Vol. 57, No. 4APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Apr. 1991, p. 1128-11330099-2240/91/041128-06$02.00/0

Adhesion of Lactobacillus amylovorus to Insoluble andDerivatized Cornstarch Granules

SYED H. IMAM'* AND R. E. HARRY-O'KURU2t

Biopolymer Research Unit, National Center for Agricultural Utilization Research, U.S. Department ofAgriculture, 1815North University Street, Peoria, Illinois 61604,1 and A. E. Staley Manufacturing Company, Decatur, Illinois 625252

Received 5 October 1990/Accepted 29 January 1991

Approximately 70% of the cells in a suspension of the amylolytic bacterium Lactobacillus amylovorus bind tocornstarch granules within 30 min at 25°C. More than 60% of the bound bacteria were removed byformaldehyde (2%) or glycine (1 M) at pH 2.0. More than 90% of the bound bacteria were removed by MgCl2(2 M; pH 7.0). Binding of L. amylovorus to cornstarch was inhibited in heat-killed cells and in cells that hadbeen pretreated with glutaraldehyde, formaldehyde, sodium azide, trypsin, or 1% soluble potato starch.Bacterial binding to cornstarch appeared to correlate with both the concentration of cornstarch in thesuspension and the amylose content in the granules. The ability of L. amylovorus to adhere to cornstarchgranules was reduced for granules that had been extracted with HCl-ethanol, HCl-methanol, HCl-propanol, or

HCl-butanol. Chemical derivatization of cornstarch resulted in a wide variety of adhesion responses by thesebacteria. For example, 2-0-butyl starch (degree of substitution, 0.09) enhanced adhesion, whereas twopalmitate starches (degree of substitution, 0.48 and 0.09) exhibited reduced adhesion activities. 2-0-(2-hydroxybutyl) starch and starch-poly(ethylene-co-acrylic acid) ester showed adhesion activities similar to thoseof the nonderivatized starch controls.

Starch and its derivatives and hydrolysis products haveimportant industrial applications (18, 27). Utilization ofstarch as a renewable raw material has increased con-siderably over the last decade (8). Specifically, starch, abiodegradable polysaccharide, has been blended with non-degradable synthetic polymers such as polyethylene, poly-ethylene-co-acrylic acid (EAA), and polymethylacrylate todevelop partially "biodegradable" plastic blends (21).To hydrolyze starch, organisms must produce amylase, an

enzyme capable of cleaving o-(1-4) glycosidic linkagesbetween a-D-glucopyranosyl residues in the starch mole-cule. Amylolytic microorganisms commonly found in soil,freshwater reservoirs, and seawater hydrolyze starch intomono- and disaccharide units.During the manufacture of starch-containing plastic films,

starch is dispersed with synthetic polymers such as polyeth-ylene, EAA, and polymethylacrylate. The properties ofthese plastic formulations can affect starch location withinthe film (10, 13, 24). The degree to which starch is accessiblein the hydrophobic plastic matrix largely influences theability of amylolytic bacteria to attach to the plastic surface,as well as availability of starch to hydrolytic enzymes (10,13). Adhesive interactions of microorganisms with insolublepolymeric surfaces is an important phenomenon implicatedin a variety of biochemical responses at both a cellular and amolecular level (2, 3, 7, 11, 12, 15, 23, 26). It has beensuggested that the binding of cellulolytic bacteria to plantcell wall and cellulose matrix facilitate cellulose fiber degra-dation (4-6, 22). Studies done in our laboratory indicatedthat starch in some plastic matrices can be readily hydro-lyzed by some amylolytic bacteria (10, 13), but little isknown about whether the adherence of amylolytic microor-ganisms to starch matrix is required for starch breakdown. It

* Corresponding author.t Present address: Biopolymer Research Unit, U.S. Department

of Agriculture, Peoria, IL 61614.

is conceivable that, in the environment, amylolytic microor-ganisms able to adhere directly to such starch-containingsubstrates will have a competitive advantage over nonadher-ent amylolytic organisms. Therefore, adhesion of amylolyticmicroorganisms to starch-containing materials may be animportant factor in determining the fate of biodegradableplastics in the environment. In this report we describe theadhesive interactions of the amylolytic bacterium Lactoba-cillus amylovorus with granular starches, including chemi-cally modified starches.

