ELSEVIER International Journal of

Food Microbiology 30 (1996) 243-253

Biochemical characteristics of Enterobacter agglomerans and related strains found in

buckwheat seeds

Kazuo Iimuraa, Akiyoshi Hosonob,*

“Motojiya Co., Ltd., Matsuntoto 390, Japan bFaculty of Agriculture, Shinshu University, Nagano-Minamiminowa 399-45, Japan

Received 9 March 1995; revised 19 June 1995; accepted 25 September 1995

Abstract

Thirty strains of bacteria were randomly isolated and identified from buckwheat seeds. The phenotypic characteristics of these strains agree well with those of the Enterobacter agglomerans-Erwinia herbicola complex. On the basis of the difference in indole production and gas production from D-glucose, the isolates were divided into 3 phenotypic groups, viz. I, II and III. Twenty two strains were in phenotypic group I, which is negative for indole production and gas production from D-glucose, and resembles Pantoea agglomerans. All six strains in phenotypic group II, which is positive for indole production and negative for gas production from D-glucose, were identified as Erwinia ananas. Two strains in phenotypic group III, which is negative for indole production and positive for gas production from D-glucose, were identified as Rahnella aquatilis.

Keywords: Enterobacter agglomerans; Erwinia ananas; Pantoea agglomerans; Rahnella aquatilis; Buckwheat seed

* Corresponding author.

0168-1605/96/$15.00 0 1996 Elsevier Science B.V. All rights reserved

PII SO1 68- 1605(96)00949-X

244 K. Iimura, A. Hosono / ht. J. Food Microbiology 30 (1996) 243-253

1. Introduction

Buckwheat (Fagopyrum esculentum), a native of central Asia, is a minor grain crols, and has traditionally been used in Japan to produce flour mainly for making buckwheat noodles. Substantial research on the bacterial flora on buckwheat seeds was performed by several investigators (Obinata et al., 1987; Miyoshi et al., 1988; Naito and Shimizu, 1986). We have reported that bacteria on buckwheat seeds harvested in Nagano and Hokkaido, Japan, were in the range of 105-10*/g, and have identified Enterobacter agglomerans as one of the dominant species in these samples (Iimura and Hosono, 1393; Iimura and Hosono, 1994).

E. agglomerans (i.e., synonymous with Erwinia herbicola, Erwinia uredo-vora and Erwinia stewartii (Dye, 1969a; Sakazaki et al., 1976; Beji et al., 1988)) is a heterogeneous species, and is considered to be associated with plant materials such as saprophytes or pathogens, and with animal as well as human reservoirs (Dye, 1969b; Muraschi et al., ,1965). Different taxonomic views have resulted in an ambiguous situation: phytopathologists prefer to use the nomenclature of Dye, 1969a, i.e., Er. herbicola, whereas clinical bacteriologists use the nomenclature of Ewing and Fife, 1972, i.e., E. agglomerans. Although many rearrangements (Sakazaki et al., 1976; Dye, 1981) have been suggested for the ‘Erwinia herbicola- Enterobacter agglomerans complex’, a consensus has not been reached (Beji et al., 1988). Ewing and Fife, 1972 investigated the biochemical characteristics of E. agglomerans and classified it into 11 biogroups. Recently, based on total DNA homology and electrophoretic protein pattern similarities, some strains of E. agglomerans and E. herbicola, including the two type strains, were proposed to form a new genus called Pantoea (Gavini et al., 1989). Referring to Ewing and Fife, 1972 and Gavini et al., 1989, we investigated the biochemical characteristics of E. agglomerans and related strains isolated from buckwheat seeds.

2. Materials and methods

2.1. Buckwheat seeds

Buckwheat seeds harvested in Nagano (4 samples) and Hokkaido (2 samples), Japan, were used for the isolation of E. agglomerans strains. The seed samples were kept in a refrigerator (4°C) and used for the isolation within one month after harvesting.

