plant pigments as biomarkers of high-molecular-weight dissolved organic carbon

Plant Pigments as Biomarkers of High-Molecular-Weight Dissolved Organic CarbonAuthor(s): Thomas S. Bianchi, Corey Lambert, Peter H. Santschi, M. Baskaran and Laodong GuoSource: Limnology and Oceanography, Vol. 40, No. 2 (Mar., 1995), pp. 422-428Published by: American Society of Limnology and OceanographyStable URL: http://www.jstor.org/stable/2838235 .

Accessed: 14/06/2014 18:19

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .http://www.jstor.org/page/info/about/policies/terms.jsp

.JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range ofcontent in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new formsof scholarship. For more information about JSTOR, please contact [email protected].

.

American Society of Limnology and Oceanography is collaborating with JSTOR to digitize, preserve andextend access to Limnology and Oceanography.

http://www.jstor.org

This content downloaded from 185.44.77.28 on Sat, 14 Jun 2014 18:19:12 PMAll use subject to JSTOR Terms and Conditions

http://www.jstor.org/action/showPublisher?publisherCode=limnoc

http://www.jstor.org/stable/2838235?origin=JSTOR-pdf

http://www.jstor.org/page/info/about/policies/terms.jsp


422 Notes

Limnol. Oceanogr., 40(2), 1995, 422-428 ? 1995, by the Amencan Society of Limnology and Oceanography, Inc.

Plant pigments as biomarkers of high-molecular-weight dissolved organic carbon

Abstract-We used plant pigments as tracers of high- molecular-weight dissolved organic C (HMW DOC) (<0.2 ,um and > 1,000 Da). Water samples were collected from four stations along a transect on cruises in June 1992 and January 1993 in the Gulf of Mexico. Samples were also collected from three stations on the continental shelf of Cape Hatteras for comparison. Chlorophylls a and b were at detectable levels in HMW DOC; concentrations of the carotenoids zeaxanthin and fucoxanthin in HMW DOC indicated that cyanobacteria, prochlorophytes, and diatoms contributed to the total HMW DOC pool across the continental margin in the gulf. The extremely low concentrations of pheopigments (except for chlorophyllide) in particulate organic C and HMW DOC suggested that direct exudation from phytoplankton and sloppy feeding by zooplankton were the major mechanisms of release. However, significant and selective pigment loss, due to light and thermal degradation in the reservoir, did occur in the absence of any cooling or light-shielding precautions. This study demonstrated that pigments can be used to acquire qualitative information on the sources of HMW DOC and to gain insight on its relative age -based on pigment turnover.

Plant pigments have been widely used as biomarkers of particulate organic C (POC) sources in freshwater and marine ecosystems (see Millie et al. 1993). Chloropig- ments such as chlorophyll a and pheopigments are com- monly used to estimate phytoplankton biomass and grazing intensity, respectively (Shuman and Lorenzen 1975; Welschmeyer and Lorenzen 1985; Bianchi et al. 1991). Carotenoids, on the other hand, are more class specific and have been used to trace specific sources of particulate carbon (Wright and Jeffrey 1987; Leavitt and Carpenter 1990; Wright et al. 1991).

The presence of pigments in the dissolved organic C (DOC) pool may also be useful in tracing sources. If present, these pigments may affect the cycling and transformation of DOC in the ocean. For example, carotenoids such as zeaxanthin can also function as photoprotectants for the living algal cell by quenching singlet oxygen which precludes peroxidation reactions (Foote et al. 1970; Foote 1974). The photochemical degradation of synthetic polymers can be reduced significantly by adding pigments (Ranby and Rabek 1975). Despite an extensive literature on the use of plant pigments as biomarkers, the emphasis has been on POC with virtually no published accounts of their application to trace sources of DOC.

The quantification and apparent age of DOC (using 14C) has received considerable attention recently, but little progress has been made on the transformation and cycling of DOC within the ocean (Hedges 1992; Lee and Henrichs 1993; Hedges et al. 1993). Plant pigments which tend to be nonpolar in nature are not likely to be effective tracers

of low-molecular-weight DOC (LMW DOC) (< 1,000 Da); however, we will show that the larger high-molecular- weight (HMW) fraction within the total DOC pool (> 1,000 Da to <0.2 ,um) has detectable pigment concentrations. A significant fraction of the total DOC in the ocean is usually represented by HMW material (Sharp 1973; Williams and Druffel 1988; Benner et al. 1992).

Pigments found in LMW DOC are likely to be com- plexed in water-soluble proteins or solubilized in deter- gentlike micelles (Harbour and Bolton 1978); the re- maining fraction of pigments in DOC would be com- plexed in hydrophobic cellular materials such as membrane materials and phytodetritus (Nelson 1993). Thus, although the actual range of molecular weights of chloropigments and carotenoids are generally <1,000 Da, their contribution to LMW DOC is likely to be minimal because of complexation with HMW membranous materials (i.e. thylakoid membranes).

