Journal of Applied Microbiology ISSN 1364-5072

ORIGINAL ARTICLE

Safety properties and molecular strain typing of lactic acid

bacteria from slightly fermented sausages
T. Aymerich1*, B. Martı́n1*, M. Garriga1, M.C. Vidal-Carou2, S. Bover-Cid2 and M. Hugas3
1 IRTA, Meat Technology Centre, Granja Camps i Armet, Monells, Girona, Spain
2 Departament de Nutrició i Bromatologia – CERTA, Facultat de Farmı́cia, Universitat de Barcelona, Barcelona, Spain
3 European Food Safety Authority (EFSA), Parma, Italy

Keywords
antibiotic resistance, biogenic amines,
fermented sausages, lactic acid bacteria,
randomly amplified polymorphic DNA-PCR.
Correspondence
T. Aymerich, IRTA, Meat Technology Centre,
Granja Camps i Armet, 17121 Monells, Spain.
E-mail: teresa.aymerich@irta.es
*These authors contributed equally to this
work.
2004/1493: received 23 December 2004,
revised 30 June 2005 and accepted 19 July
2005
doi:10.1111/j.1365-2672.2005.02772.x
Abstract
Aim: To evaluate the biodiversity of lactobacilli from slightly fermented sau-
sages (chorizo, fuet and salchichon) by molecular typing, while considering
their safety aspects.
Methods and Results: Species-specific PCR, plasmid profiling and randomly
amplified polymorphic DNA (RAPD)-PCR were used to characterize 250 lactic
acid bacteria (LAB) isolated from 21 low acid Spanish fermented sausages.
Lactobacillus sakei was the predominant species (74%) followed by Lactobacillus
curvatus (21Æ2%) and Leuconostoc mesenteroides (4Æ8%). By plasmid profiling
and RAPD-PCR 144 different strains could be differentiated, 112 belonging to
Lact. sakei, 23 to Lact. curvatus and 9 to Leuc. mesenteroides. Ion-pair high per-
formance liquid chromatography was used to detect biogenic amine produc-
tion. Tyramine and phenylethylamine were produced by 14Æ4 and 12Æ4% of the
isolates, respectively, all belonging to the species Lact. curvatus. The production
of tyramine was stronger than that of phenylethylamine. Partial sequencing of
the tyrosine decarboxylase gene from Lact. curvatus was achieved. A specific
PCR assay to detect the Lact. curvatus tyramine-producers was designed. The
disc diffusion test was used to detect antibiotic resistance among the isolates.
Most isolates displayed resistance to vancomycin and gentamicin. Only four
strains were resistant to most of the antibiotics tested. None of the isolates
were resistant to erythromycin.
Conclusions: Lactobacillus sakei would be the species of choice for further use
as starter culture in fermented sausage production. Strain typing and character-
ization of biogenic amine production together with antibiotic susceptibility
testing for the selection of starter cultures could help to increase the quality
and safety of the products.
Significance and Impact of the Study: Species-specific PCR, RAPD and plas-
mid profiling proved to be efficient at typing LAB at species and strain level.
Information on biogenic amine production and transferable antibiotic resist-
ance is important in order to avoid selection of strains with undesirable prop-
erties as starter cultures.

European Union (EC 1999). Artisanal slightly fermented

Introduction
The increasing interest in preserving biodiversity of
micro-organisms involved in the production of tradition-
ally fermented food products has been reported by the
sausages (final pH from 5Æ3 to 6Æ2) form a group of tradi-
tional Mediterranean products with a great diversity
within the different regions. The microbial population
composition of artisanal fermented sausages is influenced

