Characterisation of photo-oxidation and autoxidation products of phytoplanktonic monounsaturated fatty acids in marine particulate matter and Recent sediments


Marchand, D.; Rontani, J.F.

Organic Geochemistry 32(2): 287-304

2001


Oxidation products of phytoplanktonic monounsaturated fatty acids have been detected in particulate matter and recent sediment samples collected at the SOFI station (Gulf of Fos, Mediterranean Sea). The fluxes of these compounds are not negligible and correspond to 0.14-0.03% of the total organic carbon flux in the zone investigated. Oxidation of monounsaturated fatty acids results mainly in the production of isomeric allylic hydroperoxyacids, which are relatively stable in the water column. Saturated hydroperoxyacids, epoxyacids and dihydroxyacids, deriving probably from the addition of hydroperoxides to the double bond of monounsaturated acids, have been detected also. During early diagenesis, hydroperoxides undergo heterolytic cleavage to aldehydes and omega -oxocarboxylic or omega -hydroxycarboxylic acids, or homolytic cleavage and subsequent transformation to the corresponding alcohols or ketones. Due to the good preservation of their distribution pattern in Recent sediments, isomeric allylic hydroxyacids thus formed will constitute more useful and specific tracers of oxidative processes than omega -oxocarboxylic or omega -dicarboxylic acids (with C (sub 9) as the predominant species). These compounds will allow easy differentiation between autoxidative and photooxidative processes and could thus provide a useful indication of the residence time of particulate organic matter in the water column, the physiological state of phytoplanktonic communities and current environmental problems related to ozone depletion.

Organic
Geochemistry
PERGAMON
Organic
Geochemistry
32
(2001)
287-304
www.elsevier.nl/locate/orggeochem
Characterisation
of
photo
-oxidation
and
autoxidation
products
of
phytoplanktonic
monounsaturated
fatty
acids
in
marine
particulate
matter
and
recent
sediments
Daphne
Marchand,
Jean
-Francois
Rontani
*
Laboratoire
d'OcZilnographie
et
de
Biogeochimie
(
UMR
6535),
Centre
d'Oceanologie
de
Marseille
(OSU),
Campus
de
Luminy,
case
901,
13288
Marseille,
France
Received
12
July
2000;
accepted
23
November
2000
(returned
to
author
for
revision
9
October
2000)
Abstract
Oxidation
products
of
phytoplanktonic
monounsaturated
fatty
acids
have
been
detected
in
particulate
matter
and
recent
sediment
samples
collected
at
the
SOFI
station
(Gulf
of
Fos,
Mediterranean
Sea).
The
fl
uxes
of
these
com-
pounds
are
not
negligible
and
correspond
to
0.14-0.03%
of
the
total
organic
carbon
fl
ux
in
the
zone
investigated.
Oxidation
of
monounsaturated
fatty
acids
results
mainly
in
the
production
of
isomeric
allylic
hydroperoxyacids,
which
are
relatively
stable
in
the
water
column.
Saturated
hydroperoxyacids,
epoxyacids
and
dihydroxyacids,
deriving
prob-
ably
from
the
addition
of
hydroperoxides
to
the
double
bond
of
monounsaturated
acids,
have
been
detected
also.
During
early
diagenesis,
hydroperoxides
undergo
heterolytic
cleavage
to
aldehydes
and
at-oxocarboxylic
or
at-
hydroxycarboxylic
acids,
or
homolytic
cleavage
and
subsequent
transformation
to
the
corresponding
alcohols
or
ketones.
Due
to
the
good
preservation
of
their
distribution
pattern
in
Recent
sediments,
isomeric
allylic
hydroxyacids
thus
formed
will
constitute
more
useful
and
specific
tracers
of
oxidative
processes
than
at-oxocarboxylic
or
at-dicar-
boxylic
acids
(with
C9
as
the
predominant
species).
These
compounds
will
allow
easy
differentiation
between
aut-
oxidative
and
photooxidative
processes
and
could
thus
provide
a
useful
indication
of
the
residence
time
of
particulate
organic
matter
in
the
water
column,
the
physiological
state
of
phytoplanktonic
communities
and
current
environmental
problems
related
to
ozone
depletion.
2001
Elsevier
Science
Ltd.
All
rights
reserved.
Keywords:
Monounsaturated
fatty
acids;
Autoxidation
products;
Photo
-oxidation
products;
Hydroperoxides;
Tracers;
Particulate
matter;
Recent
sediments
1.
Introduction
In
healthy
cells
of
phytoplankton
the
primary
route
for
energy
from
the
excited
chlorophyll
singlet
state
('Chl)
is
via
the
fast
photochemical
reactions
of
photo-
synthesis
(Foote
et
al.,
1970).
In
dead
phytoplankton
cells
this
pathway
would
not
be
functional;
thus,
accel-
erated
production
of
the
longer
-lived
triplet
state
(
3
Chl,
by
intersystem
crossing)
and
toxic
oxygen
species
(singlet
*
Corresponding
author.
Tel.:
+33-4-91-82-96-23
fax:
+
33-
4-91-82-65-48.
E-mail
address:
rontani@com.univ-mrs.fr
(J.
-F.
Rontani).
oxygen,
hydrogen
peroxide,
superoxide
ion
and
hydroxyl
radical;
by
reaction
of
3
Chl
with
ground
state
oxygen)
would
be
expected
(Rontani,
1999).
The
rate
of
formation
of
these
potentially
damaging
species
can
then
exceed
the
quenching
capacity
of
the
photo
-
protective
system
of
the
cells
and
photodegradation
can
occur
(Merzlyak
and
Hendry,
1994).
Since
the
damag-
ing
effects
of
singlet
oxygen
(
1
0
2
)
primarily
involve
the
oxidation
of
membrane
lipids
(Percival
and
Dodge,
1983),
chloroplast
membrane
components
are
particu-
larly
susceptible
to
type
II
photooxidation
(i.e.
involving
1
0
2
;
Heath
and
Packer,
1968).
This
is
the
case
for
unsaturated
fatty
acids,
which
generally
predominate
in
algal
lipids,
particularly
in
the
photosynthetic
membranes
0146-6380/01/$
-
see
front
matter
©
2001
Elsevier
Science
Ltd.
All
rights
reserved.
P
II
:
S0146-6380(00)00175-3
288
D.
Marchand,
J.
-F.
Rontani
1
Organic
Geochemistry
32
(2001)
287-304
(Wood,
1974).
These
compounds,
rarely
present
in
algae
as
free
acids,
occur
more
often
as
constituents
of
more
complex
esters
(e.g.
wax
esters,
triacylglycerols,
phos-
pholipids
and
glycolipids)
and
amides
(lipopolysacchar-
ides)
(Volkman
et
al.,
1998;
Wakeham,
1999)
and
contain
cis
double
bonds,
which
distort
the
acyl
chain
and
contribute
to
membrane
fl
uidity
(Harwood
and
Russell,
1984).
Photosensitized
oxidation
of
monounsaturated
fatty
acids
involves
a
direct
reaction
of
1
0
2
with
the
carbon
-
carbon
double
bond
by
a
concerted
"ene"
addition
(Frimer,
1979)
and
leads
to
the
formation
of
hydroper-
oxides
at
each
unsaturated
carbon.
Thus,
oleic
acid
produces
a
mixture
of
9-
and
10-hydroperoxides
with
an
allylic
trans
double
bond
(Frankel
et
al.,
1979)
(Fig.
1).
These
two
hydroperoxides
may
undergo
highly
stereoselective
radical
allylic
rearrangement
respectively
to
11
-trans
and
8
-trans
hydroperoxides
(Fig.
1;
Porter
et
al.,
1995).
Recently,
it
was
shown
that
hydroperoxides
deriving
from
type
II
photosensitized
oxidation
of
C
16
0,9,
C
18
0.9
and
C
18
0.11
fatty
acids
are
produced
during
irradiation
of
killed
cells
of
Dunaliella
sp.
and
then
either
reduced
to
the
corresponding
hydroxyacids
after
homolytic
cleavage
or
cleaved
to
at-oxocarboxylic
acids
and
aldehydes
after
heterolytic
cleavage
(Rontani,
1998).
It
is
generally
considered
that
radical
autoxidation
must
play
a
role
in
destructive
biological
processes
(Porter,
1986).
Recent
observations
have
shown
that
these
processes
may
intervene
significantly
during
nat-
ural
senescence
of
phytoplankton
cells
(Marchand,
unpubl.
data).
During
the
period
1977-1984,
powerful
10
9
R
11/
\
8
R'
Autoxidation
a
hu
02
48-51%
b
49-52%
b
1
15.0%
C
10
9
R
11/
\8
R
I
Ry
1
OOH
R
=
-(CH
2
)
6
-CH
3
13.1%
c
8
10
R
-
1i
9
/
00H
R'
=
-(CH
2
)
6
-COOH
5.1%
c
R
'
10
9
2/—)L
R
IC
OOH
R
13.0%
c
.
9
10
\
8
R'
0011
22.0%
c
H00
8
00
/
11/
1
9
Radical
allylic
rearrangement
21.2%
c
11
R'
R
00H
10
9
Radical
allylic
rearrangement
a
Cis
-9
and
10
hydroperoxides
(which
represent
less
than
1%
of
autoxidation
products)
have
been
omitted
in
order
to
simplify
the
scheme.
b
From
Frankel
(1998).
c
At
14°C
(values
extrapolated
from
the
data
of
Frankel
et
al.
(1984)
and
Porter
et
al
(1994)).
Fig.
1.
Hydroperoxides
from
autoxidation
and
type
II
photo
-oxidation
of
octadec-9-enoic
acid.
8
R'
D.
Marchand,
J.
-F.
Rontani
1
Organic
Geochemistry
32
(2001)
287-304
289
analytical
methods
such
as
GC/MS
and
NMR
spectro-
metry
were
applied
to
the
problem
of
unsaturated
fatty
acid
autoxidation.
By
the
use
of
these
techniques,
Frankel
et
al.
(1977)
and
Garwood
et
al.
(1977)
reported
that
autoxidation
of
methyl
oleate
yields
mainly
a
mix-
ture
of
8
-trans,
8
-cis,
9
-trans,
10
-trans,
11
-trans
and
11
-
cis
hydroperoxides,
with
minor
amounts
of
9
-cis
and
10
-
cis
hydroperoxides
(Fig.
1).
In
the
present
work,
we
intended
to
search
for
photo
-
oxidation
and
autoxidation
products
of
mono-
unsaturated
fatty
acids
in
different
particulate
matter
and
Recent
sediment
samples
collected
in
the
Gulf
of
Fos
(Mediterranean
Sea)
in
order
to
determine:
(i)
whether
the
photodegradation
and
autoxidation
processes
parti-
cipate
efficiently
in
the
degradation
of
monounsaturated
fatty
acids
in
the
water
column
and
(ii)
whether
some
of
these
oxidation
products
are
sufficiently
stable
and
spe-
cific
to
serve
as
tracers
of
photo
-oxidation
and
aut-
oxidation
processes.
2.
Experimental
2.1.
Particulate
matter
and
sediment
sampling
The
SOFI
station
is
located
in
the
Gulf
of
Lion
(Medi-
terranean
Sea)
approximately
30
km
from
Marseilles
at
43°
04N,
08E.
This
station
is
only
exceptionally
influ-
enced
by
the
Rhone
river.
