Sources and transformations of chlorophylls and carotenoids in a monomictic sulphate-rich karstic lake environment
JOAN VILLANUEVA, l JOAN O. GRIMALT, 1 RUTGER DE WIT, l* BRENDAN J. KEELY 2 and JAMES R. MAXWELL 3
~Department of Environmental Chemistry (C.I.D.-C.S.I.C.), Jordi Girona, 18, 08034-Barcelona, Catalo- nia, Spain, 2Department of Chemistry, University of York, Heslington, York YO 1 5DD, England, U.K. and 3Organic Geochemistry Unit, School of Chemistry, University of Bristol, Cantock's Close, Bristol
BS8 ITS, England, U.K.
Abstract--A study of the pigment composition in the water column particulates and bottom sediment column of an anoxic lake, Lake Ciso (Catalonia, Spain), has been carried out. The depth-dependencies of the carotenoid and chlorophyll distributions in the water column during stratification and holomixis, as well as during the diurnal cycles, have been determined. In the sediment, the composition of the pigments has been studied over the top 25 cm. The analyses were carried out by high performance liquid chromatography (HPLC) and HPLC coupled to mass spectrometry.
The distributions of carotenoids, bacteriochlorophylls and chlorophylls in the water particulates closely follow the population dynamics of the photosynthetic organisms during stratification and holomixis. Among these, only chlorophyll a transformation products are found in the water column, namely phaeophytin a, pyrophaeophytin a and pyrophaeophorbide a. These derivatives also occur in the sediment, where they are found together with bacteriophaeophytin a and pyrobacteriophaeophytin a. This compartment-dependent occurrence corresponds to a higher degree of preservation of the bacterial vs algal pigments. The former occur in the anoxic hypolimnion or mixed water column during holomixis, and the latter in the oxic epilimnion. The presence of these compounds and the very low abundance of phaeophorbide a and absence of bacteriophaeophorbide a point to a transformation pathway where, in the absence of zooplankton grazing or high abundances of the enzyme chlorophyllase, the loss of the phytyl chain occurs predominantly after loss of the C-132 carbomethoxyl group.
Okenone, alloxanthin, ct-carotene and zeaxanthin/lutein are the major water column carotenoids. Their relative abundances contrast with their distribution in the sediments where okenone is the predominant compound and zeaxanthin/lutein are below detection limit. This effect may be in part a consequence of a more uniform okenone concentration in the water particulates over the year, leading to a higher annual sedimentation flux. Again, it may also be influenced by the higher degree of preservation of the bacterial vs algal carotenoids. A similar situation is observed for the relative abundances of chlorophyll a and bacteriochlorophyll a in water particulates and sediments.
Key words--anoxic lakes, chlorophyll a degradation products, bacteriochlorophyll a degradation prod- ucts, bacteriochlorophylls d, Cryptomonas spp., purple sulphur bacteria, green sulphur bacteria, caroten- oids, okenone, karstic lakes, Catalonia, Spain
INTRODUCTION
Photosynthetic pigment studies of present day marine and lacustrine environments have typically focused on examples having predominantly oxic water column conditions. Information about the sources and transformation processes which determine the compositions of chlorophylls and carotenoids in aquatic systems with anoxic water columns is more limited, although some recent studies have involved such aspects in the Black Sea water column and bot tom sediments (Repeta, 1993; Repeta et al., 1989; Sinninghe Damst6 et al., 1993). An understanding of
*Present address: Laboratoire d'Oceanographie Biologique, University of Bordeaux I. 2, rue du Proffeseur Jolyet, 33120 Arcachon, France.
the processes occurring in such environments requires a combined evaluation of microbial populations and pigment distributions in the water column over the most significant seasonal periods and comparison with the sedimentary distributions. A better under- standing of the pigment biogeochemistry in anoxic systems is particularly important with respect to interpreting palaeoenvironments and in the light of recent information about the transformation path- ways of chlorophyll (Keely et al., 1990) and caroten- oids (Repeta, 1989), and the incorporation of such components or their transformation products into macromolecular fractions via sulphur bridges (Schaeffer et al., 1993; Sinninghe Damst6 et al., 1993).
We report here the sedimentary and water column pigment compositions in Lake Ciso, a karstic mo- nomictic lake situated near Banyoles (Catalonia,
739
740 JOAN VILLANUEVA et al.
Spain) which is completely mixed during the winter and is thermally stratified during the rest of the year. High amounts of sulphate (10mM) give rise to a strong development of sulphate-reducing bacteria and a high concentration of hydrogen sulphide (ca 1 mM; Guerrero and Mas, 1989; Pedros-Alio et al., 1987). During stratification the epilimnion is usually oxic and the hypolimnion is anoxic. Purple sulphur bacteria grow in the metalimnion and top hypolimnion and green sulphur bacteria in the hy- polimnion. During holomixis virtually the entire water column is anoxic and the lake community is supported by anoxygenic primary production. Chromatium minus, Amoebobacter sp. and Chloro- bium spp. are the most common species utilizing the hydrogen sulphide generated by the sulphate-reduc- ing bacteria (Guerrero and Mas, 1989; Guerrero et al., 1987). These changes, e.g. vertical profiles of oxygen and hydrogen sulphide, control the nature of the populations of microorganisms and ultimately must be reflected in major qualitative and quantita- tive differences in the pigment compositions. In ad- dition to these seasonal fluctuations, there are also a diel variations. These light-dependent changes are also important because of the key role of anoxygenic phototrophic bacteria in the oxidation of reduced sulphur species (van Gemerden et al., 1985). Hence, changes in composition between day and night should be investigated to ascertain if they might reveal important variations in the physiology of the lake or vertical column migrations.
The seasonal changes in the pigment distributions in water column particulates over a one year period are described. We report the distributions in samples collected in March, July and September, correspond- ing to the early, regular and late stages of the stratification period, respectively. The holomixis period is represented by samples collected in late October. In each sampling period, pigment and bulk physico-chemical depth profiles were obtained during day and night and the predominant organisms living in the water column were identified by optical microscopy. The results are compared with the pig- ment composition in the underlying sediments.
EXPERIMENTAL
Site description
Lake Ciso is located in the karstic area of Lake Banyoles (42°08'N, 2°45'E; Catalonia, NE Spain). Lakes in this area are fed by underground water which dissolves Eocene gypsum strata, a process that may eventually result in collapsing of the surface and formation of new lakes (e.g. Lake Nou ca 150 m from Lake Ciso). Lake Ciso is almost circular (average diameter 25 m) and has a surface area of 490 m E. The maximum and average depths are 7.5 and 4.78 m, respectively, and the water volume is 2200 m 3. The mineral content is about 1.5-2g/I with sulphate
(< 12.5 mM) and calcium (< 10.2 raM) as the pre- dominant ions. Alkalinity is high (<7.6meq/l), mainly due to bicarbonate (93%). Iron and manga- nese concentrations are low (0.1 mM and 0.1 #M, respectively), presumably due to precipitation by sulphide. Ammonium concentration is about 0.18mM in the anoxic zones and soluble reactive phosphate is undetectable most of the time (Guerrero et aL, 1985).
Sampling
Water column samples were collected over the following periods, 18-20/3/1992; 9/7/1992; 17-18/9/1992; 28-29/10/1992. Because of the small size of the lake, samples were collected at the centre from a small rubber dinghy moved by means of two ropes, one anchored to the bottom and the other fastened to a tree. This allowed the boat to be positioned precisely, a maximum drift of only 1-1.5 m occurring during sampling, and avoided row- ing-associated disturbance. A delay of ca 15 min after positioning and start of sampling further minimized water column alteration effects. Prior to sampling, depth profiles of temperature, oxygen and pH were obtained.
