Determination of algal biomass with HPLC pigment ...

0 downloads 0 Views 525KB Size Report
lakes of different trophic state in comparison to microscopically measured biomass. Hans Schmid, Friedrich Bauer and Hans Bernd Stich1. Wasserwirtschaftsamt ...
Journal of Plankton Research Vol.20 no.9 pp.1651-1661, 1998

Determination of algal biomass with HPLC pigment analysis from lakes of different trophic state in comparison to microscopically measured biomass Hans Schmid, Friedrich Bauer and Hans Bernd Stich1 Wasserwirtschaftsamt, Rottachstrafie 15, D-87439 Kempten and 'Institutfur Seenforschung, Untere Seestrafie 81, D-88085 Langenargen, Germany

Introduction Determination of the biomass of phytoplankton is a classic method in limnology for evaluating the trophic state of natural waters (Wetzel and Likens, 1979; Vollenweider and Kerekes, 1980). The community composition of the phytoplankton is also an indicator of the trophic state of lakes (Desortova, 1981; Kummerlin and Biirgi, 1989). Today, biomass is usually determined by counting and measuring the algal cells microscopically, but the counting procedure is time consuming and expensive. For routine analysis, the chlorophyll a (Chi a) content is often used as a parameter (Marker et al., 1980; Nusch, 1980), in spite of the well-known problem of how accurate Chi a is in relation to algal biomass (Eppley et al., 1977; Westlake, 1980). Furthermore, Chi a does not allow any conclusion about the community composition of the phytoplankton. High-performance liquid chromatography (HPLC) is an analytical method that not only makes it possible to gain correct Chi a data, but also to obtain more detailed information about the composition of the phytoplankton (Gieskes and Kraay, 1986a; Wilhelm and Manns, 1991; Lami et al., 1992). In 1996, the Bavarian Ministry of the Environment started a project to determine the total algal biomass as well as the composition of the phytoplankton community by HPLC pigment analysis. The method is based on the fact that different algal classes can be distinguished by their content of additional pigments, the so-called marker pigments (Goodwin, 1988). In particular, thefivetaxonomically most important algal classes (diatoms, cryptophytes, chlorophytes, dinophytes and cyanophytes) are distinguishable, whereas there is no difference between diatoms and chrysophytes or between © Oxford University Press

1651

Downloaded from plankt.oxfordjournals.org by guest on October 17, 2011

Abstract. A high-performance liquid chromatography (HPLC) method is presented with which a direct determination of the phytoplankton biomass is possible. By correlating over 100 samples investigated with HPLC as well as microscopically, it has been shown that phytoplankton biomass is described very accurately by determining the concentration of accessory pigments. In contrast to earlier studies, we made a direct correlation of marker pigments with biovolume and did not consider the ratio of marker pigment to chlorophyll a. Determination of the dominating class in each lake succeeded with con-elation coefficients > 0.90, and even 0.96 against total biovolume. Because of the large number of samples, statistical proof indicates that algal marker pigments are better indicators for phytoplankton biomass than chlorophyll a. This HPLC method is ideal for large-scale monitoring because of its suitability for routine analysis and its extensive sample throughput. The procedure is applicable to different natural waters, from oligotrophic to highly eutrophic. Because of the simultaneously performed analysis of chlorophyll a, the succession of existing data is ensured.

FLSchmid, EBauer and ILStich

Method Samples were taken from 10 different lakes at different trophic states, size and depth. The data pool consisted of samples from two deep and voluminous subalpine lakes, which are now in a mesotrophic state after a period of eutrophication, two highly eutrophic small lakes and some small lakes with trophic states ranging from oligotrophic to eutrophic. All lakes are located in the south of Germany (Table I). Table I. Some hydrologic/morphometric parameters of the lakes investigated Lake

Area (km*)

Volume (m3)

Residence time (a)

Max. depth/ mean depth (m)

Total P (MgH)

No. of samples

Lake Constance Ammersee Hopfensee Sulzberger See Weissensee Niedersonth. See Griintensee Rottachsee Alpsee Elbsee

