Fluorescent reporter assays provide direct, accurate, quantitative measurements of MGMT status in human cells
Fluorescent reporter assays provide direct, accurate, quantitative measurements of MGMT status in human cells
Zachary D. NagelID 0 1 2
Andrew A. Beharry 1 2
Patrizia Mazzucato 0 1 2
Gaspar J. Kitange 1 2
Jann N. Sarkaria 1 2
Eric T. Kool 1 2
Leona D. Samson 0 1 2
0 Department of Biological Engineering, Massachusetts Institute of Technology , Cambridge, Massachusetts , United States of America, 2 Department of Chemistry, Stanford University, Stanford, California, United States of America, 3 Department of Radiation Oncology, Mayo Clinic , Rochester, Minnesota , United States of America
1 Editor: Brendan D Price, Dana Farber Cancer Institute , UNITED STATES
2 a Current address: Department of Environmental Health, Harvard T.H. Chan School of Public Health, Boston, Massachusetts, United States of America ?b Current address: Department of Chemical & Physical Sciences, University of Toronto Mississauga , Toronto, Ontario , Canada ?c Current address: Stanley Center for Psychiatric Research, Broad Institute of MIT and Harvard , Cambridge, Massachusetts , United States of America
The DNA repair protein O6-methylguanine DNA methyltransferase (MGMT) strongly influences the effectiveness of cancer treatment with chemotherapeutic alkylating agents, and MGMT status in cancer cells could potentially contribute to tailored therapies for individual patients. However, the promoter methylation and immunohistochemical assays presently used for measuring MGMT in clinical samples are indirect, cumbersome and sometimes do not accurately report MGMT activity. Here we directly compare the accuracy of 6 analytical methods, including two fluorescent reporter assays, against the in vitro MGMT activity assay that is considered the gold standard for measuring MGMT DNA repair capacity. We discuss the relative advantages of each method. Our data indicate that two recently developed fluorescence-based assays measure MGMT activity accurately and efficiently, and could provide a functional dimension to clinical efforts to identify patients who are likely to benefit from alkylating chemotherapy.
Data Availability Statement: All relevant data are
within the paper and its Supporting Information
Funding: This work was supported by grants
DP1ES022567 awarded to L.D.S. and R01-CA217809
to E.T.K. by the National Institutes of Health.
Competing interests: The authors have declared
that no competing interests exist.
O6-methylguanine DNA Methyltransferase (MGMT; also known as alkylguanine
alkyltransferase, AGT) repairs DNA damage induced by endogenous, environmental, and therapeutic
alkylating agents, and is the most important pathway for repairing O6-alkylguanine adducts in
human cells [
]. MGMT prevents cell killing by repairing O6-methylguanine (O6-MeG)
cytotoxic DNA lesions induced by the cancer chemotherapy agents Temozolomide and
Dacarbazine, and cytotoxic O6-chloroethylguanine lesions induced by cross-linking agents such as
BCNU . These chemotherapeutic agents are used to treat a variety of cancers including
glioblastoma, metastatic melanoma, and Hodgkins lymphoma. Clinical studies have
demonstrated that glioblastoma patients with MGMT deficient tumors exhibit longer overall survival
following treatment with temozolomide [
], and are more likely to respond to radiotherapy
], highlighting the potential for personalized cancer therapy based on MGMT status in
Although MGMT diminishes the effectiveness of cancer therapies, it also plays a critical
role protecting normal tissues from DNA damage. Over 10-fold inter-individual variation in
MGMT activity has been observed in normal tissues [
], and lower MGMT activity is
associated with therapy related leukemia , myelotoxicity in patients receiving temozolomide [
and lung cancer risk [
]. Since genetic variation [
], environmental exposures [
chemotherapy can each affect MGMT activity, methods that directly report MGMT function
are best suited for studies measuring inter-individual differences in both normal tissues and
cancer cells [
]. Studies investigating the relationships between MGMT activity and disease
risk and cancer therapy outcomes have been limited by cumbersome and indirect assays that
may not accurately predict MGMT activity.