40 ª 2005 The Society for Applied Microbiology, Journal of Applied Microbiology 100 (2006) 40–49
T. Aymerich et al.
by the original microbial contamination of raw materials,
temperature, redox potential, pH and water activity of the
fermentation process (Lücke 1998). The knowledge and
control of their typical in-house microflora and the pro-
duction processes are critical in terms of their microbio-
logical quality and organoleptic characteristics (Demeyer
et al. 2000). Use of reliable and fast identification meth-
ods is of great importance for the study of the microbial
composition and for monitoring inoculated starter strains
in sausage fermentation.
The species of lactobacilli most commonly found in
meat and meat products, especially in dry fermented
sausages, are Lactobacillus sakei, Lactobacillus curvatus and
Lactobacillus plantarum (Schillinger and Lücke 1987;
Hammes et al. 1990; Montel et al. 1991; Hugas et al.
1993; Santos et al. 1998; Aymerich et al. 2003). In this
context, however, some important points have to be con-
sidered for an optimal selection of strains of interest and
to define the specific microbial role in these products: (i)
rapid classification and identification of unknown isolates,
(ii) evaluation of genetic diversity among strains and the
impact of diversity on the relevant properties of micro-
organisms and (iii) strain typing to assess genetic stability
over time.
For the identification of lactobacilli, phenotypical
methods have been widely used (Schillinger and Lücke
1987; Montel et al. 1991; Hugas et al. 1993; Samelis et al.
1995; Klein et al. 1996). Taxonomic methods relying on
physiological or biochemical criteria are ambiguous and
time consuming (Kandler and Weiss 1986). Molecular
methods such as rRNA hybridization probes (Nissen and
Dainty 1995), species-specific PCR (Aymerich et al. 2003;
Kwon et al. 2004), PCR-denaturing gel electrophoresis
(Cocolin et al. 2004), real-time PCR (Furet et al. 2004)
has been developed for lactic acid bacteria (LAB) species
identification.
Randomly amplified polymorphic DNA (RAPD)-PCR
analysis has been used to estimate the biodiversity among
LAB (Cocconcelli et al. 1995; Rebecchi et al. 1998; Andri-
ghetto et al. 2001a,b). RAPD is a rapid, sensitive and effi-
cient technique that enables strain typing (Berthier and
Ehrlich 1999). The numerical analysis of the data allows
the grouping and identification of strains (Vauterin and
Vauterin 1992).
Relevant safety aspects such as bacteriocin production,
amino acid-decarboxylase activity and antibiotic resistance
profile are important properties to evaluate potential
starter cultures.
Bacteriocins from LAB are natural antimicrobials
peptides that could improve quality and safety of the
meat products by avoiding the presence of pathogens
such as Listeria monocytogenes and spoilage micro-organ-
isms, and favouring the competitiveness of their produc-
ª 2005 The Society for Applied Microbiology, Journal of Applied Microbiology 100 (2006) 40–49
Lab biodiversity in fermented sausages
ers for survival (Nes and Tagg 1996). Several bacterial
preparations such as Flora Carn LC2 (Chr Hansen) and
bacteriocinogenic LAB and/or their bacteriocins have been
used as protective cultures and their activity against
L. monocytogenes have been proved in meat products
(Foegeding et al. 1992; Hugas et al. 1995).
Biogenic amines (BA) in food, especially in fermented
products, are mainly the result of bacterial activity due to
enzymatic decarboxylation of the precursor amino acids
(ten Brink et al. 1990). Some of these molecules, mainly
tyramine and histamine, show vasoactive and/or psycho-
active properties and they have been related to food safety
(Mariné-Font et al. 1995). Some others, such as putres-
cine, cadaverine and also histamine, are associated with
the activity of contaminant bacteria (mainly enterobacte-
ria) and thus they are of concern in relation to food qual-
ity (Vidal-Carou et al. 1990; Mariné-Font et al. 1995).
Amino acid-decarboxylase activity has been reported
among LAB (Halász et al. 1994; Bover-Cid and Holzapfel
1999). Thus, the potential capability for BA production
should be taken into account in the selection and imple-
mentation of starter and protective cultures to reduce
hygienic and toxicological risks (Bover-Cid et al. 2000).
Before using selected strains as probiotics or starter cul-
tures for food production it is important to verify that
they do not present transferable antimicrobial resistance
(SCAN 2003). Antibiotic-resistant LAB have been isolated
from raw meat (Klein et al. 1998; Pavia et al. 2000;
Gevers et al. 2003), therefore fermented meat products
can be potential reservoirs for the spread of antibiotic-
resistant LAB to the consumer via the food chain (Teuber
and Perreten 2000).
The aim of this study was to analyse the biodiversity of
LAB from slightly fermented sausages at strain and species
level and to elucidate some relevant safety and hygienic
characteristics for further use as starter culture.
Materials and methods
Bacterial strains and culture conditions
Reference strains were obtained from CECT (Spanish type
culture collection) and from our own collection (IRTA-
Meat Technology Center, CTC). All strains were grown in
MRS broth (Oxoid, Basingstoke, UK) at 30
C for 24 h.
The reference strains used as control strain were: Lact.
sakei CTC494, Lact. sakei CTC232, Lact. curvatus
CTC435, Lact. curvatus CTC371, Lact. plantarum
CTC305, Lact. plantarum CTC300, Lactobacillus brevis
CECT4669, Carnobacterium piscicola CTC779, Enterococ-
cus faecium CTC492, Enterococcus faecalis CECT481,
L. monocytogenes ser. 1/2c CTC1010, L. monocytogenes ser.
1/2c CTC1011, and L. monocytogenes ser. 4b CTC1034.
41
Lab biodiversity in fermented sausages T. Aymerich et al.

PCR were identified by partial sequencing of 16S rRNA

Sausage samples and LAB strains isolation
Twenty-one slightly fermented sausages (fuet, chorizo and
salchichon) were obtained from local butchers and super-
markets. Low-acid chorizo, fuet and salchichon are differ-
ent types of fermented sausages produced with pork
minced meat and additives.
After removal of the casing, 10 g of each sample were
homogenized in 90 ml of 0Æ1% peptone (Difco Laborator-
ies, Detroit, MI, USA) and 0Æ85% NaCl (Merck, Darsms-
tadt, Germany), pH 7Æ0, in a Stomacher Lab-Blender
(model 400; Cooke Laboratories, Alexandria, VA, USA),
poured plating in MRS (Oxoid) and incubated for 72 h
at 30
C. Twelve colonies, showing typical morphology,
from each sample were randomly picked, grown over-
night in MRS and stored at )80
C after addition of 20%
of glycerol for further analysis.
Identification of isolates
Bacterial DNA was isolated using the DNeasy Tissue Kit
(Qiagen, Crawley, UK) following the instructions of the
manufacturer.
The identification at species level was carried out by
species-specific PCR in the GeneAmp PCR System 2700
(Applied Biosystems, Foster City, CA, USA) following the
modified PCR protocol of Berthier and Ehrlich (1998) and
Aymerich et al. (2003). Amplification reactions were per-
formed in a 25 ll final volume containing 2Æ5 ll of PCR
Buffer (Invitrogen, Merelbeke, Belgium), 200 lmol l)1 of
each dNTP (Promega, Madison, WI, USA), 0Æ4 lmol l)1
of each primer (Table 1), 1 U of TAQ polymerase (Roche
Molecular Biochemicals, Indianapolis, IN, USA). Those
isolates that could not be identified by species-specific
gene. PCR amplification of V1–V3 regions of 16S rRNA
gene was assessed by oligonucleotides BSF8 and BSF1541
(Table 1). The same PCR reaction described above was
used, except for an annealing temperature of 50
C. DNA
was purified with GeneClean kit II (Bio 101, La Jolla, CA,
USA). DNA was sequenced by using the BigDye Termina-
tor Version 3.1 cycle sequencing kit, the oligonucleotides
BSF8, BSF343 and BSR1541 and the sequencing device
ABI PRISM 310 (Applied Biosystems). The BLAST-W2
and ClustalW software from the European Bioinformatics
Institute (Wellcome Trust Genome Campus. Hinxton.
Cambs. CB10 1SA. UK. Web site: http://www.ebi.ac.uk)
was used for the analysis of sequences.
Strain typing
Strain typing was assessed by RAPD-PCR and plasmid pro-
filing. RAPD analysis was carried out using two random
primers (Roche Molecular Biochemicals), R5 and M13R2
(Table 1). Each 25 ll PCR mix contained 2.5 ll of 10x
PCR buffer, 1Æ5 mmol l)1 MgCl2 (2Æ5 mmol l)1 when pri-
mer R5 was used), 200 lmol l)1 of each dNTP, 20 pmol of
primer (R5 or M13R2), 1 U of TAQ polymerase and
100 ng of extracted DNA. When primer R5 was used the
amplification process consisted of 5 min of initial denatur-
ation at 94
C and 40 cycles consisting of denaturation at
94
C for 1 min, annealing at 29
C for 1Æ5 min and elonga-
tion at 72
C for 2 min, followed by a final elongation for
5 min at 74
C. With primer M13R2, 35 cycles were per-
formed, each consisting of 1 min of denaturation at 94
C,
1 min of annealing at 38
C and elongation at 72
C for
1 min. Plasmid isolation was carried out as previously des-
cribed by Anderson and McKay (1983).

