1

International Journal of Food Microbiology

February 2018, Volume 266, Pages 31-41
Achimer
http://dx.doi.org/10.1016/j.ijfoodmicro.2017.10.015 http://archimer.ifremer.fr
http://archimer.ifremer.fr/doc/00407/51858/
© 2017 Elsevier B.V. All rights reserved.

Effect of vacuum and modified atmosphere packaging on

the microbiological, chemical and sensory properties of
tropical red drum (Sciaenops ocellatus) fillets stored at 4 °C
1, 2, 3 1 2 2 3
Silbande Adèle , Adenet Sandra , Chopin Christine , Cornet Josiane , Smith-Ravin Juliette ,
1 2, *
Rochefort Katia , Leroi Françoise

1
Pôle Agroalimentaire Régional de Martinique (PARM), impasse Petit-Morne, N° 375, 97232 Lamentin,
Martinique
2
Ifremer, Laboratoire Ecosystèmes Microbiens et Molécules Marines pour les Biotechnologies (EM3B),
rue de l'Ile d'Yeu, BP 21105, 44300 Nantes, France
3
Université des Antilles, Département Scientifique Inter facultaire (DSI), EA929 AIHP-GEODE (groupe
BIOSPHERES), BP 7209, 97275 Schœlcher, Martinique

* Corresponding author : Françoise Leroi, email address : Francoise.Leroi@ifremer.fr

Abstract :

Aims

The effect of vacuum (VP – 4 °C) and CO2/N2–atmosphere (MAP – 4 °C) packaging on the quality of
red drum fillets compared with whole gutted iced fish was investigated.

Methods and results

A metagenomic approach, bacterial enumeration and isolation, biochemical and sensory analyses were
carried out. The organoleptic rejection of whole fish was observed at day 15 whereas VP and MAP
fillets appeared unacceptable only after 29 days. At these dates, total mesophilic counts reached 107–
108 CFU g− 1. According to Illumina MiSeq sequencing, Arthrobacter, Chryseobacterium,
Brevibacterium, Staphylococcus and Kocuria were the main genera of the fresh red drum fillets. At the
sensory rejection time, lactic acid bacteria (LAB), particularly Carnobacterium sp., dominated the
microbiota of both types of packaging. The pH value of fresh samples was between 5.96 and 6.37 and
did not vary greatly in all trials. Total volatile basic nitrogen (TVBN) and trimethylamine (TMA)
concentrations were low and not represent reliable indicators of the spoilage, contrary to some biogenic
amines (cadaverine, putrescine and tyramine).

Conclusion

Chilled packed fillets of red drum have an extended shelf-life compared to whole gutted iced fish.
Overall, few differences in sensory and microbial quality were observed between the VP and MAP

Please note that this is an author-produced PDF of an article accepted for publication following peer review. The definitive
publisher-authenticated version is available on the publisher Web site.
 2

samples.

Significance and impact of the study

Next-Generation Sequencing (NGS) provided data on the microbiota of a tropical fish.

Highlights

► A polyphasic approach to characterize the microbial ecosystem of red drum is used ► At day 0, less
common genera (Chryseobacterium, Brevibacterium, etc.) dominated ► Chilled packed fillets had a
longer shelf-life than whole gutted iced fish ► Packaging favored the dominance of the LAB (particularly
Carnobacterium spp.) ► Cadaverine, putrescine and tyramine could be good indicators of the fillets
spoilage

Keywords : Seafood, Biogenic amine, VP, MAP, NGS, 16S rRNA gene

Please note that this is an author-produced PDF of an article accepted for publication following peer review. The definitive
publisher-authenticated version is available on the publisher Web site.
 ACCEPTED MANUSCRIPT

1. Introduction

Red drum (Sciaenops ocellatus) culture began in the late 1970s and now represents

a worldwide production of around 70,000 tons, mainly in China and then the USA

(FAO, 2014). In France, this species is one of the main farmed marine fish with farms

being mainly located in the overseas departments and territories. Currently, the

annual production of Martinique (approximately 40 tons) remains lower than the

PT
potential maximum yield estimated at 200 tons. The usual form of commercialization

RI
is iced whole gutted and scaled red drums but farmers need to develop new products

SC
like fillets to gain the local markets (Falguière and Buchet, 2002).

In the study of Li et al. (2013b), the shelf-life of ice-stored red drum fillets was 8 days.
NU
This is shorter than that of whole fish, which was established as 13–15 days by

Fauré (2009) and Régina et al. (2014). The use of vacuum and modified-atmosphere
MA

packaging in combination with chilled storage has been found to extend the shelf-life

of meagre fillets (Genç et al., 2013; Sáez et al., 2015; Sáez et al., 2014), a fish
D

belonging to the same family (Sciaenidae) as red drum, and also fillets of sea bream,
E

sea bass and bogue (Kakouri et al., 1997; Kostaki et al., 2009; Mendes and
PT

Gonçalves, 2008), cod (Dalgaard et al., 1993), yellow grouper (Li et al., 2011),
CE

rainbow trout (Frangos et al., 2010; Rodrigues et al., 2016) and swordfish (Kykkidou

et al., 2009). However, there are no data for packed fillets of red drum.
AC

The proximate composition of fresh red drum flesh is 74–80% moisture, 0.6–2.7%

fat, 19–24% protein and 1–1.3% ash and the muscle has a pH value of 6.3–6.8 (Leon

et al., 2008; Li et al., 2013b). As for the majority of fish species, these intrinsic

properties make the flesh an extremely perishable product due to both microbial

development and biochemical reactions occurring during processing and storage

3
 ACCEPTED MANUSCRIPT

(Andrade et al., 2014). Bacterial growth is generally responsible for sensory

deterioration (Dainty, 1996; Gram and Huss, 1996; Shewan, 1971).

The free amino acid content and bacterial composition of the fish influence the

production of biogenic amines during storage and three of them (histamine,

cadaverine and putrescine) are the most significant to monitor the fish safety and

quality (Bulushi et al., 2009). In several studies, biogenic amines accumulation has

PT
been correlated with the sensory evaluation and used as chemical indicator

RI
(Jørgensen et al., 2000a, 2000b; Kim et al., 2009; Özogul et al., 2002; Veciana-

SC
Nogues et al., 1997). Rodríguez-Méndez et al. (2009) developed a multisensory

system integrating the amount of biogenic amines to assess the fish freshness.
NU
The main bacteria present in VP and MAP packed fish fillets are H2S–producing

bacteria (including Shewanella putrefaciens and Photobacterium phosphoreum),

MA

Pseudomonas sp., lactic acid bacteria (LAB) and Enterobacteriaceae (Dalgaard et

al., 1993; Frangos et al., 2010; Kostaki et al., 2009; Kykkidou et al., 2009; Li et al.,
D

2011), but also Brochothrix thermosphacta (Kakouri et al., 1997). All these
E

microorganisms are often identified as the specific spoilage organism of various

PT

fishery products (Gram and Dalgaard, 2002; Gram et al., 2002; Gram, 2009).
CE

In addition to the traditional enumeration on culture media, the microbiota

composition can be analyzed by culture-independent methods such as denaturing

AC

gradient gel electrophoresis (DGGE) or temporal temperature gradient gel

electrophoresis (TGGE). More recently, next-generation-sequencing (NGS), such as

pyrosequencing 454 and Illumina MiSeq, has been successfully used to characterize

the bacterial ecosystems of various seafoods such as cod and salmon fillets, cold-

smoked salmon, cooked shrimp and yellowfin tuna raw steaks (Chaillou et al., 2014;

Leroi et al., 2015; Silbande et al., 2016).

4
 ACCEPTED MANUSCRIPT

The first objective of this study was to investigate the effect of packaging (vacuum

and CO2/N2–atmosphere) on the shelf-life and quality of red drum fillets from

Martinique in comparison with whole gutted iced fish. The second was to monitor in

detail the quantitative and qualitative evolution of the microbiota of VP and MAP

fillets, using microbiological (culture-dependent and culture-independent techniques),

chemical, biochemical and sensory analyses.

PT
RI
2. Materials and methods

SC
2.1. Red drum sampling and storage conditions

2.1.1. First trial: comparison of whole fish and packed fillets

NU
Red drum (Sciaenops ocellatus) provided from a fish farm located in the center of the

Atlantic coast of Martinique (14°41’2.4’’N; 60°54’7.8’’W). Fish were caught with a dip-
MA

net, immediately placed under ice and prepared (scaling, gutting and filleting). Nine

whole fish (approximate weight of 1 kg per fish) and fifteen fillets with skin
D

(approximate weight of 250–300 g per fillet) were received at the PARM laboratory 6
E

h after harvesting.
PT

The whole fish were stored in a cooler box by alternating a layer of fish placed on the
CE

belly with a layer of flake ice and kept in a cold room (4°C). To maintain these

samples at 0°C, melting water was drained off and ice was replaced when
AC

necessary. Fillets were divided into 2 batches. For the first batch, fillets were

vacuum-packed (VP) in 80-µm thick plastic bags (Garcia de Pou, Girona, Spain)

made of polyamide/polypropylene with a gas-permeability of 2.78 cm3/m2/day for

water vapor, 19.95 cm3/m2/day for O2 and 164.87 cm3/m2/day for CO2 using a

packaging machine (Multivac, Lagny sur Marne, France). Fillets of the second batch

were placed in the same type of plastic bags and packed under modified atmosphere

5
 ACCEPTED MANUSCRIPT

(MAP, 50% CO2–50% N2) using a Meca 500 machine (Mecapack, Pouzauges,

France). VP and MAP samples were stored at 4°C. For each sampling date, 3 pieces

(whole fish and fillets) were tested for sensory, chemical and bacteriological quality

and a mean value of the triplicate results was used as a representative value of the

sample. The sampling times were 0, 8 and 15 days.

2.1.2. Second trial: comparison of vacuum and modified atmosphere packed

PT
fillets

RI
Twenty-seven red drum fillets were brought back to the laboratory in the same

conditions as the first trial. Fillets were VP and MAP (50% CO2–50% N2). For this

SC
trial, the plastic bags of the MAP samples were replaced by a filmed plastic tray. The
NU
properties of the polyamide/polypropylene film (Pechiney, Paris, France) were a

thickness of 90 µm and gas permeability (cm3/m2/day at 23°C, 50% RH) of 4, 30, 120
MA

and 6 for water vapor, O2, CO2 and N2, respectively. The packed fillets were stored at

4°C. Sensory, chemical, biochemical and microbiological (culture-dependent and

D

culture-independent methods) analyses were carried out just before packaging (day
E

0) until the fillets were organoleptically unacceptable. A mean value of triplicate

PT

results (3 fillets) was used as a representative value of the sample for all the
CE

sampling dates (0, 8, 15, 22 and 29 days).