571 46.6 1.94 0.36 1.34 1.35 1.23 2.96 2.47 0.20

48.5 x 10' 1.7 X 10' 8.9 X 106 2.5 X 106 16.5 X 106 13.5 X 106 4.4 X 106 253 X106 32.7 X 106 0-5X10»

4.15 2.70 0.35 0.88 1.25 0.70 0.03 0.75 0.49 0.09

252/92 81/37 10/5 15/7 25/12 21/10 11/4 35/9 22/13 4/2

30 15 70 85 26 45 65 38 35 65

49 9 17 17 3 1 1 3 2 1

1652

Downloaded from plankt.oxfordjournals.org by guest on October 17, 2011

chlorophytes and euglenophytes because they have the same major pigments. These marker pigments can easily be isolated from the algal cells (Mantoura and Llewellyn, 1983; Wright and Shearer, 1984). In recent years, there have been numerous efforts to improve the quality of HPLC separation (Wright et al., 1991; van Heukelem et al., 1994, Schmid and Stich, 1995). The actual benefits of HPLC pigment analysis of algae are: (i) qualitative estimation of the phytoplankton community (Everitt et al., 1990; Wilhelm et al., 1991; Letelier et al., 1993); (ii) estimation of the historical development of phytoplankton in a lake by sediment analysis (Ziillig, 1982; Leavitt and Findlay, 1994; Leavitt et al., 1994); (iii) water mass distinction with fingerprints of algal pigments (Gieskes, 1991); (iv) inferring the physiological status of algae from the determination of chlorophyll derivatives, e.g. grazing (Gieskes and Kraay, 1986b; Klein et al., 1986) or sedimentation of the phytoplankton (Head et al., 1994). Up to now, most published HPLC methods have used the ratio of marker pigments/Chi a (mp/Chl a) to obtain information on the distribution of the algal classes (e.g. Barlow et al., 1993), although it is known that the Chi a content of algal cells can vary widely (Bidigare et al., 1990). Because of this, it was not possible to relate the results of HPLC pigment analysis to the biovolume, except as a relative distribution (Wilhelm et al., 1995). For the same reason, we did not consider the ratio mp/Chl a for our investigations. Instead, we related the measured concentrations of marker pigments to the biovolume of the microscopically counted cells of different algal classes directly. Our results show that the marker pigments are more closely related to the dominant algal class, as well as to total biovolume, than Chi a is related to total biovolume.

Determination of algal biomass with HPLC pigment analysis

Results and discussion

Our results show that the marker pigments are more closely related to the biovolume of the corresponding algal class than Chi a is related to the total biovolume (Figure 1). There is a very close correspondence between alloxanthin and the cryptophytes (Figure IB). This relationship can be quantified by linear regression (Figure 2). There is only a small difference between the cryptophyte-dominated Lake Hopfensee (Figure 2A) and the regression representing all samples (Figure 2B). The slight fluctuation is represented in the factors plotted in the figures. The levels listed in Table II are based on 103 samples. To our knowledge, the factors in column three indicate the demonstration of a direct relationship 1653

Downloaded from plankt.oxfordjournals.org by guest on October 17, 2011

The samples were collected over a period of 3 years from Lake Constance, from annual profiles of the three smaller lakes (Lake Hopfensee, Lake Sulzberger See and Lake Ammersee), and include some individual samples from other small lakes near Kempten. The samples investigated were integrated samples from 0 to 20 m depth or from 0 to the maximum depth. An aliquot for cell counting was immediately fixed with Utermohl's solution. The algae were identified and enumerated using an inverted microscope according to Utermohl (1958). Biovolume was calculated following the algal list of the Bavarian Ministry of the Environment, which is updated every year. The accuracy of the list of algal biovolumes used was checked for each sample. For HPLC analysis, 11 of water was filtered through a glass-fibre filter (Schleicher & Schuell, No. 6). Either the filter was extracted immediately or it was stored at -20°C. Extraction was in acetone (90%, v/v) containing an ion-pairing reagent (1 g tetraethylammonium acetate, 2.5 g ammonium acetate per litre). Extraction vials were placed in a water bath (5 min, 55°C). After cooling to room temperature, the samples were sonicated for 10 min, purged through a disposable filter unit (Schleicher & Schuell, Spartan 13/20, 0.45 um) into dark-coloured autosampler vials. The HPLC system consisted of a degasser (DG 503), a low-pressure gradient pump (M 480), an autosampler (Gina 50), a column oven (Jetstream), a diode array detector (UVD 340S) and a fluorescence detector (RF 1001), all produced by Gynkotek, Germany. Chromatograms and spectra were recorded with the data system Gynkosoft. Pigments are quantified using external standards. Chlorophylls a and b were obtained from Sigma-Aldrich, Munich. Fucoxanthin, lutein, zeaxanthin, cantaxanthin, p-cryptoxanthin, echinenon, a- and p-carotene were donated by Hoffmann-La Roche, Basel, Switzerland. Pigments for which no standard substance is available were quantified according to the calibration factor for the chemically most similar pigment. For example, alloxanthin was determined using the zeaxanthin standard. The chromatographic conditions for the separation have been described in detail by Schmid and Stich (1995).

