Journ31l oi un.nder.r.3du.3- Re--search
,,Oluiie ', issue 1 - SepiemI-ier .. Oc:oer :i:u
Determination of the Extent of Atkylation in Half-Mustard Treated Albumin
John Bradford Hill
Sulfur mustard (bis-2-chloroethyl sulfide, SM) is a powerful vesicant that has been effectively used as a
chemical weapon for nearly a century [Figure 1]. SM rapidly alkylates biological molecules, resulting in damage
to epithelial tissues such as the skin, eyes and airways, commonly resulting in painful blisters, edema,
tissue necrosis, and severe conjunctivitis. Unlike many other chemical weapons, SM injuries are rarely fatal.
Rather, SM cripples a target through extensive, incapacitating and horrific injuries. However, SM damage
takes hours to manifest, so there is a window of opportunity for aggressive treatment to limit the severity of
injury1. This illustrates the critical need for rapid, diagnostic detection of SM adducts in a human system.
A component of this comprehensive goal is the development of an assay for analyzing SM alkylated
proteins. Handling of "full" SM is highly regulated, so this experiment uses alkylation with "half
mustard" (2-chloroethyl ethyl sulfide, CEES) [Figure 2].
Figure 1. Structure of "half" sulfur mustard (SM).
Figure 2. Structure of "full" mustard (CEES).
Human serum albumin (hSA) is a particularly ideal candidate for such an assay due to its systemic abundance
and extracellular presence. Furthermore, hSA possesses only one free sulfhydryl group at Cys34, which is located
in a somewhat internal region of the protein [Figure 3]. This sulfhydryl group is an ideal candidate for
alkylation upon treatment with CEES, and a method of discerning between the free and alkylated forms can be
used to indicate the extent of alkylation from CEES exposure. Additionally, hSA contains many surface residues
(his, asp, arg, gin, glu) shown to be susceptible to CEES alkylation2. The free sulfhydryl of Cys34 competes
with these surface residues for CEES. Therefore, the alkylation of the single, free sulfhydryl group indicates
a likelihood of extensive alkylation to other, more readily accessible, residues.
Figure 3. Three-dimensional structure of hSA with the cys34 amino acid shown in red and the
free sulfhydryl group in yellow. Disulfide bridges are illustrated in orange3.
The assay designed in this experiment detects the presence of free sulfhydryl groups through the use of
Ellman's reagent, 5, 5'-dithiobis (2-nitrobenzoic acid) (DTNB) [Figure 4]. DTNB reacts in solution with free
sulfhydryl groups, forming a thionitrobenzoate-protein complex and liberating a thionitrobenzoate anion [Figure
5]. This anion exhibits a vivid yellow color that absorbs maximally at 412nm4. One anion is liberated for
every reacted sulfhydryl, and each hSA protein contains one free sulfhydryl, so this 1:1:1 molar ratio provides
a simple relationship between the absorbance of the DTNB reaction product near 412nm and the presence of
hSA containing free, non-alkylated sulfhydryls.
0 0 1
Figure 4. Structure of ElIman's reagent (DTNB).
( -SH + NO,-- N-S-S / NO
0 S-S /N02+ -S NO,
COO- + H+ COO-
Figure 5. DTNB Reaction with a Free Sulfhydryl Group4.
Albumin containing alkylated sulfhydryl groups as a result of CEES exposure cannot react with DTNB and will
not produce the yellow anion. Therefore, DTNB allows for spectrophotometric quantification and comparison
between hSA and CEES-reacted hSA (CEES-hSA). It is possible, therefore, to construct a standard curve
by measuring absorptions following DTNB treatment of known solutions with varying molar ratios of pure hSA
to CEES-hSA. This curve may then serve as a guide for determining extent of hSA alkylation following
CEES exposure under various conditions, namely a dose response.
This experiment addresses the creation of and utilization of such an assay for assessing the extent of alkylation
to hSA from CEES exposure. Particularly, this assay was used to study the impact of CEES dose on the extent of
hSA alkylation as indicated by the prevalence of remaining free sulfhydryl groups.
MATERIALS AND METHODS
All CEES reactions were conducted in a fume hood with appropriate protective equipment and in the vicinity
of colleagues for assistance in the unlikely event of an accident.