2.2. Sensory analyses

AC

2.2.1. Organoleptic inspection (PARM laboratory)

The degree of freshness of the whole raw fish was assessed with the rating scale

method developed by the Ifremer station in Martinique for farmed red drum (Fauré,

2009), based on visual properties. In this study, the gills could not be evaluated

because they were removed during the evisceration step. In brief, 2 trained local

judges had to score 8 criteria on a 6-point scale, with 0 representing good quality and

6
 ACCEPTED MANUSCRIPT

6 rotten fish (Table 1). When the mean of these scores (freshness index) was equal

to or higher than 2.8, the fish was rejected.

For fillets, the appearance (color, texture, slime formation, etc.) and the odor were

described in detail and an overall spoilage score was given to each fillet on a 10-

point scale, with 0 representing fresh flesh and 10 rotten flesh.

2.2.2. Spoilage score and odor profiles (Ifremer laboratory)

PT
At each sampling date, 50 g of flesh per fillet were diced. The triplicate were pooled

RI
in a single plastic bag (150 g), frozen at -80°C and shipped to the EM3B laboratory

SC
(Ifremer, Nantes, France) under the same temperature condition. A sensory session

was organized with 13 trained panelists to describe the odors in more detail. This
NU
session was carried out in individual testing booths according to the procedure NF V

09-105 (AFNOR, 1995), equipped with a computerized system (Fizz, Biosystèmes,

MA

Couternon, France). On the morning of the test, each packet (150 g diced) was

thawed, divided into individual portions (20-25 g), placed in plastic bowls with lids and
D

maintained in an oven at 18°C during the session. All products were coded with
E

random 3-digit numbers and served to the panelists in a predefined order to avoid a
PT

bias due to the effect of the first group tested. The set of samples was scored by 2
CE

different panelists with a minimum 20-min interval. This minimized the total quantity

of red drum flesh for sensory analysis. The panelists had to score on a continuous
AC

scale from 0 to 10 (6 being the limit of acceptability) the following appropriate odor

descriptors: fish, marine, plant, floor cloth, butter/caramel, rancid, sour/fermented,

feet/cheese, red meat/blood. Data were processed by analysis of variance with 2

factors (product, panelist). Principal component analysis (PCA) was performed for

the odor profile of samples. The statistical processes were carried out using Fizz 2.50

b 37 software (Biosystèmes, Couternan, France).

7
 ACCEPTED MANUSCRIPT

2.3. Bacterial counts

A 30-g portion of dorsal muscle without skin was collected from the whole fish and

fillets with the most stringent hygienic precautions (12°C-room, disinfection of

surfaces and equipment and use of a sterile scalpel). This portion was used to

enumerate Total Mesophilic Viable Counts (TMVC), Total Psychrotrophic Viable

Counts (TPVC), lactic acid bacteria (LAB), Brochothrix sp., Enterobacteriaceae and

PT
Pseudomonas sp., as described by Silbande et al. (2016).

RI
2.4. Biochemical analyses

SC
The pH was measured with a pH-meter (Inolab, Germany) in the mother solution (1/5

diluted flesh) prepared for microbial analysis.

NU
Total volatile basic nitrogen (TVBN) and trimethylamine (TMA) were determined in

100 g of fish using the Conway micro-diffusion method (Conway and Byrne, 1933).
MA

In the second trial, biogenic amines (putrescine, cadaverine, histamine, tyramine,

spermidine and spermine) were quantified in 5 g of each sample by high pressure

D

liquid chromatography (HPLC) using a Prominence 20A System (Shimadzu, Kyoto,

E

Japan). The dansyl-chloride derivatization was realized in accordance with Duflos et

PT

al. (1999). Peaks were detected with a UV-detector (SPD-20A, Shimadzu) operating
CE

at 254 nm.

2.5. Isolation, purification and identification of bacterial isolates

AC

In the second trial, twenty-two bacteria were isolated at day 0 (fresh fillet) and at the

sensory rejection time (day 29) of VP and MAP red drum fillets. Isolates were

selected by picking colonies with various morphologies from plates: 10 colonies from

Plate Count Agar (PCA) or Long and Hammer agar (LH) and 3 colonies from Elliker

agar (ELK), Streptomycin Sulfate Thallous Acetate Agar (STAA), Violet Red Bile

8
 ACCEPTED MANUSCRIPT

Glucose Agar (VRBGA) and CHROMagar Pseudomonas. The 66 resulting isolates

were purified, characterized and identified as described by Silbande et al. (2016).

2.6. Next-generation sequencing (NGS)

2.6.1. Total bacterial DNA extraction from red drum flesh

At each sampling date of the second trial, the 3 independent mother solutions of the

triplicate were pooled in equal proportions and the bacterial DNA was extracted and

PT
purified as described by Macé et al. (2012). The concentration and purity of DNA

RI
were assessed by the Quant-iT™ PicoGreen® dsDNA assay Kit (Invitrogen,

Carlsbad, CA). DNA samples were stored at −20°C and sent to MATIS (Reykjavik,

SC
Iceland) for 16S rRNA gene amplification and sequencing.
NU
2.6.2. Bacterial 16S rRNA gene amplification and barcoded sequencing

Bacterial DNA extracted from red drum at each sampling date was analyzed by
MA

Illumina MiSeq sequencing. The 16S Tag Sequencing workflow was performed

according to the protocol provided by Illumina “16S Metagenomic Sequencing Library

D

Preparation” (www.illumina.com). Modifications of this protocol are mentioned below.

E

Amplicon PCR: The DNA samples received were diluted to 10 ng/μl. The variable V3-
PT

V4 region of the 16S rRNA gene was amplified with the primers S-D-Bact-0341-b-S-
CE

17 and S-D-Bact_0785-a-A-21 (single amplicon of approximately 460 bp). The

standard PCR Master mix x1 contained 5 μl of template DNA, 0.5 μl of 10 mM dNTP,

AC

1.25 μl of 10 μM forward primer, 1.25 μl of 10 μM reverse primer, 0.25 μl of Q5 high-

fidelity polymerase, 5 μl of 5X Q5 high GC enhancer and 11.75 μl of molecular grade

water. The thermocycler program consisted of a denaturation step of 30 s at 98°C,

followed by 30 cycles of 10 s at 98°C, 30 s at 52°C and 30 s at 72°C, and a final

elongation step of 2 min at 72°C.

9
 ACCEPTED MANUSCRIPT

PCR clean-up 1: HighPrep™ PCR beads (MAGBIO GENOMICS, Gaithersburg, USA)

were used to purify the 16S V3-V4 amplicons from free primers and primer dimers.

The PCR products were cleaned according to the MAGBIO protocol

(www.magbiogenomics.com).

Index PCR: Dual indices and Illumina sequencing adapters were attached using the

Nextera XT Index Kit. The index PCR Master mix x1 contained 23.5 μl of molecular

PT
grade water, 10 μl of buffer HF, 1 μl of dNTPs, 0.5 μl of HF Polymerase, 5 μl of Index

RI
1, 5 μl of Index 2 and 5 μl of PCR product DNA. The thermocycler program consisted

SC
of a step of 30 s at 98°C, followed by 8 cycles of 10 s at 98°C, 30 s at 52°C and 30 s

at 72°C, and a final step of 7 min at 72°C. The PCR products were cleaned as
NU
described above.

Library quantification and normalization: The DNA library was validated on a

MA

Bioanalyzer DNA 7500 chip (Agilent, Santa Clara, USA) to verify the size of the

amplicons. Library quantitation was done by fluorometry using the Quant-iT

D

PicoGreen dsDNA assay Kit (Invitrogen, Carlsbad, CA).

E

Library denaturing and MiSeq sample loading: Eight pM final loading concentration
PT

for the best cluster density and 20% PhiX control was spiked in.
CE

Sequence processing, taxonomic assignment and analysis of diversity: Sequencing

data were analyzed using the Qiime 1.9.1 pipeline (Caporaso et al., 2010a). Forward
AC

and reverse reads were joined using fastq-join (Aronesty, 2011) and quality filtered

using default Qiime parameters (Bokulich et al., 2013). Chimera were detected using

USEARCH (Edgar, 2010) and chimera-filtered reads were clustered into operational

taxonomic units (OTUs) using UCLUST (Edgar, 2010) against the SILVA SSU

database release 119 (Quast et al., 2013) at the 97% identify level, as well as de

novo. A representative sequence was selected for each OTU and aligned against the

10
 ACCEPTED MANUSCRIPT

Silva core alignment using PYNAST (Caporaso et al., 2010b). Taxonomic

assignment was performed with UCLUST (Edgar, 2010) using the SILVA taxonomy.

Beta diversity was computed using the UniFrac method (Lozupone and Knight, 2005)

and visualized in three-dimensional PCoA plots using EMPEROR (Vázquez-Baeza et

al., 2013). A phylogenetic tree was built using FASTTREE (Price et al., 2010).

Sequences have been deposited at the European Nucleotide Archive (ENA) under

PT
the project accession number PRJEB20568.

RI
2.7. Statistical analysis

SC
Chemical and cultural enumeration data were analyzed using the software R (version

2.14.0). Descriptive statistics of means, standard deviation, linear regression, two-

NU
way ANOVA and Tukey’s HSD post-hoc test were applied. A significance level of 5%

was used.
MA

3. Results
E D

3.1. Comparison of whole fish and packed fillets

PT

3.1.1. Organoleptic inspection

At day 0, the whole red drums possessed all the characteristics of fresh fish
CE

(freshness index = 0, Table 1). After 8 days of iced-storage, the index reached a
AC

value of 1.8 due to a strong change in the shape (flat) and the color of the eyes

(duller pupil and opalescent cornea), a softening of the flesh and a slight pinkish

coloration of the muscle near the spinal column. The whole fish were rejected at day

15 with a freshness index equal to the acceptability limit of 2.8 (mean of the criteria

underlined in gray in Table 1).

Fresh fish fillets (day 0) were mainly characterized by a firm texture with a uniform

and normal color and a typical marine odor. After 8 days, all samples were of very

11
 ACCEPTED MANUSCRIPT

good quality and weak changes appeared at day 15. The MAP fillets were slightly

discolored but still considered of better quality than VP fillets, which released a weak

off-odor and were greenish in color. However, despite these impacts on the sensory

quality, these fillets remained acceptable throughout the storage period (data not

shown).