Presently, MGMT status is assessed in clinical samples primarily using MGMT promoter
methylation as a proxy for MGMT expression, and in some cases using
immunohistochemistry; however, both methods can fail to reflect MGMT levels accurately. For example high levels
of MGMT expression are possible in tumors with MGMT promoter hypermethylation due to
expression of a previously unrecognized enhancer element [
]. Such observations highlight
the need for functional assays that accurately measure MGMT activity to achieve personalized
therapies based on DNA repair capacity assessments. We show here that two quantitative
fluorescence-based assays including a small molecule reporter probe [
] and a plasmid based host
cell reactivation assay [
], accurately and efficiently measure MGMT activity in human
cells. These assays are ready for use in preclinical studies, and have the potential to enable
much-needed research aimed at tailoring cancer therapy to individual patients based on DNA
repair capacity in tumor and normal tissues .
Seven B-lymphoblastoid cell lines including TK6 (RRID CVCL_0561) [
], TK6+MGMT [
and five EBV transformed cell lines available from the Coriell Cell Repository were maintained
in log phase in RPMI media with 20% FBS as previously described. The cell lines have been
previously designated as follows: #4 (GM15223; CVCL_5W53), #5 (GM15245; CVCL_5W71),
#12 (GM15385; CVCL_5Y77), #14 (GM15038; CVCL_5V37), and #16 (GM15072;
]; the same nomenclature is used here. Cells were obtained in 2001. The
Coriell Cell Repository authenticates and tests cell cultures for contamination with mycoplasma,
bacteria, and fungi; cells have not been authenticated by authors. Mycoplasma testing was
carried out and found to be negative using a commercial PCR-based kit at the time cells were last
passaged (November 2015).
Oligonucleotide cleavage assays were performed as described previously [
], and illustrated
in (S1 Fig). To generate lysates, 1.5 x 107 cells were collected, washed twice with PBS, and
suspended in 400 ?L of lysis buffer comprising 50 mM Tris, pH 7.5, 1 mM EDTA, 1 mM DTT,
5% glycerol, 50 mM NaCl, and 1 mM AEBSF (protease inhibitor). Cells were disrupted by
sonication, cell debris was pelleted by centrifugation at 14,000g for 30 minutes at 4?C,
supernatants were collected, and total protein concentration was determined using a BCA assay. Cell
2 / 13
lysates were incubated at 37?C for 30 minutes with 4 pmoles of a 32P 5?-end labeled duplex
comprising an oligonucleotide, GAACTXCAGCTCCGTGCTGGCCC, in which X represents
O6MeG, and the corresponding complementary oligonucleotide. The reaction products were
then purified by phenol/chloroform extraction and ethanol precipitation, dissolved, and finally
digested using PstI restriction enzyme. O6MeG blocks PstI cleaveage, providing the basis for a
gel-shift assay for the extent of MGMT-dependent lesion removal. Digests were analyzed by
SDS PAGE followed by autoradiographic imaging, and densitometry was used to calculate the
percentage of cleaved oligonucleotide. The linear range of the MGMT assay was established
for each cell line by varying the amount cell lysate (between 10 and 400 ?g) incubated with the
duplex. MGMT activity in each cell lysate was calculated from the slope of a linear best fit of
the percentage of oligonucleotide cleaved versus total protein concentration.
Quantitative western blotting
Preparation of cell lysates and immunoblotting were performed as described previously [
Briefly, 25 ?g of cell lysate (2.5 ?g/?L in Laemli sample buffer) were separated by SDS-PAGE
and transferred to a PVDF membrane. The membrane was blocked for 1 hour using Odyssey
Blocking Buffer and incubated with a primary mouse antibody that binds human MGMT,
followed by washing (4X) with PBS + 0.1% Tween-20 and 1 hour incubation with a secondary
antibody, Licor IRDye 680RD donkey anti-mouse. After washing (4X) with PBS + 0.1%
Tween-20, membranes were imaged in the 700 nm channel of an Odyssey imager. An actin
antibody was used as a loading control. A representative gel and accompanying quantitation is
available in S2 Fig.