Two
sediment
traps
(PPSS)
with
a
1
m
2
opening
were
deployed
on
10/12/1997
along
a
fi
xed
mooring
under
the
euphotic
layer
(56.5
m)
and
20
m
from
the
bottom
(142
m)
and
recovered
on
20/02/1998.
In
order
to
avoid
bacterial
decomposition,
the
sample
cups
were
fi
lled
before
deployment
with
fi
ltered
(0.2
um)
seawater
containing
5%
of
formaldehyde
and
1.1
g
1
-1
of
sodium
tetraborate.
The
top
layer
of
the
bottom
sediment
(10
cm)
was
collected
at
162
m
depth
with
a
multitube
corer
(Multicoror
type
MarK
IV,
Bowers
and
Connelly,
UK;
core
diameter,
15
cm).
The
oxic
layer
of
the
sediment
was
2.5
mm
deep
(Massias,
unpubl.
data).
The
samples
were
maintained
in
isotherm
bags
during
transportation
to
Marseilles.
After
elimination
of
the
swimmers,
sediment
trap
samples
were
homogenised
and
split
into
aliquots
for
gravimetric,
elemental
(CHN)
and
organic
geochemical
analyses.
All
samples
were
then
stored
at
—20°C
for
analysis.
2.2.
Treatment
of
particulate
matter
and
sediment
samples
All
manipulations
were
carried
out
with
foil
-covered
vessels
in
order
to
exclude
photochemical
artefacts.
Particulate
matter
samples
were
fi
ltered
on
pre
-weighed
GF/F
(Whatman)
paper
and
dried
at
40°C,
while
sedi-
ment
slices
were
cut
under
dim
light
and
directly
ana-
lyzed
(lyophilisation
was
avoided
since
it
might
have
induced
some
loss
of
hydroperoxides).
2.2.1.
Extraction
Particulate
matter
and
sediment
samples
were
extrac-
ted
ultrasonically
with
isopropanol/hexane
(4:1,
v/v;
de
Leeuw
et
al.,
1977).
Hexane
extracts
were
combined
and
the
isopropanol/water
phase
was
fi
ltered,
concentrated
under
vacuum
and
then
extracted
three
times
with
chloroform.
The
chloroform
and
hexane
extracts
were
combined,
dried
over
anhydrous
Na
2
SO
4
,
fi
ltered
and
concentrated
by
rotary
evaporation
at
40°C.
2.2.2.
Reduction
of
hydroperoxides
to
alcohols
Hydroperoxides
were
reduced
to
the
corresponding
alcohols
in
methanol
(25
ml)
by
excess
NaBH
4
or
NaBD
4
(10
mg/mg
of
extract)
using
magnetic
stirring
(15
min
at
0°C;
Teng
et
al.,
1973).
During
this
treat-
ment,
ketones
are
also
reduced
and
the
possibility
of
some
ester
cleavage
cannot
be
excluded.
2.2.3.
Dehydration
of
allylic
hydroperoxides
to
ketones
Extracts
were
taken
up
in
300
µ1
of
a
mixture
of
pyr-
idine
and
acetic
anhydride
(2:1,
v/v),
allowed
to
react
at
50°C
for
2
h
and
then
evaporated
to
dryness
under
nitrogen.
Under
these
conditions,
hydroperoxides
are
quantitatively
transformed
to
the
corresponding
ketones
(Mihara
and
Tateba,
1986).
2.2.4.
Alkaline
hydrolysis
Saponification
was
carried
out
on
both
reduced
and
non
-reduced
samples
(lipid
extracts,
sediments
or
particu-
late
matter).
After
reduction,
25
ml
of
water
and
2.8
g
of
potassium
hydroxide
were
added
and
the
mixture
was
directly
saponified
by
refluxing
for
2
h.
In
the
case
of
non
-
reduced
samples,
an
additional
25
ml
of
methanol
was
added
before
saponification.
After
cooling,
the
contents
of
the
fl
ask
were
extracted
three
times
with
dichloromethane
(sediments
were
fi
ltered
through
Whatman
qualitative
fi
l-
ters
before
extraction).
The
combined
dichloromethane
extracts
were
dried
over
anhydrous
NI
-
a-60
4
,
fi
ltered
and
concentrated
to
give
the
unsaponified
fraction.
The
aque-
ous
phase
was
then
acidified
with
hydrochloric
acid
to
pH
1
and
subsequently
extracted
three
times
with
dichloro-
methane.
Treatment
of
the
combined
dichloromethane
extracts
as
described
above
gave
the
saponified
fraction.
2.2.5.
Derivatization
After
solvent
evaporation,
the
residue
was
taken
up
in
400
µ1
of
a
mixture
of
pyridine
and
BSTFA
(Supelco)
(3:1,
v/v)
and
silylated
for
1
h
at
50°C.
After
evapora-
tion
to
dryness
under
nitrogen,
the
residue
was
taken
up
in
a
mixture
of
ethyl
acetate
and
BSTFA
and
analyzed
by
gas
chromatography/electron
impact
mass
spectro-
metry
(GC/EIMS).
2.2.6.
Dimethyl
disulfide
(
DMDS)
treatment
Following
the
procedure
initially
described
by
Vin-
centi
et
al.
(1987),
the
extract
was
dissolved
in
250
µ1
of
290
D.
Marchand,
J.
-F.
Rontani
1
Organic
Geochemistry
32
(2001)
287-304
hexane,
250
µl
of
DMDS
and
125
µl
of
an
iodine
solu-
tion
(60
mg
of
iodine
in
1
ml
of
diethyl
ether).
The
reaction
mixture
was
held
at
50°C
for
24
h
and
diluted
with hexane.
The
reaction
was
quenched
with
2
ml
of
an
aqueous
solution
of
Na
2
S
2
0
3
(5%)
and
the
hexane
layer
was
pipetted
off.
The
solution
was
extracted
twice
with
hexane,
the
hexane
extracts
were
combined,
dried
on
Na
2
SO
4
,
fi
ltered
and
the
solvent
was
evaporated.
The
residue
was
then
silylated
as
described
above
and
ana-
lyzed
by
GC/EIMS.
2.3.
Identification
and
quantification
of
fatty
acids
and
their
oxidation
products
These
compounds
were
identified
by
comparison
of
retention
times
and
mass
spectra
with
those
of
standards
and
quantified
(calibration
with
external
standards)
by
GC/EIMS.
For
low
concentrations
or
in
the
case
of
coelution,
quantification
was
assessed
by
selected
ion
monitoring
(SIM).
Derivatives
of
regioisomeric
hydroxy
compounds,
which
often
coelute,
can
be
distinguished
by
their
EI
mass
spectra
to
a
great
extent,
since
the
main
fragments
are
caused
by
a
-cleavage
at
the
site
of
the
functional
group.
These
isomers
could
thus
be
detected
and
quantified
by
measurement
of
the
ion
currents
of
these
specific
fragment
ions.
Characterisation
of
cis
and
trans
allylic
hydroxyacids
was
based
on
comparison
of
their
retention
times
with
these
of
standard
compounds.
Diastereoisomeric
9,10-dihydroxy
acids
show
identical
EI
mass
spectra
but
have
distinct
retention
times.
Con-
sequently,
the
characterisation
of
these
isomers
was
obtained
by
comparison
of
their
retention
times
with
these
of
standard
(RR,SS)-
and
(RS,SR)-9,10-dihydroxy
acids.
GC/EIMS
analyses
were
carried
out
with
a
HP
5890
series
II
plus
gas
chromatograph
connected
to
a
HP
5972
mass
spectrometer.
Two
fused
silica
capillary
col-
umns
were
employed:
(A):
30
mx0.25
mm
(i.d.)
column
coated
with
HP5
(Hewlett
Packard;
fi
lm
thickness,
0.25
tim)
and
(B):
30
mx0.25
mm
(i.d.)
column
coated
with
BPX35
(SGE;
fi
lm
thickness,
0.25
tim).
The
following
conditions
were
used:
oven
temperature
programmed
from
60
to
130°C
at
30°C
min
-1
and
then
from
130
to
300°C
at
4°C
min
-1
;
carrier
gas
(He)
maintained
at
1.04
bar
until
the
end
of
the
temperature
program
and
then
programmed
from
1.04
to
1.5
bar
at
0.04
bar
min
-1
;
injector
(splitless)
temperature,
300°C;
electron
energy,
70
eV;
source
temperature,
170°C;
cycle
time,
1.5
s.
2.4.
Standard
compounds
C18:1
A9,
C18:1
A11,
C
16
,
1
A9
fatty
acids
were
pur-
chased
from
Aldrich.
Palmitelaidic
acid
was
produced
by
alkaline
hydrolysis
of
the
methyl
ester
(Sigma).
Hydroperoxides
were
obtained
after
photosensitised
oxidation
(in
pyridine
in
the
presence
of
haematopor-
phyrin
as
sensitizer;
Nickon
and
Bagli,
1961)
or
Fe"
/
ascorbate
induced
autoxidation
(Loidl-Stahlhofen
and
Spiteller,
1994)
of
the
corresponding
fatty
acids.
Sub-
sequent
reduction
of
these
different
hydroperoxides
in
methanol
with
excess
NaBH
4
afforded
the
correspond-
ing
hydroxyacids.
Hydrogenation
of
these
hydroxyacids
was
carried
out
in
methanol
with
Pd/CaCO
3
as
catalyst.
Treatment
of
monounsaturated
fatty
acids
with
meta-
chloroperoxybenzoic
acid
in
dry
methylene
chloride
yielded
epoxyacids.
(RS,SR)-
and
(RR,SS)-9,10-Dihy-
droxy
acids
were
respectively
obtained
after
stereo
-
specific
oxidation
of
the
double
bond of
oleic
(or
palmitoleic)
and
elaidic
(or
palmitelaidic)
acids
with
050
4
in
pyridine-dioxane
(MacCloskey
and
MacClel-
land,
1965).
a,cit-Dicarboxylic
acids
were
purchased
from
Sigma.
at-Hydroxycarboxylic
acids
were
produced
by
reduction
of
the
corresponding
a,cit-dicarboxylic
acids
in
dry
diethyl
ether
with
half
of
the
stoichiometric
quantity
of
LiA1H
4
.
3.
Results
and
discussion
3.1.
Particulate
matter
samples
Among
the
various
sediment
trap
samples
available,
SOFI
1
and
SOFI21
samples
(Table
1)
were
selected
for
their
quite
distinct
lipid
contents.
The
saponified
frac-
tions
of
samples
SOFI21I
and
SOFI21II
(Table
1)
show
relatively
similar
chromatographic
profiles
and
contain
large
amounts
of
polyunsaturated
fatty
acids
(notably
C20.5
and
022
:6;
Fig.
2A).
It
is
difficult
to
make
definitive
statements
about
the
origin
of
the
organic
matter
solely
on
the
basis
of
the
analysis
of
the
fatty
acids.
However,
the
presence
of
relatively
high
amounts
of
022:6
fatty
acid
and
of
detectable
quantities
of
phytanic
and
4,8,12-
trimethyltridecanoic
(4,8,12-TMTD)
acids
in
these
sam-
ples
(Fig.
2A)
strongly
suggests
that
they
contain
a
sig-
nificant
amount
of
zooplankton
or
particles
that
have
been
excreted
by
zooplankton
(Sargent,
1976;
Prahl
et
al.,
1984a).