MATERIALS AND METHODS

Chemicals. Unless mentioned otherwise, all chemicalswere purchased from Sigma Chemical Co., St. Louis, Mo.Cornstarch samples containing 50 and 75% amylose werefrom American Maize Products, Hammond, Ind. Starchcontaining 25% amylose was from CPC International, En-glewood Cliffs, N.J. Waxy starch containing 2 to 4% amy-lose was a food-grade Amioca Powder obtained from Na-tional Starch and Chemical Corp., Bridgewater, N.J.Cornstarch containing 50% amylose was used in most exper-iments. Starches containing 2, 25, and 75% amylose wereused only in experiments to determine the relationshipbetween cell adhesion and amylose content (see Fig. 5).Organisms and cultures. The bacterium, L. amylovorus

NRRL B-4540, used in this study was originally isolated byNakamura from cattle waste-corn fermentations (20). Stockcultures were maintained on lactobacilli-MRS agar (DifcoLaboratories, Detroit, Mich.) slants. For experimental cul-tures, cells from the stock culture were inoculated into250-ml Erlenmeyer flasks containing 100 ml of MRS broth.Cells were allowed to grow for 2 days at 28°C with continu-ous shaking (125 rpm). At the beginning of the third day,while cells were still in mid-log phase, 50 ,uCi of 35S-Translabel (containing 70% L-[35S]methionine, 15% L-cysteine, 7%L-[35S]methionine sulfide, 3% L-[35S]cysteic acid, and 5%other 35S-labeled compounds; ICN Radiochemicals, Irvine,

1128

ADHESION OF L. AMYLOVORANS TO CORNSTARCH GRANULES 1129

Calif.) was added to the culture and the cells were allowed togrow overnight. On the next day, the labeled cells (mid- tolate-log phase) were harvested by centrifugation (12,000 x g,10 min, 4°C) and washed three to four times in 50 ml of sterilemedium. The final pellet was resuspended in sterile mediumto yield 500 to 1,000 cpmI/25 ,ul (approximately 107 cells).

Adhesion assay. A 50% (wt/vol) cornstarch suspension (50mM phosphate buffer, pH 6.0) was prepared and gentlyrocked (Hema-Tek Aliquot Mixer; Miles Laboratories) forseveral hours before use in the adhesion experiments. About0.5 ml of this suspension was dispensed into a 1.5-mlmicrocentrifuge tube to which a 25-,ul aliquot of 35S-labeledcells was added. Tubes were gently rocked for 30 min at 24°Cand then spun for 10 s (two bursts of 5 s each, 15,600 x g;Centra-M Centrifuge, International Equipment Company,Needham Heights, Mass.). This quick centrifugation pel-leted starch granules (and bacteria adhering to the granules),but retained nonadhering cells in the supernatant. It waspredetermined in earlier experiments that a similar treatmentof a suspension of labeled cells of L. amylovorus alonepelleted a negligible amount of radioactivity (<5%). Theseobservations were also confirmed by light microscopy. A50-,lI aliquot of the supernatant was added to a scintillationvial containing 3.5 ml of Ecolume Liquid scintillation fluid(ICN Biomedicals, Inc., Irvine, Calif.), and the radioactivitywas determined in a liquid scintillation spectrophotometer.Total cells bound to starch were determined by using thefollowing formula: % bound cells = {[total cpm added - totalcpm in supernatant (unbound cells)]/(total cpm added)} x100. All samples were analyzed in quadruplicate and eachexperiment was repeated several times. There was little ifany leakage of labeled material from the cells. Quenchingdue to cell suspension was negligible. Also, cornstarchgranule suspensions used in experiments were routinelyexamined for microbial contamination under a light micro-scope.To determine the optimal temperature for cell adhesion,

samples were incubated at 4, 15, 24, 30, 40, and 50°C for 30min as described above.For pH studies, buffer solution was prepared containing 50

mM each sodium citrate, potassium phosphate, Tris, andglycine. The pH was adjusted to the desired value with eitherHCl or NaOH. To study the effect of pH on cell adhesion,cells and starch granules were suspended in appropriatebuffer prior to their mixing for use in adhesion assays.To study the removal of bacteria bound to starch granules,