Quantitative real time PCR
qPCR data were described previously [
]. Total RNA was isolated using a Qiagen RNeasy kit,
and mRNA was subsequently isolated using a Qiagen Oligotex kit according to the
manufacturer?s protocols. Following DNase digest, cDNA was generated using poly-dT primers with
reverse transcriptase. TaqMan qPCR was used to quantitate MGMT transcript levels relative
to a GAPDH control. Primers and probes for MGMT (catalog number Hs.00172470) and
GAPDH (Hs.99999905) were purchased from Applied Biosystems. A 20 ?L reaction
containing TaqMan Universal PCR Master Mix (Applied Biosystems), plus probes and cDNA was
amplified by PCR using the following program: 10 minutes at 95?C, followed by 40 cycles of
denaturing at 95?C for 15s followed by annealing and extension for 1 minute at 60?C.
FM-HCR assays have been published previously [
]. Briefly, reporter plasmids were generated
by extension of O6MeG-containing oligonucleotides that were annealed to single-stranded
plasmid DNA, followed by primer extension and ligation. The DNA lesion induces
transcriptional errors that result in expression of functional mPlum fluorescent protein, unless repair
removes O6MeG, the source of transcriptional errors. As a result, cells that efficiently repair
O6MeG express relatively low levels of mPlum fluorescent protein, whereas MGMT deficient
cells express relatively high levels of mPlum fluorescent protein. Transient transfection, flow
cytometric analysis, and calculation of DNA repair capacity were described previously [
Promoter methylation assays
Methylation specific PCR assays for promoter methylation were carried out as described
]. Genomic DNA was extracted from 106 cells from cell lines using a QIAamp DNA
3 / 13
Kit, and bisulfite conversion of 1 microgram of the resulting gDNA was carried out using
an EpiTect Bisulfite Kit (Qiagen). For PCR detection of unmethylated MGMT promoter
sequences, the following primers were used: 5?TTTGTGTTTTGATGTTTGTAGGTTTTT
GT-3?, 5?AACTCCACACTCTTCCAAAAACAAAACA-3?. For detection of methylated
DNA, the following primers were used: 5?TTTCGACGTTCGTAGGTTTTCGC-3?, 5?GCAC
TCTTCCGAAAACGAAACG-3?. Approximately 100 ng of gDNA was combined with primers
at a final concentration of 400 nM, and amplified with 1 unit of AmpliTaq Gold DNA
polymerase for 35 cycles (annealing at 59 ?C, and extension at 72 ?C). PCR products were analyzed
on a 3% agarose gel visualized with ethidium bromide (S3 Fig).
Fluorogenic real-time reporter (NR-1) for repair by MGMT
Cell lysates were prepared from approximately 108 cells using the procedure described above
for MGMT assays. Cell lysates were analyzed using a recently reported DNA based fluorescent
probe (NR-1) comprising a short DNA oligomer containing a fluorophore and an
O6-benzylguanine nucleoside that is modified with a quencher dye [
]. Repair by MGMT separates the
quencher from the fluorophore, leading to an increase in fluorescence. Cell lysates (800 ?g
total protein) were combined with 50 nM of NR-1 in a 96-well plate (final volume 200 ?L), and
incubated for 2 hours at 37?C. Fluorescence at 488 nm was measured with a plate reader. We
observed that combining NR-1 with TK6 cell lysates or as purified BSA, both of which lack
MGMT, leads to an approximately 2-fold increase in fluorescent signal, indicating that a
relatively small but significant MGMT-independent increase in NR-1 fluorescence in the presence
of proteins. Thus, to calculate MGMT activity, the fluorescent signal from MGMT deficient
TK6 cell lysates combined with fluorescent probe was subtracted from the fluorescence values
measured for all other cell lysates combined with fluorescent probe.
