Vol. 37, No.3APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Mar. 1979, p. 402-4080099-2240/79/03-0402/07$02.00/0

Emulsifier of Arthrobacter RAG-1: Isolation and EmulsifyingProperties

E. ROSENBERG,* A. ZUCKERBERG, C. RUBINOVITZ, AND D. L. GUTNICKDepartment ofMicrobiology, George S. Wise Center for Life Sciences, Tel Aviv University, Tel Aviv, Israel

Received for publication 29 December 1978

The oil-degrading Arthrobacter sp. RAG-1 produced an extraceilular non-dialyzable emulsifying agent when grown on hexadecane, ethanol, or acetatemedium. The emulsifier was prepared by two procedures: (i) heptane extractionof the cell-free culture medium and (ii) precipitation with ammonium sulfate. Aconvenient assay was developed for measurement of emulsifier concentrationsbetween 3 and 75 ,tg/ml. The rate of emulsion formation was proportional to bothhydrocarbon and emulsifier concentrations. Above pH 6, activity was dependentupon divalent cations; half-maximum activity was obtained in the presence of 1.5mM Mg2e. With a ratio of gas oil to emulsifier of 50, stable emulsions were formedwith average droplet sizes of less than 1 ,um. Emulsifier production was parallel togrowth on either hydrocarbon or nonhydrocarbon substrates during the exponen-tial phase; however, production continued after growth ceased.

The growth of microorganisms on hydrocar-bons is often accompanied by the emulsificationof the insoluble carbon source in the culture-medium (2, 4, 13). In most cases, this has beenattributed to the production of extracellularemulsifying agents during the hydrocarbon fer-mentation (4). A requirement for an extracellu-lar rhamnolipid in hydrocarbon utilization bystrains of Pseudomonas aeruginosa has beenreported (7); mutants of P. aeruginosa which donot produce the rhamnolipid were found to growpoorly on the hydrocarbon substrate. Additionof rhamnolipid from the parental strain stimu-lated the growth of such mutants on hydrocar-bons. However, most extracellular emulsifyingsubstances have not been well characterized,and very little is known about their chemicalproperties, mode of action, or biological function.We have previously described an emulsifying

activity present in the culture fluid after growthof an Arthrobacter species, RAG-1, on eithercrude oil or n-hexadecane (8). This report de-scribes the isolation and partial purification ofan emulsifying agent produced by RAG-1 duringgrowth on media containing n-hexadecane, ace-tate, or ethanol. In addition, a convenient andsensitive assay for measuring emulsion forma-tion and stability is described. Subsequent re-ports will be devoted to a description of thehydrocarbon substrate specificity for emulsionsand physical and chemical properties of theemulsifying factor, as well as its immunologicalcharacteristics.

MATERIALS AND METHODSPreparation of the emulsifying substance. The

emulsifying factor ofArthrobacter sp. RAG-1, referredto as EF-RAG, was prepared by two procedures: hep-tane extraction of the cell-free culture medium aftergrowth in hexadecane medium and ammonium sulfateprecipitation of the culture fluid after growth inethanol medium. In each case, the bacteria were grownat 30°C in New Brunswick 14-liter fermentors for 4days. Emulsifying substance prepared from theethanol culture is termed EF-RAG(UET), whereasthe material from a hexadecane culture is termed EF-RAG(HD).

(i) Heptane extraction. Twenty-seven liters of ahexadecane-grown culture was cooled, and the cellswere removed by centrifugation in a Sorvall KSB flowcentrifuge. The supernatant fluid was then extractedtwice with 1/3 volume of redistilled diethyl ether.Residual ether in the aqeuous phase was removed bybubbling with filtered nitrogen gas. The combinedether phase contained no measurable emulsifying ac-tivity and was discarded. The aqueous phase wasfiltered successively through 3-, 1.2-, 0.8-, and 0.45-,ummembrane fiters (Millipore Corp.), and the clear fil-trate was then extracted four successive times with0.15 volume ofheptane. Around 10% of the emulsifyingactivity remained in the aqueous phase and was dis-carded. The heptane fractions were combined andevaporated to a yellow syrup in vacuo. After extractionwith ether, the syrup was dissolved in 100 ml of 50%aqueous methanol. The resulting viscous solution wasdialyzed against several changes of distilled water andlyophilized. The yield of lyophilized EF-RAG(HD)was 1.5 g from 27 liters, with a specific activity of 205U/mg. Similar data were obtained with the heptaneextraction procedure using gas oil instead of hexadec-ane as the growth substrate.