In this study we use plant pigments for the first time to examine the relationship between the major sources of POC (phytoplankton) and the dominant pigments found in the HMW DOC on the continental margin ofthe northwestern Gulf of Mexico. The mass concentration of HMW DOC in this region of the gulf is > 1 mg liter-' which represents, on average, -45% of the total DOC (Guo et al. 1994). Our major objectives are to determine whether plant pigments can be used as effective biomarkers to trace the dominant sources of HMW DOC in the Gulf of Mexico, to compare concentrations and composition of pigments in POC and HMW DOC collected from the Gulf of Mexico and the continental shelf off Cape Hat- teras, and to examine for artifactual effects that the ultrafiltration process may have on pigment concentrations in collected materials.

Water samples were collected for pigment analyses aboard the RV Gyre in June 1992 and January 1993 in the northwestern gulf; samples were collected at four stations (Sta. 1, 28?45'N, 94?59'W; Sta. 4, 27?30'N, 95?50'W; Sta. 5, 26040'N, 95000'W; Sta. 6, 27043'N, 95000'W). Samples were also collected at three stations (Sta. 5, 36004'N, 74043'W; Sta. 30, 36003'N, 74'46'W; Sta. 113, 36005'N, 75021'W) off Cape Hatteras aboard the RV Gyre in May 1993. Samples analyzed for this study covered a wide range of oceanographic conditions (Table 1).

Water samples were drawn from 30-liter Niskin bottles tripped at surface and bottom depths at each station. Aliquots of 500-2,000 ml were passed through Alltech anodisc filters (0.2-,um pore size) for collections of POC to be analyzed for pigments. Filters were then placed into polypropylene microcentrifuge tubes, stored in liquid nitrogen, and brought back to the lab for pigment analysis.

In the present work, DOC was fractionated according



Notes 423

Table 1. Hydrographic data.

HMW DOC

Depth Temp. Salinity DOC kDa)- SP P04 NO3 5i02 (ml Station (m) Location (OC) (37w) (AM) (AM) liter-1) (AM) liter-1)

92G7-St.1 2 28045.2'N, 94059.9'W 29.0 30.566 131 69 432 0.02 0.1 4.2 4.62 92G7-St.6 2 27043.7'N, 95000.0'W 30.0 35.639 88 37 62 0.06 0.4 1.7 4.81 92G7-St.5 2 26040.0'N, 94059.9'W 29.7 35.960 83 42 68 0.01 0.8 1.8 4.56 93Gl-St.1 2 28044.8'N, 94043.7'W 19.0 35.032 86 36 962 0.12 0.1 1.8 5.36 93Gl-St.4 2 27030.0'N, 95050.9'W 23.4 35.908 82 29 70 0.02 0.1 3.7 4.90 93Gl-St.4 355 27030.0'N, 95050.9'W 9.5 35.159 47 - 99 1.92 29.2 17.7 2.54 93Gl-St.5 2 26040.1'N, 95000.2'W 23.8 36.406 70 28 117 0.06 0.2 9.2 4.82 93Gl-St.5 1,600 26040.1'N, 95000.2'W 4.3 34.975 45 14 95 1.60 23.5 27.7 4.95 93G7-St.2 2 36003.8'N, 74046.3'W 16.0 30.846 90 40 192 0.20 0.1 0.4 6.86 93G7-St.16 24 36005.2'N, 75021.4'W 9.0 32.505 90 44 186 0.26 0.1 0.7 7.52 93G7-St.1 750 36004.6'N, 74043.8'W 4.5 34.966 47 20 55 1.22 18.9 11.7 5.70 * Suspended particulate matter.

to sizes by the cross-flow ultrafiltration techniques of Guo et al. (1994). Size was calibrated by using compounds of different molecular weights indicating that a 1,000 MW cutoff is only approximate. We used an ultrafilter membrane that had a nominal MW cutoff of 1,000 Da. An Amicon DC 1OL ultrafiltration system was used with the 1,000 K Da spiral-wound polysulfone cartridges (Ami- con, Si ON 1) (Baskaran et al. 1992). Surface seawater samples were pumped directly through an acid-rinsed 0.2- or 0.4-,um pore-size Nuclepore cartridge and subsequently through the 1,000-Da ultrafiltration system. No significant differences were observed between 0.2- and 0.4-,um pore-size filters for any parameters reported here.

In this study, HMW DOC was defined operationally as the fraction between 1,000 Da and 0.2- or 0.4-,um pore size. About 200 liters of seawater were processed, and the concentrated HMW DOC was reduced to -2 liters. The DOC recovery rate for the 1,000 Da ultrafiltration system ranged from 97 to 108% (with an average of 101%) (Guo et al. 1994). The concentrated retentate was then distrib- uted into cryo-containers, placed in a -80?C Ultralow freezer for 24 h, and then lyophilized.

Filters for HPLC analysis, which had been wrapped in aluminum foil and stored in liquid nitrogen at sea, were generally analyzed within 2 weeks of each cruise. Filters were first cut into small pieces, placed into microcentrifuge tubes (1.5 ml) containing 1 ml of 100% acetone and 50 ,ul of canthaxanthin (an internal standard), and extracted according to Bianchi et al. (1995). Preweighed lyophilized HMW colloidal materials (-20-30 mg) were also extracted for pigments with this procedure.