For each method, error bars represent standard deviation from three biological replicates
(carried out with materials independently prepared from the same cell line on different days).
One-way ANOVA with Tukey?s multiple comparisons test was used to determine the ability of
assays to distinguish MGMT levels or activity among the cell lines. All statistical analyses were
carried out in Graphpad Version 7.0c.
A panel of 24 lymphoblastoid cell lines derived from apparently healthy individuals from
diverse genetic backgrounds [
], Coriell #1?24, has been characterized previously for MGMT
levels using transcriptional profiling and MGMT activity using a fluorescence based multiplex
host cell reactivation (FM-HCR) assay [
]. Focusing on a subset of these cell lines (Coriell #4,
#5, #12, #14 and #16), together with an MGMT-deficient negative control TK6, and a
TK6-derived MGMT proficient positive control, TK6+MGMT [
], we have measured
MGMT levels or activity in the seven cell lines using six different methods (Fig 1, S1 and S2
Tables). We chose to focus on this representative subset of cell lines because previous data
from our laboratory revealed that they span the entire range of sensitivity to O6MeG
generating alkylating agents observed previously for the larger set of cell lines [
], which can be
explained in part by differences in MGMT activity [
The gold standard radiolabeled oligonucleotide-based biochemical MGMT assay revealed
an approximately 100-fold range of MGMT activity in the samples, with the following rank
order established using the biochemical MGMT assay: TK6 < Coriell #5 < Coriell
#4 < Coriell #12 < Coriell #14 < Coriell #16 < TK6+MGMT. The available quantitative
4 / 13
Fig 1. MGMT activity measured by 6 methods in 7 cell lines. A) MGMT activity reported as femtomoles of cleaved 32P labeled
O6-MeG containing oligonucleotide per microgram of protein in cell lysates. B) MGMT transcript levels normalized to TK6
+MGMT. C) MGMT activity measured by FM-HCR, reported as the inverse of reporter expression. D) MGMT protein levels in
cell lysates measured by quantitative western blotting and normalized to GAPDH protein levels. E) MGMT activity in cell lysates
measured using the NR-1 fluorescent probe and reported in arbitrary units of fluorescence. F) Results of methylation specific PCR
assays for MGMT promoter methylation; a value of 1 was assigned to the three cell lines in which promoter methylation was
detected. Cell lines were ranked in ascending order of MGMT activity measured by the biochemical assay in panel A; the order
and color scheme is preserved in each panel. Error bars represent the standard deviation of at least 3 measurements, and ?ND?
PLOS ONE | https://doi.org/10.1371/journal.pone.0208341
5 / 13
indicates that the 95% confidence interval for the measured parameter included zero. Data have been log transformed for optimal
methods for assessing MGMT status in human lymphoblastoid cell lines have been compared
with this biochemical assay (Fig 2). All six methods yielded qualitatively similar estimates of
MGMT activity in the seven cell lines, however each assay presents both unique technical
demands and unique advantages (Table 1), detailed below.
Oligonucleotide-based biochemical assays
In vitro biochemical MGMT activity assays can be regarded to be the ?gold standard? for
measuring MGMT activity because they directly measure the level of repair activity of MGMT
protein using a chemically defined substrate with a radiolabel that permits detection of very low
levels of repair activity. However, the assay has several technical drawbacks. Assay conditions
must be optimized to determine the range of cell lysate protein concentrations that produce a
linear response, which is cell-line dependent (Compare x-axes in panel C of S1 Fig).
Furthermore, relative to assays measuring activity in the highly MGMT proficient cell line TK6
+MGMT, an approximately 10-fold higher concentration of cell lysate protein was required to
distinguish the very low level of MGMT activity in cell lines #4 and #5 from the undetectable
activity in TK6 cells. Biochemical MGMT assays were the most time consuming of the four
methods studied, requiring approximately 7 hours active laboratory time for analysis of up to
4 samples in parallel.