H-Schmid, F.Bauer and H^tich

Downloaded from plankt.oxfordjournals.org by guest on October 17, 2011

Fig. 1. Comparison of Chi a with total biovolume (A) and comparison of alloxanthin with biovolume of cryptophytes (B) in Lake Hopfensee 1996. Correlation coefficients: (A) = 0.76; (B) = 0.93.

between the measured algal marker pigments and the corresponding algal classes, as well as total biovolume. This new method of calculating the biovolume of the phytoplankton is a significant step in the use of HPLC analysis of algal pigments. In Figures 3-5, the calculated biovolumes based on the concentration of marker pigments are compared to the biovolume estimation based on microscopic counts for different lakes. The microscopically determined biovolumes of the phytoplankton of Lake Hopfensee in 1996 (Figure 3A) are nearly identical to those obtained by HPLC analysis (Figure 3B). The correlation coefficients are between 0.75 for the diatoms and 0.96 for the dinophytes. Only on 1 July 1996 was there a significant difference because diatoms counted in the microscope were not detectable in terms of pigments. Apart from a sampling error, it is 1654

Determination of algal biomass with HPLC pigment analysis

OE+0 20

30 40 alloxanttiin [pg / L]

10

20

30 40 alloxanttiin [pg / L]

Fig. 2. Linear regressions of biovolume of cryptophytes against concentration of alloxanthin from Lake Hopfensee (A) and from all samples investigated (B). Correlation coefficients: (A) r = 0.92; (B) r = 0.94. Correlation coefficients without consideration of the most extreme points >20 ug 1"': (A) = 0.85; (B) = 0.85.

Table II. Factors resulting from linear regressions; straight lines forced through the origin, all samples investigated Marker pigment Fucoxanthin Alloxanthin Chlorophyll b Peridinin Cantaxanthin

Algal class Diatoms Cryptophytes Chlorophytes Dinophytes Cyanophytes

Factor (urn-1 biovolume/ug mp)

r

320000 120 000 576 000 1143 000 16 218 000

0.75 0.94 0.77 0.91 0.98

Correlation coefficient

probably due to another reason. It is almost impossible to distinguish between living and dead diatoms using light microscopy. Thus, the diatoms counted on that day may have already been dead. In this case, the HPLC result would be the more correct one. A more detailed comparison of the two methods will be published later. 1655

Downloaded from plankt.oxfordjournals.org by guest on October 17, 2011

10

ILSchmid, F.Bauer and H-Stich

3B

8 8 8 Idatofns I

I

SB

SB

SB

SB

SB SB

09

O#

O*

O)

Ov

2

g

8

g

g g

ui

3 cysnoptiytes

1 chloroptiytes

4B

| S 58 8

g g g

Fig. 4. Comparison of microscopically determined biovolume (A) and biovolume estimated by HPLC analysis (B) in Lake Sulzberger See 1996. Correlation coefficients: diatoms = 0.51; cryptophytes = 0.88; chlorophytes = 0.48; dinophytes = 0.40; cyanophytes = 0.98.