Reverse-phase HPLC analysis was conducted with the method recommended by Wright et al. (1991). A Waters 610 solvent delivery system was coupled with dual-chan- nel detection using a Waters 996 photodiode array detector and a Milton-Roy fluorescence detector with ex- citation at 440 nm and emission at >600 nm. A Rheo- dyne injector was connected via a guard column to a reverse-phase C18 Alltech Adsorbosphere column (5-gm particle size; 250 mm x 4.6 mm i.d.); injection volumes

ranged from 100 to 300 ,ul. Although the Wright et al. (1991) method has been well established for pigment analyses in POC samples, we were also able to achieve sufficient resolution for all dominant pigments in HMW DOC (Fig. 1).

High-purity HPLC standards for Chl a and b were ob- tained from Sigma Co. Fucoxanthin, zeaxanthin, canthaxanthin, and f-carotene standards were provided by Hoffman LaRoche. Pheopigments (pheophorbide, pheo- phytin, chlorophyllide) were quantified and prepared with methods described by Bianchi et al. (1993). All peaks represent methods described by Bianchi et al. (1993). All peaks were quantified with calculated response factors and the Waters Millenium software package.

A simple laboratory experiment was conducted to examine for any artifactual effects that the 0.2-,um prefilter may have on pigment concentration in the HMW DOC during ultrafiltration. This lab experiment required the construction of a two-chambered system (3-liter chambers) connected by PVC pipe (6.5-cm i.d.) which has a Poretics 0.2-,um polycarbonate membrane passing through it. The basic concept was that unfractionated DOC produced by a freshwater culture of phytoplankton (diatoms) in one chamber will diffuse (without filtering effects) into a second chamber that has an equal volume of nanopure water with no phytoplankton culture. A freshwater system was used because the unfractionated DOC can be collected by lyophilization from small volumes of water without the removal of salts. Pigment concentrations in the unfractionated DOC collected by diffusion were then compared with pigments found in the unfractionated DOC that were collected by pumping water from the chamber with the diatom culture through a 0.2-,um pore-size in- line filter (47-mm diam) using a Masterflex peristaltic pump.

The two-chambered system was first calibrated with NaCl as a tracer to determine the time required for adequate diffusion of unfractionated DOC across the membrane; equilibrium was reached in 5 h with 5 g of NaCl. Vigorous mixing was maintained in each of the chambers



424 Notes

3

6

0

45

7

O 15.00 30.00

Retention Time (min)

Fig. 1. Absorbance chromatogram of pigments in HMW DOC collected from station 93Gl-St. 4 in January 1993. The numbered peaks are identified (with known standards) as the following pigments: 1-fucoxanthin; 2-zeaxanthin; 3-canthaxanthin (internal standard); 4-chlorophyll b; 5-chlorophyll a allomer; 6-chlorophyll a; 7-:-carotene.

(using stirrers and stir-bars) to enhance rates of diffusion. A senescent culture of Navicula spp. was used in this experiment to further enhance the rate of unfractionated DOC production, assuming that senescent cultures have a greater amount of cell lysis and decay. Chlorophyllide a was used as an indicator to select a senescent diatom culture; chlorophyllide a is a degradation product of Chl a (phytol chain removed) that can be produced by chlorophyllase, an enzyme associated with senescent diatoms (Jeffrey 1974). As mentioned previously, 100% acetone was used to preclude artifacts that could result in higher concentrations of chlorophyllide a (Jeffrey and Hallegraeff 1987).

This experiment was designed to determine the percent recovery of pigments from prefiltered water within the ultrafiltration reservoir over time. In the latter stages of ultrafiltration, the reservoir holding the accumulating iso- late can warm considerably in addition to being exposed to ambient lighting. Thus, there is a potential for thermal and light degradation of pigments during this process. In fact, it is equally possible that other biochemical constituents of DOC (i.e. amino acids, fatty acids, sugars) may also undergo some thermal degradation during these stages of ultrafiltration.

To avoid the problems of diafiltration, we used freshwater collected from Clear Creek, a stream in southeast Texas (near Houston) for our artifact experiment. Water was collected in May 1994 with plastic carboys and passed through a 0.4-,um pore-size filter. In an attempt to carry out a worst-case scenario, 20 liters of filtered freshwater were recirculated through the ultrafiltration system (DC- 10) using the highest pressure (400 kPa). Recirculated water (2 liters) was collected every hour for 5 h and then lyophilized to dryness, weighed, and extracted for pig-

ment analysis using the aforementioned method. The ambient lighting and duration of time (- 5 h) used in this experiment was similar to that used in processing field samples.

