Biol Trace Elem Res (2011) 141:76–85

DOI 10.1007/s12011-010-8720-3

Chromatography and Atomic Absorption Spectrometry

for the Assessment of Heavy Metal Distribution
among Amino Acids Used in Parenteral Nutrition
Formulations—Studies with Cadmium and Lead

Paulo C. do Nascimento & Marieli S. da Marques &

Denise Bohrer & Leandro M. de Carvalho &
Claudia W. Carvalho

Received: 6 January 2010 / Accepted: 26 April 2010 /

Published online: 25 May 2010
# Springer Science+Business Media, LLC 2010

Abstract The distribution of Cd (II) and Pb (II) among amino acids in parenteral nutrition
formulations was investigated by coupling ion-chromatography (HPLC/IC) and electro-
thermal atomic absorption spectrometry. The methodology was based on ion-exchange
separation and fluorimetric amino acid detection after post-column derivatization. Cd (II)
and Pb (II) were assayed in 500-µL fractions of the column effluent. The distribution of Cd
(II) and Pb (II) in alanine (Ala), aspartic acid (Asp), glutamic acid (Glu), glycine (Gly),
histidine (His), methionine (Met), phenylalanine (Phe), serine (Ser), and threonine (Thr)
were analyzed by monitoring changes in the concentration of free amino acids by HPLC/IC.
The results indicated that Cd (II) and Pb (II) were distributed according to the following
trend: Gly–Cd > Gly–Pb > Ala–Cd > Ala–Pb > His–Cd ∼ His–Pb > Thr–Cd > Thr–Pb > Phe–
Cd∼Phe–Pb∼Asp–Cd∼Asp–Pb∼Met–Cd∼Met–Pb∼Glu–Cd∼Glu–Pb>Ser–Cd∼Ser–Pb.
The effects of amino acid concentration and stability constants of amino acid–metal
complexes are discussed.

Keywords Parenteral nutrition . Amino acids . Cadmium . Lead . Contamination

Introduction

Lead and cadmium are toxic to human health and cause deleterious effects by accumulating
in target organs [1, 2]. External exposure to these metals is often caused by pollution and
occurs in many parts of the world. Moreover, internal exposure to lead and cadmium can be
caused by industrial processing or product storage. These methods of exposure require
further investigation; specifically, pharmaceutical products containing species that form

P. C. do Nascimento (*) : M. S. da Marques : D. Bohrer : L. M. de Carvalho : C. W. Carvalho

Departamento de Química, Universidade Federal de Santa Maria, 97111-970 Santa Maria, RS, Brazil
e-mail: pcn1954@gmail.com
Chromatography and Atomic Absorption Spectrometry 77

complexes with metals should be studied in detail. Despite the strict quality control
adopted by the pharmaceutical industry, studies on the interactions between containers
and drugs, as well as the distribution of contaminants in drug formulations, are necessary
[3, 4]. Indeed, many substances used in drug formulations are good leaching agents toward
unwanted species present in containers. A prime example is the leaching of aluminum into
amino acid parenteral nutrition (PN) formulations [5–7]. These formulations are particularly
important because they contain amino acids in high concentrations. Amino acids show
different affinities toward metallic species, as well as different kinetics for complex
formation.
Some researchers have suggested that biologically important interactions can occur
between bioelements and toxic metals with similar physical and chemical properties [8].
Indeed, the ability of amino acids to complex metals can cause their spread and
accumulation into organs, which results in the development of histological abnormalities
in patients under PN treatment [5, 8–10]. Amino acids are classified as small ligands
because they have at least one amino and one carboxyl group. Stability constants for the
formation of complexes between amino acids and metallic species are available in the
literature [11]. The amino acids with the highest stability constants for Cd (II) and Pb (II),
as well as the highest concentrations in PN samples, are Ala, Asp, Glu, Gly, His, Met, Phe,
Ser, and Thr. Thus, the interactions of these amino acids with Cd (II) and Pb (II) were
investigated in this study.
Among the available methods for the simultaneous determination of amino acids, high
performance liquid chromatography (HPLC) is the only technique that offers a large
number of instrumental options [12], including separation by reversed-phase and ion-
exchange columns and the resolution of complex mixtures of amino acids in a short period
of time [13, 14]. Electrochemical detectors are frequently used to assay amino acids directly
[15, 16], while UV and fluorimetric detectors are commonly used after derivatization with
ninhydrin, ortophthalaldehyde (OPA), or dansyl chloride [17–20].
In the present work, interactions between amino acids (Ala, Asp, Glu, Gly, His, Met,
Phe, Ser, and Thr) and metallic ions were evaluated by monitoring the change in the
chromatographic signal of amino acids after reaction with Cd (II) and Pb (II). Fluorimetric
detection of derivitized amino acids with the OPA reagent and ion-exchange separation
were used to determine interactions between the metal and various amino acids.