cells were mixed with a starch suspension and incubated for30 min. At the end of the incubation, samples were spun (twotimes, 5 s), supernatant (containing unbound cells) wasseparated, and the percentage of cells bound to starch wasestimated. About 0.5 ml (approximately 2 bed volumes) ofappropriate elution buffer was added to the starch pelletcontaining bound cells, and samples were gently rocked for10 min. Subsequently, samples were spun (two times, 5 s)and the percentage of bound cells released into the superna-tant by elution buffer was estimated.For some experiments cells were heat inactivated for 3

min in a boiling water bath (a negligible amount of cell-boundlabel was lost due to this treatment) or preincubated (1 h)with 1% soluble potato starch prior to the adhesion assay. Inother experiments, cells were preincubated (1 h) with form-aldehyde (2%, wt/vol), glutaraldehyde (2%, wt/vol), sodiumazide (0.02%, wt/vol), or trypsin (1 mg/ml) and washed twice(12,000 x g, 10 min) in sterile medium prior to the adhesionassay. To determine the effect of sugars on cell adhesion,cells were incubated separately with the individual sugar (1

CH20H0

IV40

+ -I0OH

-OHNa2SO4

2-propanol

Starch n-Butyl iodide

CH20H0 N

H 3 + (CH2)14CH3 -ci

OH

Starch Palmitoyl chloride

HO 0

1) ~~~+

HOx

Polyethylene-co-acrylicacid (EAA)

0

CH2OCCH30

0 0

00<

/o--.0~\\o

Aceticanhydride

CH20H0

0 0o s

2-0-Butyl starch

[11

CH20H

OHrt[2]O-C-(CH2)4CH3

0Starch palmitate

011

H3C-CO 0

0a

EAA anhydride3

NaHDMSO

CH20H0

OH

OH

Starch

Starch EAA esterFIG. 1. Diagram showing chemical derivatization of cornstarch.

Equations 1, 2, and 3 show end product as 2-0-butyl-substituted,palmitate-substituted, and EAA-substituted starches, respectively.DMSO, Dimethyl sulfoxide.

mglml) for 1 h and subsequently mixed with starch suspen-sion and tested for adhesion activity. The sugars tested inthese experiments were glucose, fructose, galactose, malt-ose, maltoheptose, melibiose, inositol, mannose, galacto-uronic acid, glucouronic acid, methyl-a-D-galactopyranosidemonohydrate, and 6-methyl glucoside.

Chemically treated starches. Cornstarch treated with HCl-ethanol, HCl-methanol, HCl-butanol, and HCl-propanol was

prepared by the method described by Ma and Robyt (19) andwas kindly provided by John F. Robyt, Iowa State Univer-sity, Ames.

Starch derivatization. Starches were chemically modifiedby standard organic chemistry techniques. Briefly, the 2-0-butyl starch ether was prepared with n-butyliodide (99%;Aldrich Chemical Co., Milwaukee, Wis.) and commercialcornstarch (A.E. Staley Mfg. Co., Decatur, Ill.). Typically,the reaction was catalyzed by NaOH in 2-propanol saturatedwith Na2SO4. The reaction was carried out at 50°C for 40 to60 h. The product was isolated by neutralization (glacialacetic acid, pH 6.0), filtered, and washed several times withwater and then sequentially washed with 80%, 90%, andabsolute ethanol. A final wash with acetone and drying invacuo at 40°C yielded the product (Fig. 1).The 2-0-(2-hydroxybutyl) starch was synthesized by using

a modification of the method of Kesler and Hjermstad (17).Starch as described above was treated with 1,2-epoxybutaneat 45°C for 60 h. The product was isolated in a manner

VOL. 57, 1991

1130 IMAM AND HARRY-O'KURU

30 60 90

Incubation Time (min)

120 0 10 20 30 40 50 0 2 4 6 8 10 12

Temperature ( C) pH

FIG. 2. Adhesion of 35S-labeled cells of L. amylovorus incubated with starch granules in suspension for various (A) lengths of time andat various (B) temperatures and (C) pHs.

similar to that of O-butyl starch above. Starch palmitate was

prepared (Fig. 1) by reaction of palmitoyl chloride withstarch in the presence of pyridine as an acid scavenger andtoluene as diluent (9).The starch EAA ester was prepared in two steps (Fig. 1).

