You are on page 1of 11

APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Sept. 2007, p. 5453–5463 Vol. 73, No.

17
0099-2240/07/$08.00⫹0 doi:10.1128/AEM.01072-07
Copyright © 2007, American Society for Microbiology. All Rights Reserved.

Microbial Ecology of the Soppressata of Vallo di Diano, a Traditional


Dry Fermented Sausage from Southern Italy, and In Vitro and
In Situ Selection of Autochthonous Starter Cultures䌤
Francesco Villani,1* Annalisa Casaburi,1 Carmela Pennacchia,1 Luisa Filosa,1
Federica Russo,1 and Danilo Ercolini2
Department of Food Science, School of Agriculture1 and School of Biotechnological Sciences,2 University of
Naples Federico II, Via Università 100, 80055 Portici, Italy
Received 14 May 2007/Accepted 26 June 2007

The microbial ecology of “soppressata of Vallo di Diano,” a traditional dry fermented sausage from southern
Italy, was studied by using both culture-dependent and culture-independent approaches. The ripened fer-
mented sausages were characterized by high microbial loads of both staphylococci and lactobacilli. Using
PCR-denaturing gradient gel electrophoresis (PCR-DGGE) targeting the variable V3 and V1 regions of the 16S
rRNA gene and direct DNA sequencing, it was possible to identify Staphylococcus xylosus, S. succinus, and S.
equorum among the staphylococci and Lactobacillus sakei and L. curvatus within the lactobacilli. Moreover,
Debaryomyces hansenii was the main yeast species found by targeting the yeast 26S rRNA gene by PCR-DGGE.
Selected strains of S. xylosus, L. sakei, and L. curvatus were characterized for their technological properties in
the ripening conditions of the fermented sausages so as to select an autochthonous starter formulation. The
selection included the determination of nitrate reductase, lipolytic, and antioxidant activity and proteolysis
with myofibrillar and sarcoplasmic protein fractions. Such properties were evaluated in both in vitro and in
situ assays; the latter were performed by using each strain as a starter in the laboratory-scale manufacture of
soppressata of Vallo di Diano and by monitoring the microbiological and chemical changes at the end of
ripening. The results show differences between the in vitro and in situ selection results and indicate that in situ
evaluation of the technological performance of specific strains is better suited to selecting autochthonous
starter cultures for fermented-meat products than in vitro evaluation.

Italy has a long history in the production of traditional fer- terococci, and yeasts, may play a role. LAB and CNC are
mented sausages (59), and almost every Italian region offers actively involved in the development of texture, color, and
one or more of these much-appreciated products, some of flavor and exert a positive effect on the hygienic properties of
which have been awarded Protected Designation of Origin and the product, inhibiting pathogenic or spoilage flora by acidifi-
Protected Geographical Indication labels (http://europa.eu.int cation or by production of antimicrobials (2, 51, 55, 56).
/comm/agriculture/qual/en/pgi_03en.htm). In the Campania The diversity of LAB and staphylococcal populations occur-
region (southern Italy), there are many small producers mak- ring in Italian fermented sausages was recently studied (6, 8,
ing dry fermented-meat products called “soppressata” which, 29, 41, 53). The profiling of bacterial populations present in
like most fermented meats, are the result of biochemical, mi- fermented meat can be useful to identify the species involved
crobiological, physical, and sensorial changes occurring in a in the fermentative process and, eventually, to select specific
meat mixture during ripening in defined conditions of temper- strains to be employed in an autochthonous starter culture.
ature and relative humidity (RH). These products, which are The use of starter cultures is being increasingly required to
prepared without using selected starter cultures to ensure their guarantee safety and standardization of product properties,
characteristic organoleptic properties, are not usually pro- including consistent flavor and color and shorter ripening time
duced on a large-scale basis but, rather, are sold on local (14).
markets as “traditional products” (8). The physiological and technological properties of the main
According to conventional and molecular microbiological groups of bacteria involved in the fermentation of dry cured-
studies, the ripening process of fermented sausages is domi- meat products (i.e., bacteriocin production, proteolysis, lipol-
nated by lactic acid bacteria (LAB), represented mainly by ysis, and nitrite and nitrate reductase and probiotic properties)
Lactobacillus sakei, L. curvatus, and L. plantarum and by co- have often been studied (2, 8, 9, 23, 24, 30, 31, 35, 36, 42, 43,
agulase-negative cocci (CNC) represented by the Staphylococ- 50, 51, 52, 55, 58). However, in most cases the selection and
cus and Kocuria genera (14, 18, 19, 26, 34, 44). In addition, technological characterization of the cultures are based on
depending on the product, other groups, such as molds, en- assays performed in vitro, while more information could be
gathered if the same properties were tested in a real meat
fermentation system. The use of starter cultures that include
* Corresponding author. Mailing address: Dipartimento di Scienza strains isolated from local products is essential in the artisanal
degli Alimenti, Università degli Studi di Napoli Federico II, Via Uni-
versità 100, 80055 Portici (NA), Italy. Phone: 39 081 2539403. Fax: 39
manufacture of traditional fermented products, since such
081 2539407. E-mail: villani@unina.it. strains are well adapted to the particular environment and

Published ahead of print on 6 July 2007. specific manufacturing technology (26, 39). Therefore, appro-

5453
5454 VILLANI ET AL. APPL. ENVIRON. MICROBIOL.

TABLE 1. Fermented sausages analyzed in this study: microbial loads and pH and aw valuesa
CFU/g in:
Soppressata
pH aw DRBC agar ⫹ CSSc
sampleb MRS agarc MSAc
Slanetz and Bartley
VRBGAc TSN agarc
mediumc Yeasts Molds

A 6.54 0.90 7.4 ⫻ 106 3.4 ⫻ 107 3.0 ⫻ 106 5.5 ⫻ 104 1.7 ⫻ 104 1.3 ⫻ 103 ⬍10
B 6.78 0.87 3.2 ⫻ 109 1.8 ⫻ 107 1.2 ⫻ 106 6.5 ⫻ 103 5.0 ⫻ 102 2.9 ⫻ 102 ⬍10
C 6.44 0.82 1.4 ⫻ 108 1.6 ⫻ 107 1.5 ⫻ 107 1.7 ⫻ 104 2.5 ⫻ 103 1.2 ⫻ 105 ⬍10
D 6.05 0.85 1.2 ⫻ 107 2.4 ⫻ 104 2.0 ⫻ 103 2.0 ⫻ 103 ⬍103 ⬍10 ⬍10
E 5.87 0.90 1.0 ⫻ 108 1.9 ⫻ 105 ⬍103 8.4 ⫻ 103 5.0 ⫻ 103 5 ⫻ 101 ⬍10
F 6.22 0.89 1.4 ⫻ 106 ⬍104 ⬍103 ⬍102 ⬍102 ⬍10 ⬍10
G 6.27 0.88 3.0 ⫻ 108 1.5 ⫻ 105 ⬍103 2.6 ⫻ 103 ⬍102 1 ⫻ 101 ⬍10
H 5.87 0.89 4.1 ⫻ 107 1.7 ⫻ 106 1.04 ⫻ 106 3.9 ⫻ 104 ⬍102 ⬍10 ⬍10
I 6.78 0.85 1.3 ⫻ 108 1.5 ⫻ 106 7.5 ⫻ 104 4.0 ⫻ 102 2.6 ⫻ 103 ⬍10 ⬍10
L 6.19 0.90 8.0 ⫻ 107 5.0 ⫻ 106 3.5 ⫻ 105 3.0 ⫻ 103 2.0 ⫻ 103 1.3 ⫻ 103 ⬍10
a
Values are expressed as means based on results determined in duplicate experiments. Standard deviations were always lower than 20% of the means.
b
All samples were purchased in the Cilento area of Italy (Salerno-Campania region) from 10 different farms.
c
See Materials and Methods for details. CSS, chloramphenicol selective supplement; VRBGA, violet red bile glucose agar; TSN, tryptone-sulfite-neomycin.