Materials and Methods

A DX-300 gradient chromatography system (Dionex, Sunnyvale, USA) with a 1100 series
fluorescence detector (Hewlett Packard, Waldbronn, Germany), a GradiFrc fraction
collector (Pharmacia, Uppsala Sweden), and an extra LC-IOAS chromatographic pump
(Shimadzu, Kyoto, Japan) were utilized in this study. To analyze results, a C-R6A data
processor (Shimadzu, Kyoto, Japan) was used. Atomic absorption spectrometry (AAS)
measurements were conducted on a SpectrAA-200 equipped with a GTA-100 graphite
furnace and an autosampler (Varian, Melbourne, Australia). A class 100 clean bench system
(Trox, Curitiba, Brazil), BSB 939-IR sub-boiling distillation apparatus (Berghof, Eningen,
Germany), and a D-20 pH-meter (Digimed, São Paulo, Brazil) were used to prepare and
manipulate sample solutions. All glassware (pipette tips, volumetric flasks, etc.) was
immersed for at least 24 h in a 10% (v/v) HNO3/ethanol solution and washed with Milli-Q
water shortly before use. HNO3 (65%, 1.17 gmL-1), obtained from Merck (Darmstadt,
Germany) was purified by sub-boiled distillation.
78 do Nascimento et al.

Reagents

Aqueous solutions were prepared with deionized water purified by a Milli-Q high-purity
water system (Millipore, Bedford, USA). Metal stock solutions containing 1,000 mg L-1
were prepared from Cd(NO3)2·4H2O and Pb(NO3)2, while working standard were obtained
by dilutions of stock solutions. All chemicals were analytical–reagent grade and purchased
from Merck. Ala, Asp, Glu, Gly, His, Met, Phe, Ser, and Thr (biochemical grade, 99%
pure) were obtained from Sigma (St Louis, USA). To achieve separation, the HPLC column
was eluted with two mobile phases (MP) consisting of 23 mM NaOH and 7 mM Na2B4O7
(MP I) and 0.4 M CH3COONa and 1 mM NaOH in 2% (v/v) methanol (MP II). All chemicals
used in the preparation of MP were purchased from Merck. The derivatization reagent was
prepared by mixing 2.5 g o-phthaldialdehide (99% pure, Aldrich, St. Louis, USA) and 20.12 g
Na2B4O7·10H2O in 800 mL water and 50 mL methanol. The volume was completed to 1 L
with water; the solution was filtered through a cellulose acetate membrane (0.2 μm; Sartorius,
Göttingen, Germany), and 2.5 mL of ethanethiol (99% pure; Merck) was added to obtain an
o-phthaldialdehide-ethanethiol (OPA-SH) solution. This solution was prepared daily.

Procedures

Interactions between metallic ions and amino acids were studied in artificial PN solutions
following the composition observed in the commercial products Aminon 20 (JP Indústria
Farmacêutica, São Paulo, Brazil), Soramin 10% (Darrow, Rio de Janeiro, Brazil), and
Aminoplasmal L-10 (B. Braun, São Paulo, Brazil). From the amino acids present in
commercial formulations, Ala, Asp, Glu, Gly, His, Met, Phe, Ser, and Thr were selected for
this study because they take part in the majority of commercial PN solutions. Before the
addition of Cd (II) and Pb (II) to the amino acid solutions, natural contamination by
cadmium and lead was checked by AAS measurements. To avoid the contribution of kinetic
effects, cadmium and lead were added to the amino acid solutions 24 h (incubation time)
before measurements were conducted. The solutions were maintained at 24±0.5°C under
moderate stirring and were protected from direct light. A 1:1 molar ratio of metal to amino
acid was used throughout the study.