Ellman's Reagent (DTNB) Stock
A stock of 99.99% DTNB was obtained from Sigma-Aldrich as a fine, yellow powder. Sufficient DTNB was added
to 10 mL of lx phosphate buffered saline (PBS) at pH 7.0 until saturation was exceeded. The solution was
repeatedly mixed and vortexed approximately every 5 minutes for one hour to ensure complete saturation.
The DTNB solution, which displayed a slightly yellow hue, was filtered utilizing a syringe driven filter unit with a
pore size of 0.22pm to remove all solids. The DTNB solution was finally divided into imL aliquots and stored in
a freezer at -300C.
Preparation of Theoretically 100% Alkylated CEES-hSA
Five 100pL solutions of 5mg/mL hSA in pH 9.0 PBS were prepared as a CEES treatment stock. For a control
group, several 100pL solutions containing 5mg/mL hSA without CEES additions were subjected to all other
reaction conditions. 20pL of 85.85mM CEES was added to each reaction tube to attain an approximate CEES to
hSA ratio of 40pL/mg. Prior experimentation suggested that this ratio produced theoretically 100% alkylated
hSA with the most reproducible results. The CEES solution is moderately immiscible, so each sealed reaction
tube was inverted several times to ensure adequate dispersion.
The solutions were subsequently incubated at 370C for one hour using a dry block heater. The reactions
were quenched by a 300pL addition of 1M glycine at pH 9.0, and the mixtures were left to incubate at 370C for
an additional hour. After quenching, each sample was ultra-filtered and resuspended in pH 7.0 PBS three times
with a 10kDa cut-off micron column to isolate the protein from any residual CEES, glycine and CEES derivatives.
The samples were finally suspended in PBS and combined to attain a solution with a final volume of 250pL with
an approximate protein concentration of 10mg/mL. The initial reaction volumes were halved in the final
suspension in order to double the initial protein concentration and thus enhance resolution and absorption
during future DTNB additions.
Spectrophotometry for the Standard Curve
Absorption readings used a 96 well plate spectrophotometer and all measurement solutions had final volumes
of 100pL. One plate well was left to contain only pH 7.0 PBS and served as a reference. Five plate wells
contained solutions with the following volume to volume percentages of the previously prepared 10mg/mL CEES-
hSA to the control, 10mg/mL non-reacted hSA: 100%, 25%, 50%, 25% and 0% CEES-hSA. For example, the
25% solution contained 25pL of CEES-hSA and 75pL of control hSA. Absorption readings were taken at 280nm
to control for protein concentrations and at 405nm to obtain a background. The spectrophotometer was set
to 405nm rather than the thionitrobenzoate anion's maximum absorption of 412nm because the device lacked a
filter at this wavelength. Nonetheless, 405nm resides close enough to the maximum absorption of 412nm to
provide adequate absorption data. After these preliminary readings, 10uL of the DTNB solution was added to
each well, including the well containing only PBS. Several absorbance measurements were taken over regular
time intervals until a maximum absorbance was obtained for each solution.
Dose Response CEES Treatments
Eight 200pL solutions of 2mg/mL hSA in ix PBS at pH 9.0 were prepared for dose response measurements of
various CEES treatments, which were made from the 85.85mM stock CEES solution. The procedure followed that
of the "Theoretically 100% Alkylated CEES-hSA" preparation, and various additions were made according to
the schedule depicted in Table 1.
Dose response treatments
CEES Addition (pL)
CEES:hSA (molar ratio)
All reactions were incubated, quenched, and filtered as described above. The treatments were finally suspended
and stored in 200pL of ix PBS at pH 7.0.
Spectrophotometric Analysis of Dose Response Treatments
Each reaction was measured using the same 96 well plate spectrophotometer and following a similar procedure.
The noteworthy procedural changes were the use of protein concentrations of approximately 2mg/mL and
the addition of only 5pL of the DTNB solution due to the lower protein concentration.
DATA AND RESULTS
All absorbance data was controlled for the background absorbance at 405nm prior to the DTNB addition as well
as protein concentrations (based on absorbance at 280nm). The adjusted absorptions were finally
standardized against the absorbance of the solution containing 100% non-reacted hSA (Table 2).
