbon source nation pH

Glucose ............. A 5.Ob 2cEthanol ............. D 5.0 3Acetate .............. D 4.0 7Succinate............ C 4.0 3L-Malate ............ C 4.0 3

Glucose ............. F 5.6 5Ethanol ............. G 5.6 i d

Ethanol ............. H 5.6 10

a Days after inoculation required for formationof a pellicle of 0.3-cm thickness.

b The same rapid cellulose formation was alsoobserved at an initial pH of 7.0.

c With glucose, the pellicle continued to thickento 1.5 cm. Pellicles on other substrates stoppedgrowing after 14 days, at a thickness of 0.5 cm.

d A fragile sac appeared at the bottom of thesetubes after 7 days (see text).

and was retained even after five transfers onglucose. It was therefore interesting to ascertainwhether these ethanol-grown cells were also ableto grow on Hoyer's synthetic medium. Thesecells could be grown readily on Hoyer's medium(Table 1) when glucose served as the carbonsource (medium F). The classical Hoyer'smedium with ethanol as substrate (medium G)gave a fragile transparent sac at the bottom ofthe test tube, which never rose to form the char-acteristic coherent opaque pellicle at the surface.However, addition of a small amount of L-malateto this medium permitted the production of anormal pellicle. The added small amount ofL-malate gave no pellicle in the absence ofethanol. Also, after shaking down the pellicleformed in ethanol and L-malate (medium H),another pellicle appeared; this process could becontinued through the formation of three to fourpellicles in the same test tube.On the basis of these results, ethanol-grown

A. xylinum cells could be regarded as fulfillingall of Frateur's biochemical criteria used inAcetobacter classification. Another example of anAcetobacter positive for all of Frateur's criteriawas found by Carr (1958). He isolated, from

VOL. 85, 1963 287

on Decem

ber 24, 2019 by guesthttp://jb.asm

.org/D

ownloaded from

GROMET-ELHANAN AND HESTRIN

ciders, a strain of A. aceti which gave a positivecellulose reaction according to Frateur's tests,although no characteristic cellulose pellicle wasformed.

Identification of cellulose in pellicles produced onethanol and succinate. The pellicles formed duringgrowth of A. xylinum on ethanol, acetate, andsuccinate were thinner than those produced onglucose but in other respects were similar in ap-pearance to the latter. The pellicles were insolu-ble in hot 4% NaOH, and afforded productswhich reduced copper reagents and iodine afteracetolysis and hydrolysis according to Schrammand Hestrin (1954b). Unambiguous identificationof cellulose in pellicles produced by ethanol andsuccinate cultures was made on the basis of theinfrared absorption spectrum (Fig. 2). Thespectra obtained were identical with the spectrumof a pellicle produced on glucose. This cellulosehas been identified as high-polymer, native,crystalline a-cellulose (Shirk and Greathouse,1952; Hestrin and Schramm, 1954). No pelliclewas formed in the nutrient medium when thecarbon source was omitted. It is highly likely,therefore, that cultures grown on ethanol orsuccinate synthesized cellulose from these sub-strates.

z

0-

_

.Oh&S-

Cellulose synthesis by washed-cell suspensions.Further proof that the organic acids acetate,pyruvate, succinate, and L-malate can serve assubstrates for cellulose synthesis was obtained bywork with washed-cell suspensions prepared fromcultures grown on ethanol, acetate, or succinate.The washed cells were found capable of cellulosesynthesis in reaction mixtures containing onlybuffer and substrate. In each case, the synthe-sized polymer was identified as cellulose on thebasis of its chemical properties and solubility be-havior. No pellicle thick enough for measuringinfrared absorption could be obtained by theaction of these washed cells on the organic acids.However, a pellicle produced from glucose byethanol-grown washed cells was found to havethe same absorption spectrum as that of thepellicles formed in cultures (Fig. 2). It follows,therefore, that such washed cells are capable ofcellulose synthesis.Washed succinate-grown cells afforded cellulose

from glucose, suceinate, and pyruvate (1.2, 0.4,and 0.4 mg, respectively, from 10 mg of sub-strate). Addition of pyruvate to a reaction mix-ture with glucose afforded the formation of1.6 mg of cellulose. Synthesis from pyruvate wastotally inhibited by fluoroacetate at a concentra-