priate starter cultures have to be selected according to the for the absence of plasmid-mediated (tetracycline, erythromycin, oleandomycin,
specific formulation of the batter and technology of fermenta- lincomycin, and clindamycin) antibiotic resistance according to the method of
Charteris et al. (11) and for their capability to survive at low pH as previously
tion since environmental factors interact to select a limited described (42). Selected strains were identified by 16S rRNA gene sequencing as
number of strains that are competitive enough to dominate the previously reported (4) (see Table 8). Both staphylococci and lactobacilli were
process (48). Environmental factors possibly affecting strain also selected for their growth ability at pH 5.5 and at 14°C determined according
selection are wild microbial populations, ripening conditions, to Casaburi et al. (9).
Collection of bulk cells. In order to investigate the cultivable microbiota of the
temperature, RH, pH, presence of NaCl, raw materials, and
sausages, bulk cells (22) were collected from LAB, gram-positive CNC, entero-
ingredients. In order to make the ideal starter culture for any cocci, and yeast cells by using the whole content of the countable dilution plates
particular technology and recipe, it is necessary to understand used for the enumeration of these microbial groups. In particular, for each
the function we seek and to have tools to monitor the efficacy microbial group, all colonies present on the specific growth medium were resus-
of the culture (27). pended in 5 ml of tryptone soy broth (Oxoid), added to 20% glycerol, and stored
at ⫺20°C prior to further analysis.
The aims of this study were (i) to study, by culture-depen-
DNA extraction. Ten grams of each soppressata sample was diluted with 90 ml
dent and -independent molecular methods, the microbial ecol- of Ringer solution (Oxoid) and treated with a Stomacher machine for 2 min. Two
ogy of the traditional soppressata of Vallo di Diano in order to milliliters of the solution obtained was centrifuged at 17,000 ⫻ g for 10 min at
understand the microbial species occurring in this kind of fer- room temperature and washed once with 200 ␮l of TE (10 mM Tris-HCl, 1 mM
mentation and (ii) to isolate and select suitable strains of LAB EDTA, pH 8.0). After washing, the solution was again centrifuged at 17,000 ⫻
g for 10 min at room temperature. The pellet was suspended in 100 ␮l of TE and
and staphylococci, by in vitro and in situ technological charac- used for DNA extraction with a Wizard DNA purification kit (Promega, Madi-
terization, that could be used in autochthonous starter cultures son, WI) as previously described (22). After DNA precipitation with isopropanol
in manufacturing the same sausages. (0.7 volume), the resulting pellet was washed with 70% ethanol. It was then dried
and resuspended in 50 ␮l of DNA rehydration solution by incubation at 60°C for
45 min. Finally, 5 ␮l of RNase (10 mg/ml) was added and the solution was
MATERIALS AND METHODS incubated at 37°C for 30 min. The DNA extracted was stored at ⫺20°C. For
Enumeration and isolation of bacteria. Ten samples of dry fermented sau- DNA extraction from bulk cells, 500 ␮l of the bulk mixture was centrifuged at
sages, labeled A to L (Table 1), were purchased from different local factories in 17,000 ⫻ g for 10 min at room temperature. The pellet was suspended in 100 ␮l
Campania (southern Italy). The pH and water activity (aw) values for all of the of TE and used for DNA extraction as described above. Finally, 500 ␮l of yeast
samples were determined. Standard enumeration methods were used to deter- bulk cells was centrifuged at 17,000 ⫻ g for 10 min at room temperature. The
mine the microbial populations at the end of ripening. Ten grams of each sample pellet was used for DNA extraction with InstaGene matrix (Bio-Rad Laborato-
was transferred into a sterile Stomacher bag, diluted with 90 ml of Ringer ries, Hercules, CA) by following the conditions described by the supplier.
solution (Oxoid), and treated with a Stomacher machine for 2 min. Serial dec- PCR-denaturing gradient gel electrophoresis (DGGE) analysis. The primers
imal dilutions were made, and several microbial populations were targeted in V3f and V3r spanning the V3 region (positions 340 to 356 and 533 to 517,
duplicate as follows: (i) LAB on De Man, Rogosa, Sharpe (MRS) agar (Oxoid) respectively) of the 16S rRNA gene (38) were used. A GC clamp was added to
incubated in anaerobic conditions at 30°C for 48 h, (ii) gram-positive CNC on the forward primer according to the method of Muyzer et al. (38). Amplification
mannitol salt agar (MSA; Oxoid) incubated at 30°C for 48 h, (iii) enterococci on was performed in a programmable heating incubator (MJ Research Inc., Mad-
Slanetz and Bartley medium (Oxoid) incubated at 37°C for 48 h, (iv) yeasts and ison, WI). Each mixture (final volume, 50 ␮l) contained 1 ␮l of template DNA
molds on dichloran rose Bengal chloramphenicol (DRBC) agar base (Oxoid) (about 25 ng), each primer at a concentration of 0.2 ␮M, each deoxynucleoside
supplemented with chloramphenicol selective supplement (Oxoid) incubated at triphosphate at a concentration of 0.25 mM, 2.5 mM MgCl2, 5 ␮l of 10⫻ PCR
30°C for 72 h, (v) total enterobacteria on violet red bile glucose agar (Oxoid) buffer (Invitrogen), and 2.5 U of Taq polymerase (Invitrogen). Template DNA
incubated with a double layer at 30°C for 24 to 48 h, (vi) clostridia on tryptone- was denatured at 94°C for 5 min, and a touchdown PCR was performed (38). The
sulfite-neomycin agar (Sanofi Diagnostics Pasteur, Marnes, La Coquette, primers NL1 and LS2 (12) spanning the 5⬘-end region (positions 266 to 285 on
France) incubated in anaerobic conditions at 37°C for 72 h. Three to five colonies the S. cerevisiae 26S rRNA gene (GenBank accession number M19229) were
of LAB and staphylococci were randomly isolated from countable MRS and used to amplify DNA extracted from the yeast bulks. A GC clamp was added to
MSA plates, purified, and stored at ⫺20°C in MRS and tryptone soy broth with primer NL1 (38). For the amplification a mixture (final volume, 50 ␮l) was
25% glycerol. Species-specific PCR assays performed by using two set of primers prepared as described above. Reactions were run for 30 cycles: denaturation was
targeting xylulokinase (xylB) and 60-kDa heat shock protein genes (hsp60) of S. at 95°C for 60 s, annealing was at 52°C for 45 s, and extension was at 72°C for
xylosus were used as previously reported (5) to identify the staphylococcal iso- 60 s. An initial denaturation at 95°C for 5 min and a final extension at 72°C for
lates belonging to the species S. xylosus. The isolated lactobacilli were selected 7 min were used. The primers P1f and P2r spanning the V1 region (positions 41
VOL. 73, 2007 MICROBIAL ECOLOGY OF THE SOPPRESSATA OF VALLO DI DIANO 5455