Amino Acid and Metal Interaction

Solutions containing 10 µM of the selected amino acids and Cd (II) or Pb (II) were injected
onto the chromatographic system. The amino acids were separated and detected. The
effluent was collected in 500-µL fractions, and the concentration of metals was evaluated
by electrothermal atomic absorption spectrometry (ETAAS). AAS operation parameters for
both metals are provided in Table 1.
Solutions containing 10 µM of Cd (II)- or Pb (II)-amino acid complexes were
successively injected over 8 h at 2-h intervals to assess peak intensities and complex
stability during chromatographic measurements.

Chromatographic System

To separate amino acids, 25 μL of each sample was injected onto an Amino Pac PA1 (250×
4 mm; 9.0 μm) column (Dionex) under an elution gradient. For the first 3 min, the mobile
phase consisted of 100% MP I. Next, MP II was introduced and reached a concentration of
30% by 4 min. This concentration was maintained from 4 to 6 min. Over the next 11 min,
Chromatography and Atomic Absorption Spectrometry 79

Table 1 Atomic Absorption Spectrometer Operating Conditions

Instrument Cadmium Lead

Wavelength (nm) 228.8 283.3

Lamp current (mA) 4 5
Spectral slit width (nm) 0.5 0.5
Background correction Deuterium lamp Deuterium lamp
Sample volume (μL) 10 10
Temperature program
Step Cadmium Lead
Temperature Time Gas flow Temperature Time Gas flow
(°C) (s) (L min-1) (°C) (s) (L min-1)
1 85 5.0 3.0 85 5.0 3.0
2 95 40.0 3.0 95 40.0 3.0
3 120 10.0 3.0 120 10.0 3.0
4 250 5.0 3.0 400 5.0 3.0
5 250 1.0 3.0 400 1.0 3.0
6 250 2.0 0 400 2.0 0
7 1,800 0.8 0 2,100 1.0 0
8 1,800 2.0 0 2,100 2.0 0
9 1,800 2.0 3.0 2,100 2.0 3.0

Graphite furnace pyrolitic coated furnace with L´vov platform and argon as purge gas

the concentration of MP II was further increased. By 17 min, the MP was composed of 70%
MP II and 30% MP I. Next, the mobile phase was increased to 100% MP II over 1 min, and
this concentration was maintained for the next 5 min. Lastly, MP I was again introduced,
and the mobile phase was increased to 100% MP I over 1 min. The flow rate of the mobile
phase was 0.8 mL min-1. Using a three-port valve, the column effluent and OPA-SH
solution (flow rate 0.2 mL min-1) were combined and carried to the fluorescence detector.
The excitation and emission wavelengths of 330 and 455 nm, respectively, were monitored
throughout the separation.

Fluorescence Detection

To assure that modifications in peak absorption were the result of metal–amino acid
complex formation, spectra were collected with two types of solutions. Specifically, one
solution was prepared by combining 1 mL of amino acid solution (10 μM), 1 mL of water,
and 1 mL of OPA-SH solution. Alternatively, the other solution contained 1 mL of Cd (II)
or Pb (II) solution (10 μM) in place of water. Blank solutions consisted of 2 mL of water
and 1 mL of OPA-SH solution. The fluorescence intensity of solutions containing Cd (II) or
Pb (II) was measured 1 h after preparation.

Results and Discussion

Transportation, accumulation, and absorption of trace metals in animals are related to the
toxicity of the metal. Thus, the processes involved in digestion and absorption of food
(including PN formulations) may be affected by the ingestion of heavy metals as amino
80 do Nascimento et al.