Table 1 Primers used in this study and PCR conditions

Target gene/ Number Product

micro-organism Primers Sequence (5¢-3¢) References TA (
C) of cycles size (bp)

Lactobacillus sakei Ls ATGAAACTATTAAATTGGTAC Berthier and Ehrlich (1998) 45 30 293

16S GCTGGATCACCTCCTTTC
Lactobacillus curvatus Lc TTGGTACTATTTAATTCTTAG Berthier and Ehrlich (1998) 51 25 305
16S GCTGGATCACCTCCTTTC
16S rRNA gene BSF8 AGAGTTTGATCCTGGCTCAG Universal primers 50 30
BSF 343 TACGGGAGGCAGCAG
BSF1541 AAGGAGGTGATCCAGCCGCA
Random primers R5 AACGCGCAAC Random primers 29 40 –
(RAPD-PCR) M13R2 GGAAACAGCTATGACCATGA 38 35 –
tyrdc gene of LAB P1-for CCRTARTCIGGIATIGCRAARTCIGTRTG Lucas and Lonvaud-Funel 45 35 800
P2-rev GAYATIATIGGIATIGGIYTIGAYCARG (2002)
tyrdc gene of tdcC1 CTGGTGGGATTGCTATTC This study 58 35 372
Lact. curvatus tdcC2 TTCGTTCATTTCCACAAGG

TA, Temperature of annealing.

42 ª 2005 The Society for Applied Microbiology, Journal of Applied Microbiology 100 (2006) 40–49
T. Aymerich et al.
Amplification products and plasmidic DNA were elec-
trophoresed for 100 min at 80 mA in 1Æ5 and 0Æ8% ag-
arose (Bio-Rad, Hercules, CA, USA) gels, respectively.
Gels were stained with 0Æ1 lg ml)1 ethidium bromide
(Sigma, St. Louis, MO, USA). Two lines of 1 kb DNA
ladder (Invitrogen) and supercoiled DNA ladder (Invitro-
gen) were used as molecular weight and normalization gel
standards for RAPD and plasmid profiles, respectively.
The banding profiles were visualized under UV light and
digitalized by the Gelprinter photodocumentation equip-
ment (TDI, Barcelona, Spain).
Electrophoretic profiles obtained were normalized and
analysed by the software Fingerprinting II (Bio-Rad).
RAPD profiles of both primers were combined. Band
similarities between different RAPD and plasmid profiles
were analysed using Dice coefficient, and correlation coef-
ficients were calculated by the unweighted pair group
method with arithmetic averages (UPGMA).
To ensure reproducibility of the RAPD analysis we
strictly controlled all the parameters and independent
DNA isolates of the same strain were tested. Total DNA
from Lact. sakei CTC494 and Lact. curvatus CTC371 were
prepared independently four times and assayed in differ-
ent RAPD analysis. Four independent plasmid DNA
extractions from each control strain were also performed.
All electrophoretic profiles were computed and clustered
by UPGMA to obtain the value for the reproducibility of
the assay.
Antagonistic activity against Listeria monocytogenes
Antagonistic activity against L. monocytogenes was assayed
by a modified agar spot test (Tagg et al. 1976). Briefly,
1 ml of an overnight culture of each of the 250 isolates
were centrifuged 5 min at 3000 g. The supernatants were
pasteurized at 80
C for 10 min and neutralized with
5 mol l)1 NaOH to pH 6Æ5. An aliquot of 10 ll was spot-
ted onto a lawn of soft TSBYE seeded with 50 ll of an
overnight culture of a cocktail of three different strains of
L. monocytogenes. The plates were incubated overnight at
30
C. Clear halos greater than 1 mm in radius were con-
sidered as positive.
Lactobacillus sakei CTC494 and Ent. faecium CTC492
were used as control strains for bacteriocin production
(Hugas et al. 1995; Aymerich et al. 1996).
Biogenic amine production
Amino acid decarboxylase activity of isolates was assayed
through the assessment of the presence of BA in a de-
carboxylase broth as described by Bover-Cid and Holzap-
fel (1999). The medium contained the precursor amino
acids, l-tyrosine, l-triptophan, l-phenylalanine; l-ornith-
ª 2005 The Society for Applied Microbiology, Journal of Applied Microbiology 100 (2006) 40–49
Lab biodiversity in fermented sausages
ine monohydrochloride and l-lysine monohydrochloride
(Sigma Chemical Co.) at 1% each, pyridoxal-5-phosphate
(0Æ005%) as a codecarboxylase factor, thiamine (0Æ001%),
tween 80 (0Æ1%), Mg (0Æ02%), Mn (0Æ005%), Fe (0Æ004%)
and ammonium citrate (0Æ2%) as growth enhancers,
di-potassium phosphate (0Æ2%) and calcium carbonate
(0Æ01%) as buffers and bromocresol purple (0Æ006%) as
pH indicator.
BA, tyramine, tryptamine, phenylethylamine and
putrescine, were determined by ion-pair high performance
liquid chromatography and postcolumn derivatization
with ortho-phtalaldehyde according to Hernández-Jover
et al. (1996).
Additionally, a rapid assay to screen the gene encoding
tyrosine-decarboxylase (tyrdc) by PCR was carried out as
reported by Lucas and Lonvaud-Funel (2002) except for
the annealing temperature (45
C).
Partial sequencing of tyrdc gene of Lactobacillus curvatus
The tyrdc gene fragments of two different Lact. curvatus
strains (CTC6513 and CTC6677) were amplified by PCR
using the degenerated primers P1-rev and P2-for (Lucas
and Lonvaud-Funel 2002). The annealing temperature
was decreased to 45
C in order to amplify the tyrdc gene
from Lact. curvatus.
DNA was purified with the GeneClean kit II (Bio 101)
and sequenced using the primer P2. Sequencing and ana-
lysis of the results obtained were carried out as for 16S
rRNA gene sequencing.
A new pair of primers was developed to specifically
detect the tyrdc gene of Lact. curvatus. Sequences of the
newly designed primers are shown in Table 1. Amplifica-
tion reactions were performed in a 25 ll final volume con-
taining 2Æ5 ll of PCR Buffer (Invitrogen), 200 lmol l)1 of
each dNTP (Promega), 0Æ4 lmol l)1 of each primer
(Table 1) and 1 U of TAQ polymerase (Roche).
The assay was validated using five reference strains bear-
ing tyrosine-decarboxylase activity (Lact. brevis CECT4669,
C. piscicola CTC779, Ent. faecalis CECT481, Ent. faecium
CTC492 and Lact. curvatus CTC435) and four nonpro-
ducer strains (Lact. sakei CTC494 and CTC232, Lact. plan-
tarum CTC305 and CTC300) and all the isolates of Lact.
curvatus obtained from the slightly fermented sausages.
Antibiotic susceptibility testing
Susceptibility testing was based on the agar overlay disc
diffusion test described by Charteris et al. (1998) with
some modifications. LAB were grown overnight in MRS
broth at 30
C under anaerobic conditions (Oxoid jars
with AnaeroGen, Oxoid). Eight millilitres of MRS kept at
50
C were inoculated with 0Æ2 ml of the grown culture.
43
Lab biodiversity in fermented sausages T. Aymerich et al.