to 60 and 130 to 111, respectively) of the 16S rRNA gene (32) were used to pressata of Vallo di Diano. A pork meat dough was prepared by mixing for 5 min
amplify DNA extracted from bulk cells collected from MRS agar. A GC clamp the following ingredients: lean pork (76%), pork fat (4%), NaCl (2.8%), black
was added to the forward primer according to the method of Muyzer et al. (38). pepper in grains (0.6%), white wine, and potassium nitrate (250 ppm). Aliquots
Each mixture (final volume, 50 ␮l) was prepared as described above. Template (200 g) of dough were inoculated with cell suspensions of each Lactobacillus or
DNA was denatured at 94°C for 5 min, and a touchdown PCR was performed Staphylococcus strain at a final concentration of 107 CFU g⫺1; the sausages were
(38). Aliquots (3 ␮l) of all the PCR products were routinely checked on 2% prepared in natural casings, dried at 18°C for 4 h at 86% RH, and then ripened
(wt/vol) agarose gels. for 14 days at 14°C and 68% RH. Every strain was assayed in duplicate tests;
PCR products were analyzed by DGGE by using a Dcode apparatus (Bio- noninoculated sausages were prepared and ripened as a control. At time zero
Rad). Samples were applied to 8% (wt/vol) polyacrylamide gels in 1⫻ Tris- and after 14 days of ripening, sausage samples were taken for microbiological
acetate-EDTA buffer. Parallel electrophoresis experiments were performed at and chemical analyses. Microbial counts were performed on MRS agar under
60°C by using gels containing a 20%:50% and a 30%:50% urea-formamide anaerobic conditions after 48 h at 30°C and for staphylococci on MSA after 48 h
denaturing gradient (100% corresponded to 7 M urea and 40% [wt/vol] form- at 30°C. pH measurement was performed by using a Mettler-Toledo GmbH pH
amide) for the separation of V3 (or D1) and V1 amplicons, respectively. The gels meter (model MP-120-B; Novate Milanese).
were analyzed by electrophoresis for 4 h at 200 V, stained with ethidium bromide The nitrate reductase activity was determined spectrophotometrically accord-
for 5 min, and rinsed for 10 min in distilled water. ing to the method of Baldini et al. (3). Briefly, nitrates and nitrites were extracted
Sequencing of DGGE bands. Small pieces of selected DGGE fragments were from meat and treated with Carrez reagents, and the colorimetric determination
punched from the gel with sterile pipette tips. The pieces were transferred into was performed at 540 nm after reduction of residual nitrates to nitrites on a
20 ␮l of sterile water and incubated overnight at 4°C. Two milliliters of the eluted cadmium sponge. The nitrate and nitrite content was expressed in milligrams per
DNA was used for reamplification, and the PCR products were checked by kilogram of meat. For the proteolytic activity analyses the sarcoplasmic and
DGGE; DNA amplified from the respective DGGE profile was used as a control. myofibrillar proteins were extracted according to the method described by Mau-
Only products that migrated as a single band and at the same position as the riello et al. (35) and the protein concentrations were determined using a Bio-Rad
control were purified by using a QIAquick PCR purification kit (QIAGEN, protein assay with bovine serum albumin as the standard. The extracts were
Milan, Italy) and sequenced by use of primer V3r, P2r, or NL1. The DNA diluted to a final concentration of 1.77 mg ml⫺1, and SDS-PAGE was performed
sequences were determined by the dideoxy chain termination method by using a as described above. For the lipolytic activity, the total lipid content of the
DNA sequencing kit (Perkin-Elmer Cetus, Emeryville, CA) according to the samples was determined after they had been thawed and minced. The extrac-
manufacturer’s instructions. Research to identify DNA similarity was performed tion was performed using 5 g of minced meat according to the method of
using the GenBank and EMBL databases with the BLAST program to determine Folch et al. (25). The lipid phase was resuspended in ethanol:ethyl ether (1:2)
the closest known relatives of the partial 16S rRNA gene sequences obtained (1). and titrated with an alcoholic NaOH solution (0.02 N). The results are
In vitro technological characterization of Staphylococcus and Lactobacillus expressed as a percentage of oleic acid as previously described (36). A
strains. The strains belonging to the S. xylosus species were studied to evaluate lipolysis index was also calculated as the percentage of oleic acid/percentage
catalase, nitrate reductase, proteolytic, lipolytic, and antioxidant activities as of total lipids according to the method of Cattaneo et al. (10). The determi-
described in a previous work (9) in order to determine their potential for use as nation of MDA as a marker of oxidation was performed as described above
a starter for fermented sausage production. by use of 5 g of sausages.
Briefly, the nitrate reductase activity was evaluated spectrophotometrically Statistical analysis. Statistical analyses were performed using SPSS 11.5 soft-
according to Casaburi et al. (9) and lipolytic activity was assayed on pork fat ware for Windows (SPSS, Chicago, IL). The analysis of variance (Duncan test)
(using concentrations of 108 CFU ml⫺1 of each culture and incubation at 14°C was applied to determine differences in means. All the values represent the mean
for 14 days) by titration and expressed as a percentage of oleic acid as previously values obtained from three independent assays for each trait.
described (36). In order to determine the proteolytic activity, 1 ml of cell sus-
pension (108 CFU ml⫺1) in phosphate-buffered saline (Oxoid) was inoculated in
10 ml of sterile-meat sarcoplasmic (1.77 mg ml⫺1) or myofibrillar (0.12 mg ml⫺1)
extracts and incubated at 14°C for 14 days. At time zero and after 14 days the
RESULTS
cultures were analyzed by sodium dodecyl sulfate-polyacrylamide gel electro-
Enumeration of microorganisms and pH and aw measure-
phoresis (SDS-PAGE) as described by Laemmli (33) on 12% polyacrylamide gels
by use of a Mini-PROTEAN III cell (Bio-Rad) as electrophoresis equipment. ments. The results obtained by the enumeration of microor-
The electrophoretograms were run with Precision Plus Protein standards (Bio- ganisms and determination of the pH and aw values for the
Rad) of known molecular weight, and the gels were subsequently stained for 1 h soppressata samples are shown in Table 1. All the samples had
in a 0.1% (wt/vol) solution of brilliant blue R in methanol:acetic acid:water a pH far higher than 5.5 and had low aw values. The LAB load
(40:10:50) and then destained for 2 h in the staining solution without brilliant
blue. Sarcoplasmic and myofibrillar extracts were run as a control in the gel. Prior
was extremely heterogeneous, ranging from 7.4 ⫻ 106 (sample
to loading, the myofibrillar extracts were first dialyzed (3500 MWCO unit with F) to 3.2 ⫻ 109 (sample B) CFU g⫺1. Staphylococcal loads
Spectra/Por 3 tubing; Spectrum Medical Industries Inc., Los Angeles, CA) in ranged from 2.4 ⫻ 104 (sample D) to 3.4 ⫻ 107 (sample A)
distilled water overnight, lyophilized, and then loaded at a concentration of 1.77 CFU g⫺1. Enterococcal loads ranged from 2.0 ⫻ 103 (sample
mg ml⫺1 in sample buffer. The antioxidant properties of the strains were eval-
D) to 1.5 ⫻ 107 (sample C) CFU g⫺1. For samples E and G,
uated by determining the amount of malondialdehyde (MDA) produced in 10 ml
of broth containing 1% tryptone, 0.5% yeast extract, 3% NaCl, and 4% (g/vol) of enterococcal loads did not reach 103 CFU g⫺1. The numbers of
pork fat at pH 7.0 inoculated with each strain at a concentration of 108 CFU yeasts and molds did not differ greatly among all samples, with
ml⫺1 after incubation for 14 days at 14°C with shaking. The MDA concentration values ranging from 2.0 ⫻ 103 (sample D) to 7.2 ⫻ 104 (sample
was determined by using the thiobarbituric acid test described by Bolumar et al. A) CFU g⫺1. The number of Enterobacteriaceae was high only
(7). Briefly, 10 ml of broth was mixed with 10 mg of butyl hydroxide toluene and
20 ml of 5% trichloroacetic acid. The mixture was homogenized for 20 s and
in sample C (Table 1). An absence of clostridia was observed
centrifuged at 9,000 ⫻ g for 10 min at 4°C, and the supernatant was filtered for all the soppressata samples.
through Whatman paper. Four milliliters of thiobarbituric acid (0.02 M) was DGGE profiles and species identification. The total DNA
added to 4 ml of filtrate and incubated at 100°C for 1 h, and the mixture was extracted from each soppressata sample (A to L) was em-
then centrifuged at 9,000 ⫻ g for 5 min at 4°C. The supernatants obtained
ployed in PCR assays using the primers V3f and V3r to obtain
were spectrophotometrically read at 532 nm (7); the results as expressed as
milligrams of MDA per gram of pork fat. The acidifying activity of the 200-bp fragments of the V3 region of the 16S rRNA gene, and
Lactobacillus strains was evaluated by measuring growth (according to the the fragments were analyzed by DGGE. The results obtained
optical density at 600 nm) and pH every 24 h during growth in meat broth by this analysis are shown in Fig. 1; multiple bands could be
(0.1% glucose, 3% NaCl, 0.3% meat extract, 0.5% peptone, 250 ppm KNO3, detected in most of the samples. All the bands were excised
pH ⫽ 7.0) at 14°C for 14 days.
In situ technological characterization of Staphylococcus and Lactobacillus
from the gel and reamplified prior to sequencing. The results
strains. The in situ characterization of the technological properties of the se- of the identification are presented in Table 2. Members of the
lected strains was performed by simulating laboratory-scale manufacture of sop- genus Lactobacillus and the genus Staphylococcus were pre-
5456 VILLANI ET AL. APPL. ENVIRON. MICROBIOL.

FIG. 1. DGGE profiles of the V3 region of the 16S rRNA gene


obtained by PCR amplification of DNA extracted directly from the 10 FIG. 2. DGGE profiles of the V3 region of the 16S rRNA gene
soppressata samples A, B, C, D, E, F, G, H, I, and L. Identification of obtained by PCR amplification of DNA extracted from bulk cells
each band is presented in Table 2. collected from MSA for soppressata samples A, B, C, E, G, H, I, and
L. Identification of each band is reported in Table 3.

dominant in the samples analyzed. In particular, Lactobacillus


spp. were found in samples B, E, G, H, I, and L. The species detection limit. DGGE profiles obtained by the analysis of the
group L. sakei/curvatus/graminis was found in samples B, E, I, V3 region of the 16 rRNA gene of staphylococci from MSA are
and L, while L. plantarum occurred in samples G and H (Table shown in Fig. 2; the closest relatives of the sequenced frag-
2). Members of the genus Staphylococcus were found in five ments are reported in Table 3. The species mainly detected in
samples (A, B, C, D, and H). S. succinus was found in samples the eight bulk cells analyzed were S. xylosus, S. equorum, and S.
C and D, while S. equorum was found in sample B (Table 2). succinus. Each species was present in five samples, and sample
Enterococcus spp. were detected in samples C and L and mem- L was the only one containing all the three staphylococcal
bers of the genus Carnobacterium in samples E and F. Finally, species. In particular, S. xylosus was found in samples A, B, C,
Tetragenococcus halophilus was identified in sample I. I, and L; S. equorum in samples B, E, H, I, and L; and S.
PCR-DGGE was also applied to PCR products from bulk succinus in samples C, E, G, H, and L. DGGE profiles ob-
cells collected from MRS agar, MSA, Slanetz and Bartley tained by the analysis of the V3 region of the 16S rRNA gene
medium, and DRBC agar. Bulk cells were not identified when
the bacterial load for a certain group was low or below the