FM-HCR assays distinguished cells with low MGMT activity (Cell lines #4 and #5) from cells
that lack MGMT activity (TK6), and thus exhibited the same dynamic range as the gold
standard biochemical assay (A 62-fold range of activity comparing the least active sample, #5, to
the most active sample, TK6). The sensitivity of FM-HCR is achieved in part because of the
ability to detect individual cells harboring unrepaired DNA lesions that lead to transcriptional
errors and fluorescent protein expression; whereas a minor subpopulation of repair deficient
cells may be lost in ensemble measurements, they can be readily detected by FM-HCR.
Together with qPCR and the fluorescent NR-1 probe assay (below), FM-HCR required the
least amount of active laboratory time (1.5 hours) for analysis of up to 4 samples in parallel.
As has been observed by others [
], MGMT protein levels estimated from Western blots
correlated strongly with MGMT activity (R = 0.98, Fig 2D), however the low levels of MGMT in
cell lines #4 and #5, detectable by the 32P-oligonucleotide-based biochemical assay, FM-HCR
and qPCR, were below the limit of detection by western blotting. Western blots also required
approximately twice the active laboratory time (3 hours for analysis of up to 4 samples in
parallel) as the least labor-intensive assays.
Transcript levels by qPCR
MGMT transcript levels measured by qPCR analysis, reported previously [
strongly (R = 0.98) with MGMT activity measured by the biochemical MGMT assay. Analysis
by qPCR required approximately 1.5 hours for analysis of up to 4 samples in parallel.
6 / 13
Fig 2. Comparison of four quantitative MGMT assays against the biochemical MGMT assay with radiolabeled oligonucleotides. All assays
have been normalized to a control cell line (TK6+MGMT), which expresses a high level of MGMT. The Pearson correlation (R) to MGMT activity
measured using biochemical assays with 32P-labeled oligonucleotide substrates is reported for each assay. Each data point represents one of the
seven cell lines analyzed; fewer data points are reported for assays where some cell lines were below the limit of detection.
Fluorescent MGMT probe NR-1
MGMT activity as measured with the NR-1 probe correlated well with activity measured using
the biochemical MGMT assay, however two cell lines (#4 and #5), which had the lowest
activity as judged by the biochemical assay, were below the limit of detection. Analysis using the
NR-1 probe required approximately 1.5 hours for analysis of up to 4 samples in parallel.
Methylation specific PCR (MS-PCR)
Promoter methylation was detected, as expected, in TK6, previously shown to exhibit MGMT
promoter methylation, and in TK6+MGMT, which is derived from TK6 and expresses
7 / 13
1Requirements refer to the approaches used here; a range is given for methods found to require more cells to lower MGMT levels.
2Dynamic range was calculated by dividing the activity (as measured using the biochemical assay) of the most active sample (TK6+MGMT) by the activity of the least
active sample for which activity could be significantly distinguished from background.
3Intact refers to methods that can be carried out in live cells.
4Approximate cost of generating or purchasing materials needed to carry out each assay in triplicate using the approaches in methods.
MGMT constitutively under a CMV promoter [
]. Strikingly, MGMT promoter methylation
was not detected in any of the lymphoblastoid cell lines, including those that express very low
levels of MGMT, namely Coriell #5 and Coriell #4; this result highlights the potential for
MS-PCR to inaccurately identify MGMT-deficient cells as MGMT proficient. Promoter
methylation was also detected in two patient derived xenograft models of glioblastoma,
GBM12_5199 and GBM12_3080, consistent with previous findings [
]. Notably, despite the
robust MGMT promoter methylation observed in both cell lines (S3 Fig), GBM12_3080
expresses high levels of MGMT detectable by FM-HCR [
]. The conclusion that
GBM12_3080 is proficient for MGMT is also supported by previous measurements of MGMT
transcript levels and observed sensitization to TMZ upon treatment with the MGMT inhibitor
]. Analysis of MGMT promoter methylation by MS-PCR required
approximately 2 hours for analysis of up to 4 samples in parallel.