3.1.2. Chemical analyses

PT
The pH value of fresh flesh was equal to 5.96 ± 0.01. The flesh of whole fish slightly

RI
alkalinized with a final pH of 6.10 ± 0.07 while the pH of VP products fell to a value of

SC
5.87 ± 0.07 at day 15 (data not shown, significant difference with whole fish). The pH

of MAP fillets remained more stable (very weak acidification) during the storage, with
NU
a final value of 5.92 ± 0.09.

Fig. 1 represents the TVBN and TMA results. At day 0, TVBN concentrations of
MA

whole fish and fillets (VP and MAP) were 22.0 ± 4.6 mg-N 100 g-1 and 24.0 ± 2.6 mg-

N 100 g-1, respectively. A decrease was observed in the whole fish (16.6 ± 1.9 mg-N
D

100 g-1 at day 15) whereas TVBN increased in the packed fillets, particularly in VP
E

fillets where the value at the end of storage was equal to 28.4 ± 0.9 mg-N 100 g-1
PT

(p<0.05). For all batches, the TMA concentrations dropped from approximately 6 mg-
CE

N 100 g-1 at day 0 to around 2 mg-N 100 g-1 at day 8 and remained stable until day

15.
AC

3.1.3. Bacterial counts

As shown in Fig. 2, mesophilic bacteria (TMVC) were generally higher by 1 Log CFU

g-1 than the psychrotrophic flora (TPVC) in all samples and throughout storage. At

day 15, TMVC and TPVC ranged from 7.6 to 8.4 Log CFU g-1 and 6.3 to 7.3 Log CFU

g-1, respectively. LAB was the major group of bacteria enumerated in the 3 batches

(5.6 to 6.9 Log CFU g-1, depending on the storage), followed by Brochothrix sp. (4.8

12
 ACCEPTED MANUSCRIPT

to 6.2 Log CFU g-1), Enterobacteriaceae (4.0 to 5.7 Log CFU g-1) and Pseudomonas

sp. (2.2 to 3.4 Log CFU g-1 at day 15). All bacterial groups/genera grew faster and

reached higher counts in VP products than in whole fish and MAP fillets (significant

difference, p<0.05). At the end of storage, counts of LAB, Brochothrix and

Enterobacteriaceae were slightly higher than in iced whole fish.

3.2. Comparison of vacuum and modified atmosphere packed fillets

PT
3.2.1. Sensory analyses

RI
The results of the organoleptic evaluation performed at the PARM laboratory are

SC
presented in Table 2. Fresh red drum fillets (day 0) possessed a uniform bright

appearance with a typical slight pink color and a marine/fresh fish odor. Until day 15,
NU
few changes were detected and the overall score remained lower than or equal to 4

for the 2 batches. After 22 days of storage, MAP products had a duller color and a
MA

softer texture than VP products but the off-odor levels were similar. VP and MAP

fillets reached an overall spoilage score of 5.2 ± 0.3 and 5.5 ± 0.5, respectively (no
D

significant difference). VP and MAP fillets were considered unacceptable at day 29

E

(overall score > 6). A strong discoloration of the flesh was observed. VP samples had
PT

a firmer texture than MAP products but released a stronger odor on opening the
CE

packaging. The off-odor profiles were obtained in more detail by the sensory panel

from Ifremer. Fig. 3 shows the plane 1–3 (60.1% of the inertia) of the PCA performed
AC

with the mean scores of each odor descriptor. This plane was chosen because the

third axis (15.1%), related to the plant characteristic, enabled a better discrimination

of the VP and MAP fillets than the second one (19.1%). The first axis (45.0%)

discriminated unspoiled samples (days 0 and 8) with fish and marine characteristics

(left side of PCA) from more spoiled samples (days > 8) with slight off-odors such as

sour/fermented and butter/caramel (right side). VP fillets were characterized by a

13
 ACCEPTED MANUSCRIPT

plant odor whereas MAP samples released a mixture of various slight odors such as

butter/caramel, floor cloth and sour/fermented criteria.

3.2.2. Biochemical analyses

The initial pH (day 0) was equal to 6.37 ± 0.11. This value was fairly stable

throughout the storage and no significant difference (p > 0.05) was observed

between the 2 batches (data not shown). A very small production of TVBN was

PT
observed in samples with maximum concentrations of 18.5 ± 1.0 mg-N 100 g-1 and

RI
16.6 ± 1.0 mg-N 100 g-1 for VP and MAP samples, respectively (data not shown).

The TMA content was equal to 0.7 ± 0.2 mg-N 100 g-1 at the beginning and stabilized

SC
at around 2–3 mg-N 100 g-1 from the 15th day for VP products and from the 22nd day
NU
for MAP products (data not shown).

Fig. 4 shows the production of biogenic amines during storage. Putrescine and
MA

cadaverine were the major amines produced, particularly in VP samples where they

respectively reached 95 ± 15 mg kg-1 and 93 ± 16 mg kg-1 after 29 days, versus 70 ±

D

12 mg kg-1 and 48 ± 8 in MAP samples. Tyramine, absent at day 0, increased from

E

day 15 in VP and MAP products and reached 25 ± 5 mg kg-1 and 19 ± 6 mg kg-1,

PT

respectively, at the end of storage. Among the six biogenic amines, only spermidine
CE

and spermine were present in the fresh fillets (day 0) at a respective level of 8 ± 1 mg

kg-1 and 15 ± 2 mg kg-1 but no significant production was recorded (data not shown).
AC

Histamine was never detected in the red drum flesh.

3.2.4. Bacterial counts

The initial quality of fillets was poor with TMVC equal to 4.9 ± 0.1 Log CFU g-1 (Fig.

5). The TPVC was around 1 Log CFU g-1 lower (p<0.05) but increased more rapidly

than TMVC with both counts reaching their maximum concentration of 7.5–8 Log

CFU g-1 at day 15 in VP and MAP samples. The other enumerated bacteria were

14
 ACCEPTED MANUSCRIPT

present at a level of 3–4 Log CFU g-1 at the beginning of storage. The LAB count

increased rapidly up to 7.5 Log CFU g-1 at day 15 and this group was predominant in

both products (no significant difference), reaching around 8.0 Log CFU g-1 at the end

of storage. Brochothrix sp. also developed very quickly in the VP and MAP fillets and

reached 6.9 ± 0.2 Log CFU g-1 and 7.3 ± 0.1 Log CFU g-1, respectively. The

Enterobacteriaceae developed more slowly, except during the last week for VP

PT
storage when this family increased from 6.9 ± 0.4 Log CFU g-1 to 8.0 ± 1.0 Log CFU

RI
g-1. The Pseudomonas sp. count remained below 5.8 Log CFU g-1 throughout the

SC
analysis period. Overall, slightly higher counts of total mesophilic and psychrotrophic

bacteria, LAB and Brochothrix spp. were observed in MAP samples and conversely
NU
for Pseudomonas sp. and Enterobacteriaceae.

Sixty-six isolates (22 at day 0 and 22 at each sensory rejection time of VP and MAP
MA

products) were identified by the 16S rRNA sequencing gene (see Table 1 in Ref

[Silbande and Leroi, 2017]). They were mainly Pseudomonas spp. (azotoformans,
D

plecoglossicida/monteilii, fluorescens, gessardii, poae/simiae/trivialis,

E

psychrophila/fragi) (23% of the isolates), Enterobacteriaceae (Hafnia paralvei,

PT

Rahnella aquatilis, Serratia liquefaciens-like and S. myotis) (20%), B. thermosphacta

CE

(20%) and LAB (Carnobacterium divergens, C. maltaromaticum, Leuconostoc

gelidum) (12%). The others were found in smaller proportions: Shewanella spp.
AC

(baltica, morhuae/glacialipiscicola) (6%); Acinetobacter soli (3%); Aeromonas

salmonicida, Arthrobacter protophormiae, Paenibacillus glucanolyticus, Paracoccus

yeeii, Psychrobacter fozii, Sphingobacterium multivorum and Stenotrophomonas

rhizophila (1.5% each).

3.2.5. Illumina sequencing analysis

15
 ACCEPTED MANUSCRIPT

MAP products at day 15 presented very few reads (240 reads) compared with all

other samples (between 30,959 and 66,930 reads) and were excluded from the

sequencing analysis. Following this quality check, sequencing of total DNA extracted

from 8 samples (day 0, VP: days 8, 15, 22 and 29, MAP: days 8, 22 and 29) yielded

a total of 214,735 bacterial 16S rRNA sequence-read counts and 2,972 OTUs. Table

3 summarizes the number of reads and OTUs and the top 15 (total abundance)

PT
bacterial genera for the different red drum samples. There were 887 OTUs at the

RI
beginning of the experiment and this varied between 771 and 1,084 and between

SC
855 and 1,101 during the VP and MAP storage, respectively. At day 0, 128 different

genera were identified and the most prevalent were Arthrobacter sp. (including A.
NU
psychrochitiniphilus) (11.9% of abundance), Chryseobacterium sp. (10.3%),

Brevibacterium sp. (including B. linens) (8.5%), Staphylococcus sp. (6.5%) and

MA

Kocuria sp. (including K. rhizophila and K. gwangalliensis) (5.8%). The microbiota

composition of the fillets changed with the packaging and, although the number of
D

OTUs remained important, few genera dominated the ecosystem (43 for VP samples
E

and 32 for MAP samples). At day 8, Brochothrix sp. were the main bacteria with an
PT

abundance of 50.0% and 51.8% of the VP and MAP products, respectively. Other
CE

genera such as Carnobacterium sp. (VP: 19.6%, MAP: 24.0%), Pseudomonas sp.