Petri dishes containing 15 ml of MRS were overlaid with

8Æ2 ml of the inoculated MRS and allowed to solidify at
room temperature. Antibiotic discs (Oxoid) were placed
onto the overlaid plates and all plates were incubated for
20–24 h at 30
C under anaerobic conditions. All isolates
were screened for their susceptibility to ampicillin
(10 lg), chloramphenicol (30 lg), gentamicin (10 lg),
erythromycin (15 lg), linezolid (30 lg), penicillin G
(10 U), quinupristin/dalfopristin (15 lg), tetracycline
(30 lg) and vancomycin (30 lg). Breakpoints for the
interpretation of inhibition zone were those defined by
the National Committee for Clinical Laboratory Standards
(NCCLS 2002) and the British Society for Antimicrobial
Chemotherapy (Andrews and BSAC Working Party on
Susceptibility Testing 2001) for Streptococcus spp. and
those described by Charteris et al. (1998) for lactobacilli.
Strains were considered resistant if inhibition halos dia-
meters were £19 mm for ampicillin, penicillin and linezo-
lid, £18 mm for tetracycline, £15 mm for quinupristin/
dalfopristin, £14 mm for vancomycin, £13 mm for chloram-
phenicol and erythromycin and £12 mm for gentamicin.
Results
Identification of isolates
By species-specific PCR, 74% of the isolates were identi-
fied as Lact. sakei and 21Æ2% as Lact. curvatus. The
remaining isolates (4Æ8%) were identified by partial
sequencing of 16S rRNA gene as Leuconostoc mesenteroides
(GenBank accession numbers DQ092702–DQ092705 and
DQ105645–DQ105649). Lactobacillus sakei was found in
100% of the sausages and was the predominant species in
chorizo and fuet representing 89Æ3 and 76Æ3% of the iso-
lates, respectively. Lactobacillus curvatus was detected in
42Æ9% of the samples and was the predominant species in
salchichon (58Æ8% of the isolates of this type of sausages).
Leuconostoc mesenteroides was found in 14Æ3% of the sam-
ples (two samples of chorizo and one sample of fuet).
Strain typing
RAPD-PCR and plasmid profiles were used to discriminate
strains among the 250 recovered isolates. The value for the
reproducibility of both assays, estimated by analysis of
repeated DNA extracts of several control strains, was
greater than 92% for RAPD-PCR analysis and 89% for
plasmid typing (results not shown). The discriminatory
capacity varied depending on the technique used (Table 2).
By RAPD-PCR, 136 different profiles were distinguished by
combination of the results obtained with primer R5 and
M13R2. The latter primer was able to distinguish a higher
number of strains than R5, but the higher discriminatory
capacity was obtained after a combination of both primers.
Plasmid analysis differentiated 88 profiles and 14 isolates
were plasmid free. Nevertheless, the combination of the
results obtained with plasmid profiling and RAPD-PCR
allowed us to distinguish 144 different strains out of 250
isolates. Leuconostoc mesenteroides showed the higher intra-
specific variability (Table 2) since it was possible to differ-
entiate 9 profiles out of the 12 isolates of this species while
Lact. sakei showed 112 different strains out of 185 isolates.
Lactobacillus curvatus presented lower intra-specific variab-
ility than the other species (Table 2).
Bacteriocin production
All the isolates were tested for their antagonistic activity
against L. monocytogenes. Neutralized and pasteurized
supernatants obtained from the 250 isolates did not show
this capability in the in vitro assays.
Biogenic amine production
The aminogenic potential of LAB isolates is shown in
Table 3. Among them, 39 strains (14Æ8%) produced one
or more BA, although at different levels. Tyramine was
produced by 14Æ4% of the strains, followed by phenyl-
ethylamine (12Æ4% of the isolates), tryptamine (4%) and

Table 2 Comparison of the intra-species variability (IV, %) obtained with the different techniques used for strain typing

RAPD-R5 RAPD-M13R2 Combined RAPD
Plasmid profiling All techniques

Profiles* IV Profiles* IV Profiles* IV Profiles* IV Profiles* IV

Lactobacillus sakei 85 45Æ9 87 47Æ0 110 59Æ5 66 35Æ7 112 60Æ5

Lactobacillus curvatus 12 22Æ6 17 32Æ1 17 32Æ1 14 26Æ4 23 43Æ4
Leuconostoc mesenteroides 5 41Æ7 9 75Æ0 9 75Æ0 8 66Æ7 9 75Æ0

Total 102 40Æ8 113 45Æ2 136 54Æ4 88 35Æ2 144 57Æ6

IV, intra-species variability.

*Number of profiles obtained.

Combination of the profiles obtained with primer R5 and primer M13R2.
Combination of RAPD analysis (profiles obtained with primer R5 and primer M13R2) and plasmid profiles.