Initially EAA was refluxed with acetic anhydride to give themixed acetyl polyethylene-co-acrylic anhydride. The latterwas then reacted with starch to yield starch-polyethylene-co-acrylate.

RESULTSThe number of 35S-labeled L. amylovorus cells bound to

cornstarch granules increased with time, reaching a maxi-mum of 60 to 75% in 30 min, after which bound radioactivitydecreased to about 50% and remained stable for up to 3 h(Fig. 2A). In an identical experiment with heat-killed cells,only 10 to 15% of the cells bound to starch granules, yieldingno such peak as seen with the live cells in Fig. 2A (data notshown). Maximum binding of bacteria to starch granules

occurred at 24°C (Fig. 2B) and pH 6.0 (Fig. 2C). Similarresults were obtained when a suspension of L. amylovoruscells was incubated with cornstarch granules for 30 min at24°C, pH 6.0, and a portion of the incubation mixture was

Gram stained and examined by light microscopy. Resultsindicated that a substantial portion of the cells were foundassociated with the starch granules (Fig. 3).

Interestingly, binding of L. amylovorus cells to granulesincreased proportionally with the concentration of starchpresent in the incubation mixture (Fig. 4) as well as with theamylose content of the cornstarch (Fig. 5). Bound cells couldbe removed from the granules by either high ionic strength or

low pH. About 91 and 69% of the bound cells were removedfrom the granules by 2 M MgCl2 (pH 7.0) and 1 M glycine(pH 2.5), respectively (Table 1). Glutaraldehyde (2%) alsoremoved 60% of the cells bound to the starch granules (Table1).

Cells killed by heating in a boiling water bath for 3 minshowed significantly reduced binding to starch granules

FIG. 3. Binding of cells of L. amylovorus to starch granules in suspension. A portion of the sample was Gram stained and observed undera light microscope.

' 800mA 60a

o 400

20

B

I/,LJ L

APPL. ENVIRON. MICROBIOL.

A

---i

c

ll--.

ADHESION OF L. AMYLOVORANS TO CORNSTARCH GRANULES 1131

3lax

c

0

m

is

70

60-s65o40

30 -l/

20

10

0 10 20 30 40 50

% Cornstarch (W/V)FIG. 4. Adhesion of 35S-labeled cells of L. amylovorus to sus-

pension of various concentrations of cornstarch.

(Table 2). Similar results were obtained when cells were

fixed with either 2% glutaraldehyde or 2% formaldehyde(Table 2), suggesting that live cells are apparently more

liable to adhere to starch granules than heat-killed or meta-bolically inactive cells. Treatment of bacteria with trypsinalso significantly reduced their binding to cornstarch gran-

ules (Table 2), suggesting that adhesive interactions betweenbacterium and cornstarch granules may be mediated bybacterial cell surface proteins or glycoproteins.

Preincubation of L. amylovorus cells with some simplesugars did not inhibit cell binding to granules (data notshown). However, cells preincubated with 1% soluble potatostarch showed substantially reduced binding to cornstarchgranules (Table 2).

Cornstarch granules treated with HCl-ethanol, HCl-meth-anol, HCl-butanol, or HCl-propanol exhibited reduced bind-ing compared with untreated control cornstarch (Table 3).Among derivatized starches, 33% more cells were bound to2-O-butyl starch than to control cornstarch (Table 4). In thisderivatization, the hydrogen of the C-2 hydroxyl group of theanhydroglucose unit of the starch molecule is substitutedwith a butyl group (Fig. 1). On the other hand, significantlyfewer cells were bound to starch-palmitate (Table 4), inwhich the palmitate group is substituted essentially at thesame position of the starch molecule (Fig. 1). Inhibition ofcell adhesion appears to be consistent regardless of thedegree of palmitate substitution (degree of substitution 0.48or 0.09) on the starch molecule. The 2-0-(2-hydroxybutyl)

100

m 800

4B0

06

m 20

0

% Amylose In Starch

FIG. 5. Adhesion of I'S-labeled cells of L. amylovorus to sus-

pension of cornstarch granules containing variable amounts of

amylose.