This
is
quite
consistent
with
the
high
content
of
cholesta-5,22-dien-30-ol,
cholesterol
and
C
26
/C
27
sterols
observed
in
the
corresponding
unsaponified
fraction
(Volkman
et
al.,
1980;
Prahl
et
al.,
1984b).
The
abundance
of
0
20
.
5
and
022:6
fatty
acids
could
also
be
attributed
to
dinoflagellates
(Volkman,
1986;
Volkman
Table
1
Sampling
dates
of
sediment
trap
samples
Sample
Depth
(m)
Sampling
date
SOFIlI
SOFI1II
SOF121I
SOFI21II
56.5
142
56.5
142
11/12/97-14/12/97
11/12/97-14/12/97
09/02/98-12/02/98
09/02/98-12/02/98
D.
Marchand,
J.
-F.
Rontani
I
Organic
Geochemistry
32
(
2001)
287-304
291
A
Abundance
x
10'
50
40
30
20
10
0
B
4,8,12.TMTD
acid
C14:
N
0
cT
/Ac9,6,o
c,,
Phytani
acid
C
17:0
products
of
C
16
,,A9
acid
Oxidation
products
of
C
18
,,A9
acid
C
22:6
C18
:
0
9
Cis
:
0
11
4/
C
18:0
C
18:3
C
20:4
C
20:5
/
c
m
,
Abundance
x
10
5
40
30
20
10
0
10.00
15.00
C
1
,
1:0
C
12:0
C
16:0
20.00
25.00
30.00
Retention
time
(min)
-->
Oxidation
products
of
C
161
A9
acid
,
Oxidation
products
of
C
1
A9
acid
C
16
0.
9
C
15:0
C
17:0
Cts
rA
9
C
6
,,,A1
1
C
18:0
C
18:3
Oxidation
products
of
C
18
,6S11
acid
C
22:6
10
00
15.00
20.00 25.00
30.00
Retention
time
(min)
-->
Fig.
2.
Total
ion
current
chromatograms
of
the
silylated
reduced
saponified
fractions
of
(A)
SOFI21I
and
(B)
SOFIlI
samples.
et
al.,
1998);
however,
the
very
small
amounts
of
4
-
methyl
sterols
in
the
corresponding
unsaponified
frac-
tion
allowed
us
to
exclude
such
a
possibility.
The
detri-
tal
properties
of
these
samples
do
not
increase
significantly
with
depth
(Corg/N=
5.7
and
6.0
respec-
tively
for
SOFI21I
and
SOFI21II).
This
could
be
explained
by
the
rapid
sinking
of
the
zooplanktonic
material,
of
which
these
samples
appear
to
be
com-
posed.
In
contrast,
samples
SOFT
1I
and
SOFT
1II
(Table
1)
are
characterised
by
a
higher
phytoplanktonic
contribution
(low
amounts
of
4,8,12-TMTD,
phytanic
and
polyunsaturated
acids
in
the
saponified
fraction;
Fig.
2B).
The
presence
of
substantial
amounts
of
alke-
nones
in
the
corresponding
unsaponified
fraction
strongly
suggests
that
prymnesiophytes
are
important
constituents
of
this
phytoplanktonic
assemblage
(Mar-
lowe
et
al.,
1984).
The
detrital
properties
of
these
sam-
ples
increases
appreciably
with
depth
(C
org
/N=
6.4
and
8.4
respectively
for
SOFIlI
and
SOFIlII).
The
deeper
sediment
trap
sample
SOFIlII
contains
relatively
high
amounts
of
branched
fatty
acids
of
bacterial
origin
(e.g.
iso-
and
anteiso-
C15
acids;
Zegouagh
et
al.,
2000),
which
suggest
the
presence
of
significant
amounts
of
resus-
pended
particles
from
sediments
in
this
trap.
To
establish
the
position
of
the
double
bond
in
the
monounsaturated
fatty
acids,
the
saponified
fractions
were
treated
with
DMDS
(Vincenti
et
al.,
1987).
The
position
of
the
double
bond
was
determined
from
the
mass
spectra
of
the
derivatives
on
the
basis
of
the
frag-
ments
formed
by
cleavage
between
the
carbons
bearing
the
thioether
groups.
It
appeared
that
the
main
mono-
unsaturated
fatty
acids
present
in
particulate
matter
samples
were
C16
0,9,
C18
0.9
and
C18
0.11
fatty
acids
(Fig.
2).
Groups
of
unresolved
peaks
corresponding
to
iso-
meric
allylic
monohydroxyoctadecenoic
and
hex-
adecenoic
acids
were
detected
in
the
saponified
fraction
(Fig.
2).
These
compounds
were
characterized
by
GC/
EIMS
on
the
basis
of
their
retention
times
and
mass
spectra
(cleavage
at
the
carbon
bearing
—OSiMe
3
group;
292
D.
Marchand,
J.
-F.
Rontani
I
Organic
Geochemistry
32
(2001)
287-304
Rontani,
1998;
Fig.
3A
and
B).
There
are
three
groups
of
hydroxyacids
arising
respectively
from
the
oxidation
of
C
16
1
A9,
C
18
A9
and
C
18
I
A
1 1
fatty
acids.
Each
group
of
peaks
is
composed
mainly
of
six
isomeric
allylic
hydroxyacids.
It
can
be
seen
in
Fig.
4
that
the
group
of
peaks
corresponding
to
the
oxidation
products
of
octadec-9-enoic
acid
is
mainly
composed
of
9-hydr-
oxyoctadec-trans-10-enoic,
10-hydroxyoctadec-trans-8-
enoic,
1
1-hydroxyoctadec-trans-9-enoic,
1
1-hydroxyocta
dec-cis-9-enoic,
8-hydroxyoctadec-trans-9-enoic
and
8-
hydroxyoctadec-cis-9-enoic
acids.
The
amounts
of
these
hydroxyacids
increase
considerably
if
the
samples
are
reduced
with
NaBH
4
before
the
alkaline
hydrolysis.
Reduction
with
NaBD
4
instead
of
NaBH
4
was
attemp-
ted
in
order
to
determine
if
this
additional
production
of
hydroxyacids
results
from
the
reduction
of
the
corre-
sponding
hydroperoxides
or
ketoacids
(Fig.
5).
The
results
obtained
allowed
us
to
demonstrate
that
the
additional
production
of
hydroxyacids
results
mainly
(
>
80%)
from
reduction
of
the
corresponding
hydro
-
peroxides
and
to
a
lesser
extent
(
<
20%)
from
reduc-
tion
of
the
corresponding
ketoacids.
These
ketoacids
could
not
be
characterized
since
they
do
not
survive
alkaline
hydrolysis
and
are
cleaved
after
hydration
and
retro-aldol
reactions.
In
order
to
estimate
the
amounts
of
hydroperoxides
initially
present
in
the
samples,
the
lipid
extracts
of
aliquots
of
particulate
matter
samples
were
divided
into
two.
One
half
was
reduced
with
NaBH
4
and
saponified
and
the
other
acetylated
(to
dehydrate
hydroperoxides
to
ketones)
before
saponifi-
cation.
Comparison
of
the
amounts
of
hydroxyacids
present
after
acetylation
(naturally
-occurring
hydroxy-
acids)
and
after
reduction
(naturally
-occurring
-t
hydro
-
peroxide
and
ketoacid
reduction
-derived
hydroxyacids)
allowed
us
to
estimate
the
proportion
of
hydroperoxides
present
in
the
particulate
matter
samples
analyzed
(ratio
allylic
hydroperoxyacids/allylic
hydroxyacids
>--2).
Con-
sequently,
the
quantification
of
these
compounds
involved
reduction
with
NaBH
4
and
subsequent
sapo-
nification.
It
is
interesting
to
note
that
the
distribution
pattern
of
allylic
hydroxyacids
is
similar
after
acetyla-
tion
and
after
NaBH
4
reduction.
This
strongly
suggests
that
homolytic
cleavage
and
subsequent
reduction
do
not
discriminate
among
individual
allylic
hydro-
peroxyacid
s.
The
results
of
the
quantification
are
summarised
in
Table
2.
Though
oxidation
products
of
other
minor
monounsaturated
acids
(e.g.
C16.1
A7,
C16.1
A11,
020.1
A9)
were
also
detected,
only
oxidation
products
of
the
major
monounsaturated
acid
components
of
the
sam-
ples
were
quantified.
It
clearly
appears
that
mixtures
of
photooxidation
and
autoxidation
products
are
present.
A
Abundance(%)
90
80
70
60
50
40
30
73
343
0TvIS
B
Abundance
(/0)
90
80
70
60
50
40
30
73
129
241
20
10
0
55
1.I
129
Lt..111,
1
4
5
337
199
273
315
IM
-
CH
3
r
11,
?
,
7
20
10
0
5
6
5
9
1
5
111.I6A
147
LL
217
74
2
l
329
[M
-
41N
7
50
100
150
200
250
300
350
400 450
50
100
150
200 250
300
350
400
450
m/z-->
m/z-->
C
D
Abundance
%)
Abundance
(/o)
75
73
90
90
1,7.30
80
80
1
70
55
COOT.
70
60
117
60
or
50
129
155
50
40
30
14
153
[M
-
H
Z
O
-
40
30
147
129f
317
21
1
5
20
10
0
95
213
337
[M
-
185
229
245
355
20
10
0
55
L
103
301
357
[M
-
C11
3
]*
390
517
60
100 140 180
220
260
300 340
380
50
100 150
200 250
300 350
400
450
500
m/z-->
mlz-->
Fig.
3.
EI
mass
spectra
of
(A)
11-hydroxyoctadec-cis-9-enoic,
(B)
8-hydroxyoctadec-cis-9-enoic,
(C)
9,10-epoxyoctadecanoic
and
(D)
9,10-dihydroxyoctadecanoic
acids
(silylated).
D.
Marchand,
J.
-F.
Rontani
I
Organic
Geochemistry
32
(
2001)
287-304
293
A
Abundance
x
10
50
Trans
-9
B
Abundance
x
10
3
Trans
-9
Ion
227
-
Jon
343
---
Ion
241
20
40
Trans
-10
Trans
-10
16
30
12
20
Trans
-8
8
Trans
-8
N
Trans
-11
Cis
-8
Cis
-11
10
C,
s-8
Cis
-11
Trans
-11
4
0
22.20
22.40 22.60
22.80
23.00 23.20
22.20
22.40 22.60
22.80
23.00
23.20
Time
(min
ff
s/z
241
Ink
227
TMS
OTMS
OTMS
TMS
nil:
329
COOTMS
COOTMS
OOTMS
I
Time
(min)
-->
nth
343
Fig.
4.
Mass
chromatograms
of
m/z
227, 241,
329
and
343
showing
the
presence
of
six
isomeric
allylic
hydroxyacids
deriving
from
oleic
acid
in
the
silylated
reduced
saponified
fraction
of
SOFIlI
(A)
and
SOFIlII
(B)
samples.
OH
COOH
COOH
1-
NaBD
4
2-
BSTFA
3-
GC/ELMS
analysis
1-
NaBD
4
2-
BSTFA
3-
CC/EIMS
analysis
m/z
227
m/z
228
ne/z
232]
,
ni/z
318
TMS
COOTMS
7
COOTMS
-h•
COOT
7
MS
Fig.
5.
Mass
spectrometric
characterisation
of
allylic
hydroperoxy-
and
ketoacids
after
NaBD
4
reduction.