44 ª 2005 The Society for Applied Microbiology, Journal of Applied Microbiology 100 (2006) 40–49
T. Aymerich et al. Lab biodiversity in fermented sausages

Table 3 Percentage of strains with capability to produce different BA

Total Strains Lactobacillus Lactobacillus Leuconostoc

Biogenic amine Amount n = 250 sakei n = 185 curvatus n = 53 mesenteroides n = 012

Tyramine ± 2Æ0 0 9Æ4 0

+ 0Æ4 0 1Æ9 0
++ 0Æ0 0 0 0
+++ 12Æ0 0 56Æ6 0
Total 14Æ4 0 67Æ9 0
Putrescine ± 0 0 0 0
+ 0Æ4 0Æ5 0 0
++ 0Æ4 0 1Æ9 0
+++ 0 0 0 0
Total 0Æ8 0Æ5 1Æ9 0
Phenylethylamine ± 0Æ8 0 3Æ8 0
+ 2Æ4 0 11Æ3 0
++ 9Æ2 0 43Æ4 0
+++ 0 0 0 0
Total 12Æ4 0 58Æ5 0
Tryptamine ± 2Æ4 0 11Æ3 0
+ 0Æ8 0Æ5 1Æ9 0
++ 0 0 0 0
+++ 0Æ8 0Æ5 1Æ9 0
Total 4 1Æ1 15Æ1 0

±, 25–50 mg l)1; +, 50–100 mg l)1; ++, 100–1000 mg l)1; +++, >1000 mg l)1.

putrescine (0Æ8%). The production of other amines (cada-

verine or histamine) was not observed in any isolate. The
production of BA was mainly related to the species Lact.
curvatus, since all but two producer strains were identified
as this species. By contrast, only two Lact. sakei isolates
produced low levels of tryptamine or putrescine. The
presence of the gene associated to the tyramine-descarb-
oxilase (tyrdc) was assayed by PCR as described by Lucas
and Lonvaud-Funel (2002). The gene was detected in all
tyramine-producer isolates.
Partial sequencing of tyrdc gene of Lactobacillus curvatus
Partial tyrdc gene sequencing of two different isolates of
Lact. curvatus was carried out (GenBank accession num-
bers AJ871286 and AJ871287). Comparison of the tyrdc-
sequenced fragments between both Lact. curvatus strains
gave a sequence similarity of 99%. Highest homologies
were obtained with the tyrdc gene sequences of Ent. fae-
calis (GenBank accession number AF354231) and Lact.
brevis (AF446085), with a sequence similarity of 74%.
To specifically detect Lact. curvatus strains carrying the
tyrdc gene, new primers from the partial tyrdc gene
sequence of Lact. curvatus were designed. All the 53 iso-
lates of Lact. curvatus from the slightly fermented sausages
were submitted to the assay. Positive results were perfectly
correlated with the 36 tyramine-producer Lact. curvatus
isolates.
Antibiotic susceptibility testing
Antibiotic susceptibility of isolates was tested using a
modified disc diffusion technique. The prevalence of anti-
biotic resistance among isolates is shown in Table 4.
Lactobacillus sakei and Lact. curvatus isolates showed the
same results for all antibiotics except for their distinct
prevalence of resistance to b-lactams. The main difference
among lactobacilli and Leuconostoc isolates studied was
the gentamicin resistance. All isolates were resistant to
vancomycin, 98Æ4% were resistant to gentamicin and
43Æ6% to ampicillin. With regard to the other antibiotics,
resistance prevalence was always lower than 30%. All iso-
lates were susceptible to erythromycin and only two Lact.
sakei isolates showed resistance to linezolid.
All isolates displayed resistance to at least two antibiot-
ics. About half of the isolates (48%) were only resistant
to vancomycin and gentamicin whereas 42Æ8% were resist-
ant towards 3–4 antibiotics. Only 8Æ8% of isolates (mainly
Lact. sakei) displayed resistance to 5 or 6 of the antibiot-
ics tested, including vancomycin, gentamicin, ampicillin
and penicillin.
Discussion
The LAB of low-acid fermented sausages has been charac-
terized on the basis of genetic diversity and important
safety properties. Among the different genera belonging