TABLE 3. Sequence information for the DGGE bands obtained by


TABLE 2. Sequence information for the DGGE bands obtained by analyzing the V3 region of the 16S rRNA gene of the microbial
analyzing the V3 region of the 16S rRNA gene of DNA extracted bulk cells collected from MSA
directly from soppressata samples
%
Banda Closest relative Accession no.
% Identity
Banda Closest relative Accession no.
Identity
A1 Staphylococcus xylosus AY126246 100
A1 Staphylococcus xylosus/ AY688109/AM237352 100 B1 Staphylococcus xylosus AY126253 99
saprophyticus B2 Staphylococcus equorum AY126195 99
B1 Lactobacillus sakei/ AF429524/AY375292/ 100 C1 Staphylococcus succinus AY748916 99
curvatus/graminis AJ621551 C2 Staphylococcus xylosus AY126253 99
B2 Staphylococcus spp.b AY748916 99 E1 Staphylococcus succinus AY748916 100
B3 Staphylococcus equorum AY126195 99 E2 Staphylococcus equorum DQ232735 100
C1 Enterococcus faecalis AY850358 99 G1 Staphylococcus succinus AY748916.1 100
C2 Staphylococcus succinus AY748916 98 G2 Staphylococcus pulvereri/ AY126218/ 100
C3 Staphylococcus spp. AY748916 99 lentus/vitulinus AY395014/
C4 Enterococcus spp. AY865651 98 AY688104
D1 Staphylococcus succinus AY748916 97 H1 Staphylococcus spp.b AY647290 100
D2 Staphylococcus spp. AY748916 99 H2 Staphylococcus succinus AY748916 100
E1 Carnobacterium spp. AY543037 100 H3 Staphylococcus spp. AY161046 99
E2 Lactobacillus curvatus DQ336384 98 H4 Staphylococcus spp. AY647290 99
F1 Carnobacterium spp. AF425608 100 H5 Staphylococcus equorum AY126195 100
G1 Lactobacillus plantarum AB125924 100 I1 Bacillus spp. AY941803 100
H1 Lactobacillus plantarum AB125924 100 I2 Bacillus pumilus AY462210 97
H2 Staphylococcus spp. DQ530545 100 I3 Staphylococcus xylosus AY126246 100
I1 Tetragenococcus AB041349 97 I4 Staphylococcus equorum DQ232735 99
halophilus L1 Bacillus subtilis AB210949 99
I2 Lactobacillus AF429524/AB260947/ 100 L2 Bacillus spp. AY941803 100
sakei/curvatus/graminis AM113778 L3 Staphylococcus spp. DQ376925 98
L1 Enterococcus spp. AJ968602 100 L4 Staphylococcus succinus AY748916 100
L2 Lactobacillus sakei/ AF429524/AB260947/ 100 L5 Staphylococcus xylosus AY126246 100
curvatus/graminis AM113778 L6 Staphylococcus equorum DQ232735 98
a a
The letters and the numbers correspond to the bands shown in Fig. 1. The letters and the numbers correspond to the bands shown in Fig. 2.
b b
The species name is not indicated for cases in which the number of closest The species name is not indicated for cases in which the number of closest
relatives was ⬎3. relatives was ⬎3.
VOL. 73, 2007 MICROBIAL ECOLOGY OF THE SOPPRESSATA OF VALLO DI DIANO 5457

FIG. 3. DGGE profiles of the V3 region of the 16S rRNA gene


obtained by PCR amplification of DNA extracted from bulk cells
FIG. 4. DGGE profiles of the V1 region of the 16S rRNA gene
collected from MRS for soppressata samples A, B, C, D, E, F, G, H,
obtained by PCR amplification of DNA extracted from bulk cells
I, and L. Identification of each band is reported in Table 4.
collected from MRS for soppressata samples A, B, C, D, E, F, G, H,
I, and L. Identification of each band is reported in Table 5.

of LAB are shown in Fig. 3; the closest relatives of the se-


quenced fragments are reported in Table 4. ported by Cocolin et al. (14). With the sole exception of sample
Several bands of samples A, B, C, and D were identified as A, species of Lactobacillus were found in all bulk cells collected
representing non-Lactobacillus species after sequencing. Al- from MRS agar. L. sakei was found in samples D, F, and H and
most all the detected bands belonging to the genus Lactoba- L. curvatus in samples C and E, whereas both species were
cillus could not be identified at the species level but were detected in samples B, G, L, and I (Fig. 4; Table 5). Sequence
identified in groups of multiple species (Table 4). Therefore, to information for the DGGE bands obtained by analyzing the
improve the study of the LAB population, the V1 region of the V3 region of the 16S rRNA gene of the microbial bulk cells
16 rRNA gene was chosen for PCR-DGGE analysis of bulk collected from Slanetz and Bartley medium is reported in Ta-
cells from MRS agar (Fig. 4). The closest relatives of the ble 6. The bands displayed in Fig. 5 were identified as repre-
sequenced fragments are reported in Table 5. The primer set senting Enterococcus faecalis (samples A, B, C, and D) and E.
used was suitable for obtaining good differentiation among hirae (samples A and L). However, some fragments were iden-
Lactobacillus spp. without band comigration as already re- tified as harboring Staphylococcus spp. (samples A, B, and H)
or Lactobacillus spp. (sample L). Sequence information for the
DGGE bands obtained by analyzing the 26S rRNA gene of the
TABLE 4. Sequence information for the DGGE bands obtained by
analyzing the V3 region of the 16S rRNA gene of the microbial yeast communities from DRBC agar is reported in Table 7,
bulk cells collected from MRS medium and the corresponding DGGE profiles are shown in Fig. 6.
One band was detected in all the samples analyzed and was
%
Banda Closest relative Accession no. identified as representing Debaryomyces hansenii. Candida
Identity
zeylanoides was detected in samples B, C, and E, while Pichia
A1 Staphylococcus spp.b AY647303 99 guillermondii was detected only in sample H.
A2 Staphylococcus AY126226 99
saprophyticus In vitro characterization of Staphylococcus and Lactobacillus
A3 Lactobacillus spp. AY204898 98 strains. The results of the lipolytic activity analyses are shown
A4 Staphylococcus xylosus AY126253 98 in Table 8. Strong lipolytic activity was observed for the staph-
B1 Lactobacillus curvatus AY375292 96 ylococci, while weak lipolysis was observed for the lactobacilli.
B2 Lactobacillus sakei/ AF429524/AB289145/ 96
graminis/curvatus AY204894
Indeed, the oleic acid values were between 7.05% and 31.02%
C1 Enterococcus faecalis AY850358 100 for the S. xylosus strains whereas they ranged from 2.82% to
C2 Lactobacillus curvatus DQ336384 98
C3 Lactobacillus spp. AY204897 100
C4 Enterococcus spp. AY653231 97 TABLE 5. Sequence information for the DGGE bands obtained by
C5 Enterococcus spp. DQ376914 91 analyzing the V1 region of the 16S rRNA gene of the
D1 Lactobacillus curvatus DQ336384 98 microbial bulk cells collected from MRS medium
D2 Staphylococcus xylosus AY126246 99
E1 Lactobacillus graminis/ AB289303/AY204894 100 Banda Closest relative Accession no. % Identity
curvatus
F1 Lactobacillus sakei/ AF429523/AB063479/ 99 A1 Staphylococcus spp. AY189609 100
fuchuensis/curvatus AY204889 B1 Lactobacillus curvatus AY204894 98
G1 Lactobacillus sakei/ AY357581/AB289120/ 98 C1 Lactobacillus curvatus AY204894 98
fuchuensis/curvatus AY204889 D1 Lactobacillus sakei DQ631413 97
H1 Lactobacillus spp. AY204889 99 E1 Lactobacillus curvatus AY204894 98
I1 Lactobacillus sakei/ AF429524/AB289145/ 95 F1 Lactobacillus sakei DQ631413 97
graminis/curvatus AY204894 G1 Lactobacillus sakei DQ631413 97
I2 Lactobacillus spp. AY204898 95 G2 Lactobacillus curvatus AY204894 98
L1 Lactobacillus spp. AY204897 100 H1 Lactobacillus sakei DQ631413 97
L2 Lactobacillus spp. AY204897 100 I1 Lactobacillus curvatus AY204894 98
L1 Lactobacillus sakei DQ631413 97
a
The letters and the numbers correspond to the bands shown in Fig. 3. L2 Lactobacillus curvatus AY204894 98
b
The species name is not indicated for cases in which the number of closest
relatives was ⬎3. a
The letters and the numbers correspond to the bands shown in Fig. 4.
5458 VILLANI ET AL. APPL. ENVIRON. MICROBIOL.