402

on February 13, 2020 by guest

http://aem.asm

.org/D

ownloaded from

EMULSIFIER OF ARTHROBACTER RAG-1 403

(ii) Ammonium sulfate precipitation. Ammo-nium sulfate (1,760 g) was added slowly with stirringdirectly to 10 liters of cooled (40C) ethanol fermenta-tion broth (30% saturation) without prior removal ofthe cells. After standing overnight, the supernatantfluid was collected by decantation. The precipitatewas suspended in 30% saturated ammonium sulfateand centrifuged at 10,000 x g for 15 min. The combinedsupernatant fluids were further clarified by passagethrough a thin layer of kieselgel. To the cell-freesupematant fluid was added an additional portion (62g/liter) of ammonium sulfate to reach a final concen-tration of 40% saturation. The resulting precipitate,collected by centrifugation at 10,000 x g for 15 min,was dissolved in 200 ml of water, extracted with ether,dialyzed against distilled water, and lyophilized. Theyield of EF-RAG(UET) was 2.1 g from 10 liters, witha specific activity of 300 U/mg.

Standard assay for emulsifying activity. Hy-drocarbon (0.05 or 0.1 ml) was added to 7.5 ml offiltered seawater or 7.5 ml of tris(hydroxy-methyl)aminomethane (Tris)-Mg buffer (0.02 M Tris-hydrochloride, pH 7.2, supplemented with 10 mMMgSO4) containing 1 to 25 U of emulsifier per ml[about 3 to 75,ug of EF-RAG(UET) per ml] in a 125-ml flask. After reciprocal shaking (150 strokes per min,2.5-cm stroke length) for 1 h at 260C, contents of theflask were transferred to Klett tubes for measurementof turbidity in a Klett-Summerson colorimeter fittedwith a green filter. Appropriate dilutions were madein water so that the final readings were between 30and 150 Klett units. Reported values for Klett unitsare final readings times the dilution. Values for con-trols containing no emulsifier (5 to 20 Klett units)were subtracted. The average standard deviation ofany reading in the range between 30 and 1,000 Klettunits was 7%. Ten units of emulsifier is defined as thatamount of activity which yields 133 Klett units, using0.1 ml of 1:1 (vol/vol) hexadecane-2-methylnaphtha-lene mixture and 7.5 ml of Tris-Mg buffer. One unit ofactivity gives 13.3 Klett units. Specific activity is givenin units per milligram (dry weight) of EF-RAG prep-arations dialyzed against distilled water and lyophi-lized.

Materials. Agha Jari and Gach-Saran gas oils wereobtained from the Haifa Refinery, Haifa, Israel. Thechemical and physical properties of the paraffinic Ira-nian crude oils from which they were derived havebeen reported (3, 4). Prudhoe Bay crude oil used forsurface tension measurements was obtained from W.H. Wade, University of Texas. Olefin-free hexadecane(>99% purity) and Kieselgel were obtained from FlukaChemical Co., Switzerland. 2-Methylnaphthalene wasa product of Aldrich Chemical Co., Milwaukee, Wis.Microorganism and growth conditions. Strain

RAG-1, an Arthrobacter sp., has been described pre-viously (8). The organism was cultivated in a mediumcontaining 0.125% urea, 0.125% MgSO4. 7H20, 0.002%FeSO4.7H20, 0.001% CaCl2 (anhydrous), 0.025%K2HPO4, 0.2 M Tris-hydrochloride buffer (pH 7.4),and either 0.2% (vol/vol) hexadecane (hexadecane me-dium), 0.5% sodium acetate (acetate medium), or 0.1%(vol/vol) ethanol (ethanol medium). Unless otherwisestated, incubation was at 30°C with vigorous reciprocal