Samples were collected using a metal cage contain- ing a glass bottle equipped with a spring loaded Teflon plug. The device was placed at defined depth levels using a rope, with a second rope being used to lift the plug. Samples were taken from shallower to deeper levels to minimize water column disturbance and those for pigment analysis were collected in dark bottles without a space between water level and Teflon plug. Aliquots (100-400 ml) were centrifuged just after sampling and the supernatant discarded. The particles were initially stored at 4°C for 3-4 h during transport to the lab and then at -20°C in a nitrogen atmosphere. Pigment analyses were per- formed within 3-4 weeks after sampling.
Sediment core samples (40 cm × 30 cm i.d.) were collected manually in January 1993 with an in-house metallic system consisting of 1.5 m lengths that can be screwed to each other to reach total lengths of 7.5-9 m. Cores were stored at 4°C just after sampling and were frozen at - 20°C a few hours later. Pigment analysis was performed within 3-4 weeks of sampling of the thawed cores. The sections (5 cm) selected for analysis were freeze-dried prior to extraction.
Extraction
Water column particulates were extracted (x 3, 7 rain each) by sonication with acetone (2-3 ml) and the extracts combined. Further extraction showed no detectable amount of pigments by visual observation. Extracts of particulates collected in September were concentrated to 0.25-3 ml by evaporation under ni- trogen. Sediments were similarly extracted (4-5 times) with acetone (30-50 ml) until the extract was colourless. The combined extracts were filtered and vacuum evaporated to volumes of 0.5-5ml. All
Tab
le 1
. C
arot
enoi
d an
d ch
loro
phyl
l pi
gmen
ts i
dent
ifie
d in
Lak
e C
is6
Ret
enti
on
Abs
orba
uce
MS
Dia
gnos
tic
Ext
inct
ion
Sour
ce f
or
Com
pone
nt
time*
m
axim
a (n
m)
ions
co
effi
cien
tt id
entif
icat
ion
Ref
.¶]
3.--
Oke
none
28
.1
487
(515
):~
578
2320
E
ctot
hior
hodo
spir
a sp
. [7
] 4.
--B
acte
rioc
hlor
ophy
ll a
22
.2
368/
590/
770(
370)
91
0 92
3 C
hrom
atiu
m v
inos
um
DSM
185
[5
] 5.
--A
llox
anth
in
18
.8
451/
478(
440)
56
4 25
00
[5]
6.--
~hlo
roph
yll
a 25
.3
432(
430)
89
2/87
0/83
4/53
8 1
54
2
Stan
dard
[6
] 7.
--C
hlor
ophy
ll c
1
0.6
43
8(44
0)
14
00
P
haeo
dact
ilum
tri
corn
utum
[I
] &
--~
-Car
oten
e 38
.7
447/
473(
440)
53
6 27
25
Stan
dard
[9
] 9/
10.-
-Lut
ein/
Zea
xant
hin
20.2
44
3147
1(43
0)
568
2000
P
horm
idiu
m v
alde
rian
um
P4~
II
.--C
h|or
ophy
ll b
22
.6
466(
460)
22
14
Gre
an l
eave
s [6
] i 2
.--B
acte
rioe
hlor
ophy
ll d
* 1
8.1
42
7(43
0)
687
[4]
12.-
-Bac
teri
oehl
orop
hyll
d*
18
.6
427(
430)
68
7 [4
] 12
.--B
acte
rioe
hlor
ophy
ll d
* 1
9.1
42
7(43
0)
687
[4]
i 2.-
-Bac
teri
ochl
orop
hyll
d*
20.9
42
7(43
0)
687
[4]
12.-
-Bac
teri
ocbl
orop
hyll
d*
21.2
42
7(43
0)
687
[4]
! 2.-
-Bac
teri
ochl
orop
hyll
d*
21.6
42
7(43
0)
687
[4]
13.-
-Chl
orob
acte
ne
35.3
(4
40)
2000
C
hlor
obiu
m l
ir~i
cola
II
14.-
-Pha
eoph
ytin
a
42.0
40
8/50
5(41
0)
870/
812/
516
15
56
St
anda
rd
[6]
15.-
-Pyr
opha
eoph
ytin
a
53.2
40
8/50
8(41
0)
812/
516
14
52
St
anda
rd
[3]
16.-
-Pyr
opha
eoph
orbi
de a
1
6.8
(4
10)/
505]
537
548
12
76
St
anda
rd
[3]
17.-
-Bac
teri
opha
eoph
ytin
a
29.6
35
5/52
6(37
0)
888/
830/
552/
534
923
[8]
I&--
Pyr
obac
teri
opha
eoph
ytin
a
37.8
35
5/53
0(37
0)
830/
552/
534
923
[8]
21.-
-Pha
eoph
orbi
de a
1
3.8
40
8/50
5/53
5(41
0)
606/
548
14
15
St
anda
rd
[2]
23.-
-Unk
now
n 24
.4
447/
474
7A.-
-Bac
teri
opha
eoph
y tin
a'
32.0
35
5/52
6(37
0 88
8/83
0/55
2/53
4 92
3 [8
] 25
.--p
-Car
oten
e 40
. I
455/
481
(440
) 53
6 22
09
Stan
dard
[5
] 2&
--P
haeo
phyt
in a
" 45
.5
408/
505(
410)
87
0/81
2/51
6 1
55
6
Stan
dard
[6
]
*(m
in)
As
defi
ned
in t
he f
irst
chr
omat
ogra
phic
sys
tem
(se
e ex
peri
men
tal
sect
ion)
.
8 _<. H
tThe
spe
cifi
c ex
tinc
tion
coe
ffic
ient
val
ues
of s
ome
phae
opig
men
ts w
ere
calc
ulat
ed
from
the
mol
ar e
xtin
ctio
n co
effi
cien
ts o
f ph
aeop
hyti
n a.
Whe
n no
val
ues
wer
e av
aila
ble,
an
ass
umed
coe
ffic
ient
of
2000
was
tak
en f
or t
he c
arot
enoi
ds (
Man
tour
a an
d L
lew
elly
n,
1983
). :[[
Val
ues i
n pa
rent
hese
s re
fer
to t
he w
avel
engt
h se
lect
ed f
or q
uant
ific
atio
n.
§Iso
late
d fr
om c
arbo
nate
pon
ds i
n so
lar
salt
erns
of
the
Ebr
e D
elta
. I[
lsol
ated
fro
m L
ake
Cis
b (B
anyo
les,
C
atal
onia
, Sp
ain)
. ¶l J
effr
ey (
1968
); 2
, A
bach
i an
d R
iley
(197
9);
3, c
alcu
late
d fr
om p
haeo
phor
bide
a;
4, S
teen
berg
en a
nd K
orth
als
(198
2);
5, D
avie
s (1
976)
; 6,
Bro
wn
(196
9);
7, F
oppe
n (1
971)
; 8,
cal
cula
ted
from
bac
teri
ochl
orop
hyll
a;
9, g
iven
in
the
com
mer
cial
spe
cifi
catio
ns.
**T
enta
tive
assi
gnm
ent;
sup
port
ed b
y M
- at
m/z
770
, 78
4, 7
98 f
or t
he 3
maj
or c
ompo
nent
s as
pha
eoph
ytin
s in
HP
LC
-MS
Sep
t. sa
mpl
e.
742 JOAN VILLANUEVA et al.
extracts were flushed with nitrogen and stored at -20°C. The analysis was performed within 2-3 days. Test analyses on algal cultures showed that artifacts such as phaeophytins or allomers of chlorophyll a were not generated during the whole procedure.
