Download Spinning down urine samples could diminish TGF

Spinning down urine samples could diminish TGF-β1 signal
Igor Y. Pavlov, Julio C. Delgado
ARUP Institute for Clinical and Experimental Pathology, Department of Pathology,
University of Utah School of Medicine, Salt Lake City, Utah
ABSTRACT
Objective. TGF-β1 is the important mediator of immune and anti-inflammatory responses; it is involved in normal renal
function and in the development of different nephropathies. Present research was dedicated to the validation of urine
TGF-β1 measurement by Quantikine Human TGF-β1 ELISA kit (R&D Systems, Inc., Minneapolis, MN, USA).
Methods. Human recombinant TGF-β1 was spiked into normal urine at different concentrations, and frozen in aliquots.
Collection of 100 frozen urine samples from healthy donors were used to investigate distribution of urine TGF-β1 in
normal population. Creatinine concentration in urine was determined.
Results. After the thawing, considerable amount of precipitate was found in some samples. Such samples were
centrifuged, and supernatants or un-spun specimens were run with TGF-β1 assay in duplicates. TGF-β1 activation
procedure (which precedes ELISA steps) clarified all samples no matter how much precipitation they contained.
It was shown that supernatants contained only fraction of the whole sample TGF-β1 activity (77% to 29%, even less for the
normal low concentration samples).
Validation involving un-centrifuged urine samples demonstrated following TGF-β1 assay characteristics:
linearity: from 23 to 1286 pg/mL, recovery from 95 to 116%; precision: <17%; analytical specificity – Limit-Of-Blank: 3
pg/mL; reference interval: from 0 to 39 pg/mg creatinine; freeze-thaw stability: after first thawing, urine samples could
loose from 5 to 40% of TGF-β1 activity on each following freeze-thaw circle.
Conclusion: Quantikine Human TGF-β1 ELISA assay provides accurate quantitation of human TGF-β1 in urine; urine sample
preparation should not include centrifugation step; repeating freeze-thawing should be avoided.
RESULTS
Table 1. Effect of precipitate centrifugation on urine TGF-β1
determination. TGF-β1, pg/mL
un-spun
Low
TGF-β1 is an important mediator of immune and anti-inflammatory responses; it is involved in normal renal
function and in the development of different nephropathies. Human TGF-β1 is a 25 kDa, disulfide-linked,
non-glycosylated homodimer [1]. It is normally secreted as inactive, or latent, complexes, containing TGF-β1
noncovalently bound to latency associated peptide (LAP) and latent TGF-β binding protein (LTBP) disulfidelinked to LAP. LTBP may facilitate secretion or targeting of latent TGF-β. The latency proteins also contribute
stability. Biological activity requires release of TGF-β1 from the latent complex [2]. This can be done in vitro
by disruption of LAP (e.g. acidification).
General approach to the urine sample processing for clinical testing is to centrifuge samples after it is
obtained, and spin them again before testing. Some publications describe methodologies involving urine
concentrating either after clarification [3] or after activation (acidification) [4]. During our test validation
study we investigated the effect of centrifugation on TGF-β1 test results and found that urine centrifugation
could significantly diminish the test signal.