2 3 4 5 6 7 8 9 10 I 1 12 13WAVELENGTH IN MICRONS

FIG. 2. Infrared absorption spectra of cellulose pellicles produced by Acetobacter xylinum on differentcarbon sources. (A) Pellicle produced in 20 hr of incubation of washed ethanol-grown cells with glucose;(B) pellicle produced in a 7-day culture on ethanol; (C) pellicle produced in a 8-day culture on glucose;(D) pellicle produced in a 7-day culture on succinate.

288 J. BACTERIOL.

on Decem

ber 24, 2019 by guesthttp://jb.asm

.org/D

ownloaded from

SYNTHESIS OF CELLULOSE BY A. XYLINUM

tion (10-3 M) which inhibited oxidation of pyru-

vate. This is in contrast to the case with glucose,in which even 5 X 10- M fluoroacetate did notcompletely suppress cellulose synthesis (Gromet-Elhanan, 1960a).

Synthesis experiments with washed ethanol-grown cells are summarized in Tables 2 and 3.Cellulose was formed from glucose, fructose,L-malate, and pyruvate, but not from ethanol or

acetate, even in the presence of an added smallamount of L-malate.The effect of added pyruvate on cellulose syn-

thesis from sugars was tested in mixtures of glu-cose or fructose and pyruvate (Table 3). Insugar-grown cells, synthesis from glucose andfructose was found to be inhibited by pyruvateand acetate (Schramm et al., 1957a). However,in A. xylinum cultures grown on mixtures of glu-cose and ethanol or acetate, these adjuncts,though not affording cellulose, stimulated cellu-lose synthesis from glucose (Minor et al., 1954;Greathouse et al., 1954). In ethanol-grown cellsas in succinate-grown cells, under conditions inwhich pyruvate by itself afforded cellulose, theaddition of pyruvate to glucose did not signifi-cantly affect the cellulose synthesis. However,when pyruvate was added to fructose, the amountof cellulose formed greatly exceeded the sum ofthe amounts which these substrates gave sepa-

rately. This increase in cellulose was found to bedue to the stimulation by pyruvate of cellulosesynthesis from fructose, since in the absence andpresence of pyruvate the radioactivity recovered

TABLE 2. Cellulose synthesis by washedethanol-grown cells

Substratea Production of cellulose

mg

Glucoseb....... 2.0

Fructoseb....... 1.3L-Malate 0.5

Pyruvate.................... 0.4

Ethanol...... 0. ocAcetate......o.oc

a Gas phase: air.I Of the glucose, 96% was consumed during the

experiment, whereas with fructose only 43% was

consumed.c Cellulose was not formed in this system even

with an addition of a small amount (1 mg) ofL-malate.

TABLE 3. Effect of pyruvic acid on cellulosesynthesis from fructose and glucose

by ethanol-grown cells

Production ofcellulose as

Test system Amount of fraction ofsugar

___________metb-lzedAnhydro- SugarSubstrate Supplement glucose metab

Mmoles % jimoles %Glucose.... - 52.6 94 6.1 11.6Fructose .. 22.5 40 5.0 22.2Pyruvate.. - - 0.6 -

Glucose.... Pyruvateb 53.0 95 6.4 11. c

Fructosea.. Pyruvateb 23.6 42 11.4 45.8c

a Fructose T-C14 (380,000 counts/min) was usedin these experiments. In the absence and presenceof pyruvate, the decrease in soluble C14 in themedium amounted to 140,000 and 150,000 counts,min, respectively, whereas 24,000 and 58,000counts/min were recovered in cellulose.bAmount added: 10 mg.c Calculated on the difference: total cellulose

minus cellulose formed with pyruvic acid alone.