Indeed,
type
II
photooxidation
(i.e.
involving
singlet
oxygen)
results
in
the
formation
of
preponderant
trans
allylic
9-
and
10-
hydroxyacids
in
the
case
of
C16
0.9
and
C18
1
,6,9
fatty
acids
and
trans
allylic
11-
and
12-
hydroxyacids
in
the
case
of
C18
fatty
acid.
How-
ever,
part
of
these
products
also
result
from
autoxida-
tion
of
these
acids
(Fig.
1).
Autoxidative
processes
can
be
easily
characterised
based
on
the
presence
of
cis
294
D.
Marchand,
J.
-F.
Rontani
I
Organic
Geochemistry
32
(2001)
287-304
Table
2
Concentrations
(ng
mg
-1
dry
weight)
of
monounsaturated
fatty
acids
and
their
oxidation
products
in
sediment
trap
samples
from
SOFI
station
Compound
SOFI
II
SOFIlII
SOFI21I
SOFI21II
Octadec-9-enoic
acid
93.6
8.7
181.0
56.3
9-Hydroxyoctadec-trans-10-enoic
acid
13.5
1.4
3.3
1.2
10-Hydroxyoctadec-trans-8-enoic
acid
15.5
1.5
3.5
1.4
11-Hydroxyoctadec-trans-9-enoic
acid
3.0
0.4
0.6
0.3
11-Hydroxyoctadec-cis-9-enoic
acid
3.0
0.4
0.5
0.2
8-Hydroxyoctadec-trans-9-enoic
acid
2.6
0.3
0.4
0.2
8-Hydroxyoctadec-cis-9-enoic
acid
2.1
0.2
0.3
0.1
9-Hydroxyoctadecanoic
acid
0.9
0.1 0.1
Tra
10-Hydroxyoctadecanoic
acid
0.5
0.1 0.1
Tr
9,10-Epoxyoctadecanoic
acid
b
2.1
0.3
Tr
Tr
RS,SR-9,10-Dihydroxyoctadecanoic
acid'
0.5
0.1
Tr
Tr
RR,SS-9,10-Dihydroxyoctadecanoic
acid`
0.3
0.1
Tr
Tr
Octadec-11-enoic
acid
28.6
3.4
45.1
26.1
11-Hydroxyoctadec-trans-12-enoic
acid
5.2
0.7
1.2
0.7
12-Hydroxyoctadec-trans-10-enoic
acid
5.0
0.7
1.0
0.6
13-Hydroxyoctadec-
trans-11-enoic
acid
0.9
Tr Tr Tr
13-Hydroxyoctadec-cis-11-enoic
acid
1.0
Tr Tr Tr
10-Hydroxyoctadec-
trans-11-enoic
acid
0.9
Tr Tr Tr
10-Hydroxyoctadec-cis-11-enoic
acid
0.6
Tr Tr Tr
11-Hydroxyoctadecanoic
acid
0.2
0.1
Tr Tr
12-Hydroxyoctadecanoic
acid
0.2
0.1
Tr Tr
11,12-Epoxyoctadecanoic
acid
b
Tr Tr Tr
Tr
11,12-Dihydroxyoctadecanoic
acid'
Tr Tr Tr
Tr
Hexadec-9-enoic
acid
163.0
24.9
106.0
28.2
9-Hydroxyhexadec-trans-10-enoic
acid
11.1
1.1
2.4
0.8
10-Hydroxyhexadec-trans-8-enoic
acid
8.8
0.9
1.9
0.7
11-Hydroxyhexadec-trans-9-enoic
acid
1.5
0.2
Tr
Tr
11-Hydroxyhexadec-cis-9-enoic
acid
1.5
0.3
Tr
Tr
8-Hydroxyhexadec-trans-9-enoic
acid
2.2
0.2
Tr
Tr
8-Hydroxyhexadec-cis-9-enoic
acid
2.1
0.2
Tr
Tr
9-Hydroxyhexadecanoic
acid
0.6
0.1 0.1
Tr
10-Hydroxyhexadecanoic
acid
0.5
0.1 0.1
Tr
9,10-Epoxyhexadecanoic
acidb
1.5
0.5
Tr
Tr
RS,SR-9,10-Dihydroxyhexadecanoic
acid`
0.4
0.1
Tr
Tr
RR,SS-9,10-Dihydroxyhexadecanoic
acid'
0.2
0.1
Tr
Tr
a
Trace,
concentration
below
detection
limit
(
<
0.1
ng
b
Estimated
taking
into
account
amounts
of
main
degradation
products
of
epoxides
(chlorohydrins
and
(3-hydroxy
methyl
ethers)
formed
during
treatment
(see
Fig.
7).
Part
of
these
diols
derive
from
alkaline
or
acidic
hydrolysis
of
epoxides
(Fig.
7).
allylic
hydroxyacids,
which
are
specific
products
of
these
degradation
processes
(Porter
et
al.,
1994;
Frankel,
1998;
Fig.
1),
and
are
present
in
the
samples
in
sig-
nificant
amounts
(Table
2).
It
was
previously
demon-
strated
that
the
distribution
of
oleate
autoxidation
products
is
sensitive
to
the
H
-atom
availability
in
the
medium
(i.e.
starting
substrate
concentration)
and
tem-
perature
(Porter
et
al.,
1995).
On
the
basis
of
the
rela-
tively
high
localised
concentrations
of
lipid
compounds
present
in
phytodetritus
(Nelson,
1993),
we
used
the
data
of
Frankel
et
al.
(1984)
obtained
in
pure
neat
ole-
ate,
to
establish
a
linear
correlation
between
the
ratios
([cis
8-hydroxyacid]
-t
[cis
11-hydroxyacid])/[trans
9-
(or
10-)hydroxyacid]
and
the
temperature
(°C)
[Eqs.
(1)
and
(2)],
and
to
estimate
the
amounts
of
trans
9-
and
10-
hydroxyacids
resulting
from
autoxidative
processes
at
the
in
situ
temperature
(14°C).
([cis
8]
+
[cis
11])/[trans
9]
=
-0.0138T
+
1.5017
(r
2
=
0.96)
(1)
([cis
8]
+
[cis
11])/[trans
10]
=
-0.0144T
+
1.5528
(r
2
=
0.93)
(2)
D.
Marchand,
J.
-F.
Rontani
/
Organic
Geochemistry
32
(2001)
287-304
295
After
subtraction
of
the
amounts
of
trans
9-
and
10-
hydroxyacids
of
autoxidative
origin,
we
could
thus
esti-
mate
the
proportion
of
photoproducts
of
oleic
acid
in
the
different
particulate
matter
samples
analyzed
(Table
3).
The
percentage
of
these
compounds
relative
to
the
parent
fatty
acid
appears
to
be
much
higher
in
the
SOFI1I
than
in
the
SOFI21I
trap
sample.
This
is
con-
sistent
with
a
high
zooplanktonic
content
of
the
SOFI21I
sample,
since
it
is
well
known
that
photo
-
degradation
rates
of
lipidic
compounds
are
slower
in
copepod
faecal
pellets
than
in
senescent
phytoplank-
tonic
cells
(Nelson,
1993).
Since
the
two
sediment
traps
were
located
under
the
euphotic
layer,
these
percentages
do
not
change
significantly
in
the
deeper
sediment
trap
samples
(Table
3).
The
stability
of
the
ratios
also
shows
that
in
the
water
column,
degradative
processes
(biode-
gradation
and
autoxidation)
act
on
oleic
acid
and
its
photoproducts
without
selectivity.
Additional
oxidation
products
of
monounsaturated
fatty
acids
were
detected
(Table
2),
notably
cis
and
trans
9,10-epoxyacids
(Fig.
3C),
9,10-dihydroxyacids
(Fig.
3D)
and
saturated
9-,
10-,11-
and
12-hydroxyacids
(Fig.
6A).
The
amounts
of
saturated
hydroxyacids
increase
con-
siderably
if
the
samples
are
reduced
with
NaBH
4
before
the
alkaline
hydrolysis.
In
some
cases,
NaBH
4
reduction
of
allylic
ketoacids
may
directly
afford
the
correspond-
ing
saturated
hydroxyacids
(March,
1985);
an
attempt
at
reduction
with
NaBD
4
allowed
us
to
exclude
such
a
possibility
(Fig.
5).
This
manipulation
has
shown
that
half
of
the
additional
production
of
saturated
hydro-
xyacids
results
from
the
reduction
of
the
corresponding
hydroperoxides,
whereas
the
other
arises
from
the
par-
tial
reduction
of
epoxyacids
during
the
treatment
(Fig.
7).
9,10-Epoxyoctadecanoic
acids
have
been
pre-
viously
reported
in
marine
(Stephanou,
1992)
and
urban
(Stephanou
and
Stratigakis,
1993)
aerosols
and
found
to
be
labile.
Epoxyacids
are
in
fact
strongly
degraded
dur-
ing
the
treatment;
in
addition
to
the
partial
reduction
with
NaBH
4
observed
above,
they
undergo
alcoholysis
and
hydrolysis
during
alkaline
hydrolysis
and
are
con-
verted
to
chlorohydrins
and
9,10-dihydroxyacids
during
acidification
(Fig.
7).
It
is
well
known
that
some
bacteria
may
convert
A9-
fatty
acids
to
9,10-epoxyacids
(Ruet-
tinger
and
Fulco,
1981);
these
bacteria
can
further
hydrolyse
these
epoxides
to
yield
9,10-dihydroxyacids
Table
3
Percentages
of
photooxidation
and
autoxidation
products
of
oleic
acid
in
particulate
matter
samples
SOFIlI
SOFIlII
SOFI21I
SOFI21II
Photoproductsa
23 23
3
4
Autoxidation
products'
24
33
2
2
a
Relative
to
parent
fatty
acid.
(Michaels
et
al.,
1980).
Enzymatic
epoxidation
of
cis
A9
-fatty
acids
is
generally
highly
stereospecific
and
affords
only
cis
epoxides
(Croteau
and
Kolattukudy,
1975).
We
discarded
such
a
possibility
since
the
small
amounts
of
epoxyacids
surviving
the
treatment
of
par-
ticulate
matter
samples
are
made
up
of
mixtures
of
cis
and
trans
isomers,
the
initial
presence
of
the
two
isomers
being
also
supported
by
the
detection
of
two
erythro
and
threo
pairs
of
diastereoisomeric
9,10-dihydroxyacids
(Table
2).
Epoxyacids
can
also
be
formed
by
cyclization
of
alkoxyl
radicals
deriving
from
allylic
hydroperoxides
(Frankel,
1998).
However,
in
this
case,
other
epoxyacid
isomers
(e.g.
8,9-epoxyoctadecanoic
and
10,11-epoxy-
octadecanoic
acids)
should
be
formed,
but
we
failed
to
detect
these
compounds
in
the
different
samples
ana-
lyzed.
The
formation
of
epoxyacids
is
suggested
to
be
due
to
hydroperoxide-induced
autoxidation
of
mono-
unsaturated
fatty
acids.
The
mechanism
proposed
(Fig.
8)
involves
classical
addition
of
a
peroxyl
radical
to
the
double
bond
of
the
fatty
acid
(Berti,
1973).
The
radical
thus
formed
can
then:
(i)
abstract
a
hydrogen
atom
from
another
molecule
to
give
saturated
hydro-
peroxy-acids,
(ii)
lead
to
an
epoxide
by fast
intramole-
cular
homolytic
substitution
(Fossey
et
al.,
1995)
or
(iii)
react
with
molecular
oxygen
forming
a
peroxyl
radical,
which
can
abstract
a
hydrogen
atom
from
an
other
molecule
and
thus
lead
to
the
formation
of
a
9,10-dihy-
droperoxy-acid.