In data for Lake Sulzberger See, the cyanophytes were also detectable (Figure 4), but recognizing a delay in the detection of the cyanophytes, the necessity for a minimal amount of marker pigment in the extract is demonstrated. This is most typical for cyanophytes, because the content of marker pigment is very small, 1656

Downloaded from plankt.oxfordjournals.org by guest on October 17, 2011

Fig. 3. Comparison of microscopically determined biovolume (A) and biovolume calculated after HPLC analysis (B) in Lake Hopfensee in 1996. Correlation coefficients: diatoms = 0.75; cryptophytes = 0.92; chlorophytes = 0.88; dinophytes = 0.96.

Determination of algal biomass with HPLC pigment analysis

Downloaded from plankt.oxfordjournals.org by guest on October 17, 2011

Fig. 5. Comparison of microscopically determined biovolume (A) with biovolume calculated after HPLC analysis (B) in Lake Constance. Correlation coefficients: diatoms = 0.90; cryptophytes = 0.49; chlorophytes = 0.50; dinophytes = 0.78; cyanophytes = 0.87.

-10% of the Chi a content. Thus, a minimum number (abundance) of each algal class is needed for detection in the HPLC system. Below these concentrations, biovolume cannot,be estimated in a reliable fashion using HPLC (Table III). The minimum detection levels in Table III are concentrations in the algal extract calculated for an injection volume of 50 ul. The smaller the peaks in the chromatogram, the greater the likelihood of making mistakes in setting the peak delimiter correctly. To handle this problem, the optimum sampling routine would be to filter 1 1 of water through a glass-fibre filter, which should be as small as possible in order to keep the extraction volume small. If the Chi a content is expected to be below 10 ug I"1, it would be better to filter 1.5 or 2 1 of water. Figure 5 also demonstrates that in mesotrophic lakes like Lake Constance, very 1657

H-Schmid, F.Bauer and ILSticb Table III. Definition of minimal concentrations of marker pigments. The plotted biovolume is calculated for 50 ml injection volume, 8 ml extraction volume and 11filtrationvolume Pigment

Minimal area (mAU x min)

Minimal concentration (ng)

Minimal biovolume (urn-1 ml-1)

Fucoxanthin Alloxanthin Chlorophyll b Peridinin Cantaxanthin

0.30 0.45 19.2 (mV X min) 0.25 0.20

25 7 3 2 1

132 000 134 400 276 500 386 200 2 584 800

Conclusion We have developed, for the first time, a direct relationship between algal pigments and biovolume, based on more than 100 samples. We successfully determined biovolume, which is essential for the assessment of natural waters, by means of a procedure applicable to routine analysis. The method is also applicable to different types of lakes, from oligotrophic to highly eutrophic. In particular, the dominating algal classes of each lake could be determined with correlations > 0.90. For this new type of analysis and calculation, advanced equipment is required. However, because of a high degree of automation, the method has an extensive sample throughput and is therefore excellent for monitoring. Owing to the close relationship of pigment to biovolume, this method provides a large amount of information. This could not be managed with the conventional counting procedure, neither financially nor timewise. Our method could serve as a basis for comparative investigations of the water quality of lakes. 1658

Downloaded from plankt.oxfordjournals.org by guest on October 17, 2011

good correlations can be achieved. It follows from the very close relationship between biovolumes and marker pigments of the single algal classes that the comparison of the total biovolume counted with the sum of the calculated biovolumes of the single algal classes also leads to close correlations The correlation of the calculated biovolume is much better (0.96) than the correlation of measured Chi a to total biovolume (0.78) (Figure 6). In view of these results, we emphasize that the calculated biovolume on the basis of HPLC analysis can describe the phytoplankton of a lake quantitatively just as accurately as the composition of the algal community. However, only the major phytoplankton classes can be distinguished. An exact examination of species composition is only possible with microscopic determination and counting of algal cells. However, HPLC analysis is a powerful tool for basic routine lake monitoring; the more time-consuming microscopic examination remains for special problems and evaluation of the dominating species. Therefore, the HPLC method yields fundamental information for limnological investigations and the assessment of lakes. In comparison to the traditional method of assessing algal biomass using Chi a concentration, a new dimension is accessible with HPLC pigment analysis.

