Sample Preparation in Global Metabolomics of Biological Fluids and Tissues - Proteomic and Metabolomic Approaches to Biomarker Discovery - page 66

of the most widely adopted clinical metabolo-
mics applications to date. 106 In this approach,
a small drop of whole blood or other bio

such as urine and plasma and/or the in uence

of hematocrit values on spot formation. Metabo-
lite coverage of dried blood spots was higher
than for dried plasma spots and for solvent-
precipitated plasma (1,239 versus 1,039 versus
954 in positive ESI and 257 versus 122 versus
162 in negative ESI). In contrast, in a GC-MS
study by Kong et al. whole blood after solvent-
precipitation produced the highest number of
detectable features (866 features) and the highest
number of

uid is

placed on filter paper and allowed to air-dry

for several hours. This spot is subsequently
extracted in a solvent such as methanol 107 prior
to analysis. To date, only untreated and FTA
(aimed for DNA analysis, containing strong
buffers, denaturants, and free radical traps)
paper types have been examined for their utility
in metabolomic studies, 107 e 109 with untreated
paper generally preferred due to the presence
of fewer contaminant peaks.
DBS represents an attractive option for global
or untargeted metabolomic studies of bio

identi ed peaks (171 identi ed

metabolites) versus 695 (137 identi ed metabo-

lites) for dried blood spots using untreated
paper, 749 for plasma, and 614 for dried plasma
spots. 107 The authors found that DBS is not suit-
able for l-lysine, iminodiacetic acid, DL-threo-
beta-hydroxyaspartic acid, adipamide, citric
acid, or adenosine monophosphate (AMP) using
the conditions employed in the study, although
the cited list does not seem to be exhaustive,
considering that at least 34 identi

uids,
due to potential advantages such as use of
smaller sample volumes (typically 20 e 25 Lof

whole blood), ease of collection, and simpli

m

ed
transportation and storage requirements. 110
Michopolous et al. evaluated the utility of DBS
for metabolite pro

ling of whole blood, plasma,
and urine by LC-MS. 109 The pro

ed metabolites
were not observed in DBS but were successfully
detected in whole blood samples. The decrease
in the number of metabolites observed with
DBS was attributed to the losses during the
drying process and/or strong interaction with
the paper substrate. The recovery evaluation
for three compounds found irreproducible
recovery of alanine at different concentration
levels, which could potentially lead to erroneous
data interpretation, and requires further investi-
gation for a larger number of identi

les obtained
from DBS (untreated paper, 25% methanol
extraction for 20 min, evaporation, and reconsti-
tution in 1/1 methanol/water [v/v]) generally
closely resembled the pro

les after solvent
precipitation, with the exception of urine spots
acquired in negative ESI, where signi cant loss

of acidic compounds was observed after spotting
(185 versus 584 features). In terms of repeat-
ability, DBS, plasma spots, and protein-
precipitated plasma had 75%, 52%, and 73%,
respectively, of features with CV of less than or
equal to 30% in positive ESI mode, while for
negative ESI, 83%, 95%, and 86%, respectively,
of features met this criterion. Urine spots showed
very poor repeatability, with only 43% of
features showing acceptable RSD versus 81% of
features after a dilute-and-shoot strategy. This
data clearly shows that existing spotting meth-
odologies seem to perform best for whole blood
samples, which result in more concentrated
spots. Additional in-depth studies are needed
to better understand the factors leading to poorer
precision when spotting less viscous samples

ed metabo-
lites of different classes both in context of GC-MS
and LC-MS studies. Limited short-term stability
evaluation of 10 randomly selected metabolites
found no differences in up to 48 hr storage at
25 C versus e 20 C. 107 In contrast, Michopoulos
et al. showed limited long-term stability of dried
blood spot samples at room temperature, with
signi cant changes in some of the metabolites

even when stored at e 20 Cor e 80 C. 108 These
results show that dried blood spots provide
poor long-term stability for global metabolomic
studies, which invalidates one of the important
advantages of DBS: permitting easy storage