(VP: 8.7%, MAP: 3.5%) and Shewanella sp. (VP: 6.3%, MAP: 2.1%) were also
AC

detected. The proportion of Brochothrix, essentially composed of B. thermosphacta,

decreased over time but remained more abundant in MAP samples (18.1%) than in

VP products (9.6%) at day 29. LAB became the major microorganisms at the end of

storage with an abundance of 76.3% in VP samples and 72.1% in MAP samples. In

the 2 batches, this bacterial group was mainly composed of the genera

Carnobacterium (including C. maltaromaticum, C. inhibens and C. gallinarum) (VP:

16
 ACCEPTED MANUSCRIPT

35.5%, MAP: 54.5%), Vagococcus (including V. teuberi and V. fluvialis) (VP: 16.6%,

MAP: 7.6%), Lactococcus (VP: 9.6%, MAP: 8.3%) and Leuconostoc (including L.

gelidum) (VP: 13.3%, MAP: 0.7%) and to a lesser extent the genus Enterococcus

(including E. sulfureus) (VP: 1.0%, MAP: 0.9%). In both conditions, Pseudomonas

sp. (including P. lundensis) and Shewanella sp. (including S. baltica and S. morhuae)

disappeared almost completely during storage. Conversely, some

PT
Enterobacteriaceae, particularly Serratia sp. and Hafnia sp., developed in VP and

RI
MAP samples, reaching a proportion of 3.8% and 1.3% at day 29, respectively. At the

SC
genus level, unclassified bacteria represented less than 10% of sample reads, except

for fresh red drum fillets in which they were equal to 13.9%.
NU
4. Discussion
MA

In the present study, the quality of tropical farmed red drum (S. ocellatus) stored

under different conditions was compared. The preliminary study confirmed the shelf-
D

life of 15 days for the whole gutted fish stored under ice previously established by
E

Fauré (2009) and Régina et al. (2014). These authors also showed variability in this
PT

sensory quality between local fish farms, particularly influenced by the composition of
CE

the food distributed to the livestock. A shelf-life of approximately 2 weeks is often

observed for various lean to medium-fat white fish species. For example, in similar
AC

ice-storage whole fish conditions, Dicentrarchus labrax (European sea bass),

Otolithes ruber (tiger tooth croaker), Epinephelus merra (wire-netting reef cod) and

Sparus aurata (sea bream) were rejected after 15 to 18 days (Alasalvar et al., 2001;

Jeyasekaran et al., 2005; Paleologos et al., 2004; Sharifian et al., 2011). At the

rejection time of whole red drums, mesophilic and psychrotrophic counts reached 8

Log CFU g-1, which is much higher than the 6 Log CFU g-1 obtained for spoiled

17
 ACCEPTED MANUSCRIPT

sciaenids in the study of Jeyasekaran and Sugumar (1997). The decrease in TVBN

and the absence of TMA contents indicated that these indices are not reliable to

detect the sensory spoilage of whole red drums. A similar result was found for

European sea bass in which the change in volatile bases occurred after the rejection

point (Castro et al., 2006). This phenomenon may be due to the washing effect of ice

(Erkan, 2007; Ola and Oladipo, 2004). The initial pH of red drum (6.0/6.4) was lower

PT
than that of other Sciaenids, which have a pH close to neutrality (Genç et al., 2013; Li

RI
et al., 2012, 2013a, 2013b). However, the post-mortem pH often depends on various

SC
factors such as species, proximate composition of the flesh and constitution of the

microbiota (Huss, 1999).

NU
The preliminary study revealed a longer shelf-life of packed red drum fillets, which

was studied in more detail in a second trial. The shelf-life of VP and MAP fillets was
MA

almost double that of whole fish. In the study of Li et al. (2013b), raw red drum fillets

packed in air-proof polyethylene bags and stored at 4 ± 1°C were spoiled from day 8.
D

In the majority of studies, the fillets stored in ice or at chilled temperature have a
E

higher count of bacteria than whole fish and a similar or inferior shelf life (Chytiri et
PT

al., 2004a; Hernández et al., 2009; Paleologos et al., 2004; Poli et al., 2006;
CE

Taliadourou et al., 2003). However, when fillets are vacuum or modified atmosphere

packed, a shelf-life extension of several days is often observed (Arashisar et al.,

AC

2004; Ayala et al., 2011; Dalgaard et al., 1993; Genç et al., 2013; Li et al., 2011;

Mendes and Gonçalves, 2008; Sáez et al., 2014).

Just after packaging, the total bacterial count of red drum was 4–5 Log CFU g-1,

composed by diverse bacteria, mainly belonging to the phyla of actinobacteria

(45.4% of abundance), proteobacteria (24.7%), bacteroidetes (12.3%) and firmicutes

(11.1%). Gram-positive bacteria represented more than 40% of the identified OTUs

18
 ACCEPTED MANUSCRIPT

of the fresh red drum fillets. However, the genera frequently cited for fish, such as

Bacillus, Clostridium, Lactobacillus and Corynebacterium, were absent or at very low

abundance (Gram and Huss, 1996; Huss, 1999). Similarly, except for Arthrobacter

sp., the gram-negative genera usually found in the fresh muscle of marine fish were

detected at low levels (Acinetobacter, 0.8%; Photobacterium, 0.1%; Pseudomonas,

0.7%; Psychrobacter, 1.8%; Shewanella, 0.7%; Aeromonas, <0,1%; Moraxella,

PT
<0,1%) or absent or not identified (Flavobacterium; Vibrio). Enterobacteriaceae

RI
(Serratia, Hafnia, Esherichia-Shigella) represented only 2.0% of the microflora. Many

SC
strains of Pseudomonas (6 isolates, 239 reads) and Brochothrix (5 isolates, 598

reads) were isolated while their abundance were lower than Arthrobacter (4206
NU
reads) which only one isolate was recovered, certainly due to the culture media and

operator selectivity. Other strains (Enterobacteriaceae, Shewanella, Psychrobacter,

MA

etc.) were also obtained and will enable the spoilage potential of each strain to be

analyzed.
D

Less common genera, such as Chryseobacterium for gram-negative bacteria and

E

Brevibacterium, Kocuria and Staphylococcus for gram-positive bacteria, mainly

PT

composed the initial microbiota. However, these bacteria have not been isolated
CE

probably due to their difficulty to grow on the culture media used. In the study of

Chaillou et al. (2014), which detailed the bacterial diversity of various meat and
AC

seafood products by pyrosequencing, Chryseobacterium was also part of the top 3

genera found in seafood samples. Brevibacterium spp. were isolated long ago from

freshwater and sea fish and the Indian Ocean (Crombach, 1972; Johnson et al.,

1968; Kazanas, 1966), while Kocuria spp. were found more recently in marine

sediment (Kim et al., 2004). A high prevalence of Staphylococcus spp. (S.

epidermidis and S. warneri) was also observed in Chaillou et al. (2014).

19
 ACCEPTED MANUSCRIPT

As frequently found in chilled packed seafood (Dalgaard et al., 2003; Franzetti et al.,

2003; Leblanc et al., 1997; Leroi et al., 1998; Leroi, 2010; Lyhs, 2002; Paarup et al.,

2002; Paludan-Müller et al., 1998), vacuum and modified atmosphere (without O2)

conditions favored the growth of LAB. The results of Illumina MiSeq and bacterial

counts were correlated and highlighted the dominance of the LAB (8 log CFU g-1) at

the end of storage, particularly of Carnobacterium spp. (VP: 9723 reads, MAP: 12920

PT
reads). The spoilage activity of these anaerobes shows interspecies and intraspecies

RI
variations and often appears less aggressive in pure culture than in combination with

SC
Enterobacteriaceae or B. thermosphacta (Gram et al., 2002; Joffraud et al., 2006;

Leisner et al., 2007; Macé et al., 2013; Mejlholm et al., 2005; Paludan-Müller et al.,
NU
1998; Sivertsvik et al., 2002). The unpleasant odors identified on opening the

packaging were probably due to a mixture of these species. These off-odors did not
MA

persist in the products and were not strongly detected by the sensory panel, probably

due to a loss of the volatile compounds before the new packaging and storage at -
D

80°C until the sensory test. The deteriorations in appearance and texture were
E

important limiting factors for the shelf-life of red drum fillets.

PT

Another well-known spoiler of CO2-packed fish during chill storage is

CE

P. phosphoreum, responsible for a high level of TMA that contributes to the

ammonia-like odors (Dalgaard, 1995a, 1995b; Dalgaard et al., 1997; Emborg et al.,
AC

2002; Gram and Dalgaard, 2002; Hansen et al., 2009). The low abundance of this

species and others capable of reducing TMAO to TMA, such as some

Enterobacteriaceae, Vibrio and Aeromonas sp., may explain the very weak formation

of volatile bases in the VP and MAP fillets (Debevere and Boskou, 1996; Gram and

Dalgaard, 2002).

20
 ACCEPTED MANUSCRIPT

The significant production of cadaverine, putrescine and, to a lesser extent, tyramine

could be a good indicator of the spoilage of packed red drum fillets during storage.

Similar biogenic amine formation has been reported in white fish flesh such as sea

bass fillets (Paleologos et al., 2004), haddock fillets (Fernandes-Salguero and

Mackie, 1987), rainbow trout fillets (Chytiri et al., 2004b), whole hake (Ruiz-Capillas

and Moral, 2001) and whole sea bream (Koutsoumanis et al., 1999) during ice-

PT
storage. Furthermore, the tropical fish fillets (tambacu, hybrid Colossoma

RI
macropomum x Piaractus mesopotamicus) stored in the same conditions as our

SC
study (VP, 4°C) presented larger amounts of cadaverine and putrescine (around

2000-2500 mg kg-1) after 6 days (Bottino et al., 2017). Enterobacteriaceae,

NU
particularly S. liquefaciens, are often responsible for cadaverine production while

tyramine may be produced by C. maltaromaticum. Moreover, the level of putrescine

MA

was probably the result of a metabiosis between ornithine-forming LAB (precursor of

putrescine) and putrescine-forming Enterobacteriaceae (Bover-Cid and Holzapfel,

D

1999; Dainty et al., 1986; Gram et al., 2002; Jørgensen et al., 2000b; Laursen et al.,
E

2006; Lavizzari et al., 2010; Leisner et al., 1995). Other bacterial groups, identified in
PT

this study, possess the capacity to produce putrescine and cadaverine:

CE

Pseudomonas sp., particularly P. fluorescens and P. putida, and Shewanella sp.,

especially S. putrefaciens and S. baltica (Ge et al., 2017; López-Caballero et al.,

AC

2001; Özogul and Özogul, 2005; Rodriguez-Jerez et al., 1994). Conversely, B.

thermosphacta not show a production of these two biogenic amines and even tends

to attenuate the potential ability of others bacteria (Casaburi et al., 2014; Fall et al.,

2012; Mejlholm et al., 2005; Nowak and Czyzowska, 2011). The absence or

negligible production of histamine in MAP and VP samples, respectively, is due to

either a low free-histidine content in the flesh (no data available) or the absence of

21
 ACCEPTED MANUSCRIPT

histamine-producing bacteria such as Morganella morganii, Raoultella spp. and P.

phosphoreum (Drancourt et al., 2001; Kanki et al., 2002, 2004; Özoğul, 2004).