TABLE 6. Sequence information for the DGGE bands obtained by TABLE 7. Sequence information for the DGGE bands obtained by
analyzing the V3 region of the 16S rRNA gene of the microbial analyzing the 26S rRNA gene of the yeast communities
bulk cells collected from Slanetz and Bartley medium present in soppressata samples
% Banda Closest relative Accession no. % Identity
Banda Closest relative Accession no.
Identity
A1 Debaryomyces hansenii AJ749831 99
A1 Enterococcus faecalis DQ239694 99 B1 Debaryomyces hansenii AJ749831 100
A2 Staphylococcus AY126226 100 B2 Candida zeylanoides AJ749829 100
saprophyticus C1 Debaryomyces hansenii AJ749831 99
A3 Enterococcus hirae EHI18354 98 C2 Candida zeylanoides CZY15471 100
B1 Enterococcus faecalis DQ239694 99 E1 Debaryomyces hansenii AJ749831 100
B2 Staphylococcus AY126226 99 E2 Candida zeylanoides AJ749829 100
saprophyticus G1 Debaryomyces hansenii AJ749831 100
B3 Enterococcus spp.b DQ467851 100 H1 Pichia guillermondii AY894828 100
C1 Enterococcus faecalis AY850358 99 H2 Debaryomyces hansenii AJ749831 99
D1 Enterococcus faecalis DQ239694 99
a
D2 Enterococcus spp. DQ467851 100 The letters and the numbers correspond to the bands shown in Fig. 6.
H1 Staphylococcus spp. AY647290 99
H2 Staphylococcus spp. AY647290 99
H3 Staphylococcus spp. AY647290 99 and IVS37 strains were able to hydrolyze the sarcoplasmic
H4 Staphylococcus spp. AY647290 99
L1 Lactobacillus graminis/ AB289145/ 92 proteins after 14 days of incubation (Fig. 7C), showing a re-
curvatus AB260947 duction in the 22.7-kDa band and digestion of the doublet of
L2 Enterococcus hirae Y18354 99 about 20 kDa. None of the staphylococci and lactobacilli
a
The letters and the numbers correspond to the bands shown in Fig. 5. showed proteolytic activity on the myofibrillar proteins in vitro,
b
The species name is not indicated for cases in which the number of closest since the corresponding SDS-PAGE profiles were identical to
relatives was ⬎3. that of the control (data not shown).
In growth and acidification assays, all the lactobacilli showed
slow growth at 14°C in meat broth; the highest effect on pH was
4.94% for the lactobacilli (Table 8). Moreover, no significant
shown by strains L. curvatus BVL7 and CVL12 and L. sakei
antioxidant activity was observed in the strains characterized;
DVL17 and HVL36, with a decrease in pH between 2.14 and
only L. sakei IVL42 was able to produce 4.23 mg MDA per g
2.4 units compared to the control.
of fat, a concentration which was lower than the concentration
In situ characterization of Staphylococcus and Lactobacillus
of MDA found in the control (Table 8).
strains. After 14 days of ripening of meat at 14°C, viable
All the strains of S. xylosus tested in this study showed nitrate
counts showed an increase in the loads of lactobacilli ranging
reductase activity when assayed in vitro; strains AVS5, IVS38,
between 0.5 and 0.7 log cycles and an increase in 1 log cycle
and IVS39 showed better activity than the others (results not
when strain L. sakei FVL26 was tested. The loads of staphy-
shown).
lococci increased by about 1 log cycle in most cases (Table 9).
The results of the analyses of proteolysis on sarcoplasmic
The pH did not decrease during the first days of ripening, while
proteins are shown in Fig. 7. Weak proteolytic activity was
it reached values of 5.53, 5.59, and 5.60 after 14 days when the
found for the lactobacilli except for the strain of L. sakei
meat had been inoculated with L. curvatus IVL41, DVL17, and
FVL26 that did not exhibit any activity on the sarcoplasmic
FVL26, respectively (Table 9). Overall, the decrease in pH was
proteins. The electrophoretic profiles showed digestion of the
registered after the first week of ripening.
14-kDa band for all the lactobacilli (Fig. 7A). The strongest
The results of lipolytic activity analyses showed very weak
proteolytic activity was shown by L. sakei IVL42 (Fig. 7A), L.
sakei BVL6, and L. curvatus BVL7 and CVL12 (Fig. 7B),
which showed a reduction in the 22.7- and 11.5-kDa bands and
an increase in the 13-kDa band. The S. xylosus IVS39, LVS48,

FIG. 5. DGGE profiles of the V3 region of the 16S rRNA gene FIG. 6. DGGE profiles of 26S 16S rRNA gene region obtained by
obtained by PCR amplification of DNA extracted from bulk cells PCR amplification of DNA extracted from bulk cells collected from
collected from Slanetz and Bartley medium for soppressata samples A, DRBC agar base for soppressata samples A, B, C, E, G, and H.
B, C, D, H, and L. Identification of each band is reported in Table 6. Identification of each band is reported in Table 7.
VOL. 73, 2007 MICROBIAL ECOLOGY OF THE SOPPRESSATA OF VALLO DI DIANO 5459

TABLE 8. In situ and in vitro technological performance of selected Staphylococcus and Lactobacillus strains evaluated after 14 days of
ripening and incubation at 14°Ca
Lipolytic activity Antioxidant activity In vivo nitrate reductase
Microrganism strainb (% oleic acid/g of fat) (mg MDA/g of fat) activity (ppm)

In vitro In situ In vitro In situ NaNO3 NaNO2

Control 2.82A 3.69ACF 4.55A 6.28ABCH 85.99A 62.77A


S. xylosus AVS3 31.02B 3.52ABC 6.15AE 7.20BCF 63.40B 33.92B
S. xylosus AVS4 29.61B 4.02ACEF 9.47B 5.51AD 56.45BCD 51.70C
S. xylosus AVS5 26.79C 4.52CDEF 6.33AE 5.86ACDH 54.56CD 48.49C
S. xylosus BVS10 31.02B 4.01ACEF 7.30BE 7.39CFI 54.69CD 81.17F
S. xylosus IVS37 17.63D 3.07AB 9.52B 4.65D 58.82BC 52.23C
S. xylosus IVS38 21.15E 2.63BC 8.60BE 5.30AD 50.78DF 29.38B
S. xylosus IVS39 10.58F 3.65ABF 50.67C 5.07AD 78.69E 72.21D
S. xylosus LVS48 17.63D 4.05ACEF 7.99BE 6.02ABH 45.47F 85.94FG
S. xylosus LVS49 7.05G 5.11D 31.32D 11.46E 59.68BC 91.17G
L. sakei BVL6 (AY204895) 3.53AH 4.64DEF 6.00AE 7.86FI NDc ND
L. casei BVL7 (D16550) 4.23AH 4.23CDEF 6.18AE 9.28G ND ND
L. curvatus CVL12 (AY204894) 4.94H 4.56CDEF 7.73BE 8.34GI ND ND
L. sakei DVL17 (DQ631413) 4.23AH 3.03AB 7.39BE 5.13AD ND ND
L. sakei FVL26 (DQ631413) 2.82A 5.07DE 7.93BE 12.09E ND ND
L. sakei HVL36 (AY204898) 4.94H 3.90ACF 6.81ABE 7.02BCFH ND ND
L. curvatus IVL41 (AY204894) 2.82A 3.94ACF 6.18AE 8.02FI ND ND
L. sakei IVL42 (AY204898) 2.82A 4.21CDEF 4.23A 8.04FI ND ND
a
Values in columns with differing letters are significantly different (P ⱕ 0.05). The values for the results of the activities assessed in vitro and in situ were significantly
different for both lipolytic activity determinations (P ⬍ 0.01) and MDA determinations (P ⫽ 0.029).
b
S. xylosus strains were identified according to the method of Blaiotta et al. (4); the lactobacilli were identified by 16S rRNA gene sequencing (5). Accession numbers
of the closest relatives are reported in parentheses (identity ⬎ 98%).
c
ND, not determined.

lipolysis during meat ripening, with percent oleic acid values concentration, but did not exhibit nitrite reductase activity
higher for lactobacilli than for staphylococci (Table 8). The (Table 8). S. xylosus strains AVS3, AVS4, AVS5, IVS37, and
lipolysis index confirmed the presence of better lipolytic activ- IVS38 showed both nitrate and nitrite reductase activities,
ity by lactobacilli; the greatest rise in the lipolytic index was while strain IVS38 showed the best nitrite reductase activity,
registered for the sausages inoculated with S. xylosus LVS49 reducing the nitrite concentration by 53%.
and L. sakei FVL26, which showed increases of 1.42 and
1.38%, respectively, compared to the control strain results DISCUSSION
(data not shown).
After the 14 days of ripening the MDA concentrations were In this study we first determined the microbial ecology of the
always higher than that seen with the control in the sausages traditional soppressata of Vallo di Diano by using combined
inoculated with any of the Lactobacillus strains except for L. molecular and traditional tools and then isolated LAB and
sakei DVL17 (Table 8). On the other hand, six of the nine staphylococci from the sausage ecosystem in order to select
sausages inoculated with S. xylosus showed MDA concentra- strains adapted to the specific fermentation for autochthonous
tions lower than the control results, with the lowest concentra- starter formulation. The study considered only 10 premium
tions shown by strains IVS37 and IVS39. The results of the quality sausages that were collected in the specific area of
assessment of activities in vitro and in situ were found to be production.
significantly different for both lipolytic activity (P ⬍ 0.01) and The viable counts of the different bacterial groups in the
MDA determinations (P ⫽ 0.029) (Table 8). soppressata samples showed the principal role played by lac-
After 14 days of ripening, the SDS-PAGE profiles of the tobacilli and staphylococci during ripening as widely acknowl-
sarcoplasmic proteins showed a reduction in the intensity of edged in the literature (14, 26, 48). These two populations
the 43.3- and 34.7-kDa bands in all the sausages, both inocu- showed the highest viable counts, although enterococci also
lated and control (Fig. 8). The strains of S. xylosus assayed in reached high loads in some samples.
situ did not exhibit proteolytic activity on the sarcoplasmic In this work microbial diversity was assessed by a PCR-
proteins, showing SDS-PAGE profiles identical to the control DGGE approach (21) to identify the main populations in-
profiles (Fig. 8). The lactobacilli caused hydrolysis of the 14- volved in the fermentative process of the soppressata of Vallo
kDa band; moreover, strains of L. sakei BVL6 and L. curvatus di Diano. The principal PCR-DGGE protocol used in this
BVL7 caused a disappearance of the 34.7-kDa band and a work was based on analysis of the variable V3 region of the 16S
reduction in the intensity of the 43.3-kDa band. As shown in rRNA gene and allowed good differentiation to be obtained
Fig. 9, none of the staphylococci or lactobacilli was able to between the mixed populations analyzed after DNA extraction
hydrolyze myofibrillar proteins during ripening of the fer- directly from food samples and from bulk cells of staphylo-
mented sausages. cocci, enterococci, and yeasts. Analysis of the V3 region did
The S. xylosus LVS48 strain showed the strongest nitrate not allow identification at species level in the study of the bulk
reductase activity in situ, causing a reduction in the nitrate cells of lactobacilli from MRS, and study of the V1 region was
5460 VILLANI ET AL. APPL. ENVIRON. MICROBIOL.