The extracts were analyzed first by high perform- ance liquid chromatography coupled to multiwave- length UV-VIS spectrophotometry (HPLC-UV). Then, they were carried to Bristol (ca 8 h transport time) in containers full of dry ice for analysis by high performance liquid chromatography coupled to mass spectrometry (HPLC-MS). These analyses were per- formed after methylation with diazomethane and concentration to 100-300/~1 using a Speed-Vac (Savant) rotatory evaporator.
HPLC-UV
A Hewlett-Packard HP-1090 Liquid Chromato- graph equipped with an automatic injector and a diode array detector (190-600nm) was used. The detector was operated at 370, 410, 430, 440, 460 and 515 nm for selective monitoring of pigments (Table 1). Separation was performed on a C, 8 column (Spheri-5; 220 × 4.6 mm i.d.; Brownlee Labs, ODS- 224). An ion-pairing agent (IPA) prepared from 1.5 g of tetrabutylammonium acetate and 7.7g of ammonium acetate in 100 mt of water was used. Aliquots (100pl) were analyzed using a linear elution gradient similar to that described by Mantoura and Llewellyn (1983), starting from 100% methanol/water/IPA (8/1/1 v/v) to 100% methanol/acetone (8/2 v/v) in 20 min, holding the latter mixture in isocratic elution, 5 min, and sub- sequently increasing to methanol/acetone (5/5 v/v) in 15 min. Flow rate was 1 ml/min.
H P L C - M S
The system comprised a Waters MS 600 Silk tertiary delivery instrument coupled to a Finningan MAT TSQ 70 mass spectrometer via a Finnigan MAT TSP-2 thermospray interface. Before entering the spectrometer, the effluent was passed through a Waters 991 diode-array absorbance detector. Com- parison of the chromatograms, at several wave- lengths, with the profiles obtained using the HPLC-UV system allowed correlation of peaks in both chromatograms. Separation was performed with two Waters Nova-Pack Cj8 radial compression car- tridges (each 100 x 5 mm i.d.) operated at a flow rate of 1.0 ml/min. The elution gradient program is de- scribed in Table 2. Under these conditions, most of the pigments elute in the same order as in the HPLC-UV system (Table 1). Samples were injected using a Rheodyne 7125 injector valve equipped with a 25 #1 injection loop. Both injector and detector cells were chosen to withstand back-pressures up to 280 bar. The effluent was ionized in the discharge mode, with 1200 and 0V for the discharge and repeller voltages, respectively, and 65 and 250°C for the
Table 2. Solvent elution program used in HPLC-MS
Time Acetone Methanol Water (rain) (%) (%) (%)
0 0 90 10 5 0 90 10
15 70 15 t5 40 90 5 5 90 0 90 10 95 0 90 I0
vaporizer and source temperatures. Mass spectral information was obtained in the negative ion mode, scanning from 300 to 1200 Da in 1.5 s.
Identification and quantification
The carotenoid and chlorophyll-derived pigments were identified by mass spectrometric and UV-VIS absorption spectral data. In a number of cases identifications were confirmed by HPLC-UV coelu- tion with standard compounds or pigments obtained from the reference materials. These materials and the diagnostic mass and UV-VIS spectral information used for structure assignment are given in Table 1. Standards, wavelength and extinction coefficients used for the quantification of each compound are also indicated in this table. Repeated injections of stan- dards and samples over a range of dilutions were used to determine the linear range of the detector for the peak area measurements.
Reference samples
Phaeodactilum tricornutum was obtained from the Culture Collection of Algae and Protozoa (C.C.A.P., Ambleside, Cumbria, U.K.) and cultured according to the recommended procedure. Ectothiorhodospira sp. was isolated from calcite mats in Bonmati solar salterns (Santa Pola, Valencian Community, Spain) using the serial deep agar shake dilution technique (Pfennig, 1978a) following a medium and growth conditions described previously (Grimalt et al., 1992). Chlorobium limicola was isolated from lake Ciso (Banyoles, Catalonia, Spain) and Chromatium vinosum DSM 185 was kindly supplied by Dr H. van Gemerden (University of Groningen, The Netherlands).
Microscopy
Water aliquots for microscopic examination of about 50-200ml were filtered on site (Whatman G F / F glass fiber, 2.3 mm). Samples were studied on site with a portable microscope. Further examination was performed in the lab using an Axioplan optic microscope (Zeiss, Germany) equipped with Neofluar optics, phase contrast and epifluorescence with blue and green light excitations. Observations on living material were performed within a week of sampling. Samples selected for pigment analysis included those showing maximum population biomass.
Pigment composition in a
RESULTS
Physico-chemical parameters
The water depth profiles (Fig. 1) of temperature, oxygen and sulphide measured during each sampling are only plotted over the first 3 m because little change was apparent in the lower section to 7.5 m. Day samples were collected around solar midday and night samples were taken just before dawn.
The March temperature profile shows stratifica- tion, although the temperature difference between epilimnion and hypolimnion temperature is low, being 1.5°C in the day and nearly zero at night. This situation corresponds to the beginning of the stratification period in which the temperature differ- ences between the shallow and deep water bodies are small. Conversely, the temperatures for the July and September sampling periods show daytime differ- ences between epilimnion and hypolimnion of about 6:C which can decrease to ca 3.5°C in the night (July). These temperatures correspond to a well developed stratification. A specific aspect of Lake Ciso reflected in the March, July and September measurements is the shallow epilimnion (1-2m) which is a consequence of the small surface area vs depth. In this respect, an important difference be- tween July and September is the shallower epilimnion in the former case (0.6-0.8 m) whereas in the latter it is defined over 2 m. Finally, holomixis is apparent from the October profile. The seasonal changes in the four sampling periods are fully consistent with
karstic lake environment 743
the yearly fluctuations observed by others in studies involving higher frequency samplings (Pedros-Alio et al., 1987).
The oxygen profiles in March and July show an oxic epilimnion and an anoxic hypolimnion (Fig. 1). During holomixis the whole water column is anoxic (October). These features are again in agreement with previous observations by many authors (e.g. Pedros- Alio et al., 1987). The anoxia throughout the water column during holomixis defines the distinct physico- chemical characteristics of Lake Ciso. The higher epilimnetic oxygen concentration in March and July reflects intense algal photosynthetic activity. During these two periods the algal activity gives rise to well defined oxygen diel fluctuations involving higher concentrations in the presence of light. The low amounts of oxygen in the epilimnion of the September period result from the formation of a layer of sulphur at the surface which limits the water-atmosphere exchange.
The concentration profiles of sulphide concen- tration profiles show S-type curves during stratifica- tion. These correspond to high values in the hypolimnion (up to 1.4 mmol/l) and very low or zero values in the epilimnion (Fig. 1). The curves are displaced to shallower depths at night, in agreement with the oxygen decrease during the dark. Con- versely, during holomixis there is a rather constant profile, with no major differences between lake sur- face and bottom. These values and seasonal differ- ences are again consistent with the yearly fluctuations
G
EARLY (MARCH)
1 2 3
~v~ I°
°L O
O . . . . 0 1 2
0 1 2 3 DEPTH (m)
STRATIFICATION
REGULAR (JULY)
1 2
1 2 3
1 I t ~ 3 "
LATE (SEPTEMBER)
0
-0 " i 2 3 3
1"41 ~ i
• i 2 DEPTH (m)
HOLOMIXIS
OCTOBER
N D
1 2 3
1 2
1 2 DEPTH (m) DEPTH (m) Fig. 1. Water column profiles of temperature, oxygen and sulphide determined in Lake Ciso in each
sampling period. D and N refer to day and night profile curves, respectively.
744 JOAN VILLANUI'~VA et al.
observed by others in studies involving higher fre- quency sampling (e.g. Pedros-Alio et al., 1987).