TABLE 1. Removal of bound cells of L. amylovorus fromcornstarch granules, using various buffers

% Radioactivity % Bound radioactivityElution buffer bound to starch eluted off starch

granulesa granulesa

2 M MgCl2, pH 7.0 60 911 M glycine, pH 2.5 60 692% formaldehyde 60 60

a Average of duplicate samples (<5% variations).

starch and starch-EAA ester had little, if any, effect on thebacterium's ability to adhere to these starches (Table 4).

In additional experiments, cornstarch granules were di-gested for 18 h separately with commercial a-amylase prep-arations of Bacillus licheniformis type XII-A, porcine pan-creas type VI-A, barley malt type VIII-A, Aspergillus oryzaetype X-A, Bacillus sp. type II-A, and a crude enzymepreparation (concentrated and dialyzed) obtained from theculture supernatant of L. amylovorus grown in this labora-tory. The starch digestion was varified by measuring therelease of reducing sugars in the supernatants (data notshown). The light microscopic examination of starch sam-ples incubated for 18 h with enzymes revealed that granuleswere still intact, and there were no apparent morphologicalchanges in granular structure resulting from enzyme action(data not shown). When enzyme-treated starches were ex-amined for adherence of L. amylovorus cells, results indi-cated no effect on the bacterium's ability to bind to suchstarches (data not shown), suggesting that the cell bindingsites remain intact in the starch granule even after 18 h ofenzyme treatment. Previous data in which partially digestedgranules were colonized by L. amylovorus cells (16) alsoappear to substantiate these results.

DISCUSSION

Our laboratory is interested in studying bacterial attach-ment and degradative properties with respect to insolublebiopolymer matrices such as starch and cellulose (10, 13-15).As the application of starch as a raw material in biotechno-logical processes is becoming increasingly important, thereexists a need to understand more fully the initial events insurface interactions between amylolytic microorganisms andstarch matrix. Data presented here show that cells of L.amylovorus bind substantially to an insoluble cornstarchmatrix (Fig. 2). Proportionally more cells bind when theconcentration of either starch or amylose in the starch isincreased (Fig. 4 and 5, respectively). The binding interac-tions between the bacterium and starch granules appear toinvolve bonds that can be disrupted by either high ionicstrength or low pH (Table 1). Binding to starch granules was

TABLE 2. Effect of various cell treatments on cell adhesion

Treatment Concn (wt/vol) oRadioactivity

Heat killed 15.0 ± 3.0Glutaraldehyde 2% 19.0 ± 5.5Formaldehyde 2% 27.5 ± 10.6Sodium azide 0.02% 24.0 + 5.0Trypsin 1 mg/ml 21.6 ± 9.6Soluble potato starch 1% 9.0 + 5.5Control (untreated) 65.0 ± 2.0

VOL. 57, 1991

1132 IMAM AND HARRY-O'KURU

TABLE 3. Adhesion of L. amylovorus to chemically treatedcornstarch granules

Treatment % Radioactivitybound ± SD

HCl-ethanol........ 35.0 ± 1.6HCl-methanol........ 37.0 ± 6.4HCl-butanol........ 40.0 ± 1.5HCl-propanol........ 37.0 ± 3.5Untreated........ 60.0 + 7.0

significantly reduced in heat-killed cells or when the cellshad been exposed to metabolic inhibitors or fixed withformaldehyde and glutaraldehyde (Table 2), suggesting thatlive and perhaps metabolically active cells are more likely tobind cornstarch granules. Trypsin treatment of the cells alsoinhibited their ability to adhere to starch granules, indicatingthat a cell-surface protein(s) or glycoproteins may be in-volved in binding. In many systems it has been reported thatthe cell surface recognition is mediated by carbohydrate-binding proteins (3, 7, 11, 12, 25, 26).