shaking. Growth was initiated with 2-ml inocula oflate-exponential-phase cultures into 200 ml of pre-warmed media in 1-liter flasks. Viable cell number wasdetermined by spreading 0.1 ml of an appropriatedilution on ACYE agar: 0.5% sodium acetate, 0.1%yeast extract (Difco), 0.125% urea, 0.025% K2HPO4,and 1.5 % agar (Difco). Plates were incubated at 32°Cfor 3 days. Growth in ethanol and acetate media wasalso estimated by turbidity, using a Klett-Summersoncolorimeter fitted with a green filter or a Gilfordspectrophotometer, model 240. One-hundred Klettunits of exponentially growing Arthrobacter sp. RAG-1 corresponds to an absorbance at 620 nm (1-cm lightpath) of 0.816 and a biomass of 0.37 g/liter (dried at900C for 16 h).

RESULTSEmulsifying activity as a function of EF-

RAG concentration. To study the kinetics ofemulsifier production and to proceed with theisolation and purification of the active compo-nent(s), a series of simple sensitive assays foremulsifier was developed (Fig. 1). The analysiswas based on the large increase in turbidity of amixture ofwater and oil obtained from the emul-sion of the hydrocarbon in the aqueous phase.The first assay, involving emulsification of gasoil in seawater, resembles those conditions pre-

UNITS per ml

FIG. 1. Standard emulsifier assay. Emulsion (cor-rected Klett units) as a function of EF-RAG(UET)concentration was determined as described in thetext, using 0.05 ml of Gach-Saran gas oil and 7.5 mlof seawater (0), 0.05 ml of Gach-Saran gas oil and7.5 ml of Tris-Mg buffer (A), and 0.1 ml of 1:1 (vollvol) hexadecane-2-methylnaphthalene and 7.5 ml ofTris-Mg buffer (0). The preparation of EF-RAG(UET) used had a specific activity of 330 U/mg.Each point in the figure represents the average ofthree to four determinations.

VOL. 37, 1979

on February 13, 2020 by guest

http://aem.asm

.org/D

ownloaded from

404 ROSENBERG ET AL.

viously used for studying degradation and emul-sification of hydrocarbons by Arthrobacter sp.RAG-1 (8). Seawater could be replaced in theassay by dilute solutions of magnesium salts.Finally, totally defined conditions were devel-oped using a mixture of pure hydrocarbons (hex-adecane and 2-methylnaphthalene) in place ofgas oil and buffered magnesium sulfate in placeof seawater. Using the standard curves shown inFig. 1, emulsifier concentrations between 1 and25 U (corresponding to about 3 to 75,g ofpurified emulsifier) per ml can be determinedaccurately.Production of extracellular emulsifying

activity during growth. Measurement of ex-

tracellular emulsifying activity was determinedat different stages of growth in hexadecane me-

dium (Fig. 2) and ethanol medium (Fig. 3). Thefirst 12 h of growth on ethanol was accompaniedby a drop in pH from 7.4 to 7.0 (Fig. 3), afterwhich the pH remained constant for the remain-ing 36 h. In both cases, emulsifier productionwas growth associated. The maximum amountof emulsifier activities was 14 U/ml of cell-freeculture fluid in the hexadecane medium and 25

( v

I I II

i(A

U!

0 12 24 36 48

HOURS

FIG. 2. Extracellularproduction of emulsifier dur-ing growth ofArthrobacter sp. RAG- I on hexadecanemedium. Two milliliters of a late-exponential-phaseculture was inoculated into 200 ml of hexadecanemedium and incubated with reciprocal shaking at30°C. Viable cell number was determined on ACYEagar. Emulsifying activity was determined on theculture supernatant fluid after centrifugation at10,000 x gfor 15min and filtration through Whatman3 MM filter paper to remove residual hexadecane.The standard emulsifier assay (Fig. 1) was used, withgas oil as a substrate in Tris-Mg buffer. Activity inunits per milliliter of culture supernatant fluid was

calculated from the corrected Klett units after sub-tracting blanks for controls containing no emulsifier(7 Klett units) and controls containing supernatantfluid in the absence of added hydrocarbon substrate(22 Klett units).