acid complexes. The use of natural transport systems by non-essential or toxic metals such
as cadmium and lead may explain the toxicity of these metals [21]. The relative
accumulation and distribution of trace metals in different tissues is rather difficult to
model, in part, due to the difficulties in obtaining experimental data. Thus, studying
quantitative changes in free and complexed amino acids can provide the relative affinity of
amino acids and target metal ions. Among the amino acids present in PN formulations, nine
were selected as representatives for the present study. Commercially available PN
formulations possess a wide range of amino acids in varying concentrations. Identifying
all possible interactions between metallic contaminants and amino acids of PN formulations
is difficult; therefore, amino acids with high stability constants for complexation with Cd
(II) and Pb (II), as well as high concentrations in commercially available PN formulations,
were selected for the present study.
The amino acids investigated in this study were separated and detected after a 30 min of
chromatographic run with two mobile phases and gradient elution. Despite differences in
chemical composition, well-shaped chromatographic peaks were obtained. Table 2 displays
the calibration parameters including calculated limits of detection and quantification. The
latter were based on the basis line standard deviation (σ), where 3σ and 10σ criteria [22]
were considered, respectively.
Post-column derivatization of amino acids with OPA-SH solution is a useful alternative
for fluorimetric amino acid detection [13]. The reagent is stable; the reaction is rapid at
room temperature, and the sensitivity of the fluorimetric measurement is approximately ten
times higher than with ninhydrin, the most popular reagent for amino acid detection.
However, in this study, it was necessary to assess competition between Cd (II) or Pb (II)
and OPA-SH for complexation with amino acids. To this end, two sets of experiments were
conducted. In the first test, OPA-SH was added to a mixture of amino acids with Cd (II) or
Pb (II), and the resultant fluorescence was compared to similar solutions without Cd (II) or
Pb (II). In the second experiment, spectra of solutions containing Cd (II) or Pb (II), amino
acids and OPA-SH were obtained to determine if the metals altered the maximum
wavelength of the amino acid derivatives, which would lead to an erroneous quantification
of free amino acid available to react with OPA-SH. Figure 1 shows the spectra of Glu–
OPA-SH solution with and without Cd (II) or Pb (II). Due to the interactions between Cd
(II) or Pb (II) and the amino acids, a significant reduction in absorbance was observed at
330 nm. However, in presence of Cd (II) or Pb (II), changes in shapes of spectra were not
observed. Identical behavior was observed for all investigated amino acids.

Table 2 Calibration Parameters for the Amino Acid Chromatographic Determination

Amino acid Linear range (µM) Calibration function r2 LOD LOQ

Ala 0.01–50.0 y ¼ 27398x  214:6 0.9999 6×10−5 3×10−4

Asp 0.05–50.0 y ¼ 24124x  652:6 0.9997 3×10−1 1.88
Glu 0.01–50.0 y ¼ 31403x þ 122:4 0.9997 7×10−6 4×10−5
−3
Gly 0.10–50.0 y ¼ 7773:0x þ 720:0 0.9992 8×10 5×10−2
−1
His 0.10–30.0 y ¼ 6218:3x þ 576:2 0.9992 2×10 1.24
Met 0.01–50.0 y ¼ 26293x  519:2 0.9996 2×10−1 1.58
Phe 0.01–30.0 y ¼ 17291x þ 284:6 0.9997 3×10−1 2.06
Ser 0.10–30.0 y ¼ 8446:2 þ 362:4 0.9999 2×10−1 1×10−2
Thr 0.05–50.0 y ¼ 20358x  121:8 0.9999 4×10 -1
2.98
Chromatography and Atomic Absorption Spectrometry 81

Fig. 1 Absorption spectra. Curve 1: Glu–OPA-SH; curve 2: Glu–OPA-SH and Pb (II); curve 3: Glu–OPA-
SH and Cd (II)

The reduction caused by Cd(II) and Pb(II) in the fluorescence intensity of the amino
acid–OPA-SH species remained constant over the entire measurement (1 h). Indeed, OPA-
SH did not compete for the coordination sites of amino acids bound to metallic ions. Thus,
the method can be used to quantify amino acid–metal binding.