Additional water was collected from the same location in Clear Creek in July 1994 for a follow-up experiment to examine the possibility of reducing the artifactual effects of thermal and light degradation on pigments during ultrafiltration. In this experiment all conditions were as described above, except for packing ice around the entire ultrafiltration chamber, which was also wrapped with lay- ers of black plastic to further insulate and shield the filtrate from ambient lighting.

Fmax was used prior to ANOVA analyses to check for homogeneity of variances (Sokal and Rolhf 1981). A Kruskal-Wallis test was used to test for significant differences between filtered and diffused samples in the prefiltration experiment; one-way ANOVA was used to test for significant effects of time on pigment concentrations in both thermal-light experiments. A simple t-test was used to test for differences between means of pigment concentrations from two field stations. Correlation analyses on POC and HMW DOC pigment samples from the field were performed with a Pearson product moment correlation matrix. Although replicate samples were not available for HMW DOC pigment samples from the field, the reproducibility (C.V.) of selected replicates for each sample type was < 15%.

The presence of pigments in the HMW DOC fraction was not likely to be an artifact of filtration or ultrafiltration for the following reasons. First, the laboratory experiment demonstrated that there was no significant difference (P = 0.05) (after 3 d of equilibration) between Chl a and fucoxanthin concentrations in the unfractionated DOC that diffused across the 0.2-,um pore-size membrane (in the chamber containing no phytoplankton) vs. the unfractionated DOC collected after particles had been filtered out (in the chamber containing diatoms) (Table 2). The appearance of slightly lower concentrations of pigments in the unfractionated DOC (although not significant) may be due to an inadequate amount of time for total equilibration (i.e. 3 d) to occur between the chambers or the production of unfractionated DOC may have been faster than the diffusion rate across the membrane. In any event, the pigment concentrations would have been significantly lower in the diffused unfractionated DOC had there been any filtration effects.

If the filtration process had caused the breakage of cells, the ratio of pigment concentrations in HMW DOC and particulates should be similar (despite some possible frac- tionation), which however, was not the case (Table 2). For example, although the Chl a: fucoxanthin ratio in particulates (diatoms) is 2.1, the ratio is 0.52 in the filtered unfractionated DOC and 0.47 in the diffused unfractionated DOC; thus, there was a significant difference between the pigment ratio in particulate vs. both unfractionated DOC samples, with no significant difference between filtered and diffused unfractionated DOC. The Chl a: fucoxanthin ratio of diatoms (2.1) was within the range found for most other diatom species (Klein 1988; Stauber



Notes 425

Table 2. Concentrations of Chl a and fucoxanthin in particulate and unfractionated DOC collected in a laboratory experiment with diatom cultures. The unfractionated DOC was collected by filtration (47-mm diam, 0.2-Am pore-size) and diffusion (3 d) (?SD, N = 2).

Chl a (,ug g-1) Fucoxanthin (,ug g-1)

Collection Particu- Particu- procedure late DOC late DOC Filtration 1,650+201 0.43?0.15 802?42 0.83?0.23 Diffusion 0.32?0.19 0.68?0.32

and Jeffrey 1988). The same argument can also be made for the field samples, where pigment ratios in HMW DOC were very different from those in the particulates. Thus, if "pigment juices" were being produced (by filtration) in large quantities in HMW DOC, pigment ratios of HMW DOC would have been similar to that found in POC.

It should be noted that the 0.2-Am pore-size, 47-mm- diameter filter used with the vacuum filtration in the lab experiment is different from the actual prefiltration method used in the field, where water is pumped through a 0.2-,um Nuclepore cartridge filter before ultrafiltration. However, we have calculated that the greater surface area of the cartridge used in the field resulted in a pressure gradient that was significantly lower (-500 times) than that used in the lab experiment.

There was a significant (P < 0.05) loss of all dominant pigments during the 5-h ultrafiltration process (Fig. 2A). For example, 93% of Chl a, 88% of A-carotene, 83% of fucoxanthin, and 79% of zeaxanthin were lost over the 5-h period. The ratios of pigments (relative to Chl a) among sampling periods also changed over time with a 64, 69, and 58% change from the initial ratios for Chl a: fucoxanthin, Chl a: zeaxanthin, and Chl a f3-carotene (Fig. 2A). However, in the follow-up experiment, where cooling and light shielding precautions were taken, the loss of pigments was significantly reduced with losses of 63, 39, and 56% for Chl a, A-carotene, and zeaxanthin (Fig. 2B). Changes in the ratios of pigments with time were also reduced significantly with changes of 46 and 41% for Chl a: zeaxanthin, and Chl a: d-carotene. There were no significant differences between the concentrations of Chl a, zeaxanthin, and d-carotene at 2 and 5 h (Fig. 2B).