ª 2005 The Society for Applied Microbiology, Journal of Applied Microbiology 100 (2006) 40–49 45
Lab biodiversity in fermented sausages
Table 4 Prevalence of antibiotic resistant LAB from slightly fermented sausages to selected antibiotic using the disc diffusion test
Total strains
Antibiotic n = 250
Ampicillin (10 lg) 43Æ6
Chloramphenicol (30 lg) 1Æ2
Gentamicin (10 lg) 98Æ4
Erythromycin (15 lg) 0Æ0
Linezolid (30 lg) 0Æ8
Penicillin G (10 U) 29Æ2
Quinupristin/dalfopristin (15 lg) 4Æ4
Tetracycline (30 lg) 12Æ0
Vancomycin (30 lg) 100
Results are expressed as percentages.
to LAB, only Lactobacillus and Leuconostoc were recovered
from MRS plates. Lactobacillus sakei dominated over the
rest of the species (74% of isolates) and Lact. curvatus
was also present, although in a lower proportion (21Æ2%).
Few isolates were identified as Leuc. mesenteroides (4Æ8%).
These results are in agreement with those obtained using
biochemical methods by Torriani et al. (1990), Hugas
et al. (1993), Samelis et al. (1994), Santos et al. (1998)
and Parente et al. (2001) in similar types of fermented
meat products.
The combination of RAPD-PCR and plasmid profiling
allowed us to improve the differentiation among strains,
since both chromosomal and plasmidic characteristics
were considered. The isolates identified as the same strain
showed common phenotypic properties in terms of bio-
genic amine production and antibiotic susceptibility,
although isolates with specific traits could not be grouped
in single clusters, which could indicate horizontal
exchange of DNA between bacteria. Horizontal gene
transfer among LAB has been reported in vitro (Vogel
et al. 1991) and in fermented sausages (Hertel et al. 1995;
Cocconcelli et al. 2003). Some samples exhibited only one
strain type, which could suggest the addition of starter
cultures for their elaboration.
Bacteriocinogenic starter cultures have been reported to
be effective in the control of some food-borne pathogens
such as L. monocytogenes (Hugas et al. 1995) and,
although this is an important characteristic for strain
selection, none of the 250 isolates showed antagonistic
activity against this food-borne pathogen. The occurrence
of bacteriocinogenic activity among lactobacilli is gener-
ally low. Only 2Æ7 and 2Æ3% of the isolates from German
and Spanish meat products presented bacteriocinogenic
activity against L. monocytogenes (Schillinger and Lücke
1989; Garriga et al. 1993).
The formation of BA is of concern in terms of food
safety and quality. In this study, biogenic amine produc-
tion was associated with the species Lact. curvatus. More
46
T. Aymerich et al.
Lactobacillus Lactobacillus Leuconostoc
sakei n = 185 curvatus n = 53 mesenteroides n = 012
51Æ9 15Æ1 41Æ7
1Æ6 0Æ0 0Æ0
100 100 66Æ7
0Æ0 0Æ0 0Æ0
1Æ1 0Æ0 0Æ0
33Æ0 11Æ3 50Æ0
5Æ9 0Æ0 0Æ0
10Æ8 13Æ2 25
100 100 100
than half of the Lact. curvatus isolates showed a strong
tyrosine-decarboxylase activity, producing more than
1000 mg l)1 of tyramine in vitro, in agreement with
Bover-Cid et al. (2001). Moreover, most of them were
also able to simultaneously produce considerable amounts
of phenylethylamine. However, Silla-Santos (1998) did
not find any amine-producer among the lactobacilli iso-
lates from Spanish fermented meat products. It is also
important to point out that among the 21 slightly fer-
mented sausages studied, the samples showing the highest
proportion of Lact. curvatus also presented the highest
contents of tyramine and phenylethylamine. The tyramine
content of these samples ranged from 273Æ3 to
334Æ7 mg kg)1 dry weight and those of phenylethylamine
ranged from 30 to 40Æ5 mg kg)1 dry weight (unpublished
results). Tryptamine was produced by only 4% of isolates,
which is in accordance with its minor presence in fer-
mented sausages and food in general. The low putrescine
production and the lack of cadaverine production among
the LAB isolates is in agreement with the general associ-
ation of these diamines mainly with the activity of con-
taminating enterobacteria. It is noteworthy to highlight
the much lower aminogenic activity found among Lact.
sakei strains, as also reported in Bover-Cid et al. (2001).
The PCR assay developed to detect the tyrdc gene of
Lact. curvatus could be used as a rapid method to charac-
terize potential tyramine-producers within this species.
Despite PCR protocols for tyramine-producing LAB have
previously been reported (Lucas and Lonvaud-Funel
2002; Coton et al. 2004), this is the first study detecting
Lact. curvatus tyrdc gene.
The absence of transferable antimicrobial resistance
should be an important prerequisite for selection of safety
starter cultures or probiotic strains (SCAN 2003).
Although specific antibiotic resistance traits could be
desirable in situations such as antibiotic-induced diar-
rhoea (Charteris et al. 1998; D’Souza et al. 2002), trans-
ferable resistance genes may pose a risk, as they can be
ª 2005 The Society for Applied Microbiology, Journal of Applied Microbiology 100 (2006) 40–49
T. Aymerich et al.
transferred to pathogenic bacteria. Resistance to antibiot-
ics such as chloramphenicol, ampicillin, erythromycin,
tetracycline and gentamicin are generally considered
transferable acquired resistances (Danielsen and Wind
2003; SCAN 2003). In the present study, tetracycline and
chloramphenicol resistant isolates were only detected at
low percentages. It is known that certain species of Lacto-
bacillus are inherently resistant to ampicillin (SCAN
2003). In this study a high percentage of isolates, especi-
ally within Lact. sakei, presented resistance to this antibi-
otic. The susceptibility to gentamicin was studied using a
10 lg disk and most isolates were resistant, which could
indicate an intrinsic resistance to low levels of this antibi-
otic in Lact. sakei and Lact. curvatus. Several minimal
inhibition concentrations, ranging from 1 lg ml)1 (SCAN
2003) to 128 lg ml)1 (Danielsen and Wind 2003), have
been proposed for transferable gentamicin resistance. All
the isolates were inherently resistant to vancomycin in
agreement with previous reports (Handwerger et al. 1994;
Hamilton-Miller and Shah 1998; Danielsen and Wind
2003).
On the basis of competitiveness and hygienic aspects
such as biogenic amine production, Lact. sakei showed
better properties than Lact. curvatus for further use as
starter culture in slightly fermented sausages. The pro-
posed rapid PCR methods for typing and characterize