Absorbance data for the generation of a standard curve.
0.00 0.000 0.232 1.800 0.129 1.000
0.25 0.213 0.197 1.702 0.116 0.898
0.50 0.455 0.161 1.593 0.101 0.784
0.75 0.706 0.144 1.540 0.0935 0.725
1.00 1.000 0.114 1.450 0.0786 0.610
Note: The actual fraction of CEES-hSA present in each solution was calculated from the
absorption data at 280nm. The absorbance readings provided in this table have been adjusted
to account for their background readings prior to DTNB additions. The adjusted absorptions,
Abs(adj), are controlled for concentration through division by the absorption at 280nm. The
standardized absorptions are obtained from adjusted absorption values through division by the
adjusted absorption of the pure, non-CEES reacted hSA. This standardization allows for
comparison to trials subjected to different conditions and with different protein concentrations.
These standardized and adjusted absorbance values were plotted against the calculated fraction of CEES-hSA
present in each well, as determined from the absorbance at 280. The resulting plot appears below in Figure 6.
Rel Abs(adj) = (-0.3805)(Fraction CEES-hSA) + 0.9842
._ R Square = 0.9876
0.25 0.5 0.75
Calculated Fraction Composition of CEES-hSA
Figure 6. Standard curve generated from DTNB additions to solutions of varying CEES-hSA to control
hSA fractions. The demonstrated relationship is clearly linear and the regression line, with equation,
is depicted on the graph.
Dose Response CEES Treatments
The absorption data from each of the dose treatments were adjusted for background and concentration, and
then standardized as a fraction of the absorption of pure hSA, as calculated similarly for the standard
curve preparation (Table 3).
Absorbance data and calculations from CEES dose response trials, adjusted for background values,
and other conditions as in Table 2.
CEES:has (pL/ CEES:hSA Abs(405nm) Abs(280nm) Abs(ad) Standardized
mg) (molar ratio) Abs(405nm) Abs(280nm) Abs(adj) Absadj
mg) (molar ratio) Abs(adj)
0.00 0.000 0.089 1.353 0.06578 1.000
1.25 7.083 0.081 1.335 0.06067 0.9224
2.50 14.17 0.078 1.331 0.05860 0.8909
5.00 28.33 0.068 1.337 0.05086 0.7732
10.0 56.67 0.062 1.322 0.04690 0.7130
12.5 70.83 0.059 1.318 0.04477 0.6805
17.5 99.16 0.055 1.309 0.04201 0.6387
25.0 141.7 0.054 1.311 0.04119 0.6262
As demonstrated in Figure 7, the relationship between CEES dose and absorption following DTNB treatment
appears to be exponential. The strength of this relationship is supported by the strength of the
exponential regression curve in a plot where the horizontal and vertical axes are reversed (Figure 8).
Dose Response (a)
0 2 4 6 8 10 12 14 16 18 20 22 24 26
uLCEES mg hSA
Dose Response (b)
0 20 40 60 80 100
Molar Ratio [CEES hSA)
Figure 7. Scatter plot charts of dose response data. (a) Standardized absorption data plotted against
the CEES:hSA volume to weight ratio. (b) Standardized absorption data plotted against the molar ratio
of CEES to hSA. Theoretically, this depicts the number of CEES molecules per hSA protein. Note that
the absorption appears to decrease exponentially with CEES additions.
Dose Response Exponential Regression
02 0 04 05 0 0 09 1
Figure 8. Exponential regression curve for dose response data. The best fit exponential regression
curve appears on the plot along with the corresponding equation.
The equation from the standard curve illustrated in Figure 6 was used to convert the standardized and
adjusted absorption data into the theoretical fractions of CEES-hSA proteins (proteins with alkylated
sulfhydryl groups, indicating extensive CEES damage). A plot of this data is provided in Figure 9.
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
Theoretical Fraction CEES-hSA
Figure 9. Theoretical Fraction of CEES-hSA generated by various CEES dose treatments. The plot
is shown with an exponential curve of best fit accompanied by its corresponding equation. The
standard curve illustrated in Figure 5 was utilized for the conversion of absorption data to fractions
The results of these experiments suggest that an assay using DTNB and spectrophotometry is a promising method
for assessing the extent of CEES alkylation to hSA. The standard curve proved to be linear and easily
reproducible. Furthermore, the line of best fit provided reasonable estimations for the extent of alkylation in
hSA treated with varying doses of CEES.