fermented sausages analyzed in this study; the main species


were L. sakei and L. curvatus. These were probably responsible
for the physical and organoleptic characteristics of the sau-
sages tested. Of the staphylococci, besides S. xylosus the pres-
ence of S. equorum and S. succinus suggests that the latter two
species may play a role in the ripening process of traditional
fermented-meat products, as already reported (44). The fre-
quent occurrence of L. sakei and L. curvatus among the LAB
and of S. xylosus among the cocci is in agreement with the study
of other traditionally fermented sausages, whose microbial
ecology was determined by both culture-dependent and cul-
ture-independent methods (13, 14, 16, 45, 46, 47).
The most frequently recurring yeast species found in bulk
cells collected in this research was Debaryomyces hansenii. The
dominance of this species during natural fermentation of Ital-
ian sausages was already highlighted in a recent research by
Cocolin et al. (15). This microorganism has shown good po-
tential for the hydrolysis of sarcoplasmic proteins and the gen-
eration of several polar and nonpolar peptides and free amino
acids (49).
According to our results, it seems that the ripening process
of artisanal “Soppressata del Vallo di Diano” is carried out by
a limited number of species of lactobacilli and staphylococci
that are able to better compete in the environmental condi-
tions of production. As the microbial floras in many fermented
sausages produced worldwide have been shown to be charac-
terized by the same species, the differences between all the
different products may be due to specific ingredients and man-
ufacturing conditions as well as biotypes within each species
that may differently contribute to ripening.
In the second part of this work the technological perfor-
mance of selected staphylococci and lactobacilli was studied in
vitro and in situ during ripening of fermented sausages of Vallo
di Diano in order to investigate the suitability of autochtho-
nous strains to act as starter for fermentation. Nitrate reduc-
tase activity is considered the principal technological feature
for selecting staphylococci as a starter (57). Our in vitro results
allowed nitrate reductase activity to be ascertained for all the
staphylococci and showed agreement with the results of other
studies (9, 17, 52). However, nitrate reductase activity per-
formed in situ gave information that was more useful, allowing
differentiation of the strains in their ability to reduce nitrates
and/or nitrites during ripening of fermented sausages.
As pointed out above, none of the selected strains exhibited
activity on myofibrillar proteins, in contrast with our previous
FIG. 7. SDS-PAGE of sarcoplasmic protein hydrolysis by different results (35, 36), where several S. xylosus strains exhibited pro-
strains of Lactobacillus and S. xylosus. M, Precision Plus Protein stan- teolytic activity on myofibrillar proteins even though the assays
dard; C0, uninoculated control at time zero; C14, uninoculated control were then performed at 30°C and not at 14°C as in this case.
after 14 days of incubation; a to p, samples containing different strains
after 14 days of incubation. (A) a, L. sakei FVL26; b, L. sakei HVL36; Therefore, the absence of such proteolytic activity may have
c, L. curvatus IVL41; d, L. sakei IVL42. (B) e, L. sakei BVL6; f, L. been the result of nonoptimal enzyme activity of our staphy-
curvatus BVL7; g, L. curvatus CVL12; h, L. sakei DVL17. (C) i, S. lococci at 14°C. In this study, several S. xylosus strains were
xylosus LVS49; l, S. xylosus AVS3; m, S. xylosus LVS48; n, S. xylosus able to hydrolyze sarcoplasmic proteins in a synthetic sub-
IVS39; o, S. xylosus IVS38; p, S. xylosus IVS37.
strate, as already found in a previous study (35). However, the
in situ assays did not confirm the results obtained in vitro:
several S. xylosus strains exhibited no proteolytic activity on
necessary to achieve this purpose, as already reported by other sarcoplasmic proteins during sausage fermentation, although
authors (14, 26). Bacterial identification at the species level, in such activity was shown in vitro. This could have been due to a
fact, is not possible for all genera, especially when only partial lack of dominance of the staphylococci over the autochthonous
16S rRNA gene sequences are analyzed (54). microbial flora of the pork meat, as confirmed by the viable
A strong presence of lactobacilli was found in the traditional counts (Table 9). Our lactobacilli proved able to hydrolyze the
VOL. 73, 2007 MICROBIAL ECOLOGY OF THE SOPPRESSATA OF VALLO DI DIANO 5461

TABLE 9. Viable counts of staphylococci and LAB and pH evolution during ripening of soppressata of Vallo di Diano used for in
vivo technological characterization of selected strains of Staphylococcus and Lactobacillus
CFU/g at indicated time (days)a
pH at indicated time (days)
Microorganism strain Staphylococci LAB

0 14 0 14 0 1 4 5 6 7 11 14

Control 3.95 ⫻ 104 5.56 ⫻ 106 3.12 ⫻ 104 1.32 ⫻ 108 5.76 5.75 5.702 5.67 5.61 5.61 5.70 5.80

S. xylosus strains
AVS3 2.11 ⫻ 107 1.67 ⫻ 107 3.12 ⫻ 104 2.93 ⫻ 108 5.76 5.79 5.77 5.74 5.72 5.68 5.73 5.76
AVS4 2.69 ⫻ 107 2.79 ⫻ 108 3.12 ⫻ 104 3.26 ⫻ 108 5.76 5.81 5.80 5.76 5.70 5.73 5.76 5.77
AVS5 2.49 ⫻ 107 8.25 ⫻ 106 3.12 ⫻ 104 4.85 ⫻ 107 5.76 5.80 5.75 5.77 5.75 5.73 5.91 5.80
BVS10 8.70 ⫻ 106 2.08 ⫻ 107 3.12 ⫻ 104 4.23 ⫻ 108 5.76 5.81 5.82 5.77 5.73 5.71 5.82 5.77
IVS37 1.90 ⫻ 107 2.85 ⫻ 107 3.12 ⫻ 104 4.57 ⫻ 108 5.76 5.80 5.84 5.80 5.75 5.71 5.72 5.76
IVS38 1.15 ⫻ 107 4.60 ⫻ 106 3.12 ⫻ 104 3.58 ⫻ 108 5.76 5.81 5.81 5.80 5.76 5.73 5.77 5.75
IVS39 7.70 ⫻ 106 4.13 ⫻ 107 3.12 ⫻ 104 2.55 ⫻ 108 5.76 5.76 5.77 5.73 5.74 5.72 5.90 5.86
LVS48 2.07 ⫻ 107 5.72 ⫻ 107 3.12 ⫻ 104 1.95 ⫻ 108 5.76 5.81 5.83 5.80 5.83 5.78 5.83 5.94
LVS49 2.16 ⫻ 107 4.85 ⫻ 107 3.12 ⫻ 104 7.10 ⫻ 108 5.76 5.78 5.80 5.78 5.73 5.69 5.71 5.75

Lactobacillus strains
L. sakei BVL6 3.95 ⫻ 104 1.33 ⫻ 107 2.58 ⫻ 107 5.83 ⫻ 108 5.76 5.73 5.49 5.49 5.48 5.49 5.55 5.61
L. casei BVL7 3.95 ⫻ 104 4.95 ⫻ 106 1.37 ⫻ 107 6.02 ⫻ 108 5.76 5.73 5.52 5.50 5.49 5.50 5.55 5.58
L. curvatus CVL12 3.95 ⫻ 104 2.95 ⫻ 106 4.18 ⫻ 107 6.90 ⫻ 108 5.76 5.75 5.56 5.56 5.57 5.57 5.69 5.73
L. sakei DVL17 3.95 ⫻ 104 4.70 ⫻ 106 3.72 ⫻ 107 6.34 ⫻ 108 5.76 5.73 5.44 5.43 5.46 5.47 5.63 5.59
L. sakei FVL26 3.95 ⫻ 104 1.00 ⫻ 107 3.73 ⫻ 107 1.01 ⫻ 109 5.76 5.69 5.50 5.49 5.47 5.47 5.53 5.60
L. sakei HVL36 3.95 ⫻ 104 3.44 ⫻ 107 5.09 ⫻ 107 7.00 ⫻ 108 5.76 5.73 5.48 5.48 5.50 5.53 5.64 5.63
L. curvatus IVL41 3.95 ⫻ 104 4.25 ⫻ 106 1.48 ⫻ 107 8.75 ⫻ 108 5.76 5.73 5.54 5.50 5.51 5.54 5.60 5.53
L. sakei IVL42 3.95 ⫻ 104 1.58 ⫻ 105 5.30 ⫻ 107 7.35 ⫻ 108 5.76 5.73 5.50 5.50 5.50 5.53 5.70 5.73
a
Values are expressed as means based on results determined in duplicate experiments. Standard deviations were always lower than 20% of the means.

sarcoplasmic proteins both in situ and in vitro, but the sarco- A discrepancy between the results of the in situ and in vitro
plasmic fractions digested were shown to be different in the assays was also noted in the determination of lipolytic activity.
two cases except for the degradation of the 14-kDa fraction While the strains of S. xylosus showed strong lipolytic activity in
(Fig. 7A; Fig. 8). In other studies, strains of L. curvatus and L. vitro in agreement with other studies (28, 36, 37), they did not
sakei were shown to be capable of hydrolyzing 97-, 45-, 37-, and confirm their performance during ripening of the fermented
26-kDa sarcoplasmic fractions during incubation at 30°C in a sausages. This was also noted in a previous work in which S.
sarcoplasmic extract (23, 24). This evidence points to the role xylosus strains selected for good lipolytic activity in vitro proved
of temperature for such enzymatic activity and corroborates unable to release fatty acids when used as starters in the fer-
the differences between proteolytic assays performed in situ mentation of sausages (8). Our lactobacilli showed no lipolytic
and in vitro.