It is noteworthy that the sulphide concentration in the deep layers is higher when the lake is stratified than it is during holomixis. This is likely due to a combination of two factors. First, the sulphate reducing activity is highly dependent on ambient temperature, being lower in the cold season during holomixis. Second, as a consequence of the algal development in the oxygenated epilimnion during stratification, part of the organic matter synthesized by this community may reach the hypolimnic area and provide an additional carbon input for oxidation by the sulphate-reducing community. During holomixis, no algae are present, so less or- ganic carbon may arrive to the sulphate-reducing zone and no additional hydrogen sulphide is produced.
Microbial populations
The distributions of the main populations in the lake during the four sampling periods (Fig. 2) are consistent with previous studies of the microbial ecology in the lake (e.g. Guerrero and Mas, 1989; Guerrero et al., 1978, 1985; Pedros-Alio et aL, 1984, 1987). The early stratification stage represented by the March period was reflected in a growing phyto- plankton community in the metalimnion and in the presence of abundant and young populations of anoxygenic phototrophic bacteria (Fig. 2). The phytoplankton community essentially comprised Cryptomonas phaseolus (Cryptophyceae) and the phototrophic bacterial population was represented by an upper layer of purple sulphur bacteria containing mainly Amoebobacter sp., with a minor population of C. minus, and a deeper layer of the green sulphur
March ~ cf. e
.... "~. / ~ Chromatium minus ~ y "a~moebobacter purpureus
1 2
Cryptomonas
0 3
September
Chlorochromatium
x ~ ~ b a c t e r purpureus
omatium minus)~
July
/ Chlorochromatium d ~ ~ amoebobacter purpureus
j Chromatium minus (~ rr"
~ __ Cryptomona%
Ankistrodes~--~
I I 1 2
October
Very low conc. o f
Cryptomonas Chlorochromatium Chromatium
I I L I ! 2 0 l 2
Depth (m) Depth (m)
Fig. 2. Qualitative representation of the main organisms found in Lake Ciso in each sampling period.
5 3
7
12
4
MA
RC
H
150
cm
9/10
JUL
Y 2
0 cm
6
8 15
SE
PT
EM
BE
R
160
cm
OC
TO
BE
R 2
0 cm
12
3
12
6
P
13
,2 i/
12 ~
') 12
t
1 '
,,12
Fig.
3.
HP
LC
tra
ces
(440
nm
) of
the
wat
er c
olum
n pa
rtic
ulat
es w
ith
a hi
gher
pig
men
t co
nten
t in
eac
h sa
mpl
ing
peri
od.
Pea
ks a
re l
abel
led
acco
rdin
g to
com
poun
d ci
tati
ons
in t
ext.
For
ide
ntif
icat
ions
see
Tab
le
1.
IlO
8 O
o~
F," E
_<.
L~
746 JOAN VILLANUEVA et al.
bacteria Chlorobium cf. vibrioforme. Due to the lack of spirilloxanthin (1) and rhodopin (2), the Amoe- bobacter sp. identified can be assigned to A. purpureus which is the only species that contains okenone (3; Eichler and Pfennig, 1988).
The consolidated stratification period represented by the July samples was characterized by maximum phytoplanktonic biomass with a high proportion in a senescent state. The populations of phototrophic bacteria had receded. The algal populations were represented by Ankistrodesmus spp. (green algae) and C. phaseolus. The former constituted a green and thick plate in the epilimnion and the latter was located in the metalimnion. A substantial part of C. phaseolus was found in a state of senescence and was surrounded by high numbers of bacteria and fungi. Below these phytoplankton layers the hy- polimnion contained low numbers of Chromatiurn spp. and the consortium Chlorochromatium (Pfennig, 1989; Caldwell and Tiedje, 1975a, b). The September samples showed a strong depletion, near absence, in the phytoplankton population and a recovery in the phototrophic bacterial community. Ch. minus, A. purpureus and the consortium Chlorochromatium were the dominant species. The surface of the lake was covered with a film comprising sulphur, organic material and neustonic bacteria.
During holomixis (October), the microbial compo- sition was rather uniform throughout the water column, very low concentrations of C. phaseolus, Chlorochromatium consortium, Chromatium spp. and Chlorobium spp. being present.
Pigment composition
Representative HPLC-UV chromatograms (440 nm) characteristic of the water particulates from each sampling period and corresponding to the depths with higher pigment content are shown in Fig. 3. The predominant peak in the March sample corre- sponds to okenone (3), which, together with bacteri- ochlorophyll a (4), is synthesized by some species of purple sulphur bacteria, including C. minus and A. purpureus (Goodwin, 1980; Schmidt, 1978; Eichler and Pfennig, 1988; Steenbergen et al., 1987). Alloxan- thin (5) and chlorophyll a (6) are the other two main components. These, together with chlorophyll c (7) and ~t-carotene (8), are characteristic of Crypto- phyceae (Goodwin, 1980). Hence, the composition of the major pigments in the March samples reflects the presence of purple sulphur bacteria (Ch. minus and A. purpureus) and Cryptophyceae ( C. phaseolus) ident- ified by microscopic examination.
Alloxanthin (5) is the major pigment in the July period, in agreement with the presence of large amounts of C. phaseolus found in this group of samples. As indicated above, the occurrence of chlorophyll a (6), chlorophyll c (7) and or-carotene (8) in the example shown (Fig. 3) reflects the presence of this alga. The second major peak corresponds to lutein/zeaxanthin (9/10). These two compounds, to-
gether with chlorophyll b (11; present as a minor component), occur in Chlorophyceae, and among these Ankistrodesmus spp. (Goodwin, 1980), which is consistent with the predominance of this green algal species in the epilimnion.
The September distribution is dominated by the group of bacteriochlorophylls d (12). The presence of these pigments, and the small peak corresponding to chlorobactene (13), likely reflects the dominance of the Chlorochromatium consortium during this period (Goodwin, 1980; Schmidt et al., 1978). As indicated above, okenone (3) probably reflects the occurrence of A. purpureus and/or C. minus, two species which are also found in the meta- and hypolimnion during late stratification. The October example is dominated by the same pigments, consistent with the occurrence of the same microbial species, although in low numbers.
The composition of solvent extractable-free pig- ments in the underlying sediment (Fig. 4) contains alioxanthin (5) and okenone (3) as major com- ponents. These two components occur within a clus- ter of peaks that corresponds to mixtures of alloxanthin (5)-like and okenone (3)-like compounds. Two of the okenone (3)-like compounds exhibit UV spectra with absorbance maxima at 375 nm, pointing to cis-derivatives of the precursor compound (Zech- meister, 1962; Liaaen-Jensen, 1965). The comparison of the distribution with those obtained from the particulates at 3 m water depth in each sampling period (Fig. 5) shows that there is a higher preser- vation of the carotenoids vs chlorophylls in the sediments. Thus, two major water column pigments, chlorophyll a (6; July) and bacteriochlorophylls d (12; September and October), exhibit a strong depletion in their abundances relative to the carotenoids in the bottom sediment. The same trend is observed for bacteriochlorophyll a (4) in comparison with chloro- phyll a (6) in the water particulates (March) and sediment. The selective preservation of intact caro- tenoids has also been observed in studies of deposi- tional environments such as microbial mats (Villanueva et al., 1994) and is in general agreement with the good carotenoid preservation in these sys- tems (Edmunds and Eglinton, 1984; Boon et al., 1983; Palmisano et al., 1989), being related to sedimen- tation under anoxic conditions. Research is in pro- gress to determine whether some of these differences are due to incorporation of portions of these pig- ments to the macromolecular organic matter via sulphide bonding.