These experiments demonstrate that the observed concentrations of pigments in HMW DOC from our field samples are likely to be considerably higher than reported-due to artifactual losses. If this is the case, the potential for using pigments as biomarkers in HMW DOC may still remain high. Only minor alterations in our ultrafiltration procedure resulted in as much as a 49% increase in the A-carotene yield. More efficient methods of cooling are certainly needed to produce greater yields than reported here if pigments are to be used as quantitative markers of HMW DOC. The melting and replacement of ice during the 5-h duration of our modified ultrafiltration procedure was inefficient and impractical. Improvements such as a cold-finger mounted in the filtrate reservoir that

20

18 I= Fucoxanthin?

116 -p 7q~~~~~~~~~~~eaxanthiii 16

Chlorophyll a

014 -F j-carotene

- 12

ic a10-

8

V' 60 A

2 I* W.i; 1?1 xg< 4_ :_

0 1 25

Time (hours)

20- 20 - E3 Zeaxanthin

T 8

B(Chlorophyll a 16 P-Carotene

14 l

::E 12

10

88

6

L4

2 ?

Timie (hours)

Fig. 2. A. Changes in pigment concentrations in HMW DOC over 5 h of ultrafiltration. Vertical bars-SE. B. As panel A, but without fucoxanthin, which is absent in this experiment because of seasonal changes in the phytoplankton assemblage at the collection site. Vertical bars-SE.

maintains temperatures within a desired range, as well as an adequate mechanism for shielding the filtration ap- paratus from ambient light, would significantly reduce these artifactual effects. It should also be noted that the pumps in the Amicon system used in this study appear to generate more heat than the peristaltic pumps com- monly used in other ultrafiltration systems (K. Buesseler pers. comm.). Further studies are needed to determine the most efficient means of reducing ultrafiltration artifact and to assess whether there are similar effects on many of the other biochemical constituents of DOC.

It should be noted that the following discussion on pigment concentrations and turnover in HMW DOC samples collected in the field was not corrected for the aforementioned effects of artifact; however, the mere presence of detectable levels of these class-specific carotenoids still allows identification of dominant POC sources. We will show that using pigments to obtain a qualitative



426 Notes

Table 3. Plant pigment concentrations in HMW DOC and POC (?SE) from Gulf of Mexico in June 1992 (92G7) and January 1993 (93G1) and off the coast of Cape Hatteras in May 1993 (93G7). Below limits of detection-BLD.

Fucoxan- Zeaxan- Chl a Chl b thin thin Chl a Chl b Fucoxanthin Zeaxanthin

Station (sample depth) (,ug mg- 1 HMW DOC) (gg mg-1 POC)

92G7-St.1 (surface) 0.05 0.01 BLD 0.04 3.42?0.60 1.10?0.10 4.65?0.40 1.06?0.20 92G7-St.6 (surface) 0.02 0.03 0.01 BLD 93.46?0.90 1.60?0.40 4.67?0.80 10.76?2.30 92G7-St.5 (surface) BLD BLD 0.002 BLD 43.75?1.00 3.67?0.20 6.25?0.60 1.25?0.10 93Gl-St.1 (surface) BLD BLD BLD 0.01 4.92?2.50 1.32?0.14 2.29?0.20 5.75?0.08 93Gl-St.4 (surface) 0.03 0.05 0.002 0.01 100.00?2.40 19.20?1.60 36.00?4.00 24.00?1.60 93Gl-St.4 (355 m) 0.43 BLD 0.004 0.36 18.97?5.20 5.17?1.20 5.17?1.70 1.72?0.90 93Gl-St.5 (surface) 0.06 BLD BLD 0.02 60.00?13.0 12.38?2.50 8.57?2.50 16.19?0.60 93Gl-St.5 (1,600 m) BLD BLD BLD BLD 0.98?0.001 1.00?0.30 8.57?1.10 6.19?1.10 93G7-St.2 (surface) 0.05 BLD 0.006 0.03 15.12?1.51 BLD 3.03?1.70 2.23?0.30 93G7-St.16 (24 m) 0.03 0.09 0.003 0.009 44.49?3.70 36.26?5.70 29.44?1.90 4.08?0.20

assessment of HMW DOC sources is useful when inter- faced with other quantitative data.

There was a significant correlation (P < 0.05) between the lowest observed pigment concentrations in POC and HMW DOC at stations 93G1-St. 5 and 93G7-St. 1; these were the deepest samples in the study (1,600 and 750 m). However, high pigment concentrations in POC were not always correlated with high concentrations in HMW DOC. The highest concentrations of all pigments in POC, except for Chl b, occurred in the surface waters of station 9 3G 1- St. 4 (Gulf of Mexico) in January 1993 (Table 3); this was due to the upwelling effects of the outer edge of a warm- core ring (Biggs 1992; Lambert 1994). The highest concentrations of pigments (Chl a and zeaxanthin) in HMW DOC were also found at station 95G1-St. 4; however, they were in waters collected at 355 m (Table 3). Thus, the production and flux of pigment containing HMW DOC derived from phytoplankton in surface waters was most likely responsible for these high concentrations at depth. The highest total pigment concentrations in POC, at station 93G7-St. 16 off Cape Hatteras, were significantly correlated (P < 0.05) with the highest total pigment content in HMW DOC in this region.