In conclusion, chilled packed fillets of red drum present a longer shelf-life than whole

gutted iced fish. VP products retain a better appearance than MAP samples at the

end of storage. The results of Illumina sequencing provide detailed data on the

bacterial ecosystem evolution during the storage of packed red drum fillets and

PT
identify the bacterial species that do not grow on culture media. More research is

RI
needed to characterize the spoilage potential of the bacteria isolated from red drum

SC
and to develop rapid quality control methods for the local fish farming sector.
NU
Acknowledgements

This research was financially supported by the territorial community of Martinique

MA

and the European Fisheries Fund (joint convention number:

013/DM/0335/3.1.1.a/38836). The authors thank the staff of MatÍs for their

D

contribution in processing Illumina sequencing data. The authors are grateful to F.

E

Chevalier for valuable technical assistance, and Carol Robins for the English
PT

language editing of the manuscript.

CE
AC

22
 ACCEPTED MANUSCRIPT

References

AFNOR, 1995. NF V 09-105. Directives générales pour l'implantation de locaux

destinés à l'analyse sensorielle. Contrôle de la qualité des produits alimentaires -

Analyse sensorielle

Alasalvar, C., Taylor, K., Öksüz, A., Garthwaite, T., Alexis, M., Grigorakis, K., 2001.

PT
Freshness assessment of cultured sea bream (Sparus aurata) by chemical,

RI
physical and sensory methods. Food Chemistry 72, 33-40.

SC
Andrade, S.D.C.S., Mársico, E.T., de Oliveira Godoy, R.L., Franco, R.M., Junior,

C.A.C., 2014. Chemical quality indices for freshness evaluation of fish. Journal of
NU
Food Studies, 3 (1), 71-87.
MA

Arashisar, Ş, Hisar, O., Kaya, M., Yanik, T., 2004. Effects of modified atmosphere

and vacuum packaging on microbiological and chemical properties of rainbow

D

trout (Oncorynchus mykiss) fillets. International Journal of Food Microbiology 97,

E

209-214.
PT

Aronesty, E., 2011. Command-Line Tools for Processing Biological Sequencing

CE

Data, ea-utils. Expression Analysis.Durham, NC: Available online at:

AC

http://code.google.com/p/ea-utils

Ayala, M.D., Santaella, M., Martínez, C., Periago, M.J., Blanco, A., Vázquez, J.M.,

Albors, O.L., 2011. Muscle tissue structure and flesh texture in gilthead sea

bream, Sparus aurata L., fillets preserved by refrigeration and by vacuum

packaging. LWT-Food Science and Technology 44, 1098-1106.

23
 ACCEPTED MANUSCRIPT

Bokulich, N.A., Subramanian, S., Faith, J.J., Gevers, D., Gordon, J.I., Knight, R.,

Mills, D.A., Caporaso, J.G., 2013. Quality-filtering vastly improves diversity

estimates from Illumina amplicon sequencing. Nature Methods 10, 57-59.

Bottino, F. D. O., Rodrigues, B. L., de Nunes Ribeiro, J. D., Lázaro, C. A. D. L. T.,

Conte‐Junior, C. A., 2017. Influence of UV‐C Radiation on Shelf Life of Vacuum

PT
Package Tambacu (Colossoma macropomum× Piaractus mesopotamicus)

Fillets. Journal of Food Processing and Preservation, 41(4).

RI
SC
Bover-Cid, S., Holzapfel, W.H., 1999. Improved screening procedure for biogenic

amine production by lactic acid bacteria. International Journal of Food

NU
Microbiology 53, 33-41.
MA

Bulushi, I. A., Poole, S., Deeth, H. C., Dykes, G. A., 2009. Biogenic amines in fish:

roles in intoxication, spoilage, and nitrosamine formation—a review. Critical

D

reviews in food science and nutrition, 49(4), 369-377.

E

Caporaso, J.G., Kuczynski, J., Stombaugh, J., Bittinger, K., Bushman, F.D., Costello,
PT

E.K., Fierer, N., Pena, A.G., Goodrich, J.K., Gordon, J.I., 2010a. QIIME allows
CE

analysis of high-throughput community sequencing data. Nature Methods 7, 335-

336.
AC

Caporaso, J.G., Bittinger, K., Bushman, F.D., DeSantis, T.Z., Andersen, G.L., Knight,

R., 2010b. PyNAST: a flexible tool for aligning sequences to a template

alignment. Bioinformatics (Oxford, England) 26, 266-267.

24
 ACCEPTED MANUSCRIPT

Casaburi, A., De Filippis, F., Villani, F. and Ercolini, D., 2014. Activities of strains of

Brochothrix thermosphacta in vitro and in meat. Food research international, 62,

366-374.

Castro, P., Padrón, J.C.P., Cansino, M.J.C., Velázquez, E.S., De Larriva, R.M., 2006.

Total volatile base nitrogen and its use to assess freshness in European sea

PT
bass stored in ice. Food Control 17, 245-248.

RI
Chaillou, S., Chaulot-Talmon, A., Caekebeke, H., Cardinal, M., Christieans, S.,

SC
Denis, C., Desmonts, M.H., Dousset, X., Feurer, C., Hamon, E., Joffraud, J.J., La

Carbona, S., Leroi, F., Leroy, S., Lorre, S., Macé, S., Pilet, M.F., Prévost, H.,
NU
Rivollier, M., Roux, D., Talon, R., Zagorec, M., Champomier-Vergès, M.C., 2014.

Origin and ecological selection of core and food-specific bacterial communities

MA

associated with meat and seafood spoilage. International Society for Microbial

Ecology 1 (14), 1751-7362.

E D

Chytiri, S., Chouliara, I., Savvaidis, I., Kontominas, M., 2004a. Microbiological,
PT

chemical and sensory assessment of iced whole and filleted aquacultured

rainbow trout. Food Microbiology 21, 157-165.

CE

Chytiri, S., Paleologos, E., Savvaidis, I., Kontominas, M.G., 2004b. Relation of
AC

biogenic amines with microbial and sensory changes of whole and filleted

freshwater rainbow trout (Onchorynchus mykiss) stored on ice. Journal of Food

Protection 67, 960-965.

Conway, E.J., Byrne, A., 1933. An absorption apparatus for the micro-determination

of certain volatile substances: The micro-determination of ammonia. Biochemical

Journal 27 (2), 419-429.

25
 ACCEPTED MANUSCRIPT

Crombach, W., 1972. DNA base composition of soil arthrobacters and other

coryneforms from cheese and sea fish. Antonie van Leeuwenhoek 38, 105-120.

Dainty, R.H., 1996. Chemical/biochemical detection of spoilage. International Journal

of Food Microbiology 33, 19-33.

Dainty, R., Edwards, R., Hibbard, C., Ramantanis, S., 1986. Bacterial sources of

PT
putrescine and cadaverine in chill stored vacuum-packaged beef. Journal of

RI
Applied Bacteriology 61, 117-123.

SC
Dalgaard, P., Gram, L., Huss, H.H., 1993. Spoilage and shelf-life of cod fillets packed

in vacuum or modified atmospheres. International Journal of Food Microbiology

NU
19, 283-294.
MA

Dalgaard, P., 1995a. Modelling of microbial activity and prediction of shelf life for

packed fresh fish. International Journal of Food Microbiology 26, 305-317.

E D

Dalgaard, P., 1995b. Qualitative and quantitative characterization of spoilage

PT

bacteria from packed fish. International Journal of Food Microbiology 26, 319-

333.
CE

Dalgaard, P., Mejlholm, O., Christiansen, T., Huss, H.H., 1997. Importance of
AC

Photobacterium phosphoreum in relation to spoilage of modified atmosphere-

packed fish products. Letters in Applied Microbiology 24, 373-378.

Dalgaard, P., Vancanneyt, M., Euras Vilalta, N., Swings, J., Fruekilde, P., Leisner, J.,

2003. Identification of lactic acid bacteria from spoilage associations of cooked

and brined shrimps stored under modified atmosphere between 0 C and 25 C.

Journal of Applied Microbiology 94, 80-89.

26
 ACCEPTED MANUSCRIPT

Debevere, J., Boskou, G., 1996. Effect of modified atmosphere packaging on the

TVB/TMA-producing microflora of cod fillets. International Journal of Food

Microbiology 31, 221-229.

Drancourt, M., Bollet, C., Carta, A., Rousselier, P., 2001. Phylogenetic analyses of

Klebsiella species delineate Klebsiella and Raoultella gen. nov., with description

PT
of Raoultella ornithinolytica comb. nov., Raoultella terrigena comb. nov. and

Raoultella planticola comb. nov. International Journal of Systematic and

RI
Evolutionary Microbiology 51, 925-932.

SC
Duflos, G., Dervin, C., Malle, P., Bouquelet, S., 1999. Relevance of matrix effect in
NU
determination of biogenic amines in plaice (Pleuronectes platessa) and whiting

(Merlangus merlangus). Journal-Aoac International 82, 1097-1101.

MA

Edgar, R.C., 2010. Search and clustering orders of magnitude faster than BLAST.
D

Bioinformatics (Oxford, England) 26, 2460-2461.

E

Emborg, J., Laursen, B.G., Rathjen, T., Dalgaard, P., 2002. Microbial spoilage and
PT

formation of biogenic amines in fresh and thawed modified atmosphere‐ packed

CE

salmon (Salmo salar) at 2° C. Journal of Applied Microbiology 92, 790-799.

AC

Erkan, N., 2007. Sensory, chemical, and microbiological attributes of sea bream

(Sparus aurata): effect of washing and ice storage. International Journal of Food

Properties 10, 421-434.

FAO, 2014. The state of world fisheries and aquaculture 2014. Rome. 223 pp. URL:

http://www.fao.org/3/a-i3720e/index.html.

27
 ACCEPTED MANUSCRIPT

Fall, P. A., Pilet, M. F., Leduc, F., Cardinal, M., Duflos, G., Guérin, C., Joffraud, J.J.,

Leroi, F., 2012. Sensory and physicochemical evolution of tropical cooked peeled

shrimp inoculated by Brochothrix thermosphacta and Lactococcus piscium

CNCM I-4031 during storage at 8°C. International Journal of Food Microbiology,

152(3), 82-90.