EJ

tl

74

Z270

500

100

so

10

50

I

10 I

25

a)W1

0 12 24 36 48HOURS

FIG. 3. Production of extracellular emulsifier dur-ing growth of Arthrobacter sp. RAG-I on ethanolmedium. The experiment wasperformed as describedin the legend to Fig. 2 except that growth was followedby turbidity measurements. Emulsifier activity was

measured directly on the supernatant fluids aftercentrifugation to remove cells.

U/ml of cell-free culture fluid in the ethanolmedium. Emulsifier was also produced duringgrowth in acetate medium. Over 75% of theactivity was extracellular after growth in acetateor ethanol medium, whereas all measurable ac-tivity was extracellular when RAG-1 was grownon hexadecane medium (Table 1). The activityassociated with the pellet fraction was variable;in certain cases no measurable cell-bound activ-ity could be found. Disruption of the pellet frac-tions by sonic oscillation did not release addi-tional emulsifier activity.Kinetics of EF-RAG-induced emulsion

formation. The rate of emulsification of gas oilby purified EF-RAG is summarized in Fig. 4. Atfixed concentrations of EF-RAG(HD), the rateof emulsion formation as well as the final turbid-ity was proportional to a gas oil concentration ofbetween 5 and 100 mg of gas oil per ml. With 33or 100 jig of EF-RAG per ml and concentrationsof gas oil exceeding 45 mg/ml, half-maximumturbidities were reached in less than 5 min. IfEF-RAG and gas oil were allowed to interact at25°C for 2 h without shaking, half-maximumturbidities were obtained in less than 2 min ofshaking. After 60 min of shaking, turbidity con-

tinued to increase gradually for 4 h (about 10%per h). Similar data were obtained with EF-RAG(UET).Emulsion formation [60 min of shaking, 33 or

100 jig of EF-RAG(HD) per ml] as a function ofgas oil concentration is shown in Fig. 5. Emul-sions were formed over the entire gas oil concen-

II

II

-

-

I

l

APPL. ENVIRON. MICROBIOL.

on February 13, 2020 by guest

http://aem.asm

.org/D

ownloaded from

EMULSIFIER OF ARTHROBACTER RAG-1 405

tration range studied, 0.5 to 100 mg/ml. Below1.5 mg of gas oil per ml, turbidities were directlyproportional to gas oil concentration. At be-tween 8 and 30 mg of gas oil per ml, turbidityincreased about 5 Klett units/mg of gas oil.

Effect of pH and salt concentration onemulsion formation. EF-RAG-induced emul-sification of gas oil as a function of pH is shownin Fig. 6. In seawater, near-maximum emulsionswere obtained from pH 5 to at least pH 9. AbovepH 9 precipitation of salts prevented accuratemeasurements of emulsion. In aqueous solutionscontaining Tris buffer, citrate-phosphate buffer,or diluted saline, a sharp maximum was obtainedbetween pH 5 and 6. Above pH 7, activity wascompletely lost.To better understand the different results ob-

tained in seawater and freshwater, the effect of

TABLE 1. Distribution of emulsifying activity infractions ofgrowth culture

Emulsifier (U/ml) after growthon:

Fraction"Hexadec- Ethanol Acetate

ane

Pellet 0 7 2Supernatant fluid 14 23 8.5Pellicle 0

" Cultures (40 h) were centrifuged at 10,000 x g for15 min. Pellets were washed once with Tris-Mg buffer.The hexadecane pellicle formed during centrifugationwas removed and washed twice with growth mediumbefore assaying for activity. Emulsifier was assayed asdescribed in the legend to Fig. 1.

salts on EF-RAG-induced emulsification wasmeasured at pH 8.0 (Fig. 7). Maximum activitywas obtained with 5 to 40 mM MgSO4 or MgCl2.Half-maximum activity was achieved with 1.5mM Mg2+. CaCl2 (10 mM) and MnCl2 (10 mM)could substitute for MgSO4. On the other hand,NaCl (10 to 500 mM) had little effect on emul-sion formation in either the presence or theabsence of Mg2+. Thus, the ability of EF-RAGto emulsify hydrocarbons above pH 6 is depend-ent upon divalent cations.