The predominance of the carotenoids fucoxanthin (a marker for diatoms) and zeaxanthin (a marker for cyanobacteria) suggests that these classes of phytoplankton represented the dominant forms of POC and may also be the most likely sources of HMW DOC in the northwestern Gulf of Mexico and off Cape Hatteras (Table 3). More- over, it has also been shown, using lignin-phenols as biomarkers of terrestrial carbon, that this region of the gulf generally has a lower input of terrestrial C due to low river discharge; however, at certain times of year the Mis- sissippi River can have effects on this region (Hedges and Parker 1976; Lambert 1994). Prochlorophytes, which also contain zeaxanthin, are believed to be an important source of POC on the outer slope in this region of the Gulf of Mexico (Bianchi et al. 1995; Lambert 1994). However, we cannot estimate the relative importance of prochlorophytes vs. cyanobacteria because our HPLC method did not separate divinyl Chl a (found in prochlorophytes) from Chl a (Chisholm et al. 1988). The predominance of

cyanobacteria and prochlorophytes at station 93G1 -St. 5 in January 1993 may have been caused by a warm-core ring (Lambert 1994). Warm-core rings have been shown to carry oligotrophic water that is usually dominated by cyanobacteria and prochlorophytes (Biggs 1992).

The relative absence of pheopigments (i.e. pheophorbides, pheophytins, chlorophyllides) in HMW DOC (except for chlorophyllides in a few samples from this study) suggests that most pigments found in HMW DOC are likely to be derived from direct exudation products or sloppy feeding by zooplankton. Pheophorbides (loss of phytol and Mg from the tetrapyrrole) and pheophytins (loss of Mg from the tetrapyrrole) are primarily produced by heterotrophic breakdown of the parent chlorophyll molecule; pheophorbide production may be more closely associated with metazoan grazing (Daley and Brown 1973; Shuman and Lorenzen 1975; Bianchi et al. 1988, 1991). On the other hand, chlorophyllide (loss of the phytol from the tetrapyrrole) is believed to be produced by enzymatic breakdown in intact cells; it is also an intermediate com- pound in chlorophyll synthesis (Jeffrey 1974; Bidwell 1979; Jeffrey and Hallegraeff 1987).

Chlorophyllides were found in all HMW DOC and POC collected from surface waters at all our stations; HMW DOC ranged from 0.02 to 0.05 ug mg- I and POC ranged from 0.10 to 0.25 Aug mg-'. Total pheophorbides (0.02 ng mg-' HMW DOC) and pheophytins (0.01 ng mg-I HMW DOC) were found in surface water samples at only station 92G7-St. 1 in June 1992 in the gulf. Total pheophorbides and pheophytins were found at most stations (both locations); total pheophorbides ranged from 0.01 to 0.45 ng mg- POC and pheophytins ranged from 0.02 to 0.21 ng mg- POC. These low concentrations of pheophorbides and pheophytins in the POC and HMW DOC may be due to sloppy feeding by zooplankton (no gut passage of pigments). It is also unlikely that these pheopigments photodegrade considerably faster than the parent pigments (Welschmeyer and Lorenzen 1985). Thus, the relative importance of sloppy feeding vs. exudation processes in producing pigments in HMW DOC remains uncertain.

It has been estimated that it would take only 9.6 d for



Notes 427

detrital pigments to be reduced to half of their initial concentration by photooxidation under a cumulative light exposure of 3.61 x 10-21 quanta m-2 d-l (a value rep- resentative of midlatitude waters under clear skies) (Nel- son 1993). Unfortunately, most of the published accounts of pigment decay rates (excluding pheopigments) are from sediments, plant materials (i.e. macroalgae and vascular plants), and freshwater phytoplankton (Leavitt and Car- penter 1990; Bianchi et al. 1991; Sun et al. 1991). If we use these decay rates, the half-life of Chl a (in sediments) derived from phytodetritus, is 24 d compared to a half- life of 75 d for zeaxanthin; decay rates in HMW DOC from surface waters are likely to be faster.

Carotenoids that lack the 5,6-epoxide group, such as zeaxanthin, tend to be more decay-resistant than pigments that do contain it (i.e. fucoxanthin) (Repeta 1989; Bianchi and Findlay 1991). For example, although the highest concentrations of zeaxanthin and Chl a (in HMW DOC) occurred at 355 m at station 93G1 -St. 4, fucoxanthin was not present due to its rapid decay rate. However, at depths >400 m, all pigments were absent in HMW DOC despite their presence in POC (Table 3); this absence of pigments in HMW DOC suggests that the HMW DOC at these depths is older than the mean life of these pigments and that the filtration process did not produce pigments in the HMW DOC. Greater apparent 14C ages of HMW DOC at these depths further substantiates these findings (Santschi et al. 1995). Thus, it appears that pigment-containing HMW DOC is likely to be very recent, which agrees with other studies (Santschi et al. 1995; Amon and Benner 1994), especially if these pigment concentrations are lower than actual values due to an ultrafiltration artifact. The presence of these short-lived pigment compounds in HMW DOC may further support the contention that much of the HMW DOC in the ocean is cycled rapidly due to its lability, reactivity (surface activ- ity), and modem age.