2.2. Isolation of E. agglomerans and related glucose fermentative Gram-negative bacteria

Methods for the isolation of E. agglomerans and related strains were essentially the same as those described in our previous paper (Iimura and Hosono, 1993). In brief, 90 ml of sterile physiological saline was added to 10 g of ground buckwheat sample. They were treated in a blender (Stomacher Lab-Blender 400, Seward

K. Iimura, A. Hosono / ht. J. Food Microbiology 30 (1996) 243-253 245

Medical Co., Ltd., London, UK) for 1 min, followed by filtration with sterilized gauze. A decimal dilution of the filtrate was plated on ‘Eiken’ trypto-soy agar (TSA) (Eiken Chemical Co., Ltd., Tokyo, Japan) and incubated at 30°C for 48 h. About ten colonies per sample were randomly isolated. These cultures were purified, growing on the same medium at 30°C for 48 h. The isolates were examined for Gram reaction and glucose fermentation. Gram-negative glucose fermentative bacteria were identified by use of API 20E (Bio Merieux S.A., Marcy-I’Etoile, France). Biochemical properties of the strains identified as Enterobacter agglomer- ante were further examined.

2.3. Biochemical tests for E. agglomerans and related strains isolated

2.3.1. Catalase and oxidase tests Catalase production was detected by the production of bubbles in a 3% hydrogen

peroxide solution at 30°C. The oxidase test was carried out by use of cytochrome oxidase test paper (Nissui Pharmaceutical Co., Ltd., Tokyo, Japan) encompassing Pserudomonas aeruginosa IF0 12689 as a positive control and Escherichia coli JCM 1649 as a negative control.

2.3.2. Gas formation Gas formation from glucose was detected with Durham tubes in Bacto OF basal

medium (Difco Laboratories, Detroit, Mich.) incorporated at 1 .O% D-glucose. Incubation was carried out at 30°C for 4 days.

2.3.3. Motility and indole production This was determined using ‘Nissui’ lysine indole motility medium (LIM medium).

The organisms were incubated at 30°C in prepared tubes of LIM medium by stabbing. Motility was checked by observing a clouding of the medium or growth extension from the inoculation line following incubation for 48 h. The presence of indole was detected with Kovacs’ P-dimethyl-amino-benzalaldehyde reagent.

2.3.4. Pigmentation The organisms were grown on ‘Nissui’ nutrient agar (nutrient broth solidified

with 1.5% agar, pH 7.2). Growth and pigment production were observed after incubation at 30°C for 4 days.

2.3.5. Growth at 4”C, 41°C and 44°C One drop of a light suspension of organisms in distilled water was added to

‘Nissui’ nutrient broth filled to a depth of about 5 cm in test tubes and incubated in a water bath at 4”C, 41°C and 44°C for 2 to 7 days.

2.3.6. Acid production of sugars and related compounds The following compounds were incorporated at 1.0% in ‘Eiken’ brom cresol

purple (BCP) semisolid medium: D-glucose, L-arabinose, D-ribose, D-xylose, D- fructose, D-mannose, L-rhamnose, D-cellobiose, D-galactose, lactose, maltose,

246 K. Iimura, A. Hosono / ht. J. Food Microbiology 30 (1996) 243-253

melibiose, sucrose, trehalose, melezitose, raffinose, glycerol, meso-erythritol, adonitol, duJcito1, inositol, D-mannitol, D-sorbitol, or-methyl-D-glucoside, salicin, inulin and starch. Gas formation at 30°C was detected with Durham tubes over a period of 30 days.

2.3.7. Reduction of nitrate This was determined by growing the organism in ‘Nissui’ nutrient broth contain-

ing 0.1% of KNO, at 30°C for 2, 3 or 7 days and testing for the presence of nitrite with dimethyl a-naphtylamine-sulphanilic acid reagent.

2.3.8. H,S production Cultures were grown in ‘Nissui’ Kligler iron agar at 30°C for 48 h; H2S was

detected by blackening of the medium.

2.3.9. Methyl red (MR) and Voges-Proskauer (V-P) tests Cultures were grown at 30°C in Bacto MR-VP medium (Difco) containing

glucose, 0.5%; buffered peptone, 0.7%; K,HPO,, 0.5% for 48 h. Acetylmethyl- carbinol was detected using the Barritt modified method (Sakazaki et al., 1988).