Determination of algal biomass with HPLC pigment analysis

101

76 -counted btowf •

-calculated biovol

Downloaded from plankt.oxfordjournals.org by guest on October 17, 2011

81

101

-counted Mow!—*— cHoroph»8a|

Fig. 6. Comparison of calculated biovolume and counted biovolume (A) and measured Chi a and counted biovolume (B). Correlation coefficients: (A) = 0.96; (B) = 0.78; all samples investigated.

We suggest substituting conventional Chi a analysis by HPLC pigment analysis as had been recommended by Gieskes (1991) for oceanographers or by the UNESCO Pigment Manual (Jeffrey et ai, 1997). Because Chi a is also analysed with HPLC, the connection to existing data is ensured. Acknowledgements We want to thank B.Kopf, K.Konig and R.Kummerlin for providing very accurate phytoplankton data, and G.Hochsieder, A.Hamm and the reviewers of the journal for improving the manuscript. 1659

H-Schmid, F.Bauer and H5tich

References

1660

Downloaded from plankt.oxfordjournals.org by guest on October 17, 2011

Barlow,R.G., Mantoura.R.F.C, Gough,M.A. and Fileman.T.W. (1993) Pigment signatures of the phytoplankton composition in the north-eastern Atlantic during the 1990 spring bloom. Deep-Sea Res. II, 40,459-477. Bidigare,R.R., MarraJ., Dickey.T.D., Iturriaga.R., Baker,K.S., Smith.R.C. and Pakji. (1990) Evidence for phytoplankton succession and chromatic adaptation in the Sargasso Sea during spring 1985. Mar. Ecol. Prog. Ser., 60,113-122. Desortova.B. (1981) Relationship between chlorophyll-a concentration and phytoplankton biomass in several reservoirs in Czechoslovakia. Int. Rev. Ces. Hydrobioi, 66,153-169. Eppley,R.W., Harrison.W.G., Chisholm,S.W. and Stewart^. (1977) Paniculate organic matter in surface waters of Southern California and its relationship to phytoplankton./ Mar. Res., 35,671-696. Everitt,D.A., Wright,S.W., VolkmanJ.K., Thomas.D.P. and Lindstrom.EJ. (1990) Phytoplankton community compositions in the western equatorial Pacific determined from chlorophyll and carotenoid pigment distribution. Deep-Sea Res., 37, 975-997. Gieskes.W.W.C. (1991) Algal Pigment Fingerprints: Clue to Taxon-specific Abundance, Productivity and Degradation of Phytoplankton in Seas and Oceans. NATO ASI Series Vol. G27, pp. 61-99. Gieskes.W.W.C. and Kraay.G.W. (1986a) Floristic and physiological differences between shallow and the deep nanophytoplankton community in the euphotic zone of the open tropical Atlantic revealed by HPLC analysis of pigments. Mar. Bioi, 91, 567-576. Gieskes.W.W.C. and Kraay.G.W. (1986b) Analysis of phytoplankton pigments by HPLC before, during and after mass occurrence of the microflagellate Corymbellus aureus during the spring bloom in the open northern North Sea in 1983. Mar. BioL, 92,45-52. Goodwin.T.W. (1988) Plant Pigments, Vol. 2. Academic Press, London. HeadJEJ.H., Hargrave.B.T. and Rao.D.V.S. (1994) Accumulation of a phaeophorbide a-like pigment in sediment traps during late stages of a spring bloom: A product of dying algae? Limnol. Oceanogr., 39, 176-181. Jeffrey,S.W., Mantoura.R.F.C. and Wright.S.W. (1997) Phytoplankton Pigments in Oceanography. UNESCO, Paris. Klein,A.W.O., Gieskes.W.W.C. and Kraay.G.W. (1986) Digestion of chlorophylls and carotenoids by the marine protozoan Oxyrrhis marina studied by HPLC analysis of algal pigments. /. Plankton Res., 8, 827-836. Kiimmerlin.R. and Biirgi.H.R. (1989) Die langjahrige Entwicklung des Phytoplanktons im Bodensee (1961-1986). Ber. Int. Cewasserschutzkomm. Bodensee, 39.175 pp. Lami.A., Guilizzoni.P., Ruggiu,D., Simona.M. and Barbieri,A. (1992) Role of pigments on algal communities and photosynthesis. Aquat. Sci., 54, 320-330. LeavittJP.R. and Findlay.D.L. (1994) Comparison of fossil pigments with 20 years of phytoplankton data from eutrophic lake 227, experimental lakes area, Ontario. Can. J. Fish. Aquat. Sci., 51, 2286-2299. Leavitt,P.R., Schindler.D.E., Paul.AJ., Hardie.A.K. and Schindler.D.W. (1994) Fossil pigment records of phytoplankton in trout-stocked alpine lakes. Can. J. Fish. Aquat. Sci., 51,2411-2423. Letelier,R.M., Bidigare.R.R., Hebel.D.V., Ondrusek,M., Winn.C.D. and Karl.D.M. (1993) Temporal variability of phytoplankton community structure based on pigment analysis. Limnol. Oceanogr., 38,1420-1437. Mantoura.R.F.C. and Llewellyn.C.A. (1983) The rapid determination of algal chlorophyll and carotenoid pigments and their breakdown products in natural waters by reversed-phase highperformance liquid chromatography. Anal. Chim. Ada, 151, 297-314. Marker,A.F.H., Nusch,E.A., Rai,H. and Riemann,B. (1980) The measurement of photosynthetic pigments in freshwaters and standardisation of methods: Conclusions and recommendations. Arch. HydrobioL Beih. Ergebn. Limnol., 14,91-106. Nusch,E.A. (1980) Comparison of different methods for chlorophyll and phaeopigment determination. Arch. Hydrobioi. Beih. Ergebn. Limnol., 14,14-36. Schmid,H. and Stich.H.B. (1995) HPLC-analysis of algal pigments: comparison of columns, column properties and eluents. / AppL PhycoL, 7,487-494. Utermohl,H. (1958) Zur Vervollkommnung der quantitativen Phytoplankton-Methodik. Mitt. Int. Ver. Limnol., 9, 1-38. Van Heukelem.L. et al. (1994) Improved separations of phytoplankton pigments using temperaturecontrolled HPLC. Mar. Ecol. Prog. Ser, 114,303-313. VoIlenweider,R.A. and KerekesJ. (1980) The loading concept as a basis for controlling eutrophication: philosophy and preliminary results of the OECD program on eutrophication. Prog. Water Technol., 12, 5-18.