PT
Fauré, L., 2009. Évaluation de la qualité de l’ombrine ocellée (Sciaenops ocellatus)

en fonction de différents impacts. 2011, 1-75.

RI
SC
Fernandes-Salguero, J., Mackie, I., 1987. Comparative rates of spoilage of fillets and

whole fish during storage of haddock (Melanogrammus aeglefinus) and herring

NU
(Clupea harengus) as determined by the formation of non‐volatile and volatile

amines. International Journal of Food Science & Technology 22, 385-390.

MA

Frangos, L., Pyrgotou, N., Giatrakou, V., Ntzimani, A., Savvaidis, I., 2010. Combined
D

effects of salting, oregano oil and vacuum-packaging on the shelf-life of

E

refrigerated trout fillets. Food Microbiology 27, 115-121.

PT

Franzetti, L., Scarpellini, M., Mora, D., Galli, A., 2003. Carnobacterium spp. in
CE

seafood packaged in modified atmosphere. Annals of Microbiology 53, 189-198.

AC

Ge, Y., Zhu, J., Ye, X. and Yang, Y., 2017. Spoilage potential characterization of

Shewanella and Pseudomonas isolated from spoiled large yellow croaker

(Pseudosciaena crocea). Letters in applied microbiology, 64(1), 86-93.

Genç, İY., Esteves, E., Aníbal, J., Diler, A., 2013. Effects of chilled storage on quality

of vacuum packed meagre fillets. Journal of Food Engineering 115, 486-494.

28
 ACCEPTED MANUSCRIPT

Gram, L., Huss, H.H., 1996. Microbiological spoilage of fish and fish products.

International Journal of Food Microbiology 33, 121-137.

Gram, L., Dalgaard, P., 2002. Fish spoilage bacteria–problems and solutions.

Current Opinion in Biotechnology 13, 262-266.

Gram, L., Ravn, L., Rasch, M., Bruhn, J.B., Christensen, A.B., Givskov, M., 2002.

PT
Food spoilage—interactions between food spoilage bacteria. International

RI
Journal of Food Microbiology 78, 79-97.

SC
Gram, L., 2009. Microbiological spoilage of fish and seafood products. In:

Sperber,W.H., Doyle,M.P. (Eds.), Compendium of the Microbiological Spoilage of

NU
Foods and Beverages. Springer, New York, pp. 87-120.
MA

Hansen, A.Å, Mørkøre, T., Rudi, K., Rødbotten, M., Bjerke, F., Eie, T., 2009. Quality

changes of prerigor filleted Atlantic salmon (Salmo salar L.) packaged in modified
D

atmosphere using CO2 emitter, traditional MAP, and vacuum. Journal of Food
E

Science 74, M242-M249.

PT

Hernández, M.D., López, M.B., Álvarez, A., Ferrandini, E., García García, B.,
CE

Garrido, M.D., 2009. Sensory, physical, chemical and microbiological changes in

AC

aquacultured meagre (Argyrosomus regius) fillets during ice storage. Food

Chemistry 114, 237-245.

Huss, H.H., 1999. La qualité et son évolution dans le poisson frais. In: FAO (Ed.),

Document technique sur les pêches N°348. FAO, Rome, pp. 1-198.

Jeyasekaran, G., Sugumar, G., 1997. Effect of delayed icing on the shelf-life of

sciaenids. Journal of Food Science and Technology 34, 498-500.

29
 ACCEPTED MANUSCRIPT

Jeyasekaran, G., Maheswari, K., Ganesan, P., Jeya Shakila, R., Sukumar, D., 2005.

Quality changes in ice‐stored tropical wire‐netting reef cod (Epinephelus merra).

Journal of Food Processing and Preservation 29, 165-182.

Joffraud, J.J., Cardinal, M., Cornet, J., Chasles, J.S., Léon, S., Gigout, F., Leroi, F.,

2006. Effect of bacterial interactions on the spoilage of cold-smoked salmon.

PT
International Journal of Food Microbiology 112, 51-61.

RI
Johnson, R.M., Schwent, R.M., Press, W., 1968. The characteristics and distribution

SC
of marine bacteria isolated from the Indian Ocean. Limnology and Oceanography

13, 656-664.
NU
Jørgensen, L.V., Dalgaard, P., Huss, H.H., 2000a. Multiple Compound Quality Index
MA

for cold-smoked salmon (Salmo salar) developed by multivariate regression of

biogenic amines and pH. Journal of Agriculture and Food Chemistry 48, 2448–
D

2453.
E

Jørgensen, L.V., Huss, H.H., Dalgaard, P., 2000b. The effect of biogenic amine
PT

production by single bacterial cultures and metabiosis on cold‐smoked salmon.

CE

Journal of Applied Microbiology 89, 920-934.

AC

Kakouri, A., Drosinos, E., Nychas, G., 1997. Storage of Mediterranean fresh fish

(Boops boops, and Sparus aurata) under modified atmospheres or vacuum at 3

and 10°C. Developments in Food Science 38, 171-178.

Kanki, M., Yoda, T., Ishibashi, M., Tsukamoto, T., 2004. Photobacterium

phosphoreum caused a histamine fish poisoning incident. International Journal of

Food Microbiology 92, 79-87.

30
 ACCEPTED MANUSCRIPT

Kanki, M., Yoda, T., Tsukamoto, T., Shibata, T., 2002. Klebsiella pneumoniae

produces no histamine: Raoultella planticola and Raoultella ornithinolytica strains

are histamine producers. Applied and Environmental Microbiology 68, 3462-

3466.

Kazanas, N., 1966. Effect of gamma Irradiation on the Microflora of Freshwater Fish:

PT
II. Generic Identification of Aerobic Bacteria from Yellow Perch Fillets. Applied

Microbiology 14, 957-965.

RI
SC
Kim, S.B., Nedashkovskaya, O.I., Mikhailov, V.V., Han, S.K., Kim, K., Rhee, M., Bae,

K.S., 2004. Kocuria marina sp. nov., a novel actinobacterium isolated from
NU
marine sediment. International Journal of Systematic and Evolutionary

Microbiology 54, 1617-1620.

MA

Kim, M. K., Mah, J. H., Hwang, H. J., 2009. Biogenic amine formation and bacterial
D

contribution in fish, squid and shellfish. Food Chemistry, 116(1), 87-95.

E

Kostaki, M., Giatrakou, V., Savvaidis, I.N., Kontominas, M.G., 2009. Combined effect
PT

of MAP and thyme essential oil on the microbiological, chemical and sensory
CE

attributes of organically aquacultured sea bass (Dicentrarchus labrax) fillets.

Food Microbiology 26, 475-482.

AC

Koutsoumanis, K., Lampropoulou, K., & Nychas, G. J. E. (1999). Biogenic amines

and sensory changes associated with the microbial flora of Mediterranean gilt-

head sea bream (Sparus aurata) stored aerobically at 0, 8, and 15 C. Journal of

food protection, 62(4), 398-402.

31
 ACCEPTED MANUSCRIPT

Kykkidou, S., Giatrakou, V., Papavergou, A., Kontominas, M., Savvaidis, I., 2009.

Effect of thyme essential oil and packaging treatments on fresh Mediterranean

swordfish fillets during storage at 4 C. Food Chemistry 115, 169-175.

Laursen, B.G., Leisner, J.J., Dalgaard, P., 2006. Carnobacterium species: effect of

metabolic activity and interaction with Brochothrix thermosphacta on sensory

PT
characteristics of modified atmosphere packed shrimp. Journal of Agricultural

and Food Chemistry 54 (10), 3604-3611.

RI
SC
Lavizzari, T., Breccia, M., Bover-Cid, S., Vidal-Carou, M., Veciana-Nogués, M., 2010.

Histamine, cadaverine, and putrescine produced in vitro by enterobacteriaceae

NU
and pseudomonadaceae isolated from spinach. Journal of Food Protection® 73,

385-389.
MA

Leblanc, E., Stiles, M., McMullen, L., Leisner, J., 1997. Characterizationof the
D

prevailing flora of Carnobacterium spp. growing on CO2 modified atmosphere

E

packaged rainbow trout (Salmo gairdner) fillets. Development in Food Science

PT

Leisner, J., Greer, G., Dilts, B., Stiles, M., 1995. Effect of growth of selected lactic
CE

acid bacteria on storage life of beef stored under vacuum and in air. International

Journal of Food Microbiology 26, 231-243.

AC

Leisner, J.J., Laursen, B.G., Prevost, H., Drider, D., Dalgaard, P., 2007.

Carnobacterium: positive and negative effects in the environment and in foods.

FEMS Microbiology Reviews 31, 592-613.

Leon, X., Knockaert, C., Regina, F., 2008. Qualité et valorisation de l'ombrine

ocellée: l'exemple de la filière pisciculture marine martiniquaise.

32
 ACCEPTED MANUSCRIPT

Leroi, F., Joffraud, J., Chevalier, F., Cardinal, M., 1998. Study of the microbial

ecology of cold-smoked salmon during storage at 8 C. International Journal of

Food Microbiology 39, 111-121.

Leroi, F., 2010. Occurrence and role of lactic acid bacteria in seafood products. Food

Microbiology 27 (6), 698-709.

PT
Leroi, F., Cornet, J., Chevalier, F., Cardinal, M., Coeuret, G., Chaillou, S., Joffraud,

RI
J., 2015. Selection of bioprotective cultures for preventing cold-smoked salmon

SC
spoilage. International Journal of Food Microbiology 213, 79-87.

Li, X., Li, J., Zhu, J., Wang, Y., Fu, L., Xuan, W., 2011. Postmortem changes in
NU
yellow grouper (Epinephelus awoara) fillets stored under vacuum packaging at 0
MA

C. Food Chemistry 126, 896-901.

Li, T., Hu, W., Li, J., Zhang, X., Zhu, J., Li, X., 2012. Coating effects of tea polyphenol
D

and rosemary extract combined with chitosan on the storage quality of large
E

yellow croaker (Pseudosciaena crocea). Food Control 25, 101-106.

PT

Li, T., Li, J., Hu, W., 2013a. Changes in microbiological, physicochemical and muscle
CE

proteins of post mortem large yellow croaker (Pseudosciaena crocea). Food

AC

Control 34, 514-520.

Li, T., Li, J., Hu, W., Li, X., 2013b. Quality enhancement in refrigerated red drum

(Sciaenops ocellatus) fillets using chitosan coatings containing natural

preservatives. Food Chemistry 138, 821-826.