9,10-Dihydroxyacids
may
be
subse-
quently
formed
either
by
hydrolysis
of
9,10-epoxyacids
or
by
reduction
of
9,10-dihydroperoxyacids.
If
we
consider
the
percentage
of
total
autoxidation
products
of
oleic
acid
in
the
different
particulate
matter
samples
(Table
3),
we
note
a
significant
increase
with
depth
in
the
case
of
the
SOFIl
traps,
which
is
consistent
with
the
more
detritic
properties
of
the
deeper
particu-
late
matter
sample
(C
org
i
N
=
6.4
and
8.4
respectively
for
SOFI1I
and
SOFT
1II).
The
autoxidation
state
of
these
samples
is
also
well
correlated
with
the
proportion
of
the
specific
cis
-8
and
cis
-11
allylic
hydroxyacids
(Fig.
1),
which
increases
with
depth
(Fig.
4).
In
the
case
of
the
SOFI21
traps,
the
percentage
of
autoxidation
products
of
oleic
acid
does
not
change
significantly
with
depth
(Table
3).
This
result,
which
is
consistent
with
the
very
similar
C
org
i
N
ratios
obtained
for
these
two
samples
(Corg/N=
5.7
and
6.0
respectively
for
SOFI21I
and
SOFI21II)
and
with
the
presence
of
high
amounts
of
labile
polyunsaturated
fatty
acids
in
the
deeper
SOFI21II
trap,
can
be
attributed
to
the
relatively
rapid
sinking
of
the
zooplanktonic
content
of
these
samples.
Homologous
series
of
ra,cit-dicarboxylic
(Fig.
6B)
and
at-hydroxycarboxylic
acids
(Fig.
6C)
in
the
C
7
—C
li
range
have
also
been
detected
in
the
particulate
matter
sam-
ples,
C9
species
being
generally
predominant
(Table
4).
Examination
of
non
-reduced
samples
allowed
us
to
show
that
at-hydroxycarboxylic
acids
result
from
the
reduction
of
the
corresponding
at-oxocarboxylic
acids
296
D.
Marchand,
J.
-F.
Rontani
I
Organic
Geochemistry
32
(
2001)
287-304
A
Abundance
(%)
90
80
70
60
73
ma
ravl
B
Abundance
(%
90
COOTMS
80
70
60
)
75
73
TMSOOC
MS
50
187
331
50
[M
-
CH
3
]*
40
40
55
129
317
30
30
117
201
147
20
129
217
1M
-
C
11
31
+
20
10
51
103
147
204
246
302
01
i,'
385
01
10
97
171185
217
0 0
50
100 150
200 250
300
350
400
40
80
120 160
200
240 280
320
m/z-->
C
Abundance
(
°
/o)
75
90
80
70
7
+.
60
[M
-
CH
3
]'
50
317
40
55
117
30
147
[M
-
CH
3
-
TMS0H]'
20
103
129
227
10
301
0
1
1
185
285
40
80
120 160
200 240 280
320
m/z-->
Fig.
6.
EI
mass
spectra
of
(A)
10-hydroxyhexadecanoic,
(B)
oc,co-nonanedioic
and
(C)
(o-hydroxydecanoic
acids
(silylated).
H
OH
R
R'
R
R
.
D
OD
vi
OD
OH
H
OH
OCH
3
+
CH
3
OH
R'
NaBH
4
NaBD
4
Acidic
hydrolysis
OH
CI
R
)
R'
Alkaline
hydrolysis
Fig.
7.
Degradation
of
epoxyacids
during
sample
treatment.
OH OH
R
R
R
=
-(CH
2
),CH
3
R'
=
-(CH2),COOH
D.
Marchand,
J.
-F.
Rontani
I
Organic
Geochemistry
32
(2001)
287-304
297
COOH
HOO
OOH
COOH
-
HO
SH
S
*
Cis
+
trans
+
H
2
O
OH
OH
(
SH
S
0011
OOH
COOH
COOH
COOH
COOH
OOH
Heterolytic
or
homolytic
cleavage
Heterolytic
or
homolytic
cleavage
Fig.
8.
Addition
of
a
peroxyl
radical
to
the
double
bond
of
octadec-9-enoic
acid.
Table
4
Concentration
(ng
mg
-1
dry
weight)
of
C
7
-C
11
(o-oxocarboxy-
tic'
and
rx,co-dicarboxylic
acids
in
sediment
trap
samples
from
SOFI
station
Compound
SOFIlI
SOFIlII
SOFI21I
SOFI21II
(o-Oxoheptanoic
acid
(o-Oxooctanoic
acid
(o-Oxononanoic
acid
(o-Oxodecanoic
acid
(o-Oxoundecanoic
acid
rx,o)-Heptanedioic
acid
rx,co-Octanedioic
acid
rx,co-Nonanedioic
acid
rx,co-Decanedioic
acid
rx,co-Undecanedioic
acid
Tr
b
Tr Tr Tr
1.0
0.3 0.3
0.1
2.7
0.8
0.6
0.3
2.3
1.0
0.4
0.2
1.5
0.4
0.3
0.2
Tr Tr Tr Tr
1.0
0.4
0.4
0.6
7.4
3.1
1.3
1.4
2.6
1.1
0.7
0.4
2.9
1.0
0.6 0.6
a
Quantified
after
reduction
in
the
form
of
(o-hydroxycarb
oxylic
acids
(Fig.
6A).
b
Trace,
concentration
below
detection
limit
(
<
0.1
ng
during
reduction
with
NaBH
4
.
These
compounds
were
initially
discovered
in
remote
marine
aerosols
from
the
North
Pacific
(Kawamura
and
Gagosian,
1987)
and
then
detected
in
different
sediment
traps
(Kawamura
et
al.,
1990)
and
sediments
(Kawamura
et
al.,
1990;
Ste-
phanou,
1992).
They
have
been
considered
to
originate
in
the
atmosphere
by
photochemically-induced
oxida-
tive
degradation
of
biogenic
unsaturated
fatty
acids,
which
contain
double
bonds
predominantly
at
the
C-9
position
(Kawamura
and
Gagosian,
1987).
More
recently,
we
have
demonstrated
that
heterolytic
cleavage
of
allylic
hydroperoxides
(Frimer,
1979;
Fig.
9)
resulting
from
the
photodegradation
of
phytoplanktonic
mono-
unsaturated
fatty
acids
in
the
euphotic
layer
could
represent
another
source
of
these
compounds
in
the
marine
environment
(Rontani,
1998).
The
predominance
of
the
C9
species
in
the
distribution
pattern
of
these
compounds
has
been
attributed
to
the
fact
that
hetero-
lytic
cleavage
of
the
predominant
allylic
9-
and
10-
hydroperoxyacids
affords
the
at-oxononanoic
acid
in
each
case
(Rontani,
1999).
Flux
calculations
are
reported
in
Table
5.
It
appears
that
the
total
fatty
acid
oxidation
product
fl
uxes
are
small
but
significant
and
correspond
to
0.03-0.14%
of
the
total
organic
carbon
fl
ux.
The
strong
decrease
in
these
fl
uxes
observed
between
the
SOFT
1I
and
SOFIlII
sediment
traps
(Table
5)
confirms
that
the
deeper
SOFI
III
trap
is
partly
supplied
by
resuspended
particles
of
bottom
sediment,
which
contains
lower
concentra-
tions
of
fatty
acid
oxidation
products
than
particulate
matter
samples
(see
following
section).
3.2.
Sediment
samples
All
the
fatty
acid
oxidation
products
detected
in
the
particulate
matter
samples
are
present
in
the
underlying
sediment.
It
is
noteworthy
that
analysis
of
non
-reduced
samples
enabled
us
to
detect
significant
amounts
of
298
D.
Marchand,
J.
-F.
Rontani
I
Organic
Geochemistry
32
(
2001)
287-304
Table
5
Fluxes
estimated
from
analysis
of
sediment
trap
samples
collected
from
SOFI
station
SOFIlI
SOFIlII
SOFI21I
SOFI21II
Total
mass
(mg
m
-2.
d
-1
)
111
355
139
188
Total
organic
carbon
(TOC)
(mg
m
-2
d
-1
)
8.2
13.6
10.7
4.4
Oxidation
products
of
C18,1A9
fatty
acid
(ug
111
-2
d
-1
)
4.9
1.7
1.2
0.6
Oxidation
products
of
C18
:
1A
11
fatty
acid
(ug
m
-2
d
-1
)
1.5
0.5
0.3
0.2
Oxidation
products
of
Ci6
:
1A9
fatty
acid
(ug
111
-2
d
-1
)
3.4
1.3
0.6
0.3
C
7
-C
11
(o-oxocarboxylic
acids
(ug
111
-2
d
-1
)
0.8
0.9
0.2
0.2
C
7
-C
11
oc,co-dicarboxylic
acids
(ug
111
-2
d
-1
)
1.5
2.0
0.4
0.5
Total
fatty
acid
oxidation
products
(ug
111
-2
d
-1
)
12.1
6.4
2.7
1.8
Ratio:
total
fatty
acid
oxidation
products/TOC
(%)
0.14
0.05
0.03
0.04
at-hydroxycarboxylic
acids
in
the
C
7
-C
1
I
range,
which
are
lacking
in
non
-reduced
particulate
matter
samples.
Fig.
10
depicts
the
concentration
profiles
with
depth
of
monounsaturated
fatty
acids
and
their
oxidation
pro-
ducts
in
the
core
sections
investigated.
C
16
1
A9
and
C
18
I
All
fatty
acids
are
present
in
higher
proportions
(relative
to
the
C
181
,6.9
fatty
acid)
in
the
sediments
(Fig.
10)
than
in
the
particulate
matter
samples
(Table
2).
This
may
be
attributed
to
a
strong
bacterial
contribution
to
the
sedimentary
organic
matter.
In
fact,
most
bacteria
living
in
sediments
do
not
produce
sub-
stantial
amounts
of
C
18
1
A9
fatty
acid
(Zegouagh
et
al.,
2000),
whereas
C
161
,6.9
and
C
18
1
A1
1
fatty
acids
pre
-
OOH
by
or
A
dominate
in
a
number
of
species
(Lehninger,
1975;
Zegouagh
et
al.,
2000).
This
hypothesis
is
supported
by
the
presence
in
the
sedimentary
extracts
of
high
amounts
of
iso-
and
anteiso-C
15
fatty
acids.
The
con-
centrations
of
oxidation
products
of
monounsaturated
acids
in
the
sediments
of
the
SOFI
station
(Fig.
10)
are
significantly
low
compared
to
those
of
the
sediment
trap
samples
(Tables
2
and
4).
The
tests
used
in
the
case
of
particulate
matter
samples
(see
above;
Fig.
5)
allowed us
to
show
that
allylic
ketoacids
are
absent
from
the
sedi-
ments;
this
can
be
attributed
to
a
rapid
degradation
of
these
compounds
(involving
hydration
and
subsequent
retro-aldol
reactions)
in
the
sedimentary
environment
Mlylie
rearrangement
COON
ROOH
HOOC
Heterolytic
cleavage
COOH
ROOH
Homolytic
cleavage
S
.
k
S
.
OH
Bacteria
00011
COON
0011
COOH
.**
OH
COON
Bacteria
Mineralisation
(13
-oxidation)
COOH
Hydratation
and
retroaldolisation
Fig.