Thomas S. Bianchi' Corey Lambert'

Center for Coastal and Marine Studies Department of Biology Lamar University Beaumont, Texas 77710

i Present address: Department of Ecology, Evolution, and Or- ganismal Biology, Tulane University, New Orleans, Louisiana 70118-5698.

Acknowledgments We thank the crew and officers of the R V Gyre, Matt Quigley,

Kent Warnken, Sarah Oktay, and M. Ravichandran for assis- tance in sample collection.

This research is based in part on work supported by the Na- tional Science Foundation (grant OCE 90-12103), the Depart- ment of Energy (grant DE-FG 05-92ERG 1421), and the Texas Institute of Oceanography.

Peter H. Santschi M. Baskaran Laodong Guo

Department of Oceanography Texas A&M University Galveston 77553

References

AMON, R. M. W., AND R. BENNER. 1994. Rapid cycling of high-molecular-weight dissolved organic matter in the ocean. Nature 369: 549-552.

BASKARAN, M., P. H. SANTSCHI, G. BENOIT, AND B. D. HoNEYMAN. 1992. Scavenging of Th isotopes by colloids in seawater of the Gulf of Mexico. Geochim. Cosmochim. Acta 56: 3375-3388.

BENNER, R., B. J. EADIE, AND P. B. HATCHER. 1992. Bulk chemical characteristics of dissolved organic matter in the ocean. Science 255: 1561-1564.

BLANCHI, T. S., R. DAWSON, AND P. SAWANGWONG. 1988. The effects of macrobenthic deposit-feeding on the degradation of chloropigments in sandy sediments. J. Exp. Mar. Biol. Ecol. 122: 243-255.

, AND S. FINDLAY. 1991. Decomposition of Hudson estuary macrophytes: Photosynthetic pigment transfor- mations and decay constants. Estuaries 14: 65-73.

- , , AND R. DAWSON. 1993. Organic matter sources in the water column and sediments of the Hudson River estuary: The use of plant pigments as tracers. Estuarine Coastal Shelf Sci. 36: 359-376.

- 1 , - ,AND D. FONTVIEILLE. 1991. Experimental degradation of plant materials in Hudson River sediments. 1. Heterotrophic transformation of plant pigments. Biogeo- chemistry 12: 171-187.

, C. LAMBERT, AND D. C. BIGGS. 1995. Distribution of chlorophyll-a and phaeopigments in the northwestern Gulf of Mexico: A comparison between fluorimetric and high performance liquid chromatography measurements. Bull. Mar. Sci. 56: 25-32.

BIDWELL, R. G. S. 1979. Plant physiology. MacMillan. BIGGS, D. C. 1992. Nutrients, plankton, and productivity in

a warm-core ring in the western Gulf of Mexico. J. Geophys. Res. 97: 2143-2154.

CHISHOLM, S. W., AND OTHERS. 1988. A novel, free-living prochlorophyte abundant in the oceanic euphotic zone. Na- ture 334: 340-343.

DALEY, R. J., AND S. R. BROWN. 1973. Experimental char- acterization of lacustrine chlorophyll diagenesis. 1. Physi- ological and environmental effects. Arch. Hydrobiol. 72: 277-304.

FOOTE, C. S. 1974. Photosensitized oxidation and singlet oxygen: Consequences in biological systems, p. 85-133. In W. A. Pryor [ed.], Free radicals in biology. V. 2. Academic.

, Y. C. CHANG, AND R. W. DENNEY. 1970. Chemistry of singlet oxygen. 10. Carotenoid quenching parallels biological protection. J. Am. Chem. Soc. 92: 5216-5218.

Guo, L., C. H. COLEMAN, AND P. H. SANTscHI. 1994. The distribution of colloidal and dissolved organic carbon in the Gulf of Mexico. Mar. Chem. 45: 105-119.

HARsouR, J. R., AND J. R. BOLTON. 1978. The involvement of the hydroxyl radical in the destructive photooxidation of chlorophylls in vivo and in vitro. Photochem. Photobiol. 28: 231-234.



428 Notes

HEDGES, J. I. 1992. Global biogeochemical cycles: Progress and problems. Mar. Chem. 39: 67-93.

, B. A. BERGAMASHI, AND R. BENNER. 1993. Compar- ative analyses of DOC and DON in natural waters. Mar. Chem. 41: 121-134.

, AND P. L. PARKER. 1976. Land-derived organic matter in the surface sediments from the Gulf ofMexico. Geochim. Cosmochim. Acta 40: 1019-1029.

JEFFREY, S. W. 1974. Profiles of photosynthetic pigments in the ocean using thin-layer chromatography. Mar. Biol. 26: 101-110.

, AND G. M. HALLEGRAEFF. 1987. Chlorophyllase distribution in 10 classes of phytoplankton-a problem for chlorophyll analysis. Mar. Ecol. Prog. Ser. 35: 293-304.

KLEIN, B. 1988. Variations of pigment content in two benthic diatoms during growth in batch cultures. J. Exp. Mar. Biol. Ecol. 115: 237-248.