DISCUSSION
Water column processes
The concentration profiles of the main carotenoids in the water column are shown in Fig. 6. The algal carotenoids [alloxanthin (5), lutein/zeaxanthin (9/10) and ~-carotene (8)] were dominant (individually, up to 1200 #g/l), being distributed in thin layers located
Pigment composition in a karstic lake environment 747
3
5
. . . . ~_._.._A~- • 1 3 18A
Fig. 4. HPLC trace (440 nm) from the top section (0--5 cm) of the bottom sediment. Peaks are labelled according to compound citations in text. For identifications see Table 1.
in the epilimnion or the metalimnion, corresponding to the plates formed by the precursor organisms. These layers were present in significant concen- trations only when the precursor microorganisms exhibited high biomass. Thus, alloxanthin (5) was only found in the periods represented by March and, in particular, July, in agreement with the observed populations of C. phaseolus (Fig. 2) and with the seasonal dynamics of this algal species (Pedros-Alio et al., 1987). In the July period, the day profiles exhibited higher concentrations and shallower depth maxima than the night profiles, which is again in agreement with the diel fluctuation of the algae, A similar seasonal and diel trend was observed for the other cryptomonad carotenoid, ~t-carotene. Lutein/zeaxanthin (9/10) also maximized in July and exhibited higher concentrations in the day, which corresponded to the occurrence of high biomass of the precursor organism, Ankistrodesmus spp.
Conversely, okenone (3) occured in lower concen- trations (up to 140/~g/l), was present in the hy- polimnion and showed more uniform depth profiles and seasonal patterns. This distribution also followed that of the biomass of the source organisms, the purple sulphur bacteria. The concentrations during holomixis were significantly lower than those found in the hypolimnion during stratification (Fig. 6) which is in close agreement with the lower biomass of purple sulphur bacteria in the holomixis period. During stratification okenone (3) generally exhibited minimal concentrations at the surface, maxima at the top of the hypolimnion and a smooth decrease with depth. Another distinct feature concerns the differ- ences between the day and night profiles, the latter
tending to show higher concentrations and a displace- ment towards shallower depths, both profiles tending to follow the H2S concentration changes. This is in agreement with the motile character of Chromati- aceae that can move in search of their H2S optimum (Sorokin, 1970; Pedros-Alio and Sala, 1990). This type of diel vertical movements has also been ob- served in microbial mats (Jorgensen et aL, 1983; Revsbech et al., 1983).
The concentration gradients of chlorophyll a (6) and the bacteriochlorophylls (4 and 12) are displayed in Fig. 7. Chlorophylls b and c (11 and 7, respectively) were only found in significant amounts during the July sampling and their concentration profiles are compared in Fig. 8 with the chlorophyll a (6) profile for the same period. They showed higher concen- trations and shallower depth maxima during the day period. These distributions paralleled those of lutein/zeaxanthin (9/10) and alloxanthin (5), respect- ively, which is in agreement with the common algal precursors, Ankistrodesmus spp. and C. phaseolus, respectively. Conversely, the profile for chlorophyll a (6) showed higher concentrations during the night. There is no obvious explanation for this discrepancy which could be due to a sampling artifact (i.e. the chlorophyll a (6) maximum was between the sampled depths 0.2 and 0.4 m and was missed).
The profiles for bacteriochlorophyll a (4; Fig. 7) were very similar to those for okenone (3; Fig. 6) which is in accordance with the common source of these two pigments. The series of bacteriochloro- phylls d 02) was present in higher abundance and, during stratification, always showed concentration profiles maximizing a few centimeters (ca 10-20 cm)
• 3 5
MA
RC
H
¸SE
PTE
MB
ER
JUL
Y
23
3
OU
FOB
FIi
< Z >,
e~
12
12 1
2
I1
12
Fig.
5. H
PL
C t
race
s (4
40 n
m)
of t
he 3
m d
epth
wat
er p
arti
cula
tes
coll
ecte
d du
ring
the
day
. P
eaks
are
labe
lled
acco
rdin
g to
com
poun
d ci
tati
ons
in t
ext.
For
ide
ntif
icat
ions
se
e T
able
1.
Pigment composition in a karstic lake environment 749
EARLY t~aARCI-I)
0 1 2
0 1 2
200
N ~ . 0 ! 2
DEPTH (m)
STRATIFICATION R E G ~ (JULY)
0 1 2
1200
3 0 1 2 DEPTH (m)
LATE (SEYrEMBES 14(I
o I 2
hi
DEPTH (m)
HOLOMIXIS OCTOBER
13
DFFrtt (m)
Fig. 6. Water column profiles of the main carotenoids in each sampling period. D and N refer to day and night profile curves, respectively. Missing boxes correspond to concentrations below detection limit. Units
in/t g/l.
below the bacteriochlorophyll a (4) maxima (Fig. 7). This again reflects the biomass distribution of the precursor organisms, Chlorobiaceae (or the Chlorochromatium consortium), which typically maximizes under more reducing conditions (lower water column depths) than does the purple sulphur bacterial community (Pfennig, 1978b). The pigment composition of these organisms is adapted to light harvesting under conditions of shading by algae or by purple sulphur bacteria (Abella et al., 1980), or by high contents of humic substances (Parkin and Brock, 1980).
In short, the general trend during the stratification period was a distribution of algal pigments [chloro- phylls a (6), b (11) and c (7), alloxanthin (5) and lutein/zeaxanthin (9/10); Figs 6 and 7] restricted to the oxic epilimnion, with almost undetectable concen- trations in the anoxic hypolimnion. On the other hand, the pigments of the photosynthetic bacteria [okenone (3), bacteriochlorophylls a (4) and d (12); Figs 6 and 7] were found in the hypolimnion, includ- ing the depths below the photic zone where photosyn- thesis is not possible (Guerrero et aL, 1985). Thus, the pigments formed within the anoxic bottom waters were better preserved as intact components than the pigments present in the upper layers.
The main transformation product pigments in the water column were chlorophyll a (6) derivatives:
phaeophytin a (14), pyrophaeophytin a (15) and pyrophaeophorbide a (16) (Fig. 8). They co-occurred with chlorophyll a (6), being found in higher concen- trations in July, when a senescent bloom of cryp- tomonae dominated the epilimnion populations. One, pyrophaeophytin a (15), was found in concentrations (up to 1200 pg/1) that are higher than those of the precursor chlorophyll a (up to 900/~g/1). No py- rochlorophyll a, phaeophorbide a or steryl chlorin esters were found. The profiles of the products were essentially coincident with that of chlorophyll a (6), the concentrations in the hypolimnion being minimal. Since zooplankton activity is minimal in the lake, this distribution, and the high concentrations of the de- rivatives relative to the chlorophyll a (6) content, indicate that the transformations involving loss of Mg and, subsequently, loss of the C-132 carbomethy- oxyl and phytyl substituents, are rapid and result from algal senescence. In contrast, no bacteriochloro- phyll derivatives were identified in the water column.
Transport to sediments and post-depositional trans- formations
The carotenoids okenone (3) and alloxanthin (5) are the major components and occur in high concen- trations (6 and 4 mg/g, respectively) in the top sedi- ment (Fig. 9). The high sedimentary concentration of okenone (3) and relatively low level in the water
OG 22/3-5~AA
750 JOAN VILLANUEVA et al.
90(
=
>.