33
 ACCEPTED MANUSCRIPT

López-Caballero, M. E., Sánchez-Fernández, J. A. and Moral, A., 2001. Growth and

metabolic activity of Shewanella putrefaciens maintained under different CO2 and

O2 concentrations. International journal of food microbiology, 64(3), 277-287.

Lozupone, C., Knight, R., 2005. UniFrac: a new phylogenetic method for comparing

microbial communities. Applied and Environmental Microbiology 71, 8228-8235.

PT
Lyhs, U., 2002. Lactic acid bacteria associated with the spoilage of fish products. In:

RI
Anonymous (Eds), University of Helsinki, Faculty of Veterinary Medicine,

SC
Department of Food and Environmental Hygiene,

Macé, S., Cornet, J., Chevalier, F., Cardinal, M., Pilet, M.F., Dousset, X., Joffraud,
NU
J.J., 2012. Characterisation of the spoilage microbiota in raw salmon (Salmo
MA

salar) steaks stored under vacuum or modified atmosphere packaging combining

conventional methods and PCR–TTGE. Food Microbiology 30 (1), 164-172.

D

Macé, S., Joffraud, J.J., Cardinal, M., Malcheva, M., Cornet, J., Lalanne, V.,
E

Chevalier, F., Sérot, T., Pilet, M.F., Dousset, X., 2013. Evaluation of the spoilage
PT

potential of bacteria isolated from spoiled raw salmon (Salmo salar) fillets stored
CE

under modified atmosphere packaging. International Journal of Food

Microbiology 160 (3), 227-238.

AC

Mejlholm, O., Bøknæs, N., Dalgaard, P., 2005. Shelf life and safety aspects of chilled

cooked and peeled shrimps (Pandalus borealis) in modified atmosphere

packaging. Journal of Applied Microbiology 99, 66-76.

34
 ACCEPTED MANUSCRIPT

Mendes, R., Gonçalves, A., 2008. Effect of soluble CO2 stabilisation and vacuum

packaging in the shelf life of farmed sea bream and sea bass fillets. International

Journal of Food Science & Technology 43, 1678-1687.

Nowak, A. and Czyzowska, A., 2011. In vitro synthesis of biogenic amines by

Brochothrix thermosphacta isolates from meat and meat products and the

PT
influence of other microorganisms. Meat science, 88(3), 571-574.

RI
Ola, J.B., Oladipo, A.E., 2004. Storage life of croaker (Pseudotholitus senegalensis)

SC
in ice and ambient temperature. African Journal of Biomedical Research 7, 13-

17.
NU
Özogul, F., Taylor, K. D. A., Quantick, P., Özogul, Y., 2002. Biogenic amines
MA

formation in Atlantic herring (Clupea harengus) stored under modified

atmosphere packaging using a rapid HPLC method. International journal of food

D

science & technology, 37(5), 515-522.

E

Özoğul, F., 2004. Production of biogenic amines by Morganella morganii, Klebsiella

PT

pneumoniae and Hafnia alvei using a rapid HPLC method. European Food
CE

Research and Technology 219, 465-469.

AC

Özogul, F., Özogul, Y., 2005. Formation of biogenic amines by Gram-negative rods

isolated from fresh, spoiled, VP-packed and MAP-packed herring (Clupea

harengus). European Food Research and Technology, 221(5), 575-581.

Paarup, T., Sanchez, J.A., Peláez, C., Moral, A., 2002. Sensory, chemical and

bacteriological changes in vacuum-packed pressurised squid mantle (Todaropsis

eblanae) stored at 4 C. International Journal of Food Microbiology 74, 1-12.

35
 ACCEPTED MANUSCRIPT

Paleologos, E., Savvaidis, I., Kontominas, M., 2004. Biogenic amines formation and

its relation to microbiological and sensory attributes in ice-stored whole, gutted

and filleted Mediterranean Sea bass (Dicentrarchus labrax). Food Microbiology

21, 549-557.

Paludan-Müller, C., Dalgaard, P., Huss, H.H., Gram, L., 1998. Evaluation of the role

PT
of Carnobacterium piscicola in spoilage of vacuum-and modified-atmosphere-

packed cold-smoked salmon stored at 5 C. International Journal of Food

RI
Microbiology 39, 155-166.

SC
Poli, B.M., Messini, A., Parisi, G., Scappini, F., Vigiani, V., Giorgi, G., Vincenzini, M.,
NU
2006. Sensory, physical, chemical and microbiological changes in European sea

bass (Dicentrarchus labrax) fillets packed under modified atmosphere/air or

MA

prepared from whole fish stored in ice. International Journal of Food Science &

Technology 41, 444-454.

E D

Price, M.N., Dehal, P.S., Arkin, A.P., 2010. FastTree 2–approximately maximum-
PT

likelihood trees for large alignments. PloS one 5, e9490.

CE

Quast, C., Pruesse, E., Yilmaz, P., Gerken, J., Schweer, T., Yarza, P., Peplies, J.,

Glockner, F.O., 2013. The SILVA ribosomal RNA gene database project:
AC

improved data processing and web-based tools. Nucleic Acids Research 41,

D590-6.

Régina, F., Eugène, S., Rinna, R., 2014. Etude de la qualité des productions

aquacoles en fonction des conditions d'élevage et de conservation.

36
 ACCEPTED MANUSCRIPT

Rodrigues, B.L., da Silveira Alvares, T., Sampaio, G.S.L., Cabral, C.C., Araujo,

J.V.A., Franco, R.M., Mano S.B., Junior C.A.C., 2016. Influence of vacuum and

modified atmosphere packaging in combination with UV-C radiation on the shelf

life of rainbow trout (Oncorhynchus mykiss) fillets. Food Control, 60, 596-605.

Rodriguez-Jerez, J. J., Lopez-Sabater, E. I., Roig-Sagues, A. X. and Mora-Ventura,

PT
M. T., 1994. Histamine, cadaverine and putrescine forming bacteria from ripened

Spanish semipreserved anchovies. Journal of Food Science, 59(5), 998-1001.

RI
SC
Rodríguez-Méndez, M. L., Gay, M., Apetrei, C., De Saja, J. A., 2009. Biogenic

amines and fish freshness assessment using a multisensor system based on

NU
voltammetric electrodes. Comparison between CPE and screen-printed

electrodes. Electrochimica Acta, 54(27), 7033-7041.

MA

Ruiz‐Capillas, C., Moral, A., 2001. Production of biogenic amines and their potential
D

use as quality control indices for hake (Merluccius merluccius, L.) stored in ice.
E

Journal of Food Science 66, 1030-1032.

PT

Sáez, M.I., Martinez, T.F., Cardenas, S., Suarez, M.D., 2014. Effects of vacuum and
CE

modified atmosphere on textural parameters and structural proteins of cultured

meagre (Argyrosomus regius) fillets. Food Science and Technology International

AC

= Ciencia y tecnologia de los alimentos internacional 21, 467-478.

Sáez, M.I., Martínez, T.F., Cárdenas, S., Suárez, M.D., 2015. Effects of different

preservation strategies on microbiological counts, lipid oxidation and color of

cultured meagre (Argyrosomus regius, L.) fillets. Journal of Food Processing and

Preservation 39, 768-775.

37
 ACCEPTED MANUSCRIPT

Sharifian, S., Zakipour, E., Mortazavi, M.S., Arshadi, A., 2011. Quality assessment of

tiger tooth croaker (Otolithes ruber) during ice storage. International Journal of

Food Properties 14, 309-318.

Shewan, J.M., 1971. The microbiology of fish and fishery products—a progress

report. Journal of Applied Bacteriology 34 (2), 299-315.

PT
Silbande, A., Adenet, S., Smith-Ravin, J., Joffraud, J.J., Rochefort, K., Leroi, F.,

RI
2016. Quality assessment of ice-stored tropical yellowfin tuna (Thunnus

SC
albacares) and influence of vacuum and modified atmosphere packaging. Food

Microbiology 60, 62-72.

NU
Silbande, A. and Leroi, F., 2017. Informations on the origin and the identification of
MA

the bacterial strains isolated from the red drum (Sciaenops ocellatus).

International Journal of Food Microbiology Data in Brief “submitted”.

D

Sivertsvik, M., Jeksrud, W.K., Rosnes, J.T., 2002. A review of modified atmosphere
E

packaging of fish and fishery products - significance of microbial growth,

PT

activities and safety. International Journal of Food Science and Technology 37,
CE

107-127.
AC

Taliadourou, D., Papadopoulos, V., Domvridou, E., Savvaidis, I.N., Kontominas,

M.G., 2003. Microbiological, chemical and sensory changes of whole and filleted

Mediterranean aquacultured sea bass (Dicentrarchus labrax) stored in ice.

Journal of the Science of Food and Agriculture 83, 1373-1379.

Vázquez-Baeza, Y., Pirrung, M., Gonzalez, A., Knight, R., 2013. EMPeror: a tool for

visualizing high-throughput microbial community data. Gigascience 2, 1.

38
 ACCEPTED MANUSCRIPT

Veciana-Nogues, M. T., Mariné-Font, A., Vidal-Carou, M. C., 1997. Biogenic amines

as hygienic quality indicators of tuna. Relationships with microbial counts, ATP-

related compounds, volatile amines, and organoleptic changes. Journal of

Agricultural and Food Chemistry, 45(6), 2036-2041.

www.illumina.com, assessed in April 2017. Full URL: http://web.uri.edu/gsc/files/16s-

PT
metagenomic-library-prep-guide-15044223-b.pdf.

RI
www.magbiogenomics.com, assessed in April 2017. Full URL:

SC
http://www.magbiogenomics.com/image/data/Literature/Protocols/HighPrep%20

PCR%20Protocol.pdf.
NU
MA
E D
PT
CE
AC

39
 ACCEPTED MANUSCRIPT

List of figures

Fig. 1: TVBN (light) and TMA (dark) production (mg-N 100 g-1) in whole iced red drum

(ICE_Whole) and in fillets vacuum packed at 4°C (VP_Fillet) or modified-atmosphere

packed at 4°C (MAP_Fillet) after 0, 8 and 15 days. Values with different superscript

letters are significantly different (p > 0.05). Bars represent standard deviations.