9.
Main
degradative
pathways
of
allylic
hydroperoxyacids
in
recent
sediments
(modified
from
Rontani,
1998).
D.
Marchand,
J.
-F.
Rontani
I
Organic
Geochemistry
32
(
2001)
287-304
299
0
50
0
2
10
12
ila
ng
dry
weight
100 150
200
250
300 350
400 450
C18:1
6
.
9
acid
Allylic
hydroxyacids
-.--
Saturated
hydroxyacids
-cc-
9,10-Epoxyacid
9,10-Dihydroxyacid
2
6
c7
10
12
ng
gl
dry
weight
400
800
1200
1600
C
16
,
1
49
acid
Allylic
hydroxyacids
Saturated
hydroxyacids
9,10-Dihydroxyacids
ng
g
dry
weight
100
200
300
400
500
600
700
o
o
2
4
ng
gl
dry
weight
100
200
300
400
500
600
700
O
O
1
6
0
=
6
3
8
10
C
18
,
1
411
acid
10
C
7
-C
11
w-Oxoacids
Allylic
hydroxyacids
C
7
C
11
w-Hydroxyacids
Saturated
hydroxyacids
C
7
-C
11
w-Diacids
12
12
Fig.
10.
Depth
profiles
of
the
main
monounsaturated
fatty
acids
and
their
oxidation
products
in
sediment
core
sections
of
the
SOFI
station.
The
structures
of
these
isomeric
compounds
are
give
in
Table
2
and
to
the
fact
that
their
formation
involves
oxidation
of
the
corresponding
secondary
alkoxyl
radical
by
molecular
oxygen
(Fig.
9),
a
process
which
is
limited
to
the
oxic
zone
of
the
sediment.
A
significant
part
of
allylic
hydroperoxyacids,
which
were
relatively
well
preserved
in
particulate
matter
samples,
is
degraded
in
the
fi
rst
centimetres
of
the
sedi-
ment
(ratio
allylic
hydroperoxyacids/allylic
hydroxy-
acids
1
at
11
cm
depth).
This
is
probably
due
to
the
fact
that
heterolytic
cleavage
must
be
strongly
favoured
in
the
sedimentary
environment.
This
hypothesis
is
sup-
ported
by
the
strong
proportion
of
at-dicarboxylic
and
at-
oxocarboxylic
acids
(with
C9
as
the
most
abundant;
Table
6)
detected
in
the
sediments
(Fig.
10).
It
is
impor-
tant
to
note
that
the
distribution
pattern
of
allylic
hydroxyacids
is
well
preserved
in
the
sediments
(Table
7).
This
strongly
suggests
that
during
early
diagenesis
degradation
does
not
selectively
discriminate
among
individual
allylic
hydroperoxy-
or
hydroxyacids.
Thus,
the
autoxidative
and
photochemical
signatures
will
be
Table
6
Relative
percentages
of
C
7
—C
11
co-dicarboxylic,
co-oxo-
carboxylic
and
co-hydroxycarboxylic
acids
detected
in
2-4
cm
slice
of
sediments
from
SOFI
station
Compound
C7
C
8
C9
C10
C11
co-Dicarboxylic
acids
2
9
59
15
15
co-Oxocarboxylic
acids
Tra
18
52
12
18
co-Hydroxycarboxylic
acids
Tr
14
31
33
22
a
Trace.
300
D.
Marchand,
J.
-F.
Rontani
I
Organic
Geochemistry
32
(2001)
287-304
Table
7
Relative
percentages
of
allylic
hydroxyacids
resulting
from
oxidation
of
C18
:
1
A9
acid
detected
in
upper
sediment
of
SOFI
station
Sediment
slice
(cm)
%
of
cis
-8
and
-11
allylic
hydroxyacids
%
of
trans
-9
and
-10
allylic
hydroxyacids
%
of
trans
-8
and
-11
allylic
hydroxyacids
0-2
13
70
17
2-4
11
69
20
4-6
10
72
18
6-8
10
70
20
8-10
13
63
24
10-12
NDa
ND ND
a
Not
determined
(concentrations
too
low).
uniquely
imprinted
and
preserved
in
the
sediment
regardless
of
early
diagenesis
of
organic
matter.
As
pre-
viously
observed
by
Kawamura
et
al.,
(1990)
in
sedi-
ments
of
the
North
Pacific,
the
predominance
of
the
9-
oxononanoic
acid
is
also
preserved
throughout
the
core
analysed.
The
proportion
of
saturated
hydroxyacids
(relative
to
allylic
hydroxyacids)
appears
to
be
much
higher
in
the
sediment
(Fig.
10)
than
in
the
particulate
matter
samples
(Table
2).
This
result
may
be
attributed
to:
(i)
the
reduction
of
the
double
bond
of
allylic
hydroperoxy-
acids,
(ii)
the
relative
recalcitrance
of
saturated
hydro-
peroxyacids
towards
degradation,
(iii)
the
production
of
these
saturated
species
in
the
sedimentary
environment
or
(iv)
the
presence
of
a
higher
proportion
of
epoxyacids
in
the
sediments.
In
fact,
it
has
previously
been
demon-
strated
that
reduction
processes
play
a
part
in
the
early
diagenesis
of
unsaturated
fatty
acids
(Rhead
et
al.,
1971).
This
could
also
be
the
case
for
allylic
hydroxy-
acids;
however,
the
profiles
of
saturated
hydroxyacids
obtained
(pairs
of
9-,
10-
and
11-,
12-
isomers)
allowed
us
to
discard
such
a
possibility
(which
must
lead
to
the
formation
of
a
higher
number
of
regioisomeric
hydroxy
compounds).
The
production
of
C
7
—C
11
at-hydroxy-
carboxylic
acids
observed
in
the
sediment
probably
results
from
heterolytic
cleavage
of
saturated
hydro-
peroxyacids
(Fig.
11).
Owing
to
the
considerably
weaker
migratory
aptitude
of
alkyl
groups
(relative
to
the
vinyl
group;
Frimer,
1979),
this
process
must
be
much
slower
than
in
the
case
of
allylic
hydroperoxyacids.
This
can
explain
the
relatively
good
preservation
of
saturated
hydroperoxides
observed
in
the
sediment
(ratio
satu-
rated
hydroperoxyacids/saturated
hydroxyacids
3
at
11
cm
depth)
and
the
lack
of
at-hydroxycarboxylic
acids
in
particulate
matter
samples.
This
hypothesis
is
also
well
supported
by
the
absence
of
C9
predominance
in
the
distribution
pattern
of
at-hydroxycarboxylic
acids
(Table
6),
since
heterolytic
cleavage
of
each
pair
of
saturated
hydroperoxyacids
affords
two
distinct
at-
hydroxycarboxylic
acids
(Fig.
11).
Owing
to
the
rela-
tively
good
preservation
of
hydroperoxides
in
recent
sediments,
which
constitute
potential
sources
of
peroxyl
radicals,
production
of
saturated
hydroperoxides
within
intact
biological
debris
(e.g.
well
-silicified
diatoms;
Rontani
and
Marchand,
2000)
cannot
be
totally
exclu-
ded.
Under
anaerobic
conditions,
the
pathway
proposed
above
for
the
formation
of
these
compounds
in
the
water
column
(Fig.
8)
must
be
strongly
favoured.
In
fact,
the
formation
of
saturated
hydroperoxides
and
dihydroxyacids
being
competitive
processes
(Fig.
8),
the
likelihood
that
the
radical
resulting
from
the
addition
of
a
peroxyl
radical
to
the
double
bond
of
mono-
unsaturated
acids
abstracts
a
hydrogen
atom
on
another
molecule
must
be
much
more
higher
in
the
absence
of
oxygen.
A
significant
part
of
saturated
hydroxyacids
also
results
from
the
reduction
of
epoxyacids
during
the
treatment.
Comparison
of
the
amounts
of
dihydrox-
yacids
(which
may
derive
from
alkaline
or
acidic
hydrolysis
of
the
corresponding
epoxides;
Fig.
7)
in
sediments
(Fig.
10)
and
particulate
matter
samples
(Table
2)
well
supports
the
hypothesis
of
the
presence
of
a
higher
proportion
of
epoxyacids
in
the
sediments.
3.3.
Tracer
potential
of
different
oxidation
products
The
detection
of
homologous
series
of
at-oxo-
carboxylic
acids
(with
C9
as
the
most
abundant)
and
their
corresponding
ra,co-dicarboxylic
acids
in
marine
sediments
(Kawamura
et
al.,
1990;
Stephanou,
1992)
and
sediment
traps
(Kawamura
et
al.,
1990)
was
inter-
preted
as
indicative
of
atmospheric
transport
of
unsatu-
rated
fatty
acid
photooxidation
products
rather
than
of
an
autochthonous
formation
(Kawamura
et
al.,
1990;
Stephanou
and
Stratigakis,
1993).
The
production
of
similar
compounds
was
recently
observed
during
the
irradiation
of
senescent
phytoplanktonic
cells
(Rontani,
1998;
Rontani
et
al.,
1998)
and
attributed
to
heterolytic
cleavage
of
allylic
hydroperoxides
resulting
from
the
photo
-degradation
of
phytoplanktonic
unsaturated
acids
(Rontani,
1998).
These
results
call
into
question
the
use
of
these
compounds
as
tracers
of
atmospheric
input
to
the
sea,
since
the
photosensitised
oxidation
of
unsaturated
fatty
acids
in
senescent
phytoplanktonic
cells,
which
can
act
throughout
the
euphotic
layer
of
the
D.
Marchand,
J.
-F.
Rontani
/
Organic
Geochemistry
32
(
2001)
287-304
301
COON
HOO
Heterolytic
cleavage
COOH
_
COOH
+
H
2
O
HOyCOOH
CI
COOH
OHC
ROOH
HOOC
-
COOH
ROOH
Fig.
11.
Heterolytic
cleavage
of
10-hydroperoxyoctadecanoic
acid.
oceans,
must
constitute
a
significant
source
of
at-oxo-
carboxylic
and
ra,cit-dicarboxylic
acids
in
the
marine
environment.
Moreover,
it
is
important
to
note
that
het-
erolytic
cleavage
of
allylic
hydroperoxides
produced
by
autoxidation
of
A9
unsaturated
acids
also
results
in
the
production
of
at-oxocarboxylic
and
ra,cit-dicarboxylic
acids
(with
C9
as
the
more
abundant
species;
Table
8).
It
is
clear
that
these
compounds
cannot
be
considered
any
longer
as
specific
tracers
of
photooxidative
degradation
of
biogenic
unsaturated
fatty
acids
but
only
of
their
oxidation.
Though
selective
biodegradation,
(acid
catalyzed)
migration
of
double
bonds
and
cis/trans
isomerisation
has
been
shown
to
occur
in
sediments,
the
preservation
of
the
distribution
pattern
of
the
different
isomeric
allylic
hydroxyacids
observed
in
the
core
(see
above)
means
Table
8
co-Oxocarboxylic
and
oc,co-dicarboxylic
acids
produced
by
het-
erolytic
cleavage
of
allylic
hydroperoxides
resulting
from
oleic
acid
autoxidation
co-Oxocarboxylic
acidsa
oc,co-Dicarboxylic
acidsa
C7
3
C8
21
C9
39
C
10
30
C
11
2
C12
5
2
8
49
33
8
Tr
b
a
Relative
percentages.
b
Trace.
that
the
use
of
these
compounds
as
tracers
of
oxidative
processes
in
the
marine
environment
may
be
envisaged.