Amino Acid and Metal Interaction

After the addition of Cd (II) or Pb (II) to amino acid solutions, the samples were injected
onto the column, and the effluent was collected in 500-µL fractions over the entire
chromatographic run. The concentration of metallic ions in each fraction was measured
with ETAAS. The recovery of metals in the fractions collected before the elution of the first
analyte was 100.2%±0.3 and 102.3%±1.1 of Cd (II) and Pb (II), respectively. This
behavior was expected because the amino acids were separated on an anionic column, and
the metal–amino acid species are predominantly neutral, which results in weak interactions
with the stationary phase. The observed reduction in the amino acid peak height was
attributed to the formation of a non-fluorescent amino acid–metal–OPA-SH compound;
thus, the reduction in the amino acid signal corresponds to the fraction of amino acid bound
to the metal. Moreover, the injection of samples with and without Cd (II) (or Pb (II))
showed that the complexes were not dissociating during HPLC separation.
Figure 2a shows the signals of standard samples containing the amino acids, while
Fig. 2b, c show the chromatogram of samples spiked with Cd (II) and Pb (II). A comparison
of the chromatograms indicates that the presence of metal decreased the signals of the
amino acids.

Amino Acid and Metal Interaction in PN Samples

Artificial PN solutions containing the investigated amino acids were prepared covering the
compositions of commercial PN formulations selected for this study (see Procedures). To
quantify amino acids without overloading the chromatographic system, six working PN
82 do Nascimento et al.

Fig. 2 Chromatographic signals a amino acid OPA-SH; b Pb (II) and amino acid OPA-SH; c Cd (II) and
amino acid OPA-SH. 1 Ala; 2 Gly; 3 Ser; 4 Met; 5 Thr; 6 His; 7 Phe; 8 Glu; 9 Asp. All solutions were
10 μM in metal and amino acid. Chromatographic conditions: see text

samples (Table 3) were prepared by diluting stock solutions with water. The pH of the
solutions was between 6 and 7. The calibration functions displayed in Table 2 were used for
the quantification of amino acids in working PN samples. The chromatographic signals of
the amino acids were identified by comparing the retention times of amino acids in standard
and spiked solutions. These samples were previously evaluated by ETAAS for trace
amounts of Cd (II) and Pb (II).
To assess the distribution of Cd (II) and Pb (II) among the amino acids, the metals were
added separately to working solutions of PN at a 1:1 molar ratio to follow the reductions in
the chromatographic peaks caused by interactions of Cd (II) and Pb (II) with the amino
acids The ratio of amino acids in the working PN samples was identical to those of the
commercial PN solutions. The aim of this study was not to follow the real contamination of
PN samples by cadmium and lead but instead to assess the distribution of metals among

Table 3 Amino Acid Parenteral Nutrition Working Solutions and Stability Constants

Amino acid concentration (µM)

Sample Ala Asp Glu Gly His Met Phe Ser Thr

A 13.53 0.00 0.00 19.50 0.00 1.37 1.37 0.00 9.27

B 37.39 2.57 7.55 25.87 8.57 6.00 7.96 5.92 8.25
C 29.26 5.35 4.27 26.64 7.69 9.04 7.81 5.99 18.18
D 38.61 2.45 8.07 26.31 8.78 6.58 7.51 5.51 8.79
E 39.48 0.00 0.00 48.11 5.08 7.82 7.82 0.00 9.56
F 20.28 0.00 0.00 13.09 6.75 6.87 6.87 22.14 10.64
Average 29.76 1.73 3.32 26.59 6.15 6.28 6.56 6.59 10.78
Log kstb Cd 3.88 4.30 4.02 4.28 5.39 3.69 4.51 4.33 3.54
Log kstb Pb 5.43 6.08 5.57 5.63 5.95 4.38 4.01 4.66 4.74
Chromatography and Atomic Absorption Spectrometry 83