Stability of EF-RAG-induced emulsions.

V 400

I

GAS-OIL (mg/mI)

FIG. 5. EF-RAG-induced emulsion formation as afunction ofgas oil concentration. The experiment wasperformed as described in the legend to Fig. 3, using33 (0) and 100 (0) pg ofEF-RAG(HD) per ml. Emul-sion formation was determined after 60 min.

MINUTES

FIG. 4. Kinetics ofEF-RAG-induced emulsion ofgas oil. Side-armed flasks (100 ml) containing (A) 0.25 mgofEF-RAG(HD) or (B) 0. 7mg ofEF-RAG(HD) in 7.5 ml ofseawater were shaken with various amounts ofgasoil: (A) 4.5 mg; (l) 1I mg; (U) 45 mg; (0) 176 mg; (0) 582 mg. The numbers on the figure indicate the weightratios ofgas oil to EF-RAG.

VOL. 37, 1979

on February 13, 2020 by guest

http://aem.asm

.org/D

ownloaded from

406 ROSENBERG ET AL.

Z200 -

c

i

z

-JO100 1

0

0~~~~~2 4 6 8 10 12

pH

FIG. 6. Emulsification ofgas oil by EF-RAG(HD)as a function of pH. One hundred-milliliter flaskscontained 33 pg of EF-RAG per ml, 6 mg of AghaJari gas oil per ml, and 7.5 ml of seawater (0), 10mM NaCl (0), 100 mM citrate-phosphate buffer (A),or 50 mM Tris-NaOH buffer (O). The pH values ofseawater and 10mM NaCl were adjusted by additionof HCI or NaOH. Turbidities were determined afterreciprocal shaking (150 strokes per min) at 25°C for60 min.

Gas oil emulsions formed in the presence of EF-RAG slowly separate into two phases when al-lowed to stand undisturbed: a lower, clearaqueous phase and a turbid upper phase con-taining concentrated oil droplets, bound EF-RAG, and water. As observed with a phasemicroscope, emulsion breakage (demulsifica-tion) was a result of "creaming" due to densitydifferences between the two phases and was notaccompanied by droplet coalescence or aggre-gation. The rate of phase separation was fol-lowed by turbidity measurements in a Klett tube(Fig. 8). Emulsion stability depended more uponthe ratio of gas oil to EF-RAG than on theabsolute concentration of EF-RAG or gas oilused to form the emulsion. With gas oil to EF-RAG ratios of less than 25, over 24 h of standingwas required for a 50% decrease in turbidity.With ratios between 25 to 200 and 200 to 1,000,half-maximum turbidities were reached in 1 to24 h and 10 to 60 min, respectively. In all cases,the upper "cream" immediately dispersed inaqueous media, indicating the presence of oil-in-water emulsion. Emulsion breakage was en-hanced by divalent cations. The rate of flotationof oil droplets as a function of the gas oil to EF-RAG ratio is shown in Fig. 9. The average radiiof the droplets were calculated from Stokes'equation (10), using 0.90 g/cm3 as the density ofgas oil. The calculated droplet sizes were in good

agreement with measurement of droplet size byphase microscopy (using a calibrated eyepiecemicrometer). With a ratio of gas oil to EF-RAGof 50, the droplets were barely visible by lightmicroscopy.Lowering of interfacial tensions between

petroleum fractions and seawater by EF-RAG. The ability of EF-RAG to lower the in-terfacial tension between a series of n-alkanesand seawater is shown in Fig. 10. Using similartechniques, the interfacial tensions between Pru-dhoe Bay crude oil and seawater was measured,using 1 and 10 mg of EF-RAG(HD) per ml,yielding 8.3 and 6.9 dyn/cm, respectively. (Sur-factants are available which lower the tensionsto less than l0' dyn/cm) (1).