PT
Fig. 2: Changes in bacterial enumerations (Log CFU g-1) in whole iced red drum

(ICE_Whole) and in fillets vacuum packed at 4°C (VP_Fillet) or modified-atmosphere

RI
packed at 4°C (MAP_Fillet) after 0, 8 and 15 days. Bars represent standard

SC
deviations. NU
Fig. 3: Standardized Principal component analysis (PCA) performed with the mean

scores of profiling odors: simultaneous representation of samples and odor

MA

descriptors on plane 1–3 (60.1% of inertia). Ellipses indicate groups of samples with

similar odors. Sample nomenclature: D0, fresh fillets; VP, fillets vacuum packed;
D

MAP, fillets modified-atmosphere packed (50% CO2, 50% N2). Numbers in labels of
E
PT

samples express duration of storage (in days).

Fig. 4: Changes in biogenic amines concentrations (mg kg-1) in fillets vacuum packed
CE

at 4°C (VP_Fillet) or modified-atmosphere packed at 4°C (MAP_Fillet) after 0, 8, 15,

AC

22 and 29 days. Bars represent standard deviations.

Fig. 5: Changes in bacterial enumerations (Log CFU g-1) in fillets vacuum packed at

4°C (VP_Fillet) or modified-atmosphere packed at 4°C (MAP_Fillet) after 0, 8, 15, 22

and 29 days. Bars represent standard deviations.

40
 ACCEPTED MANUSCRIPT

Table 1: Score and sensory criteria evaluated during the storage under ice (0°C) of whole red drums.

Score
Criteria
0 1 2 3 4 5 6

Color shiny black pupil shiny black pupil

duller pupil and
opalescent cornea
P T
gray pupil and milky white pupil and white pupil and

Eye

Form convex, bulging convex, bulging

transparent cornea

less bulging flat

R I cornea

concave in the center

milky cornea

concave
milky cornea

very concave

silvery, slightly colored silvery, slightly colored silvery, slightly colored

S C
browning and blood excessive
Operculum Color
with red or brown with red or brown with red or brown

N U
seepage around eye
yellowish yellow
yellow

Peritoneum Integrity

Flesh
intact

pre-rigor
adhesion

firm
no adhesion

M
elastic
A cracked

springy
deteriorated

soft
lysed

flaccid
totally lysed

very flaccid
Belly cavity
Wall intact intact

E D soft fragile perforated perforated perforated

Flesh near the Adhesion

column breaks and

flesh does not come off

P T
column breaks and

flesh does not come off

adhesion less adhesion no adhesion
flesh comes off

easily
flesh comes off

very easily
spinal column
Color normal
C E normal slight color pink red brown very brown

C
* Gray criteria indicate the unpleasant effects found at the rejection time (day 15).

41
 ACCEPTED MANUSCRIPT

Table 2: Spoilage score and sensory characteristics (appearance, texture, odor) of fresh red drum fillets and each storage condition

(VP* and MAP**, 4°C) after 8, 15, 22 and 29 days.

D0 D8 D15 D22 D29

Fresh fillet VP MAP VP MAP VP

T
MAP

P
VP MAP

Spoilage score 0.0 ± 0.0 1.5 ± 0.5 2.3 ± 0.3 3.2 ± 0.3 4.0 ± 0.0 5.2 ± 0.3

R I 5.5 ± 0.5 8.0 ± 0.0 7.5 ± 0.0

SC
weak greening strong
uniform color, weak weak overall strong
overall of the flesh, darkening and strong

uniform
slightly pink,

bright, weak
whitening/yellowing

of the flesh, less

darkening of

the flesh, slime

darkening of

N U slime weak greening

discoloration

with an overall
discoloration

Appearance
color,

slightly pink,
browning of

the flesh near

bright, weak

browning of the
formation,

browning of the M A
the flesh, dull,

browning of the
formation,

strong
of the flesh,

very dull, strong

greenish

appearance, a
with an overall

greenish

bright
the spinal flesh near the spinal
E D
flesh near the
flesh near the browning of the browning of the
lot of slime
appearance,

column column

P T spinal column
spinal column flesh near the

spinal column
flesh near the

spinal column
formation
very dull

Texture firm firm

C
firm
E firm soft firm softer slightly soft disintegrated

Odor on

opening the
marine and

fresh fish
very weak
A C
very weak marine
cut grass and

slightly pungent
moderate

sweet, sour and

moderate cut moderate sweet strong cut grass moderate

marine odor odor grass odor and meat odors odor amine odor
packaging odors odors amine odors

*VP: Vacuum Packed fillets stored at 4°C;

**MAP: Modified Atmosphere (50% CO2–50% N2) Packed fillets stored at 4°C

42
 ACCEPTED MANUSCRIPT

Table 3: Sequencing information (reads, OTUs, genera) and sequence-read counts

of bacterial genera identified by Illumina MiSeq sequencing for fresh fillets and each

storage condition (VP* and MAP**, 4°C) after 8, 15, 22 and 29 days.

Fresh fillet VP MAP

day 0 day 8 day 15 day 22 day 29 day 8 day 22 day 29

SEQUENCING INFORMATION

Total number of reads 35265 20155 17628 22151 27398 31808 36661 23669

PT
Total number of OTUs 887 801 771 960 1084 927 1101 855

RI
Number of different identified genera 128 58 40 40 43 77 41 32

IDENTIFIED OTUs (genera-level)

SC
Carnobacterium 133 3960 5910 10320 9723 7623 14778 12910

Brochothrix 598 10075 6899 3344 2641 16478 11357 4295

NU
Lactococcus 0 180 469 2275 2637 555 3170 1964

Vagococcus 8 375 904 1603 4555 491 1517 1789

MA

Arthrobacter 4206 544 30 7 3 1447 100 16

Shewanella 249 1270 964 93 50 672 1297 280

Leuconostoc 0 6 22 78 3643 51 224 171

D

Pseudomonas 239 1757 324 86 71 1111 296 64

E

Chryseobacterium 3642 18 0 0 0 116 0 0

PT

Brevibacterium 3006 12 3 1 0 109 1 0

Serratia 380 51 435 907 759 26 92 263

CE

Staphylococcus 2283 1 0 1 0 38 0 0

Kocuria 2054 10 6 0 1 93 3 0
AC

Deinococcus 1721 4 0 0 0 52 1 0

Enterococcus 18 56 72 206 277 79 433 210

Psychrobacter 623 68 13 5 0 480 23 2

Planomicrobium 6 185 216 228 171 95 197 111

Rhodovulum 1189 2 0 0 0 3 0 0

Geobacillus 2 139 207 220 207 50 111 95

Dermacoccus 969 5 3 0 1 37 1 0

Lactobacillus 11 1 1 786 25 3 47 17

43
 ACCEPTED MANUSCRIPT

Dietzia 634 1 0 0 0 4 0 0

Rhodococcus 547 7 0 0 0 75 6 0

Iodobacter 0 2 49 71 88 3 266 100

Streptococcus 469 2 6 19 34 2 10 9

Microbacterium 476 3 0 0 0 12 1 0

Propionibacterium 460 0 1 0 0 20 0 0

Acinetobacter 299 17 1 3 34 83 12 0

Halomonas 404 0 1 3 0 11 1 0

PT
Paracoccus 409 1 0 0 0 6 0 0

RI
Pelomonas 385 2 0 0 0 1 0 0

Bradyrhizobium 274 1 0 0 0 12 1 0

SC
Zymomonas 274 0 0 0 0 0 0 0

Epilithonimonas 189 1 0 0 0 55 0 0
NU
Hafnia 88 0 4 13 73 0 0 12

Rhodanobacter 180 0 0 0 0 4 0 0
MA

Aeromonas 14 29 55 11 32 11 19 11

Myroides 0 4 12 5 46 56 45 11

Granulicatella 44 3 4 34 37 2 31 20
D

Stenotrophomonas 163 1 0 0 0 5 0 0
E

Brevundimonas 159 0 0 0 0 6 0 0
PT

Proteiniclasticum 157 0 1 0 0 3 0 0

Enhydrobacter 158 0 0 0 0 1 0 0
CE

Silanimonas 150 0 0 0 0 1 0 0

Rothia 143 1 0 0 0 1 0 0
AC

Citricoccus 136 2 0 0 0 6 0 0

Niabella 137 1 0 0 0 4 0 0

Corynebacterium 116 2 0 0 0 7 0 0

Terrabacter 111 0 0 0 0 3 0 0

Luteimonas 108 0 0 0 0 0 0 0

Others (total abundance < 0.1%) 2339 36 48 93 173 155 58 34

Unassigned 4905 1320 968 1739 2117 1650 2563 1285

*VP: Vacuum Packed fillets stored at 4°C;

**MAP: Modified Atmosphere (50% CO2–50% N2) Packed fillets stored at 4°C
44
 ACCEPTED MANUSCRIPT

Fig. 1

30 a
abc ab ab ab
27 abc
mg-N 100g-1

24 bc bc
21
c
18
15
12
d d

PT
9 de
6 ef ef
f f f f
3

RI
0
ICE_Whole VP_Fillet MAP_Fillet

SC
NU
MA
E D
PT
CE
AC

45
 ACCEPTED MANUSCRIPT

Fig. 2

Ice_Whole
9
8
Log CFU g-1

7
6
5
4
3
2
1
0
0 3 6 9 12 15

PT
Storage time (days)

VP_Fillet
9

RI
8
Log CFU g-1

7
6

SC
5
4
3
2
NU
1
0
0 3 6 9 12 15
Storage time (days)
MA

MAP_Fillet
9
8
Log CFU g-1

7
D

6
5
E

4
3
PT

2
1
0
0 3 6 9 12 15
CE

Storage time (days)

AC

46
 ACCEPTED MANUSCRIPT

Fig. 3

PT
RI
SC
NU
MA
DE
PT
CE
AC

47
 ACCEPTED MANUSCRIPT

Fig. 4

VP_Fillet
120
mg kg-1

100
80
60
40
20
0
0 3 6 9 12 15 18 21 24 27 30

PT
Storage time (days)

MAP_Fillet
120

RI
mg kg-1

100
80

SC
60
40
20
NU
0
0 3 6 9 12 15 18 21 24 27 30
Storage time (days)
MA
E D
PT
CE
AC

48
 ACCEPTED MANUSCRIPT

Fig. 5

VP_Fillet
9
8
Log CFU g-1

7
6
5
4
3
2
0 3 6 9 12 15 18 21 24 27 30

PT
Storage time (days)

MAP_Fillet
9

RI
8
Log CFU g-1

SC
6
5
4
3
NU
2
0 3 6 9 12 15 18 21 24 27 30
Storage time (days)
MA
E D
PT
CE
AC

49