Determination of algal biomass with HPLC pigment analysis Westlake.D.F. (ed.) (1980) Primary production. In The Functioning of Fresh-water Ecosystem. IBP Springer-Verlag, Berlin, Vol. 22, pp. 588-611. Wetzel,R.B. and Likens.G.E. (1979) Limnological Analysis. Saunders, Philadelphia, 357 pp. Wilhelm.C. and Manns.L. (1991) Changes in pigmentation of phytoplankton species during growth and stationary phases—consequences for the reliability of pigment based methods of biomass determination. / Appl. Phycol., 3, 305-310. Wilhelm.C., Rudolph,K. and Renner.W. (1991) A quantitative method based on H.P.L.C. aided pigment analysis to monitor structure and dynamics of the phytoplankton assemblage—A study from Lake Meerfelder Maar (Eifel, Germany). Arch. Hydrobiol., 123, 21-35. Wilhelm.C. et al. (1995) The HPLC-aided analysis of phytoplankton cells as a powerful tool in water quality control. / Water SRT-Aqua, 44,132-141. Wright.S.W. and ShearerJ.D. (1984) Rapid extraction and high-performance liquid chromatography of chlorophyll and carotenoids from marine phytoplankton. /. Chromatogr., 294,281-295. Wright,S.W. et al. (1991) Improved HPLC method for the analysis of chlorophylls and carotenoids from marine phytoplankton. Mar. EcoL Prog. Ser., TJ, 183-196. Zullig.H. (1982) Untersuchungen iiber die Stratigraphie von Carotinoiden im geschichteten Sediment von 10 Schweizer Seen zur Erkundung friiherer Phytoplankton Entfaltungen. Schweiz. Z. Hydrol., 44,1-98. Received on August 3, 1997; accepted on April 16, 1998 Downloaded from plankt.oxfordjournals.org by guest on October 17, 2011

1661