DISCUSSIONThe assay for the emulsifying factor produced

by RAG-1 is both convenient and sensitive. Fur-thermore, since the turbidity of the emulsion isdirectly proportional to increasing concentra-tions of EF-RAG, the procedure can be usedeither to follow the purification of the materialfrom the culture fluid or to assay cultures di-rectly in order to monitor production. Finally,since the emulsifying factor can be assayed witha defined substrate such as a mixture of hexa-

300

200z0

15

mM

FIG. 7. Emulsification ofgas oil by EF-RAG as afunction of salt concentration. EF-RAG(HD) was di-alyzed extensively against cold 0.01 N HCI and thendistilled water. After adjustment ofthepH to 7.0 withTris base, emulsification was assayed as described inthe legend to Fig. 5 in the presence of MgCl2 (0) orNaCl (0).

APPL. ENVIRON. MICROBIOL.

on February 13, 2020 by guest

http://aem.asm

.org/D

ownloaded from

EMULSIFIER OF ARTHROBACTER RAG-1 407

U,

z

eM-

MINUTES

FIG. 8. Stability of emulsion as a function of the ratio ofgas oil to EF-RAG. Emulsions formed after 120min as described in the legend to Fig. 5, using 33 (A) and 100 (B) pg ofEF-RAG(HD) per ml, were allowed tostand without shaking (zero time). Percent Klett units (KU) [(KUt,,/KUt-o x 1X00] are plotted as a function ofstanding time. The numbers on the figure indicate the weight ratios ofgas oil to EF-RAG.

w 3co ia .5on 0fi

_'a

. a0,

0

! Aa E

GAS-OIL:EF- RAG

FIG. 9. Rate at which emulsified oil droplets riseas a function of the gas oil to EF-RAG ratio. Exper-iments were performed as described in the legend toFig. 5: (0) 100 jg ofEF-RAGper ml; (0) 33 pg ofEF-RAG per ml. The average radii of droplets, F, were

calculated from Stokes' equation, V= 21,800 r2, whereV is the velocity at which oil droplets rise in centi-meters per second and r is the radius in centimeters.

decane and 2-methylnaphthalene in a systemwhich does not require seawater, the conditionsfor measuring activity can be reproduced in anylaboratory. This feature is useful in light of thelarge variability in different petroleum fractionsas well as in seawater itself. The demonstrationof substrate specificity for EF-RAG-mediated

Eu

10

, 10

6 8 10 12 14 16

CARBON NUMBERFIG. 10. Interfacial tensions of n-alkanes in sea-

water containing 0.1% EF-RAG(HD). Values for in-terfacial tension (y) were determined at 27°C, usingthe spinning-drop interfacial tensiometer (1).

emulsification is the subject of an accompanyingpaper (9).The rate of emulsion formation and the final

turbidities were found to depend on the concen-trations of both EF-RAG and hydrocarbon,whereas emulsion stability was found to be afunction of the EF-RAG-hydrocarbon ratio inthe assay. This suggests that the activity of theemulsifier involves a direct interaction with thehydrocarbon itself rather than simply exerting

VOL. 37, 1979

on February 13, 2020 by guest

http://aem.asm

.org/D

ownloaded from

408 ROSENBERG ET AL.

an effect on the surface tension of the medium.In fact, several surfactants which brought abouta much larger drop in the surface tension werefound to be much less active in emulsification ofgas oil (unpublished data). This apparent affin-ity of ER-RAG for hydrocarbon substratesformed the basis of the heptane extraction stepused in one of the purification procedures.A second feature of emulsion formation is

associated with the sharp decline in activitybelow pH 4. This may be attributed to theprotonation of a carboxyl group in EF-RAG (14).Above pH 7.0 the activity is strongly inhibitedin the absence of divalent cations.Although insufficient evidence has been ob-

tained to explain the function of divalent cationsin the emulsification process at alkaline. pH,metal ions may be involved either in maintainingthe appropriate configuration of the emulsifieror in stabilizing the EF-RAG-oil complex.The time course of EF-RAG production was