2.3.10. Liquefaction of gelatin A medium containing beef extract (Kyokuto Pharmaceutical Co., Ltd., Tokyo,

Japan), 1%; peptone (Nissui), 1%; NaCl, 0.5%; gelatin (Difco), 12%; pH 7.0; sterilized at 121°C for 15 min, was dispensed to a depth of 5 cm in test tubes and bacterial strains were inoculated by stabbing. Cultures were incubated at 30°C for 30 days. Liquefaction was observed after chilling tubes at 4°C after incubation.

2.3.1 I. Urease test Christensen’s urea agar (Christensen, 1946) was used with incubation for 48 h.

The same medium without urea was used as a control for avoiding misjudgment.

2.3.12. Phenylalanine deaminase test Phenylalanine agar (Ewing et al., 1957) was used with incubation at 30°C for 48

h. The production of phenylpyruvic acid (by deaminase) was indicated by the formation of the characteristic green color upon the addition of an acidified ferric chloride solution to the culture.

2.3.13. Decarboxylase test Moeller’s medium (Moeller, 1955) containing 0.3% agar was employed, with

bromothymol blue replacing bromocresol purple indicator. Results were taken after 4 days of incubation at 30°C.

2.3.14. Deoxyribonuclease (DNase) test Plates .of ‘Eiken’ DNA agar were heavily spotted with loopfuls of organisms from

nutrient agar and incubated at 30°C for 48 h. DNase production was detected by flooding plates with n-HCl and observing zones of clearing around growth.

K. Iimura, A. Hosono / Int. J. Food Microbiology 30 (1996) 243-253 241

2.3.15. Malonate utilization test Malonate utilization was determined using ‘Eiken’ malonate broth. Observations

were made at 30°C over 48 h.

2.3.16. Citrate utilization test Citrate utilization was determined using Bacto Simmons citrate agar (Difco).

Observations were made at 30°C over 4 days.

2.3.17. Growth in KCN Test tubes (16-150 mm), filled to a depth of 5 cm with ‘Eiken’ KCN broth base

containing the same volume of 0.015% KCN solution were each inoculated with a loopful of a 24 h nutrient broth culture, tightly closed with rubber stoppers and observed for growth at 30°C over 48 h. Growth was indicated by a clouding of the medium.

2.3.18. Digestion of es&in This was examined by the methods described by Sakazaki (Sakazaki et al., 1988).

2.3.19. Kauffmann-Petersen’s organic acid utilization test Kauffmann-Petersen’s test was examined by the methods described by Sakazaki

(Sakazaki et al., 1988). p-Galactosidase was determined using an ONPG (o-nitro- phenyl-l)-galactosidase) Disc (Nissui).

3. Results and discussion

We used the API 20E system to clarify the taxonomic position of the strains isolated from buckwheat seed samples. Table 1 shows the API 20E system codes of 30 strains isolated. The API 20E system codes 1005132 (strain T-33), 1005133 (strains T-3, T-6, T-8, T-12, T-17, T-18, T-21, T-24 and T-34) and 1205132 (strains T-11 and T-35) were interpreted with the API data base (API system 3rd edition) as E. agglomerans. Codes 1005332 (strain T-22), 1005333 (T-l, T-2, T-15 and T-19) and 1045573 (strain T-9) were assigned to E. agglomerans or to Erwinia sp. Codes 1045773 (strain T-14) and 1245773 (strain T-4, T-10, T-23 and T-32) were assigned to E. agglomerans, Erwinia sp., Klebsiella oxytoca or Klebsiella planticola, whereas code 1205573 (strains T-5, T-36) was assigned to Rahnella aquatilis. However, the API 20E system code 1205133 (strains T-7, T-13, T-16, T-20 and T-31) was not in this data base.

Table 2 shows phenotypic characteristics of the 30 strains isolated. These strains showed positive reactions for the following productions: catalase, anaerobic growth, motility at 30°C gelatin liquefaction, acetoin from D-glucose and a-galac- tosidase, and acids from D-glucose, L-arabinose, D-ribose, D-xylose, D-fructose, D-mannose, L-rhamnose, D-galactose, maltose, sucrose, trehalose and D-mannitol. The negative results were for cytochrome oxidase, urease, arginine dihydrolase, lysine and ornithine decarboxylase, and H,S. These results agree well with those of the E. agglomerans-E. herbicola complex. (Brenner et al., 1984).