amino acids; thus, the concentration of cadmium and lead were adjusted to maintain a 1:1
molar ratio of amino acids to metals. Figure 3 displays the relative reduction in the
concentration of free amino acids (amino acid–OPA-SH species) caused by interactions
with cadmium and lead. The original concentrations obtained from the chromatograms were
normalized to the highest value of each amino acid and scaled by a factor of 1/SD, where
SD is the standard deviation of the normalized values of each amino acid concentration.
The transformation of the original data set [23] was made to allow for comparisons between
samples because the concentration of amino acids varied widely. The bars in Fig. 3
represent the extent of interaction between metals and amino acids present in each sample.
The distribution of Cd (II) and Pb (II) among the amino acids displayed the following trend:
Gly–Cd>Gly–Pb>Ala–Cd>Ala–Pb>His–Cd∼His–Pb>Thr–Cd>Thr–Pb>Phe–Cd∼Phe–
Pb∼Asp–Cd∼Asp–Pb∼Met–Cd∼Met–Pb∼Glu–Cd∼Glu–Pb>Ser–Cd∼Ser–Pb.
The stability constants (kstb) of metal–amino acid complexes [11] cover ca. three decades
(Table 3). Specifically, the log kstb ranges from 3.54 to 5.39 for Cd (II)–amino acid
complexes and 4.01 to 6.08 for Pb (II) complexes. It was assumed that the more stable
complexes would be related to a larger reduction in the concentration of free amino acid
(Fig. 3). However, the Cd (II) and Pb (II) distributions among the amino acids did not fit the
relative order of stability constants. For instance, the highest log kstb (6.08) was observed
for the Pb (II)–Asp complex, and the lowest log kstb (3.54) was observed for the Cd (II)–Thr
complex. These complexes, however, did not provide the largest or smallest change in free
amino acid concentration as showed by the bars 6 and 15 (Fig. 3) for all samples. The
relationship between the concentration of metal and amino acid was related to the
distribution of metal in amino acid solutions. Considering the average amino acid
concentrations in the six working PN samples and the stability constants of metal–amino
acid complexes (Table 3), metal distribution is mainly dependent on the amino acid

Fig. 3 Relative decrease in the concentration of free amino acids. 1 Cd–Met; 2 Pb–Met; 3 Cd–Phe; 4 Pb–
Phe; 5 Cd–Asp; 6 Pb–Asp; 7 Cd–Gly; 8 Pb–Gly; 9 Cd–Glu; 10 Pb–Glu; 11 Cd–Ser; 12 Pb–Ser; 13 Cd–His;
14 Pb–His; 15 Cd–Thr; 16 Pb–Thr; 17 Cd–Ala; 18 Pb–Ala. Composition of amino acid samples (A–F): see
Table 3
84 do Nascimento et al.

concentration, while values of kstb showed a secondary effect. On average, Gly and Ala
were present in the highest concentration and showed the largest reduction in signal as showed
by the bars 7, 8 and 17, 18, respectively (Fig. 3). However, both Gly and Ala did not possess
the highest values of kstb. Similarly, Asp and Glu possessed high values of kstb for Cd (II) and
Pb (II) complexes; however, low reductions in the concentration of free amino acid were
observed. Thus, the observed reduction in amino acid concentration may have been the result
of a low initial concentration in PN samples. The kstb of Thr complexes were similar to Met
for Cd (II) but were higher than the kstb of Pb (II)–Met. However, the initial concentration of
amino acid appeared to be the determining factor in the reduction of free Thr due to
interactions with metal ions. His, Met, Phe, and Ser were present in similar concentrations;
thus, the kstb values could be used to explain the observed reductions in free amino acid
concentrations. Considering complexes of Cd (II) and Pb (II) with an identical amino acid, the
values of kstb were generally higher for Pb (II) than Cd (II). However, the reduction in free
amino acids caused by interactions with Cd (II) was higher or similar to Pb (II).

Conclusion

Anion-exchange chromatography of amino acids coupled with fluorescence detection is a

useful tool for the investigation of interactions between metals and amino acids. The
reduction in chromatographic peaks of amino acids caused by interactions with Cd (II) and
Pb (II) was used to assess the distribution of metal ions among amino acids. The stability
constants of metal–amino acid complexes did not determine the distribution of metal among
amino acids. The concentration of amino acids in solution was the most important factor for
the distribution of metal. Thus, the results of this study indicated that Cd (II) and Pb (II) can
interact with the amino acids of PN solutions under conditions that they are delivered to the
patients.

Acknowledgements The authors express their gratitude to CAPES (Probral N° 240/06) and CNPq for
financial support.