in cellulose amounted to 16 and 42%, respec-tively, of the fructose metabolized. On the otherhand, the rate of sugar consumption was notaltered in the presence of pyruvate, as can beseen from the chemical and isotopic data inTable 3. Also, the synthesis from pyruvate itselfwas unaltered in the presence of fructose.The observed different stimulatory effects on

glucose and fructose can be explained by differ-ences in their metabolic patterns: more than 95%of the glucose was consumed during the reactionas compared with about 42% of the fructose(Tables 2 and 3). The reason for this differenceis the fact that the cells used here stem originallyfrom cells grown on glucose at an initial pH 7.0.These glucose-grown cells were adapted to fruc-tose because of the small amount of fructoseformed from glucose when autoclaved at pH 7.0,but the transfers on ethanol in the total absenceof fructose caused partial deadaptation.Washed acetate-grown cells synthesized cellu-

lose from glucose, fructose, succinate, and pyru-vate. Only traces of cellulose were formed fromacetate, but the addition of a small amount ofglucose raised the synthesis to a level whichsignificantly exceeded that found with eithersubstrate alone (Table 4). With C"4-acetate in the

VOL. 85, 1963 289

on Decem

ber 24, 2019 by guesthttp://jb.asm

.org/D

ownloaded from

GROMET-ELHANAN AND HESTRIN

presence of glucose, the recovery of counts incellulose corresponded to 1.1% of the total C14.On the basis of this result, it could be concludedthat cellulose synthesis from acetate was not af-fected by the presence of glucose, and that thehigher yield of cellulose observed in the presenceof added glucose was due to stimulation of cellu-lose synthesis from glucose by the acetate.From the foregoing observations, it is evident

that, in A. xylinum cells able to form cellulosefrom acetate or pyruvate, these acids could stimu-late cellulose synthesis from sugars, but onlywhen the sugar metabolism proceeded at a lowrate or when small amounts of sugar were used.The acids did not affect the rate of sugar con-sumption, and under these conditions the energyformed by oxidation of part of the sugar mightnot be enough to maintain synthesis at its maxi-mal rate. The stimulation might then be ex-plained by a sparing mechanism whereby theacids are preferentially oxidized, thus sparing thesugar substrate for cellulose synthesis.

Oxidation of substrates by cells grown on ethanol,acetate, or glucose. In view of the demonstratedeffects of the carbon source in the growth mediumon the substrate spectrum of cellulose synthesis,it was of interest to learn whether the observedvariations in synthetic ability were correlatedwith differences in the oxidative capacity of thecells. The results (Table 5) indicate that growthon glucose resulted, in fact, in loss of ability tooxidize acetate and in a delay in the onset ofoxidation of dicarboxylic acids, which were not

TABLE 4. Cellulose synthesis by washedacetate-grown cells

Test system Production ofSubstrate Supplement cellulose

mg

Glucose.. 1.0Fructose -.1.5Succinate o 0.3Pyruvate ........ 0.3Acetate.......... - 0.1Acetatea......... Glucoseb 0.5

- Glucoseb 0.2

b Acetate J-C14 (126,000 counts/min) was usedin this experiment; 1,273 counts/min were re-covered in cellulose.

b Amount added: 1 mg.

TABLE 5. Oxidation of citric-cycle intermediatesby Acetobacter xylinum cells grown

on various substratesa

Substrates ofoxidation

Pyruvate ......Ethanol .......Acetate........Succinate......Fumarate ......

L-Malate ......a-Ketogluta-

rate.........Citrate ........

