Methodology from Pickney, J.L.: The USC HPLC Method - Technical Description.pdf: www.researchgate.net/profile/James_Pinckney/publication/265999187_The_USC_HPLC_Method_-Technical_Description/links/55266d0d0cf2628d5afdf397/The-USC-HPLC-Method-Technical-Description.pdf

1. INTRODUCTION

The HPLC method used at USC for photopigment separations is derived from the Van Heukelem et al. (1992, 1994) and Pinckney et al. (1996) protocols. Two different reverse-phase C18 columns are connected in series. A single monomeric guard column is followed by a monomeric reverse-phase C18 column and a polymeric reverse-phase C18 column. This column configuration was originally devised to enhance photopigment separations from sediment samples containing numerous (>150) photopigment and pigment degradation products. Monomeric columns provide strong retention and high efficiency while polymeric columns select for similar compounds with minor differences in molecular structure (Van Heukelem et al. 1992, Jeffrey et al. 1997). In addition to increasing the number of theoretical plates, the combination of both monomeric and polymeric columns optimizes photopigment separations based on two different molecular properties (coarse and fine structure). This method allows for the baseline separation of most major pigments including lutein/zeaxanthin and chlorophyll c3. However, chlorophylls c1 and c2 are not completely separated. Divinyl chlorophylls a an b are not completely resolved but occur as shoulders on the monovinyl chlorophylls a and b and can be visually identified in chromatograms.

2. EXTRACTION

SeaHARRE 4 samples were immediately frozen at -80 deg C upon receipt. For HPLC analysis, filters were placed in disposable polypropylene microfuge tubes (2 ml) and lyophilized (-50 deg C, 0.57 mbar, 12 h; Labconco FreeZone 2.5) to remove all water from the filters. After lyophilization, filters were cut into 6 equal sections and placed in microfuge tubes. Samples were extracted in 90% acetone (600 ul), and stored at -20 deg C for 18 - 20 h. Each sample also received 50 ul of the synthetic carotenoid trans β-apo-8'-carotenal (Sigma, cat. no. 10810) in 90% acetone as an internal standard using a gas-tight syringe (Hamilton) and click dispenser (Hamilton PB600-1). After extraction, the extract was clarified using a 0.45 um PTFE filter (Gelman Acrodisc). A known volume of the extract (400 ul) was then dispensed into amber glass autosampler vials (2.0 ml) and sealed with PTFE-silicone caps.

3. HPLC ANALYSIS

The instrumentation consisted of a binary gradient pump (Shimadzu dual LC10-AT vp and Controller SCL-10A vp), temperature-controlled autosampler (Shimadzu SIL10-A vp) with a 500 ul injection loop, column oven (Shimadzu CTO-10AS vp), and photodiode array detector (PDA, Shimadzu SPD-M10A vp; 200 to 800 nm range). For the PDA, spectra (380 - 700 nm) were obtained at 2.00 sec intervals for the duration of each run and photopigment peaks were quantified at 440 nm (± 4 nm). Two different reverse-phase C18 columns were connected in series. A single monomeric guard column (Rainin Microsorb, 0.46 x 1.5 cm, 3 um packing) was followed by a monomeric reverse-phase C18 column (Varian Microsorb-MV 100 - 3, 0.46 x 10 cm, 3 um packing) and a polymeric reverse-phase C18 column (Vydac 201TP54, 0.46 x 25 cm, 5 um packing). The column oven maintained a constant 40 deg C for the duration of the gradient. A non-linear binary gradient, adapted from Van Heukelem et al. (1992), was used for pigment separations (Table 1). Solvent A consists of 80% methanol : 20% ammonium acetate (0.5 M adjusted to pH 7.2) and solvent B is composed of 80% methanol : 20% acetone (Table 1). Solvents were degassed with an in-line degasser (Shimadzu DGU 14A). All solvents were HPLC-grade and chemicals were analytical grade.

