Documentation of Procedures for Dissolved Free Amino Acids Analysis

Most sampling and all analyses were performed by Joan E. Sheldon (now in
Department of Marine Sciences, University of Georgia - jsheldon@arches.uga.edu).
Samples were collected in clean 5-liter Niskin samplers, using surgical
silicone tubing for closure.  Samples were taken directly from the Niskin
samplers either by attaching a new 50 ml syringe to the hosecock with a
short length of acid-cleaned tygon tubing or by removing the plunger,
filling the syringe, and replacing the plunger in the syringe barrel.  The
first 10 ml were expelled through a 0.2 µm Anotec filter cartridge to wash
the filter.  Three 10 ml subsamples were then expelled through the Anotec
cartridge into fired scintillation vials.  The vials were capped with
cone-lined caps that had been acid-washed and rinsed with HPLC-grade water.
The sample vials were placed in a freezer and were returned without thawing
to the home laboratory for analysis by HPLC, following the general
precolumn derivatization procedure of Lindroth and Mopper (1978) and Mopper
and Lindroth (1982).  We followed the specific procedures of Henrichs and
Williams (1985), except that samples were thawed at 4°C, 2 ml subsamples
were derivatized, and 300 µl were injected.  Derivatized primary amino
acids were separated in a 250 x 4.6 mm cartridge-style column (Alltech)
packed with spherisorb ODS-1, 5 µm (Phase Separation) and detected using a
Kratos FS970 spectrofluoromonitor.  The amino acids quantified were
aspartic acid, glutamic acid, serine, histidine+threonine+glycine
(co-eluted), alanine, tyrosine, arginine, methionine, valine,
phenylalanine, isoleucine, leucine, and lysine.  Triplicates were usually
run in the HPLC.  Areas under the curve of each amino acid were integrated.
In making a best estimate of the concentration of each amino acid, the
variance around both the samples and the standards was taken into account
in reporting upper and lower 95% confidence limits.  See estimation of X
from Y in Sokal and Rohlf (1981).

While most analytical values fell within a quite narrow range, an
occasional set of samples was much higher in DFAA content.  Such
disparities are usually written off as failures of clean procedure.
However, in view of the fact that the replicates agree, we report these
values.  Future users of the data can make of them what they wish.

On the Blue Fin cruise of January 1991, we experimented with other kinds of
filter cartridges.  Replicate samples with the suffix N are filtered
through a 0.2 µm Nuclepore filter.  Replicate samples with the suffix M are
filtered through a 0.45 µm Millipore cartridge.

References

Henrichs, SM, Williams, PM 1985.  Dissolved and particulate amino acids and
carbohydrates in the sea surface microlayer.  Mar. Chem. 17: 141-163.

Lindroth, P, Mopper, K 1978.  High performance liquid chromatographic
determination of subpicomole amounts of amino acids by precolumn
fluorescence derivatization with o-phthalaldehyde.  Analyt. Chem. 51:
1667-1674.

Mopper, K., Lindroth, P. 1982.  Diel and depth variations in dissolved free
amino acids and ammonium in the Baltic Sea determined by shipboard HPLC
analysis.  Limnol. Oceanogr. 27: 336-347.

Sokal, R. R., Rohlf, F. J. 1981.  Biometry, 2nd Edition.  W. H. Freeman &
Co., San Francisco.