Due
to
their
specificity
(Fig.
1),
cis
hydroxyacids
will
constitute
good
tracers
of
autoxidative
processes.
In
oxic
zones,
the
proportion
of
these
isomers
could
thus
provide
useful
indications
regarding
the
residence
time
of
organic
matter
in
the
water
column.
Cis
hydroxyacids
also
provide
a
basis
for
estimating
the
part
of
trans
hydroxyacids
resulting
from
photoxidative
processes
(see
above),
which
could
be
associated
with
other
widely
distributed
photoproducts
of
the
chlorophyll
phytyl
side
chain
(Cuny
and
Rontani,
1999)
and
sterols
(Rontani
and
Marchand,
2000,
Marchand,
unpubl.
data)
to
con-
stitute
a
"pool"
of
useful
indicators
of
photooxidative
alterations
of
phytoplankton.
Atmospheric
scattering
is
most
pronounced
in
the
blue
and
UV
region
of
solar
light.
Consequently,
ozone
depletion
results
in
the
exposure
of
phytoplankton
to
enhanced
UV
-B
and
blue
light
doses.
If
UV
-B
has
been
shown
to
reduce
the
sur-
vival,
growth and
production
of
phytoplankton
(Sker-
ratt
et
al.,
1998),
an
enhancement
of
the
light
doses
in
the
blue
region
must
result
to
the
production
of
higher
amounts
of
excited
chlorophylls
and
thus
to
an
enhanced
photodegradation
of
the
lipid
components
of
phytoplankton.
Consequently,
photo
-oxidation
pro-
ducts
of
lipid
components
of
phytoplankton
could
give
useful
information
about
ozone
depletion.
Saturated
9-
and
10-
or
11-
and
12-hydroxyacid
pairs
(resulting
from
reduction
of
the
corresponding
hydro
-
peroxides
or
epoxyacids)
also
constitute
good
tracers
of
302
D.
Marchand,
J.
-F.
Rontani
/
Organic
Geochemistry
32
(2001)
287-304
unsaturated
fatty
acid
autoxidation.
In
contrast,
although
bacterial
hydration
of
the
A9
double
bond of
fatty
acids
is
well
known
(Hammond,
1988;
Koritala
et
al.,
1989),
this
process
yields
only
10-hydroxyacids
(Ratledge,
1994).
Since
autoxidative
and
photoxidative
degradation
of
lipid
components
of
phytoplankton
intervene
mainly
in
senescent
cells,
the
percentage
of
monounsaturated
fatty
acid
oxidation
products
relative
to
their
parent
fatty
acids
could
be
indicative
of
the
physiological
state
of
phytoplanktonic
communities.
4.
Conclusions
Autoxidation
and
photooxidation
products
of
mono-
unsaturated
fatty
acids
have
been
characterised
and
quantified
in
particulate
matter
and
recent
sediment
samples
collected
at
the
SOFI
station
(Gulf
of
Lion,
Mediterranean
Sea).
The
results
confirm
previous
observations
obtained
in
vitro
(Rontani,
1998)
and
show
that
the
oxidation
of
monounsaturated
fatty
acids
in
senescent
phytoplanktonic
cells
results
mainly
in
the
formation
of
isomeric
allylic
hydroperoxyacids,
which
are
relatively
stable
in
particulate
matter.
Other
oxida-
tion
products
(saturated
hydroperoxides,
epoxyacids
and
dihydroxyacids),
deriving
probably
from
the
addi-
tion
of
hydroperoxides
to
the
double
bond
of
mono-
unsaturated
fatty
acids,
have
also
been
detected.
The
fatty
acid
oxidation
product
fl
uxes
correspond
to
0.03-
0.14%
of
the
total
organic
carbon
fl
ux
in
the
zone
investigated.
During
early
diagenesis,
hydroperoxides
undergo
either
heterolytic
cleavage
to
yield
aldehydes,
alcohols,
at-oxocarboxylic
and
at-hydroxycarboxylic
acids,
or
homolytic
cleavage
and
subsequent
transformation
to
the
corresponding
alcohols
and
ketones.
Isomeric
allylic
hydroxyacids
thus
formed
constitute
much
more
useful
and
specific
markers
of
oxidative
processes
than
at-oxo-
carboxylic
or
at-dicarboxylic
acids
(with
C9
as
the
pre-
dominant
species).
The
distribution
pattern
of
allylic
hydroxyacids,
which
is
well
preserved
in
recent
sedi-
ments,
allows
differentiation
between
autoxidative
and
photooxidative
processes.
These
compounds
could
thus
provide
useful
information
about
the
residence
time
of
particulate
organic
matter
in
the
water
column,
the
physiological
state
of
phytoplanktonic
communities
and
current
environmental
problems
related
to
ozone
deple-
tion.
Heterolytic
cleavage
seems
to
play
an
important
role
in
the
degradation
of
light
-induced
or
autoxidative
hydroperoxides
during
early
diagenesis.
This
could
explain
the
good
preservation
of
saturated
hydroper-
oxides
observed
throughout
the
core
analysed.
These
results
support
well
our
previous
hypothesis
(Rontani
and
Marchand,
2000),
i.e.
that
some
hydroperoxides
must
be
sufficiently
stable
in
marine
sediments
to
parti-
cipate
in
the
degradation
of
unsaturated
organic
com-
pounds
under
anoxic
conditions.
Taking
into
account
the
high
amounts
of
photo
-
products
of
monounsaturated
fatty
acids
detected
in
the
particulate
matter
samples
and
the
well
known
increas-
ing
rates
of
autoxidation
(Frankel,
1998)
and
photo
-
oxidation
(Rontani
et
al.,
1998)
of
fatty
acids
with
their
degree
of
unsaturations,
it
can
be
concluded
that
con-
siderable
amounts
of
polyunsaturated
fatty
acids
must
be
oxidized
in
the
marine
environment.
However,
dur-
ing
this
study
we
failed
to
detect
oxidation
products
of
this
kind
of
fatty
acid.
This
is
possibly
due
to:
(i)
the
instability
of
the
hydroperoxides
formed,
or
(ii)
the
involvement
of
cross
-linking
reactions
leading
to
the
formation
of
macromolecular
structures
(Neff
et
al.,
1988)
non
-amenable
to
gas
chromatography.
The
char-
acterisation
and
quantification
of
such
structures
(which
could
play
a
role
in
the
formation
of
humic
substances;
Harvey
et
al.,
1983)
in
particulate
matter
or
senescent
phytoplanktonic
samples
constitutes
an
exciting
chal-
lenge
for
the
future.
Acknowledgements
We
wish
to
thank
Dr.
J.
K.
Volkman
for
his
kind
and
helpful
suggestions
regarding
the
origin
of
organic
mat-
ter
in
sediment
trap
samples.
We
also
thank
Drs.
V.
Grossi,
P.
Raimbault,
N.
Garcia
and
C.
Grenz
for
the
sampling
and
initial
treatment
of
the
particulate
matter
and
sediment
samples
used
in
this
work.
We
are
very
grateful
to
Dr.
P.
Raimbault
for
providing
the
total
fl
uxes,
C,
g
/N
and
TOC
values
of
the
particulate
matter
samples
and
to
Mr.
M.
Paul
for
his
careful
reading
of
the
English.
Many
thanks
are
due
to
the
anonymous
referees
for
their
useful
and
constructive
comments.
Associate
Editor
J.R.
Maxwell
References
Berti,
G.,
1973.
Stereochemical
aspects
of
the
synthesis
of
1,2-
epoxides.
In:
Allinger,
N.L.,
Eliel,
E.L.
(Eds.),
Topics
in
Ste-
reochemistry.
Interscience
Publishers,
New
York,
pp.
95-234.
Croteau,
R.R.,
Kolattukudy,
P.E.,
1975.
Enzymatic
epoxida-
tion
of
18-hydroxy-cis-9,10-epoxystearic
acid
by
a
particulate
preparation
from
spinach.
Archives
of
Biochemistry
and
Biophysics
170,
61-72.
Cuny,
P.,
Rontani,
J.
-F.,
1999.
On
the
widespread
occurrence
of
3-methylidene-7,11,15-trimethylhexadecan-1,2-diol
in
the
marine
environment:
a
specific
isoprenoid
marker
of
chlor-
ophyll
photodegradation.
Marine
Chemistry
65,
155-165.
Foote,
C.S.,
Chang,
Y.C.,
Denny,
R.W.,
1970.
Chemistry
of
singlet
oxygen
X-
Carotenoid
quenching
parallels
biological
protection.
Journal
of
the
American
Chemical
Society
92,
5216-5218.
D.
Marchand,
J.
-F.
Rontani
I
Organic
Geochemistry
32
(2001)
287-304
303
Fossey,
J.,
Lefort,
D.,
Sorba,
J.,
1995.
Free
Radicals
in
Organic
Chemistry.
Masson,
Paris
(pp.
1-307).
Frankel,
E.N.,
1998.
Lipid
Oxidation.
The
Oily
Press,
Dundee
(pp.
1-302).
Frankel,
E.N.,
Garwood,
R.F.,
Khambay,
B.P.S.,
Moss,
G.,
Weedon,
B.C.L.,
1984.
Stereochemistry
of
olefin
and
fatty
acid
oxidation.
III.
The
allylic
hydroperoxides
from
the
antioxidation
of
methyl
oleate.
Journal
of
the
Chemical
Society
Perkin
Transactions
I,
2233-3340.
Frankel,
E.N.,
Neff,
W.E.,
Bessler,
T.R.,
1979.
Analysis
of
autoxidized
fats
by
gas
chromatography
-mass
spectrometry:
V.
Photosensitized
oxidation.
Lipids
14,
961-967.
Frankel,
E.N.,
Neff,
W.E.,
Rohwedder,
W.K.,
Khambay,
B.P.S.,
Garwood,
R.F.,
Weedon,
B.C.L.,
1977.
Analysis
of
autoxidized
fats
by
gas
chromatography
-mass
spectrometry:
I.
Methyl
oleate.
Lipids
12,
901-907.
Frimer,
A.A.,
1979.
The
reaction
of
singlet
oxygen
with
olefins:
the
question
of
mechanism.
Chemical
Reviews
79,
359-387.
Garwood,
R.F.,
Khambay,
B.P.S.,
Weedon,
B.C.L.,
Frankel,
E.N.,
1977.
Allylic
hydroperoxides
from
the
autoxidation
of
methyl
oleate.
Journal
of
the
Chemical
Society
Chemical
Communications.
364-365.
Hammond,
R.C.,
1988.
Enzymatic
modification
at
the
mid
-
chain
of
fatty
acids.
Fat
Science
Technology
90,
18-27.
Harvey,
G.R.,
Boran,
D.A.,
Chesal,
L.A.,
Tokar,
J.M.,
1983.
The
structure
of
marine
fulvic
and
humic
acids.
Marine
Chemistry
12,
119-132.
Harwood,
J.L.,
Russell,
N.J.,
1984.
Lipids
in
Plants
and
Microbes.
George
Allen
&
Unwin,
London
(pp.
1-162).
Heath,
R.L.,
Packer,
L.,
1968.
Photoperoxidation
in
isolated
chloroplasts
II
Role
of
electron
transfer.
Archives
of
Bio-
chemistry
and
Biophysics
125,
850-857.