FIG. 9. SDS-PAGE of myofibrillar proteins extracted from differ-


FIG. 8. SDS-PAGE of sarcoplasmic proteins extracted from differ- ent sausages. M, Precision Plus Protein standard; C0, uninoculated
ent sausages. M, Precision Plus Protein standard; C0, uninoculated sausages (control) at time zero; C14, uninoculated sausages (control)
sausages (control) at time zero; C14, uninoculated sausages (control) after 14 days of ripening; a to e, sausages inoculated with different
after 14 days of ripening; a to e, sausages inoculated with different strains after 14 days of ripening (a, L. curvatus BVL7; b, L. sakei BVL6;
strains after 14 days of ripening (a, S. xylosus IVS39; b, S. xylosus c, L. sakei FVL26; d, S. xylosus IVS39; e, S. xylosus LVS48; f, L. sakei
LVS48; c, L. sakei FVL26; d, L. sakei BVL6; e, L. curvatus BVL7). HVL36).
5462 VILLANI ET AL. APPL. ENVIRON. MICROBIOL.

activity either in situ or in vitro, although lipolytic strains of lipolisi in salami industriali e in produzioni artigianali. Atti del XIII Con-
vegno Nazionale AIVI, Montesilvano, Italy, 6 June 2003.
Lactobacillus were selected in other works (28, 40). However, 11. Charteris, W. P., P. M. Kelly, L. Morelli, and J. K. Collins. 1998. Antibiotic
the absence of lipolysis by lactobacilli in situ seen in this work susceptibility of potential probiotic Lactobacillus species. J. Food Prot. 61:
is in agreement with other studies (20, 31) where LAB with 1636–1643.
12. Cocolin, L., L. F. Bisson, and D. A. Mills. 2000. Direct profiling of the yeast
lipolytic enzymes inactive in sausage-ripening conditions were dynamics in wine fermentations. FEMS Microbiol. Lett. 189:81–87.
described. When assays using synthetic media were performed, 13. Cocolin, L., M. Manzano, C. Cantoni, and G. Comi. 2000. Development of
no antioxidant activity was shown by our strains. However, a rapid method for the identification of Lactobacillus spp. isolated from
naturally fermented Italian sausage using a polymerase chain reaction-tem-
decreased MDA concentrations, indicating antioxidant activ- perature gradient gel electrophoresis. Lett. Appl. Microbiol. 20:126–129.
ity, were detected after 14 days of ripening of the fermented 14. Cocolin, L., M. Manzano, C. Cantoni, and G. Comi. 2001. Denaturing gra-
dient gel electrophoresis analysis of the 16S rRNA gene V1 region to mon-
sausages started with several staphylococci and L. sakei IVL42. itor dynamic changes in the bacterial population during fermentation of
In conclusion, different results can be obtained when testing Italian sausages. Appl. Environ. Microbiol. 67:5113–5121.
technological performance in vitro and in situ, suggesting that 15. Cocolin, L., R. Urso, K. Rantsiou, C. Cantoni, and G. Comi. 2006. Dynamics
and characterization of yeasts during natural fermentation of Italian sau-
the most suitable conditions need to be determined in advance sages. FEMS Yeast Res. 6:692–701.
for better selection of a starter mixture. Our results obtained 16. Comi, G., R. Urso, L. Iacumin, K. Rantsiou, P. Cattaneo, C. Cantoni, and L.
by inoculating the strains directly into Vallo di Diano sausages Cocolin. 2005. Characterization of naturally fermented sausages produced in
the north east of Italy. Meat Sci. 69:381–392.
showed that the background microbial population and the food 17. Coppola, R., M. Iorizzo, R. Saotta, E. Sorrentino, and L. Grazia. 1997.
matrix (raw materials, endogenous enzymes, etc.) chosen could Characterization of micrococci and staphylococci isolated from soppressata
affect the technological activities of the selected strains. There- molisana, a Southern Italy fermented sausage. Food Microbiol. 14:47–53.
18. Coppola, S., G. Mauriello, M. Aponte, G. Moschetti, and F. Villani. 2000.
fore, the results indicate that the most reliable selection is Microbial succession during ripening of Naples-type salami, a southern Ital-
achieved by testing the strains in the real meat system, where ian fermented product. Meat Sci. 56:321–329.
19. Coventry, J., and M. W. Hickey. 1991. Growth characteristics of meat starter
most of the ripening conditions are simulated. On the basis of cultures. Meat Sci. 30:41–48.
their technological performance during the ripening of the 20. Demeyer, D., M. Raemaekers, A. Rizzo, A. Holck, A. De Smedt, B. ten Brink,
fermented sausages, we consider S. xylosus AVS5 and L. sakei B. Hagen, C. Montel, E. Zanardi, E. Murbrekk, F. Leroy, F. Vanden-
driessche, K. Lorentsen, K. Venema, L. Sunesen, L. Stahnke, L. De Vuyst, R.
DVL17 to be the most suitable strains for use as the autoch- Talon, R. Chizzolini, and S. Eerola. 2000. Control of bioflavour and safety in
thonous starter in the manufacture of the specific product fermented sausages: first results of a European project. Food Res. Int.
considered in this study. Real manufacture of Vallo di Diano 33:171–180.
21. Ercolini, D. 2004. PCR-DGGE fingerprinting: novel strategies for detection
fermented sausages is in progress using this combination of of microbes in food. J. Microbiol. Methods 56:297–314.
strains as the starter. 22. Ercolini, D., G. Moschetti, G. Blaiotta, and S. Coppola. 2001. The potential
of a polyphasic PCR-DGGE approach in evaluating microbial diversity of
natural whey cultures for water-buffalo mozzarella cheese production: bias of
ACKNOWLEDGMENT “culture dependent” and “culture independent” approaches. Syst. Appl.
Microbiol. 24:610–617.
This study was supported by the Ministry of University, Rome, Italy 23. Fadda, S., Y. Sanz, G. Vignolo, M. C. Aristory, G. Oliver, and F. Toldrà.
(action PRIN 2004). 1999. Hydrolysis of pork muscle sarcoplasmatic proteins by Lactobacillus
curvatus and Lactobacillus sakei. Appl. Environ. Microbiol. 65:578–584.
24. Fadda, S., Y. Sanz, G. Vignolo, M. C. Aristory, G. Oliver, and F. Toldrà.
REFERENCES 1999. Characterization of muscle sarcoplasmic and myofibrillar protein hy-
1. Altschul, S. F., T. L. Madden, A. A. Schaffer, J. Zhang, Z. Zhang, W. Miller, drolysis caused by Lactobacillus plantarum. Appl. Environ. Microbiol. 65:
and D. J. Lipman. 1997. Gapped BLAST and PSI-BLAST: a new generation 3540–3546.
of protein database search programs. Nucleic Acids Res. 25:3389–3402. 25. Folch, J., M. Lees, and G. H. S. Sloane-Stanley. 1957. A simple method of
2. Aymerich, M. T., M. Hugas, and J. M. Monfort. 1998. Review: bacterioci- isolation and purification of total lipids from animal tissues. J. Biol. Chem.
nogenic lactic acid bacteria associated with meat products. Food Sci. Tech. 26:497–507.
Int. 4:141–158. 26. Fontana, C., P. S. Cocconcelli, and G. Vignolo. 2005. Monitoring the bacte-
3. Baldini, M., F. Fabietti, S. Giammarioli, R. Onori, L. Orefice, and A. Stac- rial population dynamics during fermentation of artisanal Argentinean sau-
chini. 1996. Metodi di analisi utilizzati per il controllo chimico degli alimenti. sages. Int. J. Food Microbiol. 130:131–142.
In Rapporti ISTISAN 96/34, p. 176. Istituto Superiore della Sanità, Rome, 27. Hansen, E. B. 2002. Commercial bacterial starter cultures for fermented
Italy. foods of the future. Int. J. Food Microbiol. 78:119–131.
4. Blaiotta, G., C. Pennacchia, D. Ercolini, G. Moschetti, and F. Villani. 2003. 28. Hugas, M., and J. M. Monfort. 1997. Bacterial starter cultures for meat
Combining denaturing gradient gel electrophoresis of 16S rRNA gene V3 fermentation. Food Chem. 59:547–554.
region and 16S-23S rRNA gene spacer region polymorphism analyses for the 29. Iacumin, L., G. Comi, C. Cantoni, and L. Cocolin. 2006. Ecology and dy-
identification of staphylococci from Italian fermented sausages. Syst. Appl. namics of coagulase-negative cocci isolated from naturally fermented Italian
Microbiol. 26:423–433. sausages. Syst. Appl. Microbiol. 29:480–486.
5. Blaiotta, G., C. Pennacchia, E. Parente, and F. Villani. 2003. Design and 30. Johansson, G., J. L. Berdague, M. Larsson, N. Tran, and E. Borch. 1994.
evaluation of specific primers for rapid and reliable identification of Staph- Lipolysis, proteolysis and formation of volatile components during ripening
ylococcus xylosus strains isolated from dry fermented sausages. Syst. Appl. of a fermented sausage with Pediococcus pentosaceus and Staphylococcus
Microbiol. 26:601–610. xylosus as starter cultures. Meat Sci. 38:203–218.
6. Blaiotta, G., C. Pennacchia, F. Villani, A. Ricciardi, R. Tofalo, and E. 31. Kenneally, P. M., R. G. Leuschner, and E. K. Arendt. 1998. Evaluation of the
Parente. 2004. Diversity and dynamics of communities of coagulase-negative lipolytic activity of starter cultures for meat fermentation purposes. J. Appl.
staphylococci in traditional fermented sausages. J. Appl. Microbiol. 97:271– Microbiol. 84:839–846.
284. 32. Klijn, N., A. H. Weerkamp, and W. M. de Vos. 1991. Identification of
7. Bolumar, T., Y. Sanz, M. Flores, M. C. Aristoy, F. Toldrà, and J. Flores. mesophilic lactic acid bacteria by using polymerase chain reaction-amplified
2006. Sensory improvement of dry-fermented sausages by the addition of variable regions of 16S rRNA and specific DNA probes. Appl. Environ.
cell-free extracts from Debaryomyces hansenii and Lactobacillus sakei. Meat Microbiol. 57:3390–3393.
Sci. 72:457–466. 33. Laemmli, U. K. 1970. Cleavage of structural proteins during the assembly of
8. Casaburi, A., M. C. Aristoy, S. Cavella, R. Di Monaco, D. Ercolini, F. Toldrá, the head of bacteriophage T4. Nature 227:680–685.
and F. Villani. 2007. Biochemical and sensory characteristics of traditional 34. Lücke, F. K., and H. Hechelmann. 1987. Starter cultures for dry sausages and
fermented sausages of Vallo di Diano (Southern Italy) as affected by use of raw ham. Composition and effect. Fleischwirtschaft 67:307–314. (In Ger-
starter cultures. Meat Sci. 76:295–307. man.)
9. Casaburi, A., G. Blaiotta, G. Mauriello, O. Pepe, and F. Villani. 2005. 35. Mauriello, G., A. Casaburi, and F. Villani. 2002. Proteolytic activity of
Technological activities of Staphylococcus carnosus and Staphylococcus simu- Staphylococcus xylosus strains on pork myofibrillar and sarcoplasmic proteins
lans strains isolated from fermented sausages. Meat Sci. 71:643–650. and use of selected strains in the production of “Naples-type” salami.
10. Cattaneo, P., C. Bersani, and S. Stella. 2003. Indice di maturazione e di J. Appl. Microbiol. 92:482–490.
VOL. 73, 2007 MICROBIAL ECOLOGY OF THE SOPPRESSATA OF VALLO DI DIANO 5463