References

1. Harraki B, Guiraud P, Rochat MH et al (1995) Interactions related to trace elements in parenteral

nutrition. Pharm Acta Helv 70:269–278
2. Buchman AL, Neely M, Bruce Grossie V et al (2001) Organ heavy-metal accumulation during parenteral
nutrition is associated with pathologic abnormalities in rats. Nutrition 17:600–606
3. Nascimento PC, Marques MS, Hilgemann M et al (2006) Simultaneous determination of cadmium,
copper, lead and zinc in amino acid parenteral nutrition solutions by anodic stripping voltammetry and
sample digestion by UV irradiation. Anal Lett 39:777–790
4. Petroianu GA, Kosanovic M, Shehatta IS (2004) Green coconut water for intravenous use: trace and
minor element content. J Trace Elem Exp Med 17:273–282
5. Pluhator-Murton MM, Fedorak RN, Audette RJ et al (1999) Trace element contamination of total
parenteral nutrition. 1. Contribution of component solutions. JPEN J Parenter Enteral Nutr 23:222–227
6. Bohrer D, Nascimento PC, Pomblum S et al (2002) Contribution of the raw material to the aluminum
contamination in parenterals. J PEN 26:383–388
7. Brzóska MM, Moniuszko-Jakoniuk J (2001) Interactions between cadmium and zinc in the organism.
Food Chem Toxicol 39:967–980
8. Leung FY, Grace DM, Alfieri MAH et al (1995) Abnormal trace-elements in a patient on total parenteral-
nutrition with normal renal-function. Clin Brioche 28:297–302
Chromatography and Atomic Absorption Spectrometry 85

9. Leung FY, Grace DM, Alfieri MAH (1998) Selenium and zinc levels in surgical patients receiving total
parenteral nutrition. Biol Trace Elem Res 61:33
10. Mahaffey KR (1984) Toxicity of lead, cadmium and mercury-considerations for total parenteral
nutritional support. Bull N Y Accad Med 60:196–209
11. Martell A E, Smith R M, Motekaitis R J (1998) Critically selected stability constants of metal complexes
database. Version 5.0, National Institute of Standards and Technology NIST, Gaithersburg
12. Holme DJ, Peck H (1998) Analytical Biochemistry, 3rd edn. Prentice Hall, England
13. Herraiz T (1997) Sample preparation and reversed phase-high performance liquid chromatography
analysis of food-derived peptides. Anal Chim Acta 352:119–139
14. Castelain S, Kamel S, Picard C et al (1995) A simple and automated HPLC method for determination of
total hydroxyproline in urine. Comparison with excretion of pyridinolines. Clin Chim Acta 235:81–90
15. Leroy P, Barbaras M, Colin JL et al (1995) Ion-exchange liquid chromatography method with indirect
UV detection for the assay of choline in pharmaceutical preparations. Pharm Biomed Anal 13:581–588
16. Marrubini G, Caccialanza G, Massolini G (2008) Determination of glycine and threonine in topical
dermatological preparations. J Pharm Biomed Anal 47:716–722
17. Fekkes DJ (1996) State-of-the-art of high-performance liquid chromatographic analysis of amino acids in
physiological samples. J Chromatogr B 682:3–22
18. Bohrer D, Nascimento PC, Mendonça JK et al (2004) Interaction of aluminium ions with some amino
acids present in human blood. Amino Acids 27:75–83
19. Rigobello-Masini M, Penteado JCP, Liria CW et al (2008) Implementing stepwise solvent elution in
sequential injection chromatography for fluorimetric determination of intracellular free amino acids in
the microalgae Tetraselmis gracilis. Anal Chim Acta 628:123–132
20. Zhao WC, Liu LJ, Zhao XE et al (2008) Application of 2-(11H-benzo[a]carbazol-11-yl) ethyl
carbonochloridate as a precolumn derivatization reagent of amino acid by high performance liquid
chromatography with fluorescence detection. Chin J Anal Chem 36:1071–1076
21. King LM, Banks WA, George WJ (2000) Differential zinc transport into testis and brain of cadmium-
sensitive and -resistant murine strains. J Androl 21:656–663
22. Massart DL, Vandginste BGM, Deming SN, Michotte Y, Kaufman L (1988) Chemometrics: a textbook,
1st edn. Elsevier, Amsterdam
23. Esbesen K, Schönkopf S, Midtgaard T, Guyot D (1998) Multivariate analysis in practice, 3rd edn. Camo
ASA, Norway