MATERIALS and METHODS
34
42 %
342
263
77 %
1388
402
29 %
Ur1
24
9
39 %
Ur2
39
7
19 %
Ur3
22
12
52 %
Ur4
9
0
0%
Ur5
16
0
0%
Ur6
24
4
17 %
High
Table 2. Reproducibility study. TGF-β1, pg/mL
Run # 1 Run # 2
# replicas:
Normal urine
Low
During validation study of the Quantikine Human TGF-β1 ELISA kit (R&D Systems, Inc., Minneapolis,
MN, USA, Cat. #SB100B), recombinant TGF-β1 (R&D Systems, Cat. # 240-B-002) was spiked into normal
urine sample at low, medium and high concentrations, and frozen in aliquots. One hundred urine
samples from ARUP healthy donor collection (21 to 70 years old, median age 41 years, 58 males and 42
females) were used to investigate distribution of urine TGF-β1 in normal population. Creatinine
concentrations in tested samples were determined at ARUP Automated Core Lab, and final results
were reported as a ratio of TGF-β1 over the creatinine. During specimen processing experiments, urine
samples were clarified on Fischer Scientific mini-centrifuge (6600 rpm, 2 minutes). TGF-β1 assay was
performed according to the manufacturer recommendations [6] with one exception: urine samples
were centrifuged neither before the freezing nor after the thawing. Data was analyzed with R software
[5]: outlier detection was performed with Grubbs’ test, 95 % reference interval with 90 % confidence
limits was calculated by bootstrapping. Linearity test was verified with EP Evaluator software (EP
Evaluator® Release 8, Data Innovations, LLC, S. Burlington, VT)
recovery
82
Medium
INTRODUCTION
spun
Medium
High
4
4
23
Run # 3
6
17
23
Run-to-run
Within run (#3)
Average
CV
CV
21
17 %
17 %
71
61
59
64
11 %
10 %
112
116
103
110
6%
10 %
1073
1147
1040
1087
5%
5%
After the thawing, considerable amount of precipitate was found in those aliquots. During the
reference interval study, 6 out of 100 healthy donor urine samples were also found visibly turbid. Turbid
samples with visible precipitate were centrifuged, and supernatants, self-resuspended and un-spun
specimens were run with TGF-β1 assay in duplicates. Effect of precipitate removing on urine TGF-β1
determination is shown in Table 1. Self-resuspended and un-centrifuged samples revealed almost
identical results (data not shown).
It was observed that TGF-β1 activation procedure (acidification following by neutralization) clarified all
samples no matter how much precipitation they contained. Following validation procedures were
performed with un-centrifuged urine specimens.
Linearity. Recombinant human TGF-β1 was spiked in fresh normal human urine and titrated 1:2 with the
same urine sample. Results were found linear in measured TGF-β1 range from 23 to 1286 pg/mL.
Recovery was ranging from 95 % to 116 %.
Intra- and inter-assay precision studies were performed with normal urine and 3 urine samples spiked
with different concentrations of recombinant TGF-β1. Average, standard deviations, CV and number of
replicas which were run in reproducibility study are shown in Table 2.
Limit-of-Blank was determined as a concentration corresponding to Average plus 2 Standard Deviations
of zero analyte sample’s optical density at 450 nm. Twelve replicas of R&D Systems Calibrator Diluent
were measured with the TGF-β1 ELISA Kit. Limit-Of-Blank was found to be 3 pg/mL TGF-β1.
For reference interval study, 2 highest TGF-β1/creatinine measurements were considered to be outliers
(Grubbs’ test p-value <0.005). Remaining data was highly skewed, so reference interval was calculated
by bootstrapping the data. Lower level of the reference interval was zero; upper level was found to be
equal to 39 pg/mg creatinine (90 % confidence interval from 31 to 44 pg/mg creatinine).
Samples prepared for the reproducibility study along with normal urine sample were frozen – thawed
one (as for the regular test), 3 or 5 times. Results of TGF-β1 testing of those specimens showed that
each extra freezing-thawing circle decreased marker level from 5 % to 20 %.
CONCLUSIONS
Quantikine TGF-β1 kit Manual recommends spinning down urine samples to remove sediments; many
publications describing TGF-β1 urine test procedure do the same thing. Our data shows that centrifugation
of the urine samples could considerably diminish TGF-β1 content. As it was shown, freezing – thawing of
urine sample decreases TGF-β1 level. Therefore, the uniformity of the urine sample preparation is very
important for TGF-β1 assay. We recommend that urine samples should be collected, frozen without
centrifugation, thawed once and applied to the test.
REFERENCES
1. Lawrence AD. Transforming growth factor-β: an overview. Kidney Int 1995; 47:S19-S23
2. Taylor AW. Review of activation of TGF-β in immunity. J of Leukocyte Biol 2009; 85(1):29-33.
3. Ellis D, Forrest K, Erbey J, et al. Urinary measurement of transforming growth factor-β1 and type IV collagen as new markers of renal injury:
application in diabetic nephropathy. Clin Chem1998; 44(5):950-956.
4. Huan Y, DeLoach S, Daskalakis C, et al. Regulation of transforming growth factor -β1 by insulin in prediabetic African Americans. Kidney Int 2010;
78:318-324.
5. R Development Core Team (2010). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna,
Austria. ISBN 3-900051-07-0, URL http://www.R-project.org/
6. Quantikine® Human TGF-β1 Immunoassay, Package Insert, R&D Systems, Cat, Num. SB100B.