248 K. Iimura, A. Hosono / ht. J. Food Microbiology 30 (1996) 243-253

On the basis of the differences in indole production and gas production from D-glucose, the isolates were divided into 3 phenotypic groups, viz., I, II and III. Out of the 30 strains isolated, 22 strains (73.3% of the total isolates) were negative for indole production and gas production from D-glucose, which were grouped in phenotype I. The API 20E system codes of these strains were 1005132, 1005133, 1005332, 1005333, 1205132 and 1205133. These codes were characterized by the modes of citrate utilization, and of acid production from amygdalin and inositol as illustrated in Fig. 1. However, all the strains involved in this group were negative for acid production on the BCP semisolid medium and positive for citrate utiliza- tion on the Bacto Simmons citrate agar (Table 2). Many taxonomical data have been published on the E. agglomerans-E. herbicola complex (Gavini et al., 1983; Brenner et al., 1984; Verdonck et al., 1987; Beji et al., 1988). Recently, Gavini et al.,

Table 1 Enterobucter ugglomerans isolated from buckwheat seeds and their API 20E profile numbers

Strain number Buckwheat harvested API 20E profile

Place Year

T-33 T-3 T-6 T-8 T-12 T-17 T-18 T-21 T-24 T-34 T-22 T-l T-2 T-15 T-19 T-11 T-35 T-7 T-13 T-16 T-20 T-31 T-9 T-14 T-4 T-10 T-23 T-32 T-5 T-36

Hata, Nagano 1991 1005132 Azumi, Nagano 1991 1005133 Kawahigasi, Hokkaido 1991 1005133 Kawahigasi, Hokkaido 1991 1005133 Uryu, Hokkaido 1991 1005133 Uryu, Hokkaido 1991 1005133 Uryu, Hokkaido 1991 1005133 Matsumoto, Nagano 1991 1005133 Matsumoto, Nagano 1991 1005133 Hata, Nagano 1991 1005133 Matsumoto, Nagano 1991 1005332 Azumi, Nagano 1991 1005333 Azumi, Nagano 1991 1005333 Uryu, Hokkaido 1991 1005333 Uryu, Hokkaido 1991 1005333 Kawahigasi, Hokkaido 1991 1205132 Hata, Nagano 1990 1205132 Kawahigasi, Hokkaido 1991 1205133 Uryu, Hokkaido 1991 1205133 Uryu, Hokkaido 1991 1205133 Uryu, Hokkaido 1991 1205133 Saku, Nagano 1990 1205133 Kawahigasi, Hokkaido 1991 1045573 Uryu, Hokkaido 1991 1045773 Azumi, Nagano 1991 1245773 Kawahigasi, Hokkaido 1991 1245773 Matsumoto, Nagano 1991 1245773 Saku, Nagano 1990 1245773 Azumi, Nagano 1991 1205573 Hata, Nagano 1991 1205573

Tab

le

2 Pe

noty

pic

char

acte

rist

ics

of E

nter

ohac

ter

uggl

omer

uns

and

rela

ted

stra

ins

isol

ated

fr

om

buck

whe

at

seed

s

Cha

ract

eris

tics

Gro

wth

at

4°C

Gro

wth

at

41°

C

Gro

wth

at

44°

C

Prod

uctio

n of

yel

low

pig

men

t M

otili

ty

Cat

alas

e C

ytoc

hroo

mox

idas

e K

CN

(g

row

th)

Gel

atin

liq

uefa

ctio

n In

dole

pr

oduc

tion

Vog

es-P

rosk

auer

re

actio

n N

itrat

e re

duce

d to

nitr

ite

H,S

pr

oduc

tion

Hyd

roly

sis

of e

scul

in

Gas

pro

duct

ion

from

D

-glu

cose

A

rgin

ine

dihy

drol

ase

(Moe

ller)