12(
EARLY (MARCH)
N D
1 2 3
STRATIFICATION
P~GULA~ (JULY)
0 1 2
t A ~ ( S ~ . ~ F ~ )
. . . . . N
1 2 3
HOLOMIXIS OCTOBER
3 0 1 2 3
o,:J / " i
DEPTH (m)
2O
1 2 3 1 2 DEPTH (m) DEPTH (m)
~ D
0 1 2 3 DEPTH (m)
Fig. 7. Water column profiles of the main chlorophylls and bacteriochlorophylls in each sampling period. D and N refer to day and night profile curves, respectively. Missing boxes correspond to concentrations
below detection limit. Units in #g/1.
column (up to 140 #g/l; Fig. 6) contrasts with the situation for alloxanthin (5) and also for lutein/zeaxanthin (9/10), both having high water column values (up to 1200 pg/l and 700/lg/1, respect- ively). However, the occurrences of alloxanthin (5) and lutein/zeaxanthin (9/10) in the water column showed a strong seasonal dependence whereas
okenone (3) was more uniformly distributed over the year. This may reflect the importance of accumulated sedimentation effects for the incorporation of the pigments into the sedimentary record. In this context, a high biomass of Cryptomonads [the source of alloxanthin (5)] has only been apparent since 1985 (Gasol et al., 1990) whereas the presence of
9 O 0
Or . ~ - ' . . 0 1 2
REGULAR STRATIFICATION (JULY)
0 i 2 3 0 1
0 1 2 DEPTH (m)
iot , DEPTH (m)
!,
D
i 2 DEPTH (m)
Fig. 8. Water column profiles of the main chlorophylls and chlorophyll derivatives in July, 1992. D and N refer to day and night profile curves, respectively. Units in/~g/l.
Pigment composition in a karstic lake environment 751
CHLOROPHYLL a 450,
o $ lo
BACTERIOCHLOROPHYLL a
0 10 DEPTH (cm)
o
l
BOTI'OM SEDIMENT ALLOXANTHIN a-CAROTENE
1 PHAEOPHYTIN a
BACTERIOPHAEOPHYTIN a
0 $ 10 DEPTH (~m)
141: PYROPItAEOPItYTIN a
PYROBACTERIOPHAEOPH YTIN a
DEPTH (cm)
Fig. 9. Depth concentration profiles of the major carotenoids and chlorophyll bottom sediments. Units in/~g/g.
B'CAROTENE 700~
Ol . o 5 10
DEP'm 0:~)
pigments identified in
l
Chromatiaceae is a uniform feature of the lake (Guerrero and Mas, 1989; Guerrero et al., 1987).
These differences in sedimentary preservation may also reflect the distinct water column distribution of algal and bacterial carotenoids, okenone (3) mainly occurring in the anoxic hypolimnion and alloxanthin (5) and lutein/zeaxanthin (9/10) in the oxic epilimnion or suboxic epilimnion/metalimnion (Fig. 6) where photochemical or microbiological degradation may be enhanced. Studies on fossil pigments of meromic- tic lake ecosystems have shown that intact caroten- oids from sulphur phototrophic bacteria can be found in sedimentary records spanning more than 8000 yr (Brown and Mclntosh, 1987).
In the present case, the difference between alloxan- thin (5) and okenone (3) cannot be explained in terms of lower sedimentation rates of the remains of the precursor organisms because the Cryptomonads have been observed to exhibit higher sinking speeds than the purple sulphur bacteria (Guerrero and Mas, 1989). Furthermore, these differences cannot be ex- plained simply as a result of differences in chemical stability. The two triple bonds and two non-aromatic rings of alloxanthin (5) would suggest a higher reac- tivity than for okenone (3). However, ~t-carotene (13), whose structure would suggest a similar or lower reactivity than for okenone (13), also showed the same trend as alloxanthin (5) in comparing its distri- bution in the water column and sediments.
Chlorophyll a (6) and bacteriochlorophyll a (4) are the only chlorophylls found in the sediments. Bacte-
riochlorophylls d (12) were not identified. As in the case of carotenoids, the comparison between the water column and sediment concentrations gives in- formation on the preservation of these pigments. Thus, chlorophyll a (6) showed concentration max- ima of up to 900/~g/l (Fig. 7) in the water column and about 450 #g/g (Fig. 9) in the sediment (3200/~g/g when the chlorophyll a derivatives are included). This situation is in contrast with that for bacteriochloro- phyll a (4) which had a lower concentration in the water column (180/~g/l) than in the sediment, 300/zg/g (4700#g/g including the bacteriochloro- phyll a derivatives). Hence, the differences in preser- vation are analogous to those for the carotenoids, chlorophyll a (6) being produced in the oxic layer and bacteriochlorophyll a (4) in the anoxic zone.
As indicated above, chlorophyll a derivatives were abundant in the water column whereas bacteri- ochlorophyll a derivatives were not present. In the sediment, however, both chlorophyll a (6) and bacteriochlorophyll a (4) derivatives [including phaeophytin a (14), pyrophaeophytin a (15), bacte- riophaeophytin a (17) and pyrobacteriophaeophytin a (18)] occur. This indicates a transformation path- way in the sediment that, as in the water column, involves loss of Mg and subsequent loss of the C-13 2 carbomethoxyl substituent. The differences between the two sets of transformation products suggest different stages in the transformation pathways. Thus, the high concentration of bacteriophaeophytin a (17) vs pyrobacteriophaeophytin a (18) suggests
752 :' JOAN VILI.,ANUEVA et aL
/
14
~ 02Me
/ /
-Phytyl """ O
1G Fig. 10. Observed transformation pathway of chlorophyll a in Ciso Lake. The same transformation
pathway has been observed for bacteriochlorophyll a.
Str
uct
ure
s ci
ted
in
tex
t 0
0t
3
1 C
H3
0
2 H
O
OH
5 •
i O
H
R =
Et
R ~
- Vin
yl
HO
7
O
.°°-
O
4
....o~x'~'
~N
-" -\N~
Fig,
lla
. St
ruct
ures
1-
12.
11
o..°
O
0
R 8
~o
HO
• ,~
o ,>
u R2
R I ~ Et, Pr, iB
n
R 2 -- Me Et
12
"-4
Str
uct
ure
s ci
ted
in
text
,~OH
.°..
...°
°..-
0~--
-~
0 0
' O
Me
0
14
16
15
O
O
O
"°
°'
~O
~ "'
'' ...
° O
O
--'~
OM
e O
~
O
O"
XO
Phyt
y 1 0
18
20
"-~O
Me
22
17
19
21
Fig.
li
b.
Stru
ctur
es
13-2
2.
X
< t" Z t~ ta
Pigment composition in
that the bacteriochlorophyU a series is at an earlier stage than in the chlorophyll a series, in keeping with the similar amounts of phaeophytin a (14) and pyrophaeophytin a (15). In the context of the scheme proposed by Keely et al. (1990) for the transformation of chlorophyll pigments in another lake environment, the transformations in Lake Ciso involve a simpler pathway (Fig. 10). Accordingly, in the absence of zooplankton grazing (e.g. Head and Harris, 1992) or algae containing high abun- dances of the enzyme chlorophyllase, i.e. diatoms (Jeffrey and Hallegraeff, 1987), the loss of the phytyl chain occurs after the loss of the C-132 carbomethoxy substituent.
The depth concentration profiles of the major pigments in the sediments (0-5, 5-10 and 10-25 cm sections) generally show (Fig. 9) a dramatic decrease irrespective of the particular component. The total pigment contents drop from 18 to 6.5 and to 0.2 mg/g for the 0-5, 5-10 and 10-25 cm intervals, respectively. In this respect, several carotenoid transformation products such as loliolide, iso-loliolide and dihy- droactinidiolide (Repeta, 1989; Klok et al., 1984) were minor sedimentary components. However, a possible explanation for this major depth-related decrease may be that there is a progressive removal of both types of pigments by way of incorporation via sulphide bridges into macromolecular fractions, a process known to occur for both carotenoids and chlorophyll-derived components (Schaeffer et al., 1993; Sinninghe-Damst6 et al., 1993). Such a process may also be a factor in altering the abundances of carotenoids vs chlorins. Studies relating to these aspects are in progress.