Kawamura,
K.,
Gagosian,
R.B.,
1987.
Implications
of
(o-oxo-
carboxylic
acids
in
the
remote
marine
atmosphere
for
photo
-
oxidation
of
unsaturated
fatty
acids.
Nature
325,
330-332.
Kawamura,
K.,
Handa,
N.,
Nozaki,
Y.,
1990.
(o-Oxocarboxylic
acids
in
the
sediment
trap
and
sediment
samples
from
the
North
Pacific:
implication
for
the
transport
of
photo
-oxida-
tion
products
to
deep-sea
environments.
Geochemical
Jour-
nal
24,
217-222.
Koritala,
S.,
Hosie,
L.,
Hou,
C.T.,
Hessletine,
C.W.,
Bagby,
M.O.,
1989.
Microbial
conversion
of
oleic
acid
to
10-hydro-
xystearic
acid.
Applied
Microbiology
and
Biotechnology
32,
299-304.
de
Leeuw,
J.W.,
Simoneit,
B.R.T.,
Boon,
J.J.,
Rijpstra,
W.I.C.,
de
Lange,
F.,
van
der
Leeden,
J.C.W.,
Correia,
V.A.,
Bur-
lingame,
A.L.,
Schenck,
P.A.,
1975.
Phytol
derived
com-
pounds
in
the
geosphere.
In:
Campos,
R.,
Goni,
J.
(Eds.),
Advances
in
Organic
Geochemistry
1977.
Enadimsa,
Madrid,
pp.
61-79.
Lehninger,
A.L.,
1975.
Biochimie.
Bases
Moleculaires
de
la
Structure
et
des
Fonctions
Cellulaires.
Flammarion,
Bourges
(pp.
1-1088).
Loidl-Stahlhofen,
A.,
Spiteller,
G.,
1994.
rx-Hydroxyaldehydes,
products
of
lipid
peroxidation.
Biochimica
et
Biophysica
Acta
1211,
156-160.
MacCloskey,
J.A.,
MacClelland,
M.J.,
1965.
Mass
spectra
of
O-isopropylidene
derivatives
of
unsaturated
fatty
esters.
Journal
of
the
American
Chemical
Society
87,
5090-5093.
March,
J.,
1985.
Advanced
Organic
Chemistry,
3rd
Edition.
John
Wiley
and
Sons,
New
York
(p.
694).
Marlowe,
B.I.T.,
Green,
J.C.,
Neal,
A.C.,
Brassell,
S.C.,
Eglin-
ton,
G.,
Course,
P.A.,
1984.
Long
-chain
(nC
37
-C
39
)
alke-
nones
in
the
prymnesiophyceae
(Distribution
of
alkenones
and
other
lipids
and
their
taxonomic
significance).
British
Phycology
Journal
19,
203-216.
Merzlyak,
M.,
Hendry,
G.A.F.,
1994.
Free
radical
metabolism,
pigment
degradation
and
lipid
peroxidation
in
leaves
during
senescence.
Proceedings
of
the
Royal
Society
of
Edinburgh
102B,
459-471.
Michaels,
B.C.,
Ruettinger,
R.T.,
Fulco,
A.J.,
1980.
Hydration
of
9,10-epoxypalmitic
acid
by
a
soluble
enzyme
from
Bacillus
megaterium.
Biochemistry
and
Biophysics
Research
Com-
munications
92,
1189-1195.
Mihara,
S.,
Tateba,
H.,
1986.
Photosensitized
oxygenation
reactions
of
phytol
and
its
derivatives.
Journal
of
Organic
Chemistry
51,
1142-1144.
Neff,
W.E.,
Frankel,
E.N.,
Fujimoto,
K.,
1988.
Autoxidative
dimerization
of
methyl
linolenate
and
its
monohydroper-
oxides,
hydroperoxy
epidioxides
and
dihydroperoxides.
Journal
of
the
American
Oil
Chemical
Society
65,
616-623.
Nelson,
J.R.,
1993.
Rates
and
possible
mechanism
of
light
-
dependent
degradation
of
pigments
in
detritus
derived
from
phytoplankton.
Journal
of
Marine
Research
51,
155-179.
Nickon,
A.,
Bagli,
J.F.,
1961.
Reactivity
and
geochemistry
in
allylic
systems
I
(Stereochemistry
of
photosensitized
oxyge-
nation
of
monoolefins).
Journal
of
the
American
Chemical
Society
83,
1498-1508.
Percival,
M.P.,
Dodge,
A.D.,
1983.
Photodynamic
damage
to
chloroplast
membranes,
photooxidized
oxidation
of
chlor-
oplast
acyl
lipids.
Plant
Science
Letters
29,
255-264.
Porter,
N.A.,
1986.
Mechanisms
for
the
autoxidation
of
poly-
unsaturated
lipids.
Accounts
in
Chemical
Research
19,
262-
268.
Porter,
N.A.,
Mills,
K.A.,
Carter,
R.L.,
1994.
A
mechanistic
study
of
oleate
autoxidation:
competing
peroxyl
H
-atom
abstraction
and
rearrangement.
Journal
of
the
American
Chemical
Society
116,
6690-6696.
Porter,
N.A.,
Caldwell,
S.E.,
Mills,
K.A.,
1995.
Mechanisms
of
free
radical
oxidation
of
unsaturated
lipids.
Lipids
30,
277-
290.
Prahl,
F.G.,
Eglinton,
G.,
Corner,
E.D.S.,
O'Hara,
S.C.M.,
1984a.
Copepod
fecal
pellets
as
a
source
of
dihydrophytol
in
marine
sediments.
Science
224,
1235-1237.
Prahl,
F.G.,
Eglinton,
G.,
Corner,
E.D.S.,
O'Hara,
S.C.M.,
Forsberg,
T.E.V.,
1984b.
Changes
in
plant
lipids
during
pas-
sage
through
the
gut
of
Calanus.
Journal
of
Marine
Biology
64,
317-334.
Ratledge,
C.,
1994.
Biodegradation
of
oils,
fats
and
fatty
acids.
In:
Ratledge,
C.
(Ed),
Biochemistry
of
Microbial
Degrada-
tion.
Kluwer
Academic
Publishers,
Dordrecht,
pp.
89-141.
Rhead,
M.M.,
Eglinton,
G.,
Draffan,
G.H.,
1971.
Conversion
of
oleic
acid
to
saturated
fatty
acids
in
Severn
estuary
sedi-
ments.
Nature
232,
327-330.
Rontani,
J.
-F.,
1998.
Photodegradation
of
unsaturated
fatty
acids
in
senescent
cells
of
phytoplankton:
photoproduct
structural
identification
and
mechanistic
aspects.
Journal
of
Photochemistry
and
Photobiology
A114,
37-44.
Rontani,
J.
-F.,
1999.
Photodegradation
of
lipidic
compounds
during
the
senescence
of
phytoplankton.
In:
Boule,
P.
(Ed.),
The
Handbook
of
Environmental
Chemistry.
Springer
Ver-
lag,
Berlin
Heidelberg,
pp.
264-282.
304
D.
Marchand,
J.
-F.
Rontani
I
Organic
Geochemistry
32
(2001)
287-304
Rontani,
J.
-F.,
Marchand,
D.,
2000.
Photoproducts
of
phyto-
planktonic
sterols:
a
potential
source
of
hydroperoxides
in
marine
sediments?
Organic
Geochemistry
31,
169-180.
Rontani,
J.
-F.,
Cuny,
P.,
Grossi,
V.,
1998.
Identification
of
a
pool
of
lipid
photoproducts
in
senescent
phytoplanktonic
cells.
Organic
Geochemistry
29,
1215-1225.
Ruettinger,
R.T.,
Fulco,
A.J.,
1981.
Epoxidation
of
unsatu-
rated
fatty
acids
by
a
soluble
cytochrome
P
-450
-dependent
system
from
Bacillus
megaterium.
The
Journal
of
Biological
Chemistry
256,
5728-5734.
Sargent,
J.R.,
1976.
The
structure,
function
and
metabolism
of
lipids
in
marine
organisms.
In:
Malins,
D.C.,
Sargent,
J.R.
(Eds.),
Biochemical
and
Biophysical
Perspectives
in
Marine
Biology.
Academic
Press,
New
York,
pp.
149-212.
Skerratt,
J.H.,
Davidson,
A.D.,
Nichols,
P.D.,
MacMeekins,
T.A.,
1998.
Effect
of
UV
-B
on
lipid
content
of
three
antartic
marine
phytoplankton.
Phytochemistry
49,
999-1007.
Stephanou,
E.G.,
1992.
rx,co-Dicarboxylic
acid
salts
and
rx,o)-
dicarboxylic
acids
photooxidation
products
of
unsaturated
fatty
acids
present
in
marine
aerosols
and
marine
sediments.
Naturwissenschaften
79,
128-131.
Stephanou,
E.G.,
Stratigakis,
N.,
1993.
Oxocarboxylic
and
a,(3-
dicarboxylic
acids:
photooxidation
products
of
biogenic
unsaturated
fatty
acids
present
in
urban
aerosols.
Environ-
mental
Science
and
Technology
27,
1403-1407.
Teng,
J.I.,
Kulig,
M.J.,
Smith,
L.L.,
Kan,
G.,
van
Lier,
J.E.,
1993.
Sterol
metabolism
XX
(Cholesterol
7(3-hydroperoxide).
Journal
of
Organic
Chemistry
38,
119-123.
Vincenti,
M.,
Guglielmetti,
G.,
Cassani,
G.,
Tonini,
C.,
1987.
Determination
of
double
bond
position
in
diunsaturated
compounds
by
mass
spectrometry
of
dimethyl
disulphide
derivatives.
Analytical
Chemistry
59,
694-699.
Volkman,
J.
K.,
Corner,
E.
D.
S.,
Eglinton,
G.,
1980.
Trans-
formation
of
lipids
in
the
marine
food
web
and
in
underlying
bottom
sediments.
In:
Colloques
Internationaux
du
CNRS
293,
Biogeochimie
de
la
Matiere
Organique
a
l'Interface
Eau
-Sediment
Marin,
Editions
CNRS,
Paris,
pp.
185-197.
Volkman,
J.K.,
1986.
A
review
of
sterol
markers
for
marine
and
terrigeneous
organic
matter.
Organic
Geochemistry
9,
83-100.
Volkman,
J.K.,
Barrett,
S.M.,
Blackburn,
S.I.,
Mansour,
M.P.,
Sikes,
E.L.,
Gelin,
F.,
1998.
Microbial
biomarkers:
a
review
of
recent
research
developments.
Organic
Geochemistry
29,
1163-1179.
Wakeham,
S.G.,
1999.
Monocarboxylic,
dicarboxylic
and
hydroxy
acids
released
by
sequential
treatments
of
suspended
particles
and
sediments
of
the
Black
sea.
Organic
Geochem-
istry
30,
1059-1074.
Wood,
B.J.B.,
1974.
Fatty
acids
and
saponified
lipids.
In:
Steward,
W.D.
(Ed.),
Algal
Physiology
and
Biochemistry.
University
of
California
Press,
Berkeley,
pp.
236-265.
Zegouagh,
Y.,
Derenne,
S.,
Largeau,
C.,
Saliot,
A.,
2000.
A
geochemical
investigation
of
carboxylic
acids
released
via
sequential
treatments
of
two
surficial
sediments
from
the
Chagjiang
delta
and
East
China
sea.
Organic
Geochemistry
31,
375-388.