L

ysin

e de

carb

oxyl

ase

Om

ithin

e ,d

ecar

boxy

lase

Ph

enyl

alan

ine

deam

inas

e D

eoxy

ribo

nucl

ease

p-

Gal

acto

sida

se

Ure

ase

Util

izat

ion

of:

Citr

ate

(Sim

mon

s)

Mal

onat

e D

-Tar

trat

e (K

-P)

Mut

ate

(K-P

) A

cid

prod

uced

fr

om:

L-A

rabi

nose

Phen

otyp

e I

Pon

toea

ag

glom

eran

s (2

2 Ph

enot

ype

II E

rwin

ia

anan

as

(6

stra

ins)

R

eact

ion

stra

ins)

R

eact

ion

(+)

(33)

_

_ _

_

+

+

+

+

+

+

_ -

95

83

+

+

_ +

+

+

95

_

_ -

91

61

- _

_ _

_ _

_ _

+

- -

17

+

+

_ _

+

+

91

_

5 _

_ _

+

+

Phen

otyp

e II

I R

ahne

lla

aqua

tili

s (2

str

ains

) R

eac-

tio

n h

Tab

le 2

(co

ntin

ued)

Cha

ract

eris

tics

Phen

otyp

e I

Pun

toea

ag

glom

eran

s (2

2 Ph

enot

ype

II E

rwin

ia a

nanu

s st

rain

s)

Rea

ctio

n (6

str

ains

) R

eact

ion

Phen

otyp

e II

I R

ahne

lla

aqua

tili

s (2

str

ains

) R

eac-

tio

n 3

D-R

ibos

e D

-Xyl

ose

D-F

ruct

ose

D-G

luco

se

D-t

iann

ose

L-R

ham

nose

D

-Cel

lobi

ose

D-G

alac

tose

L

acto

se

Mal

tose

M

elib

iose

Su

cros

e T

reha

lose

M

elez

itose

R

afko

se

Gly

sero

l m

eso-

Ery

ther

itol

Ado

nito

l D

ulci

tol

Inos

itole

D

-Man

nito

l D

-sd’

rbito

l a-

Met

hyl-

D-g

luco

side

Sa

licin

In

ulin

St

arch

+

+

+

k?

+

+

+

2 “3

+

+

+

+

+

.k

+

+

+

+

9 +

+

+

P 0

- +

+

.

+

f +

$

t +

4

+

+

+

3 -

83

+

B

+

+

+

+

+

+

3 3 -

- -

&

-..

5 67

s’

o

a ij-

18

f P

- -

- ti

- -

T=

-

-t

8 -

83

-

+

+

+

k -

+

+

‘;-

- -

!z

86

83

+

- -

- -

- -

+,

all

stra

ins

posi

tive;

-,

all

stra

ins

nega

tive;

nu

mbe

r in

dica

ted

perc

enta

ge

of p

ositi

ve

stra

ins;

0,

de

laye

d re

actio

n.

K. Iimura, A. Hosono / ht. J. Food Microbiology 30 (1996) 243-253 251

Entervtmcter zggkmem idated from buckvdheat Seeds

zk-

T-5 indole production

756, + _ _ _ _ - _,

phenwpic group III

CIT + CIT - CIT - IN0 + IN0 + IN0 -

1245773 1045773 1045573

T:?0 T-14 T-9

T-23 T-32

_----~--~~~_______, phenotypic group II

1205133 1205132 1005333 1005332 1005133 1005132

T-7 T-l 1 T-l T-22 T-3 T-33 T-l 3 T-35 T-2 T-6 T-l 6 T-l 5 T-8

T:E T-19 T-12

T-1’:, T-21 T-24 T-34

c_______________________________________-_____, phenotypic group I

Fig. 1. API 20E profile and phenotype of Enterobacter agglomerans isolated from buckwheat seeds. API 20E reaction: CIT, utilization of citrate; AMY, acid production from amygdalin; INO, acid production from inositol. API 20E was performed at 30°C for 24 h, and results were coded by seven-digit profile numbers as described by the manufactuer.