CONCLUSIONS
During stratification and holomixis the distri- butions of carotenoids and chlorophylls in the water column particulates of Lake Ciso closely followed the population dynamics of the photosynthetic organ- isms. Thus, the major pigments, lutein/zeaxanthin (9/10) and alloxanthin (5), were found in the epi- limnion and their concentration changes were directly related to the abundances of their algal precursors, Ankistrodesmus spp. and C. phaseolus, respectively. Likewise, okenone (3) and bacteriochlorophyll a (4) were distributed in the hypolimnion and corre- sponded closely to the occurrence of purple sulphur bacteria, particularly A. purpureus and Ch. minus. The same was true for the bacteriochlorophylls d (12), which paralleled the abundances of Chl. vibrioides and the Chlorochromatium consortium. The close interdependence between the temporal and spatial trends of these pigments and their precursor organ- isms was also reflected in the vertical distributions in the water column and the diel fluctuations. No transformation derivatives of these bacterial com- pounds were found in the water column.
a karstic lake environment 755
The chlorophyll a (6) concentrations in the water column also corresponded closely to the abundances of the precursor algae, showing layers situated in the epilimnion which maximize when stratification was coincident with warmer ambient temperatures. How- ever, in contrast with the bacteriochlorophylls, vari- ous chlorophyll a transformation products were already present in the water column. These products, phaeophytin a (14), pyrophaeophytin a (15) and pyrophaeophorbide a (16), exhibited concentration profiles which paralleled the chlorophyll a (6) distri- bution. This feature, and the lack of the derivatives in the hypolimnion, indicates that loss of Mg and, subsequently, loss of C- 132 carbomethoxyl and phytyl substituents, occur rapidly during algal senescence. Analogous bacteriochlorophyll a (4) transformation products (17, 18) also resulting from loss of Mg and, subsequently, of the C-132 carbomethoxyl group only occurred in the sediment. Differences in the relative abundances of the sedimentary chlorophyll a and bacteriochlorophyll a derivatives suggest the latter to be at an earlier stage in the transformation pathway. This is consistent with the occurrence of bacteri- ochlorophyll a (4) in the anoxic water column layers where none of its derivatives were present. In the absence or very low abundance of pyrochlorophyll a (19), pyrobacteriochlorophyll a (20), phaeophorbide a (21) and bacteriophaeophorbide a (22), the occur- rence of phaeophytin and pyrophaeophytin com- ponents points to a transformation mechanism showing that, in the absence of zooplankton grazing or high abundances of the enzyme chlorophyllase, loss of the phytyl side chain occurs only after loss of the C-132 carbomethoxyl group in this lake environment.
Another distinct feature of the sedimentary pig- ment composition of Lake Ciso is the good preser- vation of certain carotenoids, both of algal (alloxanthin, 5) or purple sulphur bacterial (okenone, 3) origin. However, their relative concentrations in the sediment markedly contrast with those found in the water column, showing an enrichment of the former with respect to the latter. This difference may be a consequence of a more uniform concentration of okenone (3) in the water column particulates over the year, which may be reflected in a higher total sedi- mentation flux. However, it may also relate to a higher degree of preservation of the phototrophic bacterial vs algal carotenoids in the anoxic and oxic lake waters, respectively. In the absence of light and oxygen, okenone (3) appears to be better preserved. A similar difference between water column and sedi- ment composition was observed in the relative con- tents of chlorophyll a (6) and bacteriochlorophyll a (4). Bacteriochlorophylls d (12) and their transform- ation products therefrom were not identified in the sediment.
In addition to the carotenoids biosynthesized orig- inally, other sedimentary components with the same molecular weight and similar spectra, but eluting as
756 JOAN 'VILLANUEVA et al.
other resolved peaks in the 440 nm H P L C traces, were also found. It is likely tha t some of these componen t s are cis isomers of the precursor carotenoids.
Acknowledgements--We are grateful to Mr Jordi Lopez (Department of Environmental Chemistry. C.I.D.-C.S.I.C.) for his useful comments. Thanks are also due to Drs J. de Leeuw and R. Ocampo for useful suggestions and review. The present work has been financed in part by SCIENCE Project No. SC1-CT91-0736 (EEC) and PGC Project PB90- 0094 (DGICYT; MEC), J. V. and R. d. W. have been supported by grants from the Spanish Ministry of Edu- cation and the European Environmental Research Organiz- ation (E.E.R.O.), respectively. We thank the N.E.R.C. for HPLC-MS facilities (GR3/6619) and the Royal Society and the NERC (GR9/548) for PDA facilities.
REFERENCES
Abaychi J. K. and Riley J. P. (1979) The determination of phytoplankton pigments by HPLC. Anal. Chim. Acta 107, 1-11.
Abella C., Montesinos E. and Guerrero R. (1980) Field studies on the competition between purple and green sulfur bacteria for available light. Dev. Hydrobiol. 3, 173-181.
Boon J. J., Hines H., Burlingame A. L., Klok J., Rijpstra W. I. C., de Leeuw J. W., Edmunds K. E. and Eglinton G. (1983) Organic geochemical studies of Solar Lake laminated cyanobacterial mats. In Advances in Organic Geochemistry 1981 (Edited by Bjoroy M.), pp. 207-227. Wiley, Chichester.
Brown S. R. (1969) Absorption coefficients of chlorophyll derivatives. J. Fish. Res. B. Can. 25, 523-540.
Brown S. R. and Mclntosh H. J. (1987) The history of sulphur phototrophs in a meromictic lake ecosystem. Acta Academ. Aboen. 47, 83-95.
Caldwell D. E. and Tiedje J. M. (1975a) A morphological study of anaerobic bacteria from the hypolimnia of two Michigan Lakes. Can. J. Microbiol. 21, 362-376.
Caldwell D. E. and Tiedje J. M. (1975b) The structure of anaerobic communities in the hypolimnia of several Michigan lakes. Can. J. Microbiol. 21, 377-385.
Davies T. W. (1976) Carotenoids. In Chemistry and Biochemistry of Plant Pigments (Edited by Goodwin T. W.), Vol. 2, pp. 36-56. Academic Press, Washington, D.C.
Edmunds K. L. H. and Eglinton G. (1984) Microbial lipids and carotenoids and their early diagenesis in the Solar Lake laminated microbial mat sequence. In Microbial Mats: Stromatolites (Edited by Cohen, Y., Castenholz R. W. and Halverson H. O.), pp. 343-389. Liss, New York.
Eichler B. and Pfennig N. (1988) A new purple sulfur bacterium from stratified freshwater lakes, Amoebobacter purpureus sp. nov. Arch. Microbiol. 149, 395-400.
Foppen F. H. (1971) Tables for the identification of caroten- oid pigments. Chromatogr. Rev. 14, 133-298.
Gasol J. M., Mas J., Pedros-Alio C. and Guerrero R. (1990) Ecologia microbiana y limnologia en la laguna Ciso: 1976-1989. Scientia Gerundensis 16/2, 155-178.
van Gemerden H., Montesinos E., Mas J. and Guerrero R. (1985) Diel cycle of metabolism of phototrophic purple sulfur bacteria in Lake Ciso (Spain). Limnol. Oceanogr. 30, 932-943.
Goodwin T. W. (1980) The Biochemistry o f the Carotenoids Vol. 1, Plants, 2nd Edition. Chapman and Hall, London.
Grimalt J. O., de Wit R., Teixidor P. and Albaiges J. (1992) Lipid biogeochemistry of Phormidium and Microcoleus mats. Org. Geochem. 26, 509-530.