LAMBERT, C. 1994. Chemical biomarkers as tracers of organic matter in the Gulf of Mexico. M.S. thesis, Lamar Univ. 70 p.

LEAVITT, P. R., AND S. R. CARPENTER. 1990. Aphotic pigment degradation in the hypolimnion: Implications for sedimen- tation studies and paleolimnology. Limnol. Oceanogr. 35: 520-535.

LEE, C., AND S. M. HENRICHS. 1993. How the nature of dissolved organic matter might affect the analysis of dissolved organic carbon. Mar. Chem. 41: 105-120.

MILLIE, D. F., H. W. PAERL, AND J. P. HURLEY. 1993. Mi- croalgal pigment assessments using high-performance liquid chromatography: A synopsis of organismal and eco- logical applications. Can. J. Fish. Aquat. Sci. 50: 2513- 2527.

NELSON, J. R. 1993. Rates and possible mechanism of light- dependent degradation of pigments in detritus derived from phytoplankton. J. Mar. Res. 51: 155-179.

RANBY, B., AND J. F. RABEK. 1975. Photodegradation, photooxidation, and photostabilization of polymers. Wiley.

REPETA, D. J. 1989. Carotenoid diagenesis in recent marine

sediments: 2. Degradation of fucoxanthin to loliolide. Geo- chim. Cosmochim. Acta 53: 699-707.

SANTSCHI, P. H., AND OTHERS. 1995. Isotopic and biochemical evidence for the recent origin of colloidal organic matter in the ocean. Geochim. Cosmochim. Acta 59: 625-631.

SHARP, J. 1973. Total organic carbon in seawater-comparison of measurements using persulfate oxidation and high temperature combustion. Mar. Chem. 1: 211-229.

SHUMAN, F. R., AND C. J. LORENZEN. 1975. Quantitative degradation of chlorophyll by a marine herbivore. Limnol. Oceanogr. 20: 580-586.

SoKAL, R. R., AND F. J. ROLHF. 1981. Biometry, Freeman. STAUBER, J. L., AND S. W. JEFFREY. 1988. Photosynthetic pig-

ments in fifty-one species of marine diatoms. J. Phycol. 24: 158-172.

SUN, M., R. C. ALLER, AND C. LEE. 1991. Early diagenesis of chlorophyll-a in Long Island Sound sediments: A measure of carbon flux and particle reworking. J. Mar. Res. 49: 379- 401.

WELSCHMEYER, N. A., AND C. J. LORENZEN. 1985. Chlorophyll budgets: Zooplankton grazing and phytoplankton growth in a temperate fjord and the central Pacific gyre. Limnol. Oceanogr. 30: 1-2 1.

WILLIAMS, P. M., AND E. R. M. DRUFFEL. 1988. Dissolved organic matter in the ocean: Comments on a controversy. Oceanography 1: 14-17.

WRIGHT, S. W., AND S. W. JEFFREY. 1987. Fucoxanthin pigment markers of marine phytoplankton analyzed by HPLC and HPTLC. Mar. Ecol. Prog. Ser. 38: 259-266.

, AND OTHERS. 1991. Improved HPLC method for the analysis of chlorophylls and carotenoids from marine phytoplankton. Mar. Ecol. Prog. Ser. 77: 183-196.

Submitted: 20 January 1994 Accepted: 14 September 1994 Amended: 17 November 1994

Limnol. Oceanogr., 40(2), 1995, 428-433 ? 1995, by the American Society of Limnology and Oceanography, Inc.

On the chlorophyll a retention properties of glass-fiber GF/F filters

Abstract-Extensive comparisons of the retention properties of glass-fiber GF/F and 0.2-,um membrane filters show that these two filter types retain equivalent amounts of Chl a. The experiments conducted were in the open ocean waters of the Pacific Ocean, from 46?N to 28?S, including waters from the equatorial divergence, the low latitude subtropical gyres, and higher latitudes. These results contradict a recent report that suggests that in some cases GF/F filters underestimate chlorophyll concentration by 4-fold when compared with 0.2-,um membrane filters. The data set also allowed examination of latitudinal gradients in integrated chlorophyll. Previously observed latitudinal gradients in the North Pacific were present but were much weaker than those found in the South Pacific.

Dickson and Wheeler (1993) reported that particulate matter collected by vacuum filtration of water samples from the North Pacific yielded significantly higher concentrations of chlorophyll a (Chl a) for 0.2-,um Nuclepore filters compared to Whatman GF/F glass-fiber filters. In the most extreme case, at 28?N, 1 55?W, the 0.2-,um filters retained four times the amount of Chl a retained on GF/F filters. If these results truly reflect the Chl a retention properties of GF/F filters, then the oceanographic com- munity may have seriously underestimated Chl a and primary productivity in the sea, as most biological ocean- ographers routinely make measurements of Chl a and primary productivity with GF/F filters. We report here



plant pigments as biomarkers of high-molecular-weight dissolved organic carbon

Documents