Just prior to the HPLC run, an ion-pairing (IP) solution (1.00 M ammonium acetate) was added to the vial in a ratio of 4 parts extract: 1 part ammonium acetate. Prior work has shown that there is negligible pigment degradation within 12 hours of adding the IP solution if the sample is placed in a refrigerated autosampler rack (4.0 deg C). However, the IP solution should not be added to the sample if the time until sample analysis is greater than 18 hours.

4. CALIBRATION

Peaks were identified based on retention time and spectral matches with pigment spectra obtained from liquid standards (DHI, Hørsholm, Denmark) (Table 2). Peak areas for chromatograms are quantified using Shimadzu Client/Server 7.2.1 SP1 software. The PDA was calibrated using a multi-point calibration procedure for a range of injection volumes (25 - 300 ul) of pigment standards (DHI, Denmark). Regressions were performed using known pigment concentration (Y) vs. the integrated peak area (X) and were of the form Y = mX + b; where m is the slope and b is the Y intercept. All regressions had an r2 > 0.98. The slope of the fitted line was used as the response factor for pigment concentration calculations. The concentrations of pigments for which standards were unavailable were estimated using the ratio method outlined in Jeffrey et al. (1997, p. 443-4)

5. VALIDATION

Carotenal blanks (trans β-apo-8'-carotenal in 90% acetone) were run after every 10 samples to verify peak time reproducibility, peak area precision, and instrument performance during the sequence run. Peaks were identified based on retention time and comparison of absorbance spectra with a spectral library derived from pure pigment standards (DHI, Denmark). Long-term quality control is achieved by analyzing pure standards for chlorophyll a and the Mixed Standard supplied by DHI, Denmark at monthly intervals. Instrument performance is measured and compared with previous measures to determine changes in performance metrics. Volumetric measuring devices are checked weekly. The performance metrics for this method are shown in Table 3.

6. DATA PRODUCTS

Pigment concentrations were calculated for each identifiable peak using the following equation:

where CPi is the pigment concentration in ug l-1, APi pigment peak area, FPi is the response factor, Vc is the injection volume (ul), Vm is the total extract volume (volume of added acetone + volume of internal standard in ml), R is the ratio of the volume of IP solution + Vm divided by Vm, Vf is the volume of seawater filtered (liters), Ac is the average peak area for carotenal standards, and As is the peak area of carotenal in the sample.

7. CONCLUSIONS

This method has been employed by USC for ca. 15 years to analyze a broad spectrum of sample types from marine and freshwater habitats. The execution of the method is straightforward and involves minimum manipulation of the samples and extracts, relatively inexpensive, and does not generate hazardous waste products. The primary weakness of the method is the inability to completely separate chlorophylls c1 and c2 and divinyl chlorophylls.

REFERENCES

Jeffrey, S.W., Mantoura, R. F. C., Wright, S. W., [Eds.] 1997. Phytoplankton Pigments in Oceanography: Guidelines to Modern Methods. UNESCO, Paris.

Pinckney, J.L., Millie, D.F., Howe, K.E., Paerl, H.W., Hurley, J.P. 1996. Flow scintillation counting of 14C-labeled microalgal photosynthetic pigments. Journal of Plankton Research 18:1867-1880.

Van Heukelem, L., Lewitus, A.J., Kana, T.M. 1992. High-performance liquid chromatography of phytoplankton pigments using a polymeric reversed-phase C18 column. Journal of Phycology 28:867-872.

Van Heukelem, L., Lewitus, A.J., Kana, T.M., Craft, N.E. 1994. Improved separations of phytoplankton pigments using temperature-controlled high-performance liquid chromatography. Marine Ecology Progress Series 114:303-313.

BCO-DMO Data Processing Notes:

-Reformatted column names to comply with BCO-DMO standards.
-Data were originally separated by month (September and November) into two spreadsheets. Data were combined into one file and the columns "month" and "year" were added.
-Colors and headers were removed.