Cells of L. amylovorus pretreated with several simplesugars prior to their exposure to starch granules did notresult in inhibition of binding to starch, but cells pretreatedwith 1% soluble potato starch were unable to adhere to cornstarch granules (Table 2). These results suggest that starchbinding sites on the bacterial cell surface recognize poly-meric carbohydrate substrates rather than simple sugars.That the L. amylovorus cells bind normally to cornstarchgranules that have been acted upon by commercial ot-amy-lases for 18 h indicates that the cell binding site on thegranular surface is distinct from that of enzyme active sitesand the two activities may be different from each other.Furthermore, because no enhancement of cell adhesion wasobserved in starch granules predigested by amylase, limitedenzymatic digestion of cornstarch granule did not exposeany additional cell-binding sites on the granule surface. Inthis regard, Anderson and Salyers (1) showed that starchbreakdown by Bacteroides thetaiotaomicron involves boththe starch-binding sites localized on the bacterial outermembrane and the starch-degrading enzymes located in theperiplasm of the cell. It was suggested that, in Bacteroidessp., amylolytic enzymes are not secreted extracellularly, andthe binding of the starch molecule to the bacterial cellsurface appears to be the first step in passing the moleculethrough the outer membrane into the periplasmic space.Because chemical treatments of starch (HCl-ethanol, HCl-

methanol, HCl-butanol, and HCl-propanol) can (i) desiccateand precipitate the starch polymer, (ii) disrupt the intermo-lecular hydrogen bonding between the starch molecules, and(iii) reduce the polymer chain length, it is not surprising tofind that L. amylovorus cells showed reduced binding tochemically treated starches (Table 3).

Replacing the hydroxyl hydrogen at C-2 of the starchmolecule with an n-butyl group enhanced the binding of L.amylovorus cells (Table 4). This starch derivative has beenobserved to exhibit a tendency to swell in cold water,producing an increased surface area for binding. The loca-tion of the n-butyl group would also disrupt intramolecularhydrogen bonding between contiguous anhydroglucoseunits, exposing more cell-binding sites on the starch mole-cule for bacterial attachment. Palmitate substitution ofstarch, on the other hand, inhibited cell adhesion (Table 4),probably because the hydrophobic palmitate chain interferessterically with binding sites on the surface of the starch

TABLE 4. Adhesion of L. amylovorus to derivatized starches

Derivatized starch' o% RadioactivityDerivatizedstarch" ~~~~~~bound ±SD

2-0-Butyl............... 82.0 ± 0.52-0-(2-Hydroxybutyl) ................,,,,,,,,,,.58.0 ± 1.32-Palmitoyl (DS, 0.48)............... 32.7 ± 172-Palmitoyl (DS, 0.09)............... 21.0 ± 3.9EAA-ester............... 48.6 ± 4.9Cornstarch (control) ............... 60.0 ± 5.0

a DS, Degree of substitution.

granule. The differences in the binding activities of the L.amylovorus cells to raw cornstarch and to chemically treatedas well as some derivatized cornstarch may in part be due tothe resulting substrates having significantly different surfacemorphology, in contrast to the amylase-treated granuleswhich have the same binding activity as raw cornstarchgranules.

ACKNOWLEDGMENTS

We thank Timothy D. Leathers and Frederick R. Dintzis forcritically reviewing the manuscript and Mary Kinney, ChristineKowalczyk, and Holly Dickinson for technical assistance. We aregrateful to Deborah Bitner for clerical assistance

REFERENCES1. Anderson, K. L., and A. A. Salyers. 1989. Biochemical evidence

that starch breakdown by Bacteroides thetaiotaomicron in-volves outer membrane starch-binding sites and periplasmicstarch-degrading enzymes. J. Bacteriol. 171:3192-3198.

2. Baier, R. E., and L. Weiss. 1975. Demonstration of involvementof absorbed proteins in cell adhesion and cell growth. Adv.Chem. Ser. 145:300-307.

3. Bar-Shavit, Z., I. Ofek, R. Goldman, and N. Sharon. 1977.Mannose residues on phagocytes as receptors for attachment ofE. coli and Salmonella typhi. Biochem. Biophys. Res. Com-mun. 78:455-460.

4. Bhat, S., R. J. Wallace, and E. R. 0rskov. 1988. Study ofrelationship between straw quality and its colonization byrumen microorganisms. J. Agric. Sci. 110:561-565.

5. Bhat, S., R. J. Wallace, and E. R. 0rskov. 1990. Adhesion ofcellulolytic ruminal bacteria to barley straw. Appl. Environ.Microbiol. 56:2698-2703.

6. Cheng, K.-J., C. S. Stewart, D. Dinsdale, and J. W. Costerton.1983. Electronmicroscopy of bacteria involved in the digestionof plant cell walls. Anim. Feed Sci. Technol. 10:93-120.