Carbon source in growth medium

Glucoseb Ethanol' Acetated

02

14.09.001.91.33.1

C02

32.0003.93.76.6

02

12.510.51.1

2.8

0.8 1.5 -0.3 0.5 0.4

C02

27.51.22.5

6.0

02

9.011.04.18.08.48.8

_ 1.00.6 0.3

C02

17.01.56.513.617.517.7

2.00.5

a Rates are expressed relative to glucose = 10,for the time interval 0 to 30 min.

b Mixture contained 8 mg of lyophilized cells,10 jAmoles of substrate, and 100,umoles of sodiumcacodylate; pH 5.0.

c Mixture contained 2 mg (dry wt) of fresh cells,10,4moles of substrate, and 100,umoles of potas-sium phosphate; pH 6.0.

d Mixture contained 6 mg (dry wt) of fresh cells,15 ,umoles of substrate, and 50 Amoles of sodiumcacodylate; pH 5.5.

attacked at all during the first 20 min. Ethanol-grown cells occupied, in these respects, an inter-mediate position between that of the glucose andthe acetate cells.

DISCUSSION

The experiments here reported demonstratethe possibility of obtaining a mutant (in Shim-well's definition: Shimwell, 1959) of A. xylinumwhich could be grown on Hoyer's medium.Hence, even this biochemical criterion, which wasfound to be stable by Shimwell (1957b, 1959), isin fact mutable. De Ley (1961) did not acceptShimwell's view of the abnormally high muta-bility of the acetic acid bacteria, since most ofhis strains showed "unchanged sets of propertiesover a period of three years." However, De Leyand Stouthamer (1959) observed a spontaneousloss of pigment formation (change of Glucono-bacter (Acetobacter) melanogenum into G. sub-oxydans). Tosic and Walker (1946) reported onthe spontaneous loss and gain of the ability ofA. kuetzingianum to form a leathery (cellulose,

290 J. BACTERIOL.

on Decem

ber 24, 2019 by guesthttp://jb.asm

.org/D

ownloaded from

SYNTHESIS OF CELLULOSE BY A. XYLINUM

see Kaushal and Walker, 1951) pellicle. Kulkaand Walker (1954) observed a spontaneous gainin ability to produce gluconic acid from glucoseby A. ascendens (changes into A. rancens), andSchramm and Hestrin (1954a) obtained cellulose-less mutants of A. xylinum (changes into A.mesoxydans). Also, new isolates corresponding toa certain species, but differing in one biochemicalproperty, were reported by Frateur and Simonart(1952), who isolated catalase-negative A. rancensand A. mesoxydans, and by Carr (1958), isolatinga cellulose-producing A. aceti.On the basis of the foregoing observations, all

the biochemical criteria used in Acetobacterclassification (Frateur, 1950; Breed, Murray,and Smith, 1957) seem to be mutable. Shimwell(1957b) and De Ley (1961) pointed out thatAcetobacter species exhibit a stepwise gradationin their biochemical properties, with no abruptchange between neighboring species. It follows,therefore, that even one mutation might beenough to shift one species into another, or causethe appearance of new species, as indicated above.Therefore, the demonstrated mutability of allthe biochemical criteria, whether high or low,leads one to accept Shimwell's conclusion that theacetic acid bacteria cannot be classified at all(Shimwell, 1959).Growth of A. xylinum on ethanol, acetate, and

succinate was accompanied by formation of acellulose pellicle. Moreover, washed-cell suspen-sions prepared from such cultures transformedcitric-cycle intermediates into cellulose. Resultsobtained in sugar-grown cells, from experimentswith specifically labeled hexoses, suggested thatcellulose arises in A. xylinum; as in green plants,from an intracellular pool of hexose phosphate,formed by way of a pentose cycle (Schramm,Gromet, and Hestrin, 1957b). If this postulate isapplied also to nonsugar-grown cells, then cellu-lose synthesis from citric acid-cycle intermediateswould require formation of hexose phosphatefrom these intermediates. Possible pathways forcellulose synthesis from organic acids by succi-nate-grown A. xylinum cells are discussed byBenziman and Burger-Rachamimov (1962).