the potential for tyramine production together with
antibiotic testing is of major importance in order to
select appropriate strains to increase the safety of the
products.
Acknowledgements
This research was funded by the Spanish Inter-Ministerial
Commission of Science and Technology (CICYT ALI99-
0308) and the EU project TRADISAUSAGE QLK1-
CT-2002-02240. We thank Y. Beltran, A. Claret and
D. Tibau for technical assistance. B. Martı́n was the
recipient of a predoctoral grant awarded by the Ministry
of Science and Technology. S. Bover-Cid acknowledges
the funding support from the Ministry of Science and
Technology (Ramón y Cajal Program).
References
Anderson, D.G. and McKay, L.L. (1983) Simple and rapid
method for isolating large plasmid DNA from Lactic
Streptococci. Appl Environ Microbiol 46, 549–552.
Andrews, J.M. for the BSAC Working Party on Susceptibility
Testing (2001) BSAC standardized disc susceptibility test-
ing method. J Antimicrob Chemother 48, 43–57.
Andrighetto, C., Knijff, E., Lombardi, A., Torriani, S.,
Vancanneyt, M., Kersters, K., Swings, J. and Dellaglio, F.
ª 2005 The Society for Applied Microbiology, Journal of Applied Microbiology 100 (2006) 40–49
Lab biodiversity in fermented sausages
(2001a) Phenotypic and genetic diversity of enterococci
isolated from Italian cheeses. J Dairy Res 68, 303–316.
Andrighetto, C., Zampese, L. and Lombardi, A. (2001b)
RAPD-PCR characterization of lactobacilli isolated from
artisanal meat plants and traditional fermented sausages of
Veneto region (Italy). Lett Appl Microbiol 33, 26–30.
Aymerich, T., Holo, H., Havarstein, L.S., Hugas, M., Garriga,
M. and Nes, I.F. (1996) Biochemical and genetic character-
ization of enterocin A from Enterococcus faecium, a new
antilisterial bacteriocin in the pediocin family of bacterioc-
ins. Appl Environ Microbiol 62, 1676–1682.
Aymerich, M.T., Martı́n, B., Garriga, M. and Hugas, M. (2003)
Microbial quality and direct PCR identification of lactic acid
bacteria and nonpathogenic staphylococci from artisanal
low-acid sausages. Appl Environ Microbiol 69, 4583–4594.
Berthier, F. and Ehrlich, S.D. (1998) Rapid species identifica-
tion within two groups pf closely related lactobacilli using
PCR primers that target the 16S/23S rRNA spacer region.
FEMS Microbiol Lett 161, 97–106.
Berthier, F. and Ehrlich, S.D. (1999) Genetic diversity within
Lactobacillus sakei and Lactobacillus curvatus and design of
PCR primers for its detection using randomly amplified
polymorphic DNA. Int J Syst Bacteriol 49, 997–1007.
Bover-Cid, S. and Holzapfel, W.H. (1999) Improved screening
procedure for biogenic amine production by lactic acid
bacteria. Int J Food Microbiol 53, 33–41.
Bover-Cid, S., Hugas, M., Izquierdo-Pulido, M. and
Vidal-Carou, M.C. (2000) Reduction of biogenic amine
formation using a negative amino acid-decarboxylase star-
ter culture for fermentation of fuet sausages. J Food Prot
63, 237–243.
Bover-Cid, S., Hugas, M., Izquierdo-Pulido, M. and Vidal-
Carou, M.C. (2001) Amino acid-decarboxylase activity of
bacteria isolated from fermented pork sausages. Int J Food
Microbiol 66, 185–189.
ten Brink, B., Damink, C., Joosten, H.M. and Huis in’t Veld,
J.H. (1990) Occurrence and formation of biologically
active amines in foods. Int J Food Microbiol 11, 73–84.
Charteris, W.P., Kelly, P.M., Morelli, L. and Collins, J.K.
(1998) Antibiotic susceptibility of potentially probiotic
Lactobacillus species. J Food Prot 61, 1636–1643.
Cocconcelli, P.S., Cattivelli, D. and Gazzola, S. (2003) Gene
transfer of vancomycin and tetracycline resistances among
Enterococcus faecalis during cheese and sausage fermenta-
tions. Int J Food Microbiol 88, 315–323.
Cocconcelli, P.S., Porro, D., Galandini, S. and Senini, L.
(1995) Development of RAPD protocol for typing of
strains of lactic acid bacteria and enterococci. Lett Appl
Microbiol 21, 376–379.
Cocolin, L., Rantsiou, K., Iacumin, L., Urso, R., Cantoni, C.
and Comi, G. (2004) Study of the ecology of fresh sau-
sages and characterization of populations of lactic acid
bacteria by molecular methods. Appl Environ Microbiol 70,
1883–1894.
47
Lab biodiversity in fermented sausages
Coton, M., Coton, E., Lucas, P. and Lonvaud, A. (2004) Iden-
tification of the gene encoding a putative tyrosine decarb-
oxylase of Carnobacterium divergens 508. Development of
molecular tools for the detection of tyramine-producing
bacteria. Food Microbiol 21, 125–130.
Danielsen, M. and Wind, A. (2003) Susceptibility of Lactobacil-
lus spp. to antimicrobial agents. Int J Food Microbiol 82,
1–11.
Demeyer, D., Raemaekers, M., Rizzo, A., Holck, A., De Smedt,
A., Ten Brink, B., Hagen, B., Montel, C. et al. (2000) Con-
trol of bioflavour and safety in fermented sausages: first
results of a European project. Food Res Intern 33, 171–180.
D’Souza, A., Rajkumar, C., Cooke, J. and Bulpitt, C. (2002)
Probiotics in prevention of antibiotic associated diarrhoea:
meta-analysis. BMJ 324, 1361.
European Commission (1999) Europe’s Agenda 2000. Streng-
thening and widening the European Union. Brussels:
European Commission.
Foegeding, P.M., Thomas, A.B., Pilkington, D.H. and
Klaenhammer, T.R. (1992) Enhanced Control of Listeria
monocytogenes by in situ-produced pediocin during dry
fermented sausage production. Appl Environ Microbiol 58,
884–890.
Furet, J.P., Quenee, P. and Tailliez, P. (2004) Molecular quan-
tification of lactic acid bacteria in fermented milk products
using real-time quantitative PCR. Int J Food Microbiol 97,
197–207.
Garriga, M., Hugas, M., Aymerich, M.T. and Monfort, J.M.
(1993) Bacteriocinogenic activity of lactobacilli from fer-
mented sausages. J Appl Bacteriol 75, 142–148.
Gevers, D., Danielsen, M., Huys, G. and Swings, J. (2003)
Molecular characterization of tet(M) genes in Lactobacillus
isolates from different types of fermented dry sausage. Appl
Environ Microbiol 69, 1270–1275.
Halász, A., Barath, A., Simon-Sakardi, L. and Holzapfel, W.
(1994) Biogenic amines and their production by micro-
organism in food. Trends Food Sci Technol 5, 42–49.
Hamilton-Miller, J.M.T. and Shah, S. (1998) Vancomycin
susceptibility as an aid to the identification of lactobacilli.
Lett Appl Microbiol 26, 153–154.
Hammes, W.P., Bantleon, A. and Min, S. (1990) Lactic acid
bacteria in meat fermentation. FEMS Microbiol Rev 87,
165–174.
Handwerger, S., Pucci, M.J., Volk, K.L., Liu, J. and Lee, M.S.
(1994) Vancomycin-resistant Leuconostoc mesenteroides and