1989 analyzed the E. agglomerans-E. herbicola complex on the basis of phenotypic characteristics and DNA-DNA hybridization data and proposed the transfer of E. agglomerans to Pantoea gen. nov. as Pantoea agglomerans comb. nov. Referring to the phenotypic characteristics of P. agglomerans (Gavini et al., 1989), we recognized that most of the phenotypic characteristics of the strains of phenotypic group I were very similar to P. agglomerans, but with a few exceptions, viz. in production of ornithine decarboxylase and in acid production from lactose and D-cellobiose.

As shown in Table 1 and Fig. 1, tests in the API 20E system gave the results that the strains (T-4, T-9, T-10, T-14, T-23 and T-32) in phenotypic group II, which is positive for indole production and negative for gas production from D-glucose, were identified either as E. agglomerans or as Erwinia sp. Mergaert et al., 1984 reported that a strain having the code 1245773 could be interpreted with the API data base as E. ananas or E. uredovora, taking into account that the strain was positive for indole production and negative for gas production from D-glucose. Furthermore, Beji et al., 1988 and Gavini et al., 1989 reported that E. ananas showed positive reactions for the following: yellow pigment and indole, and acids from D-cellobiose, lactose, maltose, melibiose, raffinose, sucrose, sorbitol and salicin, but not for adonitol, dulcitol and a-methyl-D-glucoside. In reference to the phenotypic characteristics of E. ananas (Beji et al., 1988; Gavini et al., 1989), we

252 K. limura, A. Hosono / Int. J. Food Microbiology 30 (19%) 243-253

have recognized that most of the phenotypic characteristics of the strains of phenotypic group II closely resembled E. ananas.

All the strains (T-5 and T-36) in phenotypic group III, which are negative for indole production and positive for gas production from D-glucose, were identified as R. aquatilis with the API 20E system. Colonies were smooth, translucent, and convex with an entire margin (data not shown). Acetoin was positive but lysine and ornithine decarboxylase, arginine dihydrolase and DNase were negative. Acid production was observed from D-glucose, L-arabinose, D-ribose, D-xylose, D-fruc- tose, D-mannose, L-rhamnose, D-cellobiose, D-galactose, lactose, maltose, meli- biose, sucrose, trehalose, raffinose, dulcitol, D-mannitol, D-sorbitol and salicin but not from melezitose, erythritol, adonitol, inositol, cr-methyl-D-glucoside, inulin and esculin (Table 2). These characteristics fit well with those of R. aquatilis strains described by Izard et al., 1979 E. agglomerans. Berge et al., 1991 reported that R. aquatilis is a rhizosphere-associated bacteria and also a nitrogen fixer. Hamze et al., 1991 reported that R. aquatilis has also been isolated mostly from water, soil, lager beer breweries and in a few cases, from human clinical specimens.

References

Beji, A., Mergaert, J., Gavini, F., Izard, D., Kersters, K., Leclerc, H. and De Ley, J. (1988) Subjective synonymy of Erwinia herbicola, Erwinia milletiae and Enterobacter agglomerans and redefinition of the taxon by genotypic and phenotypic data. Int. J. Syst. Bacterial. 38, 77-88.

Berge, O., Heuhn, T. and Achouak, W., (1991) Rahnella aquatilis, a nitrogen fixing enteric bacterium associated with the rizosphere of wheat and maize. Can. J. Microbial. 37, 1955203.

Brenner, D.J., Fanning, G.R., Knutson, J.K.L., Steigerwalt, A.G. and Krichevsky, M.I. (1984) Attempts to classify herbicola group Enterobacter agglomerans strains by deoxyribonucleic acid hybridization and phenotypic tests. Int. J. Syst. Bacterial. 34, 45-55.

Christensen, W.B. (1946) Urea decomposition as a means of differentiating Proteus and paracolon

cultures from each other. J. Bacterial. 52, 461-466. Dye, D.W. (1969a) A taxonomic study of the genus Erwiniu. II. The ‘carotovora’ group. New Zeal. J.

Sci. 12, 81-97. Dye, D.W. (1969b) A taxonomic study of the genus Erwinia. IV. ‘A typical Erwinias’. New Zeal. J. Sci.