ACKNOWLEDGMENT

This investigation was supported in part bygrant E-1494 from the National Institute ofAllergy and Infectious Diseases, National Insti-tutes of Health, U.S. Public Health Service.

LITERATURE CITEDBEIYERINCK, M. W. 1898. Uber die Arten der

Essigbakterien. Zentr. Bakteriol. Parasitenk.Abt. II 4:209-216.

BENZIMAN, M., AND H. BURGER-RACHAMIMOV.1962. Synthesis of cellulose from pyruvateby succinate-grown cells of Acetobacter xyli-num. J. Bacteriol. 84:625-630.

BOURNE, E. J., AND H. WEIGEL. 1954. C14-cellulosefrom Acetobacter acetigenum. Chem. Ind.(London) 1954:132.

BREED, R. S., E. G. D. MURRAY, AND N. R. SMITH.1957. Bergey's manual of determinative bac-teriology, 7th ed. The Williams & WilkinsCo., Baltimore.

CARR, J. G. 1958. A strain of A. aceti giving apositive cellulose reaction. Nature 182:265-266.

DE LEY, J. 1961. Comparative carbohydratemetabolism and a proposal for a phylogeneticrelationship of the acetic acid bacteria. J.Gen. Microbiol. 24:31-50.

DE LEY, J., AND J. STOUTHAMER. 1959. The mech-anism and localization of hexonate metabo-lism in Acetobacter suboxydans and Acetobactermelanogenum. Biochim. Biophys. Acta 34:171-183.

DUBOIS, M., K. A. GILLES, J. K. HAMILTON,P. A. REBERS, AND F. SMITH. 1956. Colori-metric method for determination of sugarsand related substances. Anal. Chem. 28:350-356.

FORZIATI, F. H., AND J. W. ROWEN. 1951. Effectof changes in crystalline structure on theinfrared absorption spectrum of cellulose. J.Res. Nat. Bur. Std. 46:38-42.

FRATEUR, J. 1950. Essai sur la systematique deAcetobacters. Cellule Rec. Cytol. Histol. 53:287-392.

FRATEUR, J., AND P. SIMONART. 1952. CongressoInternazionale Industrie Agrarie, Roma.

GREATHOUSE, G. A., H. G. SHIRK, AND F. W. MI-NOR. 1954. Cellulose production by Acetobacterxylinum from unlabelled glucose and C14-acetate and C'4-ethanol. J. Am. Chem. Soc.76:5157-5158.

GROMET, Z., M. SCHRAMM, AND S. HESTRIN.1957. Synthesis of cellulose by Acetobacterxylinum. 4. Enzyme systems present in acrude extract of glucose-grown cells. Biochem.J. 67:679-689.

GROMET-ELHANAN, Z. 1960a. Oxidative pathwaysin Acetobacter xylinum. Ph.D. thesis, TheHebrew University, Jerusalem.

GROMET-ELHANAN, Z. 1960b. Synthesis of celluloseby A. xylinum cells from citric acid cycleintermediates. Bull. Res. Council IsraelSect. A 9:94.

291VOL. 85, 1963

on Decem

ber 24, 2019 by guesthttp://jb.asm

.org/D

ownloaded from

GROMET-ELHANAN AND HESTRIN

HALL, A. N., J. HUSAIN, K. S. TIWARI, AND T. K.WALKER. 1956. Nutritional requirements ofAcetobacter species: inorganic ammoniumsalts as sources of nitrogen. J. Appl. Bacteriol.19:31-35.

HENNEBERG, W. 1898. Weitere Untersuchungenuber Essigbakterien. Centr. Bakteriol. Par-asitenk. Abt. II. 4:14-20.

HESTRIN, S., AND M. SCHRAMM. 1954. Synthesisof cellulose by Acetobacter xylinumn. 2. Prepa-ration of freeze-dried cells capable of poly-merizing glucose to cellulose. Biochem. J.58:345-352.