parallel to the time course of growth on bothhydrocarbon and nonhydrocarbon substratesduring the exponential growth phase; however,emulsifier production continued after growthceased. Although not presented in this report,the production of EF-RAG was found to begrowth associated even at very low cell densities(103 to 106 cells per ml), using a highly sensitiveradioimmunoassay system. Additional supportfor growth-associated production stems fromcontinuous-culture experiments in which pro-duction was observed during steady-stategrowth conditions (unpublished data). It re-mains to be determined whether production ofthe bioemulsifier involves de novo synthesis orsimply the excretion of the polymer into themedium. Preliminary experiments with resting-cell suspensions indicate that both utilizable car-bon and nitrogen sources are required for EF-RAG production.A number of different microorganisms have

been shown to produce hydrocarbon emulsifiers(2, 4-7, 11-13). These emulsifiers appear to differfrom EF-RAG with respect to chemical compo-sition, substrate specificity, response to divalentcations, pH optima, and temperature sensitivity.However, since previous reports do not includea quantitative estimation of emulsifying activityon a weight basis, a more detailed comparison of

the different preparations with respect to thisimportant characteristic is currently not possi-ble.

ACKNOWLEDGMENTWe thank W. H. Wade of the Department of Chemistry,

University of Texas, Austin, for interfacial tension measure-ments and for useful discussions.

LITERATURE CITED

1. Cayias, J. L., R. S. Schechter, and W. H. Wade. 1975.The measurement of low interfacial tension via thespinning drop technique, p. 234-247. In ACS Sympo-sium Ser. no. 8, Adsorption at interfaces. AmericanChemical Society, Washington, D.C.

2. Erickson, L. E., and T. Nakahara. 1975. Growth incultures with two liquid phases: hydrocarbon uptakeand transport. Proc. Biochem. 10:9-13.

3. Ferrero, E. P., and D. T. Nidiols. 1972. Analyses of 169crude oils from 122 foreign oil fields, p. 53. In Bureau ofMines Information Circular no. 8542. U.S. Departmentof the Interior, Washington, D.C.

4. Gutnick, D. L., and E. Rosenberg. 1977. Oil tankersand pollution: a microbiological approach. Annu. Rev.Microbiol. 31:379-396.

5. Holdom, R. S., and A. G. Turner. 1969. Growth ofMycobacterium rhodochrous on n-decane: a newgrowth factor and emulsifying agent. J. Appl. Bacteriol.34:448-456.

6. Iguchi, T., I. Takeda, and H. Ohsawa. 1969. Emulsify-ing factor of hydrocarbon produced by a hydrocarbon-assimilating yeast. Agric. Biol. Chem. 33:1657-1658.

7. Itoh, S., and T. Suzuki. 1972. Effect of rhamnolipids ongrowth of Pseudomonas aeruginosa mutant deficientin n-paraffin-utilizing ability. Agric. Biol. Chem 36:2233.

8. Reisfeld, A., E. Rosenberg, and D. Gutnick. 1972.Microbial degradation of crude oil: factors affecting thedispersion in sea water by mixed and pure cultures.Appl. Microbiol. 24:363-368.

9. Rosenberg, E., A. Perry, D. T. Gibson, and D. L.Gutnick. 1978. Emulsifier of Arthrobacter RAG-1:specificity of hydrocarbon substrate. Appl. Environ.Microbiol. 37:409-413.

10. Shaw, D. J. 1970. Introduction to colloid and surfacechemistry, p. 17. Butterworth & Co., London.

11. Zajic, J. E., H. Guignard, and D. F. Gerson. 1977.Emulsifying and surface active agents from Corynebac-terium hydrocarboclastus. Biotechnol. Bioeng. 19:1285-1301.

12. Zajic, J. E., H. Guignard, and D. F. Gerson. 1977.Properties and biodegradation of a bioemulsifier fromCorynebacterium hydrocarboclastus. Biotechnol.Bioeng. 19:1303-1320.

13. Zajic, J. E., and C. J. Panchal. 1976. Bio-emulsifiers.CRC Crit. Rev. Microbiol. 5:39-66.

14. Zuckerberg, A., A. Diver, Z. Peeri, D. L. Gutnick, andE. Rosenberg. 1978. Emulsifier of Arthrobacter RAG-1: chemical and physical properties. Appl. Environ.Microbiol. 37:414-420.

APPL. ENVIRON. MICROBIOL.

on February 13, 2020 by guest

http://aem.asm

.org/D

ownloaded from