Guerrero R. and Mas J. (1989) Multilayered microbial communities in aquatic ecosystems: growth and loss factors. In Physiological Ecology of Benthic Microbial Communities (Edited by Cohen Y. and Rosenbcrg E.), pp. 37-51. American Society for Microbiology, Washington D.C.
Guerrero R., Abella C. and Miracle M. R. (1978) Spatial and temporal distribution of bacteria in a meromictic karstic lake basin: relationships with physicochemical parameters and zooplankton. Int. Ver. Theor. Angew. Limnol. Verb. 20, 2264-2271.
Guerrero R., Montesinos E., Pedros-Alio C., Esteve I., Mas J., van Gemerden H., Hofman P. A. G. and Bakker J. F. (1985) Phototrophic sulfur bacteria in two Spanish lakes: vertical distribution and limiting factors. Limnol. Oceanogr. 30, 919-931.
Guerrero R., Pedr6s-Ali6 C., Esteve I. and Mas J. (1987) Communities of phototrophic sulfur bacteria in lakes of the Spanish Mediterranean region. Acta Academiae Aboensis 47, 125-151.
Head E. J. H. and Harris L. R. (1992) Chlorophyll and carotenoid transformation and destruction by Calanus spp. grazing on diatoms. Mar. Ecol. Prog. Ser. 86, 229-238.
Jeffrey S. W. (1968) Properties of two spectrally different components in chlorophyll c preparations. Biochim. Bio- phys. Acta 177, 456-467.
Jeffrey S. W. and Hallegraeff G. M. (1987) Chlorophyllase distribution in ten classes of phytoplankton: a problem for chlorophyU analysis. Mar. Ecol. Prog. Ser. 35, 293-304.
Jorgensen B. B., Revsbech N. P. and Cohen Y. (1983) Photosynthesis and structure of benthic microbial mats: Microelectrode and SEM studies of four cyanobacterial communities. Limnol. Oceanogr. 28, 1075-1093.
Keely B. J., Prowse W. G. and Maxwell J. R. (1990) The Treibs hypothesis: an evaluation based on structural studies. Energy and Fuels 4, 628-634.
Klok J., Baas M., Cox H. C., de Leeuw J. W. and Schenck P. A. (1984) Loliolides and dihydroactinidiolide in a recent marine sediment probably indicate a major trans- formation pathway of carotenoids. Tetrahedron Lett. 25, 5577-5580.
Liaaen-Jensen S. (1965) Bacterial carotenoids. XVIII. Aryl- carotenes from Phaeobium. Acta Chem. Scand. 19, 1025-1030.
Mantoura R. F. C. and Llewellyn C. A. (1983) The rapid determination of algal chlorophyll and carotenoid pig- ments and their breakdown products in natural waters by reverse-phase high-performance liquid chromatography. Anal. Chim. Acta. 151, 297-314.
Palmisano A. C., Cronin S. E., D'Amelio E. D., Mufioz E. and des Marais D. J. (1989) Distribution and survival of lipophilic pigments in a laminated microbial mat com- munity near Guerrero Negro, Mexico. In Microbial Mats: Physiological Ecology of Benthic Microbial Com- munities (Edited by Cohen Y. and Rosemberg E.), pp. 138-152. American Society for Microbiology, Washington, D.C.
Parkin T. B. and Brock T. D. (1980) The effects of light quality on the growth of photosynthetic bacteria in lakes. Arch. Microbiol. 125, 19-27.
Pedros-Alio C. and Sala M. (1990) Microdistribution and d i e l vertical migration of flagellated vs. gas vacuolate
purple sulfur bacteria in a stratified water body. Limnol. Oceanogr. 35, 1637-1644.
Pedros-Alio C., Abella C. and Guerrero R. (1984) Influence of solar radiation, water flux and competition on biomass phototrophic bacteria in Lake Ciso, Spain. Int. Ver. Theor. Angew. Limnol. Verh. 22, 1097-1101.
Pedros-Alio C., Gasol J. M. and Guerrero R. (1987) On the ecology of a Cryptomonas phaseolus population forming a metalimnetic bloom in Lake Ciso, Spain: annual
Pigment composition in a karstic lake environment 757
distribution and loss factors. Limnol. Oceanogr. 32, 285-298.
Pfennig N. (1978a) Rhodocyclus purpureus gen. nov. and spec. nov. a ring-shaped, vitamin Bz2 requiring member of the family Rhodospirillaceae. Int. J. Syst. Bacteriol. 2,8, 283-288.
Pfennig N. (1978b) General physiology and ecology of photosynthetic bacteria. In The Photosynthetic Bacteria (Edited by Clayton R. K. and Sistrom W. R.), pp. 3-18. Plenum Publishing Corporation, New York.
Pfennig N. (1989) Addendum to the green sulfur bacteria. In Bergeys Manual of Systematic Bacteriology (Edited by Staley J. T., Bryant M. P., Pfennig N. and Holt J. G.), Vol. 3, pp. 1696-1697. Williams and Wilkins, Baltimore.
Repeta D. J. (1989) Carotenoid diagenesis in recent marine sediments: II. Degradation of fucoxanthin to loliolide. Geochim. Cosmochim. Acta 53, 699-707.
Repeta D. J. (1993) A high resolution historical record of Holocene anoxygenic primary production in the Black Sea. Geochim. Cosmochim. Acta 57, 4337-4342.
Repeta D. J., Simpson D. J., Jorgensen B. B. and Jannasch H. W. (1989) Evidence for anoxygenic photosynthesis from the distribution of bacteriochlorophylls in the Black Sea. Nature 342, 69-72.
Revsbech N. P., Jorgensen B. B., Blackburn T. H. and Cohen Y. (1983) Microelectrode studies in the photosyn- thesis and O~, H2S, and pH profiles of a microbial mat. Limnol. Oceanogr. 28, 1062-1074.
Schaeffer P., Ocampo R., Callot H. J. and Albrecht P.
(1993) Extraction of bound porphyrins from sulphur-rich sediments and their use for reconstruction of palaeoenvi- ronments. Nature 364, 133-136.
Schmidt K. (1978) Biosynthesis of carotenoids. In The Photosynthetic Bacteria (Edited by Clayton R. K. and Sistrom W. R.), pp. 729-750. Plenum Press, New York.
Sinninghe Damst6 J. S., Wakeham S. G., Kohnen M. E. L., Hayes J. M. and de Leeuw J. W. (1993) A 6000-year sedimentary molecular record of chemocline excursions in the Black Sea. Nature 362, 827-829.
Sorokin Y. I. (1970) Interrelations between sulphur and carbon turnover in meromictic lakes. Arch. Hydrobiol. 66, 391-446.
Steenbergen C. L. M. and Korthals H. J. (1982) Distribution of phototrophic microorganisms in the anaerobic and microaerophilic strata of Lake Vechten (The Netherlands). Pigment analysis and role in primary pro- duction. Limnol. Oceanogr. 27, 883-895.
Steenbergen C. L. M., Korthals H. J. and van Nes M. (1987) Ecological observations on phototrophic sulfur bacteria and the role of these bacteria in the sulfur cycle of monomictic Lake Vechten (The Netherlands). Acta Academ. Aboen. 47, 97-115.
Villanueva J., Crimalt J. O., de Wit R., Keely B. J. and Maxwell J. R. (1994) Chlorophyll and carotenoid pig- ments in solar saltern microbial mats. Geochim. Cos- mochim. Acta In press.
Zechmeister L. (1962) Cis-trans Isomeric Carotenoids, Vitamins A and Arylpolienes. Springer-Verlag, Wien.