The exponential relationship between CEES dose and sulfhydryl group alkylation corresponds with underlying
theory and expectations. CEES is known to react rapidly in solution, so alkylation of the single, free
sulfhydryl competes with both the alkylation of other hSA surface residues as well as the hydrolysis of CEES
in buffer. The rate of CEES hydrolysis is likely comparable to that of SM, whose half-life in water is approximately
16 minutess5. As CEES additions increase, the alkylation of the free sulfhydryl becomes more probable and the
impact of CEES hydrolysis diminishes. With respect to sulfhydryl competition with other residues, researchers in
the Netherlands have successfully isolated SM derivatized hemoglobin with alkylated aspartate, glutamine
and histidine residues as well as alkylated N-terminal amines2. The isolation of these products suggests
CEES alkylation of corresponding residues in hSA. Most importantly, these alkylated residues no longer compete
with the sulfhydryl for CEES, further supporting the observed exponential relationship between CEES dose
and sulfhydryl alkylation.
The data generated from this assay also provides a rough estimate for the number of CEES alkylation sites in
hSA. The free sulfhydryl group resides in a relatively internal region of hSA, so complete alkylation of this residue
in a solution likely corresponds with comprehensive alkylation of the entire protein. For example, examination
of Figure 8 suggests that maximum sulfhydryl alkylation occurs approximately at a CEES to hSA molar ratio
of 300:1. This indicates an approximate theoretical maximum of 300 alkylation sites per protein, assuming no
CEES hydrolysis occurs in the reaction buffer.
A noteworthy observation in the absorbance data is the decrease in protein concentration with increasing
CEES doses. This is illustrated by the diminishing absorption values at 280nm (Table 3), a wavelength where
protein absorbs maximally. This loss of protein may be attributed to a variety of factors. Protein solubility
depends heavily on the presence of surface charges. CEES alkylation replaces charged residues with uncharged
alkyl groups, thereby decreasing the solubility of the protein. Sufficient alkylation of these charged residues
my cause protein to precipitate from solution, explaining the relationship between protein loss and CEES dose. It
is also possible that cross-linking may have occurred between proteins as a result of extensive alkylation. This
would also increase the likelihood of protein precipitation and, therefore, a diminished protein concentration
This assay for determining the extent of CEES alkylation in hSA has shown initial promise. The results of
this experiment also provide valuable data for the estimation of hSA alkylation resulting from exposure to
various levels of CEES. This method may prove to be a starting point for an effective measure of a human's level
of SM exposure through the examination of albumin isolated from a blood sample. Most importantly, this
experiment serves as a contributor and precedent for future research and progress toward the goal of developing
an effective detection method and treatment for one of the most horrific weapons in human history.
1. Papirmeister B, Feister AJ, Robinson SI, Ford RD. Medical defense against mustard gas: toxic mechanisms
and pharmacological implications. CRC Press; 1991. p 1-3.
2. Noort D, Hulst AG, Trap HC, de Jong LPA, Benschop HP. Synthesis and mass spectrometric identification of the
major amino acid adducts formed between sulphur mustard and hemoglobin in human blood. Arch Toxicol, 1997
vol. 71. p 171-178.
3. Ghuman J, Zunszain PA, Petitpas I, Bhattacharya AA, Otagiri M, Curry S. Structural basis of the drug-
binding specificity of human serum albumin. J Mol Biol,vol. 353 p 38-52.
4. Means GE, Feeney, RE. Chemical modification of proteins. Holden-Day; 1971. p 155-157.
5. Mustard gas emergency response card. Centers for Disease Control and Prevention. 6 Apr. 2005. US Dept. of
Health and Human Services. 3 Mar. 2007 .
Back to the Journal of Undergraduate Research
College of Liberal Arts and Sciences I University Scholars Program I University of Florida I
UPI UNIVERSITY of
The n!trr~ali'?i fc'r T hwr CfrN C
ï¿½ University of Florida, Gainesville, FL 32611; (352) 846-2032.