12, 833-839. Dye, D.W. (1981) A numerical taxonomic study of the genus Erwiniu. New Zeal. J. Agric. Res. 24,

223-229. Ewing, W.H., Davis. B.R. and Reavis, R.W. (1957) Phenylalanine and malonate media and their use in

enteric bacteriology. Publ. Hlth. Lab. 15, 153-161. Ewing, W.H. and Fife, M.F. (1972) Enterobacter aggfomerans (Beiierinck) comb. nov. (the herbicola-

lathry bacteria). Int. J. Syst. Bacterial. 22, 4-l 1. Gavini, F., Lefebvre, B. and Leclerc, H. (1983) Etude taxonomiqu de souches appartenant ou

apparentees au genre Erwinia group Herbicola, et a I’espece Enterobacter ugglomeruns. Syst. Appl. Microbial. 4, 218-235.

Gavini, F., Mergaert, J., Beji, A., Mielcarek, C., Izard, D., Kersters, K. and De Ley, J. (1989) Transfer of Enterobucter ugglomerans (Beijerinck 1988) Ewing and Fife, 1972 to Puntoeu gen. nov. as Pantoeu

agglomerans comb. nov. and description of Puntoeu dispersu sp. nov. Int. J. Syst. Bacterial. 39, 337-345.

Hamze, M., Mergaert, J., Van Vuuren, H.J.J., Gavini, F., Beji, A.. Izard, D. and Kersters, K. (1991) RuhneNu aquatilis. a potential contaminant in lager beer breweries. Int. J. Food Microbial. 13. 63-68.

K. Iimura, A. Hosono / Int. J. Food Microbiology 30 (1996) 243-253 253

Iimura, K. and Hosono, A. (1993) Flora of contaminated bacteria in buckwheat seeds. J. Food Hygien. Sot. Jpn. 34, 524-528.

Emma, K. and Hosono, A. (1994) Identification of Gram negative bacteria isolated from buckwheat seeds. J. Food Hygien. Sot. Jpn. 35, 525-529.

Izard, D., Gavini, F., Trinel, P.A. and Leclerc, H. (1979) Rahnella aquatilis, nouveau membre de la famille des Enterobacteriuceae. Ann. Microbial. 130A, 163- 177.

Mergaert, J., Verdonk, L., Kersters, K., Swings, J., Boeufgras, M. and De Ley, J. (1984) Numerical taxonomy of Erwiniu species using API systems. J. Gen. Microbial. 130, 189331910.

Miyoshi, A., Takeshima, W., Varanyanond, W., Sawa, M. and Fuji, H. (1988) Studies on the growth control of microorganisms in new local processed food development. Examination of micro-flora and microbiological putrefaction on ‘Nama Nihon Soba’ (Fresh Japanese Buckwheat Noodle). Bull. Hyogo Prefec. Agric. Center Experim., Extens. Educ. 36, 101-110.

Moeller, V. (1955) Simplified tests for some amino acid decarboxylases and for the arginine dihydrolase system. Acta. Path. Microbial. Stand. 36, 1588172.

Muraschi, T.F., Friend, M. and Bolles, D. (1965) Erwiniu-like microorganisms isolated from animal and human hosts. Appl. Microbial. 13, 128-131.

Naito, H. and Shimizu, Y. (1986) Studies on microflora and microbial putrefaction on Japanese buckwheat. New Food Ind. 28, 8-10.

Obinata, H., Ohara, T., Muramatsu, N. and Matsuhashi, T. (1987) Microbiological infection of buckwheat grains and a method of microbial reduction. Res. Rep. Nagano State Lab. Food Technol. 15, 119-121.

Sakazaki, R., Tamura, K., Johnson, R. and Colwell, R.R. (1976) Taxonomy of some recently described species in the family Enterobacteriaceae. Int. J. Syst. Bacterial. 26, 158- 179.

Sakazaki, R., Yoshizaki, E. and Miki, K. (1988) New Bacterial Culture Media 2nd I. Kindai shuppan, Tokyo, Japan pp. 95-127.

Verdonck, L., Mergaert, J., Rijckaert, C., Swings, J., Kersters, K. and De Ley, J. (1987) Genus Erwinia: numerical analysis of phenotypic features. Int. J. Syst. Bacterial. 37, 4418.