36. Mauriello, G., A. Casaburi, G. Blaiotta, and F. Villani. 2004. Isolation and the microbial ecology of Italian fermented sausages. Appl. Environ. Micro-
technological properties of coagulase negative staphylococci from fermented biol. 71:1977–1986.
sausages of Southern Italy. Meat Sci. 67:149–158. 48. Rebecchi, A., S. Crivori, P. G. Sarra, and P. S. Cocconcelli. 1998. Physio-
37. Montel, M. C., F. Masson, and R. Talon. 1997. Bacterial role in flavour logical and molecular techniques for the study of bacterial community de-
development. Meat Sci. 49:111–123. velopment in sausage fermentation. J. Appl. Microbiol. 84:1043–1049.
38. Muyzer, G., E. C. de Waal, and A. G. Uiterlinden. 1993. Profiling of complex 49. Santos, N. N., R. C. Santos-Mendosa, Y. Sanz, T. Bolumar, M. C. Aristoy,
microbial populations by denaturing gradient gel electrophoresis analysis of and F. Toldrà. 2001. Hydrolysis of pork muscle sarcoplasmic proteins by
polymerase chain reactions-amplified genes coding for 16S rRNA. Appl. Debaryomyces hansenii. Int. J. Food Microbiol. 68:199–206.
Environ. Microbiol. 59:650–700. 50. Sanz, Y., S. Fadda, G. Vignolo, M. C. Aristoy, G. Oliver, and F. Toldrà. 1999.
39. Papa, F., and L. Grazia. 1990. Produzione ed. utilizzazione di colture starter Hydrolysis of muscle myofibrillar proteins by Lactobacillus curvatus and
fresche nell’industria dei salumi. Industrie Alimentari XXIX:537–540. Lactobacillus sake. Int. J. Food Microbiol. 53:115–125.
40. Papamanoli, E., N. Tzanetakis, E. Litopoulou-Tzanetaki, and P. Kotzekidou. 51. Stiles, M. E. 1996. Biopreservation by lactic acid bacteria. Antonie Leeu-
2003. Characterization of lactic acid bacteria isolated from a Greek dry- wenhoek 70:331–345.
fermented sausage in respect of their technological and probiotic properties. 52. Talon, R., D. Walter, S. Chartier, C. Barriere, and M. C. Montel. 1999. Effect
Meat Sci. 65:859–867. of nitrate and incubation conditions on the production of catalase and nitrate
41. Parente, E., S. Grieco, and M. A. Crudele. 2001. Phenotypic diversity of lactic reductase by staphylococci. Int. J. Food Microbiol. 52:47–56.
acid bacteria isolated from fermented sausages produced in Basilicata 53. Urso, R., G. Comi, and L. Cocolin. 2006. Ecology of lactic acid bacteria in
(Southern Italy). J. Appl. Microbiol. 90:943–952. Italian fermented sausages: isolation, identification and molecular charac-
42. Pennacchia, C., D. Ercolini, G. Blaiotta, O. Pepe, G. Mauriello, and F. terization. Syst. Appl. Microbiol. 29:671–680.
Villani. 2004. Selection of Lactobacillus strains from fermented sausages for 54. Vandamme, P., B. Pot, M. Gillis, P. De Vos, K. Kersters, and J. Swings. 1996.
their potential use as probiotics. Meat Sci. 67:309–317. Polyphasic taxonomy, a consensus approach to bacterial systematics. Micro-
43. Pennacchia, C., E. E. Vaughan, and F. Villani. 2006. Potential probiotic biol. Rev. 60:407–438.
Lactobacillus strains from fermented sausages: further investigations on their 55. Villani, F., L. Sannino, G. Moschetti, G. Mauriello, O. Pepe, R. Amodio-
probiotic properties. Meat Sci. 73:90–101. Cocchieri, and S. Coppola. 1997. Partial characterization of an antagonistic
44. Rantsiou, K., and L. Cocolin. 2006. New development in the study of the substance produced by Staphylococcus xylosus 1E and determination of the
microbiota of naturally fermented sausages as determined by molecular effectiveness of the producer strain to inhibit Listeria monocytogenes in Ital-
methods: a review. Int. J. Food Microbiol. 108:255–267. ian sausages. Food Microbiol. 14:555–566.
45. Rantsiou, K., E. H. Drosinos, M. Gialitaki, R. Urso, J. Krommer, J. 56. Villani, F., O. Pepe, G. Mauriello, G. Salzano, G. Moschetti, and S. Coppola.
Gasparik-Reichardt, S. Toth, I. Mataxopoulos, G. Comi, and L. Cocolin. 1994. Antimicrobial activity of Staphylococcus xylosus from Italian sausages
2005. Molecular characterization of Lactobacillus species isolated from nat- against Listeria monocytogenes. Lett. Appl. Microbiol. 18:159–161.
urally fermented sausages produced in Greece, Hungary and Italy. Food 57. Weber, H. 1994. Dry sausage manufacture, the importance of protective
Microbiol. 22:19–28. cultures and their metabolic products. Fleischwirtsch 74:278–282. (In Ger-
46. Rantsiou, K., G. Comi, and L. Cocolin. 2004. The rpoB gene as a target for man.)
PCR-DGGE analysis to follow lactic acid bacteria population dynamics 58. Wolf, G., and W. P. Hammes. 1988. Effect of haematin on the activities of
during food fermentation. Food Microbiol. 21:481–487. nitrate reductase and catalase in lactobacilli. Arch. Microbiol. 149:220–224.
47. Rantsiou, K., R. Urso, L. Iacumin, C. Cantoni, P. Cattaneo, G. Comi, and L. 59. Zambonelli, C., F. Papa, P. Romano, G. Suzzi, and L. Grazia. 1992. Micro-
Cocolin. 2005. Culture-dependent and -independent methods to investigate biologia dei salumi. Edagricole, Bologna, Italy.