HOYER, D. P. 1898. Bijdrage tot de kemis van deazynbacterien. Dissertation, University ofLeiden, Waltman, Delft.

KAUTSHAL, R., AND T. K. WALKER. 1951. Formationof cellulose by certain species of Acetobacter.Biochem. J. 48:618-621.

KULKA, D., AND T. K. WALKER. 1954. The keto-genic activities of Acetobacter species in aglucose medium. Arch. Biochem. Biophys.50:169-179.

MACLEOD, M., AND R. ROBISON. 1929. The appli-cation of the iodometric method to the esti-mation of small amounts of aldoses. Biochem.J. 23:517-523.

MINOR, F. W., G. A. GREATHOUSE, H. G. SHIRK,A. M. SCHWARTZ, AND M. HARRIS. 1954.Biosynthesis of C14-specifically labelled cellu-lose by Acetobacter xylinum. I. From D-glu-cose-1-Cl4 with and without ethanol. J. Am.Chem. Soc. 76:1658-1661.

NELSON, N. 1944. A photometric adaption of theSomogyi method for the determination ofglucose. J. Biol. Chem. 153:375-380.

RAO, M. R. R., AND J. L. STOKES. 1953. Nutritionof the acetic acid bacteria. J. Bacteriol. 65:405-412.

ROE, J. H., J. H. EPSTEIN, AND N. P. GOLDSTEIN.1949. A photometric method for the determi-nation of inulin in plasma and urine. J. Biol.Chem. 178:839-845.

SCHRAMM, M., Z. GROMET, AND S. HESTRIN.

1957a. Synthesis of cellulose by Acetobacterxylinum. 3. Substrates and inhibitors. Bio-chem. J. 67:669-679.

SCHRAMM, M., Z. GROMET, AND S. HESTRIN.1957b. Role of hexose-phosphate in synthesisof cellulose by Acetobacter xylinum. Nature179:28-29.

SCHRAMM, M., AND S. HESTRIN. 1954a. Factorsaffecting production of cellulose at the air-liquid interface of a culture of Acetobacterxylinum. J. Gen. Microbiol. 11:123-129.

SCHRAMM, M., AND S. HESTRIN. 1954b. Synthesisof cellulose by Acetobacter xylinum. 1. Micro-method for the determination of cellulose.Biochem. J. 56:163-166.

SHIMWELL, J. L. 1956. Transmutation of speciesin the genus Acetobacter. J. Inst. Brewing62:339-343.

SHIMWELL, J. L. 1957a. The true significance ofHoyer's medium in the differentiation ofAcetobacter species. J. Inst. Brewing 63:44-45.

SHIMWELL, J. L. 1957b. A pattern of evolution inthe genus Acetobacter. J. Inst. Brewing 63:45-56.

SHIMWELL, J. L. 1959. A re-assessment of thegenus Acetobacter. Antonie van LeeuwenhoekJ. Microbiol. Serol. 25:49-67.

SHIMWELL, J. L., AND J. G. CARR. 1961. The typeculture concept in Acetobacter and othergenera. Antonie van Leeuwenhoek J. Micro-biol. Serol. 27:65-75.

SHIRK, H. G., AND G. A. GREATHOUSE. 1952.Infrared spectra of bacterial cellulose. Anal.Chem. 24:1774-1775.

STACEY, M. 1954. Enzymic synthesis of polysac-charides. Advan. Enzymol. 15:301-315.

STEEL, R., AND T. K. WALKER. 1957. Celluloselessmutants of certain Acetobacter species. J.Gen. Microbiol. 17:12-18.

ToSIc, J., AND T. K. WALKER. 1946. A procedurefor the characterization of the acetic acidbacteria. Part II. J. Soc. Chem. Ind. 65:180-184.

292 J. BACTERIOL.

on Decem

ber 24, 2019 by guesthttp://jb.asm

.org/D

ownloaded from