Lactobacillus casei syntetize cytoplasmic peptidoglycan
pecursors that terminate in lactate. J Appl Bacteriol 176,
260–264.
Hernández-Jover, T., Izquierdo-Pulido, M., Veciana-Nogués,
M.T. and Vidal-Carou, M.C. (1996) Ion-pair high-per-
formance liquid chromatographic determination of biogen-
ic amines in meat and meat products. J Agric Food Chem
44, 2710–2715.
Hertel, C., Probst, A.J., Cavadini, C., Meding, E. and Hammes,
W.P. (1995) Safety assessment of genetically modified
48 ª 2005 The Society for Applied Microbiology, Journal of Applied Microbiology 100 (2006) 40–49
T. Aymerich et al.
microorganisms applied in meat fermentation. Syst Appl
Microbiol 18, 469–476.
Hugas, M., Garriga, M., Aymerich, T. and Monfort, J.M.
(1993) Biochemical characterization of lactobacilli from
dry fermented sausages. Int J Food Microbiol 18, 107–113.
Hugas, M., Garriga, M., Aymerich, M.T. and Monfort, J.M.
(1995) Inhibition of Listeria in dry fermented sausages by
the bacteriocinogenic Lactobacillus sake CTC494. J Appl
Bacteriol 79, 322–330.
Kandler, O. and Weiss, N. (1986) Regular, nonsporing gram-
positive rods. In Bergey’s Manual of Systematic Bacteriology,
Vol. 2 ed. Holt, J.G. and Sneath, P.H.A. pp. 1208–1234.
Baltimore: Williams & Wilkins.
Klein, G., Dicks, L.M.T., Pack, A., Hack, B., Zimmermann,
K., Dellaglio, F. and Reuter, G. (1996) Emended descrip-
tion of Lactobacillus sake (Katagiri, Kitahara, and
Fukami) and Lactobacillus curvatus (Abo-Elnaga and
Kandler): numerical classification revealed by protein fin-
gerprinting and identification based on biochemical pat-
terns and DNA-DNA hybridizations. Int J Syst Bacteriol
46, 367–376.
Klein, G., Pack, A., Bonaparte, C. and Reuter, G. (1998) Tax-
onomy and physiology of probiotic lactic acid bacteria.
Int J Food Microbiol 41, 103–125.
Kwon, H.S., Yang, E.H., Yeon, S.W., Kang, B.H. and Kim, T.Y.
(2004) Rapid identification of probiotic Lactobacillus spe-
cies by multiplex PCR using species-specific primers based
on the region extending from 16S rRNA through 23S
rRNA. FEMS Microbiol Lett 239, 267–275.
Lucas, P. and Lonvaud-Funel, A. (2002) Purification and
partial gene sequence of the tyrosine decarboxylase of
Lactobacillus brevis IOEB 9809. FEMS Microbiol Lett 211,
85–89.
Lücke, F.K. (1998) Fermented sausages. In Microbiology of
Fermented Foods, Vol. 2 ed. Wood, B.J.B. pp. 441–483.
London: Blackie Academic & Professional.
Mariné-Font, A., Vidal-Carou, M.C., Izquierdo-Pulido, M.,
Venciana-Nogués, M.T. and Hernández-Jover, T. (1995)
Les amines biogènes dans les aliments: leur signification,
leur analyse. Ann Fals Exp Chim Toxicol 88, 119–140.
Montel, M.C., Talon, R., Fournaud, J. and Champomier, M.C.
(1991) A simplified key for identifying homofermentative
Lactobacillus and Carnobacterium spp. from meat. J Appl
Bacteriol 70, 469–472.
NCCLS (2002) Performance standards for antimicrobial sus-
ceptibility testing; Twelfth Informational Supplement
M100-S12 NCCLS, Wayne, Pennsylvania.
Nes, I.F. and Tagg, J.R. (1996) Novel lantibiotics and their
pre-peptides. Antonie van Leeuwenhoek 69, 89–97.
Nissen, H. and Dainty, R.H. (1995) Comparison of the use of
rRNA probes and conventional methods in identifying
strains of Lactobacillus sake and Lactobacillus curvatus iso-
lated from meat. Int J Food Microbiol 25, 311–315.
Parente, E., Griego, S. and Crudele, M.A. (2001) Phenotypic
diversity of lactic acid bacteria isolated from fermented
T. Aymerich et al.
sausages produced in Basilicata (Southern Italy). J Appl
Microbiol 90, 943–952.
Pavia, M., Nobile, C.G.A., Salpietro, L. and Angelillo, I.F.
(2000) Vancomycin resistance and antibiotic suscep-
tibility of enterococci in raw meat. J Food Prot 63,
912–915.
Rebecchi, A., Crivori, S., Sarra, P.G. and Cocconcelli, P.S.
(1998) Physiological and molecular techniques for the
study of bacterial community development in sausage fer-
mentation. J Appl Microbiol 84, 1043–1049.
Samelis, J., Maurogenakis, F. and Metaxopoulos, J. (1994)
Characterisation of lactic acid bacteria isolated from natur-
ally fermented Greek dry salami. Int J Food Microbiol 23,
179–196.
Samelis, J., Tsakelidou, E., Metaxopoulos, J. and Kalantzo-
poulos, G. (1995) Differentiation of Lactobacillus sake and
Lactobacillus curvatus isolated from naturally fermented
Greek dry salami by SDS-PAGE of whole-cell proteins.
J Appl Bacteriol 78, 157–163.
Santos, E.M., González-Fernández, C., Jaime, I. and Rovira, J.
(1998) Comparative study of lactic acid bacteria house
flora isolated in different varieties of ‘chorizo’. Int J Food
Microbiol 39, 123–128.
SCAN (2003) Opinion of the Scientific Comittee on Animal
Nutrition (SCAN) on the criteria for assessing the safety of
microorganisms resistant to antibiotics of human clinical and
veterinary importance. Brussels, Belgium: Health and Con-
sumer Protection Directorate-General.
ª 2005 The Society for Applied Microbiology, Journal of Applied Microbiology 100 (2006) 40–49
Lab biodiversity in fermented sausages
Schillinger, U. and Lücke, F.K. (1987) Identification of lactoba-
cilli from meat and meat products. Food Microbiol 4, 199–
208.
Schillinger, U. and Lücke, F.K. (1989) Antibacterial activity of
Lactobacillus sake isolated from meat. Appl Environ Micro-
biol 55, 1901–1906.
Silla-Santos, M.H. (1998) Amino acid decarboxylase capability
of microorganisms isolated in Spanish fermented meat
products. Int J Food Microbiol 39, 227–230.
Tagg, J.R., Dajani, A.S. and Wannamaker, L.W. (1976) Bacteri-
ocins of gram-positive bacteria. Bacteriol Rev 40, 722–756.
Teuber, M. and Perreten, V. (2000) Role of milk and meat
products as vehicles for antibiotic-resistant bacteria. Acta
Vet Scand 93, 75–87.
Torriani, S., Dellaglio, F. and Palummeri, M. (1990) Charac-
terization of lactobacilli isolated from Italian salami. Ann
Microbiol 40, 225–233.
Vauterin, L. and Vauterin, P. (1992) Computer-aided objective
comparison of electrophoretic patterns for grouping and
identification of microorganisms. Eur Microbiol 1, 37–41.
Vidal-Carou, M.C., Izquierdo-Pulido, M., Martin-Morro, M.C.
and Mariné-Font, A. (1990) Histamine and tyramine in
meat products: relationship with meat spoilage. Food Chem
37, 239–249.
Vogel, R.F., Lohmann, M., Weller, A.N., Hugas, M. and
Hammes, W.P. (1991) Structural similarity and distribu-
tion of small cryptic plasmids of Lactobacillus curvatus and
Lactobacillus sake. FEMS Microbiol Lett 84, 183–190.
49