7. Dazzo, F. B. 1981. Bacterial attachment as related to cellularrecognition in the rhizobium-legume symbiosis. J. Supramol.Struct. Cell Biochem. 16:91-103.

8. Doane, M. W. 1988. New industrial uses of corn, p. 87-99. InProceedings of the 42nd Annual Corn & Sorghum ResearchConference, part II. Corn.

9. Farbenind, I. G., and A.-G. German. 1923. Starch esters. Chem.Abstr. 24,962.

10. Gould, J. M., S. H. Gordon, L. B. Dexter, and C. L. Swanson.1990. Starch-containing biodegradable plastics. Am. Chem.Soc. Symp. Ser. 433:65-75.

11. Imam, S. H., R. F. Bard, and T. R. Tosteson. 1981. Involvementof carbohydrate-binding macromolecules in the intraspecificmarine microbial association, p. 580-584. In Proceedings of theOceans 81 Conference, vol. 1. I.E.E.E. Publication no.81chl685-7. Institute of Electrical and Electronic Engineers,New York.

12. Imam, S. H., R. F. Bard, and T. R. Tosteson. 1984. Specificity ofmarine microbial surface interactions. Appl. Environ. Micro-biol. 48:833-839.

13. Imam, S. H., and J. M. Gould. 1990. Adhesion of an amylolyticArthrobacter sp. to starch-containing plastic films. Appl. Envi-ron. Microbiol. 56:872-876.

APPL. ENVIRON. MICROBIOL.

ADHESION OF L. AMYLOVORANS TO CORNSTARCH GRANULES

14. Imam, S. H., R. V. Greene, and H. L. Griffin. 1990. Adhesiveproperties of a symbiotic bacterium from a wood-boring marineshipworm. Appl. Environ. Microbiol. 56:1317-1322.

15. Imam, S. H., and W. J. Snell. 1987. Degradation of the frame-work of the Chlamydomonas cell wall by proteases present in a

commercially available a-amylase preparation. Appl. Environ.Microbiol. 53:1701-1704.

16. Imam, S. H., A. Burgess-Cassler, G. L. C6te, S. H. Gordon, andF. L. Baker. Curr. Microbiol., in press.

17. Kesler, C. C., and E. T. Hjermstad. 1964. Hydroxyethyl- andhydroxypropyl starch. Methods Carbohyd. Chem. IV(71):304-306.

18. Koch, H., and H. von Ropper. 1988. New industrial productsfrom starch. Stareke 40:121-131.

19. Ma, W.-P., and J. F. Robyt. 1987. Preparation and characteri-zation of soluble starches having different molecular sizes andcompositions, by acid hydrolysis in different alcohols. Carbo-hydr. Res. 166:283-297.

20. Nakamura, L. K. 1981. Lactobacillus amylovorus, a new starch-hydrolyzing species from cattle waste-corn fermentations. Int.J. Syst. Bacteriol. 31:56-63.

21. Otey, F. H., R. P. Westoff, and W. M. Doane. 1987. Starch-based blown films 2. Ind. Eng. Chem. Res. 26:1659-1663.

22. Patterson, H., R. Irvin, J. W. Costerton, and K.-J. Cheng. 1975.Ultrastructure and cell adhesion properties of Ruminococcusalbus. J. Bacteriol. 122:278-287.

23. Pearl, H. W. 1976. Specific association of the blue-green algaeAnabaena aplanizomenon with bacteria in freshwater blooms.J. Phycol. 12:431-435.

24. Shogren, R. L., A. R. Thompson, R. V. Greene, S. H. Gordon,and G. L. Cote. J. Appl. Polymer Sci., in press.

25. Simpson, D. L., S. D. Rosen, and S. H. Barondes. 1974.Discoidin, a developmentally regulated carbohydrate-bindingprotein from Dictyostelium discoidin. Purification and charac-terization. Biochemistry 13:3487-3493.

26. Tosteson, T. R., and W. A. Corpe. 1976. Enhancement ofadhesion of marine Chlorella vulgaris to glass surface. Can. J.Microbiol. 21:1025-1031.

27. von Roper, H., and H. Koch. 1988. New carbohydrate deriva-tives from biotechnical and chemical processes. Stareke 40:453-464.

VOL. 57, 1991 1133