The time-sensitive nature of cancer treatment as well as the serious side effects and risk of
therapy-related cancers in patients treated with radiotherapy and chemotherapy have motivated a
search for biomarkers that can predict whether specific therapies will work for individual
]. The success of chemotherapy hinges upon the existence of a therapeutic window
in which cancer cells can be killed without severe normal tissue toxicity. Acquired DNA repair
defects drive genomic instability, a hallmark of cancer [
], and can sensitize cancer cells to
]. Notably, MGMT defects occur in many cancers including glioblastoma
], colorectal cancer [
], leukemia [
], lymphoma , small cell lung cancer [
breast cancer [
], pancreas cancer [
], and melanoma [
], and in some cases increases the
effectiveness of therapeutic agents that generate DNA lesions that are MGMT substrates [
Thus, the ability to determine MGMT status accurately and efficiently in cancer cells could
allow clinicians to tailor therapies to the needs of individual patients.
Although the biochemical MGMT assay is considered a gold standard because it provides a
sensitive quantitative measure of functional MGMT levels in cell lysates, it was by far the most
time consuming (7 hours active time for up to 4 samples processed in parallel, Table 1) due to
the need for testing multiple conditions for establishing the linear range of the assay (S1 Fig).
In addition, the biochemical assays required a radiolabeled oligonucleotide, and the largest
8 / 13
number of cells (up to 108) of the six assays used. These considerations render
oligonucleotide-based assays too labor intensive for clinical use.
Three indirect assays, namely western blotting for MGMT protein levels, qPCR for MGMT
transcript levels, and methylation specific PCR for MGMT promoter methylation status, were
less labor intensive and required fewer cells. Indeed, promoter methylation assays, and to a
lesser extent, immunohistochemistry, are currently used in the analysis of clinical samples;
however, immunohistochemical approaches can fail to predict MGMT activity consistently
], promoter methylation analysis can fail to predict transcript levels [
], and MGMT
transcript levels can fail to predict protein levels due to post-transcriptional regulation .
Furthermore, MGMT activity is affected by posttranslational modifications [
], and the
repair protein is inactivated by its substrates following a single turnover [
]; these important
contributions to MGMT activity cannot be detected by indirect assays. Thus, despite their
promise for patient stratification in the context of alkylating chemotherapy, the challenges
associated with existing MGMT assays has limited their potential for guiding therapy
The fluorescent probe NR-1, and FM-HCR assays overcome the problems associated with
subjective histopathological scoring and the lack of correlation between promoter methylation
and MGMT activity by providing a direct functional assessment of MGMT activity, without
the need for radiolabeled probes or extensive sample processing. For example, methylation
specific PCR indicated a total lack of MGMT promoter methylation in Coriell #4 (S3 Fig),
which, in fact, expresses very low levels of MGMT according to the biochemical assay.
FM-HCR assays detected the low level of MGMT activity present in Coriell #4, which was
below the limit of detection by western blotting (S2 Fig), and required optimization of
conditions for detection by the biochemical assay.
Two glioblastoma xenograft lines, GBM12_3080 and GBM12_5199, both exhibit MGMT
promoter methylation (S3 Fig), but the FM-HCR assay correctly assigns GM12_3080 to be
MGMT proficient [
], consistent with previous independent characterization [
advantage of promoter methylation assays is that they are generally regarded to be specific for
cancer cells, since the MGMT promoter is not methylated in normal cells. Although the
fluorescence-based assays are not inherently specific for cancer cells, cell-type specificity could be
achieved by carrying them out in conjunction with cell surface markers.
While both fluorescent reporter assays directly measure MGMT activity, their differing
sample requirements and modes of detection endow them with complementary strengths. The
FM-HCR assay must be used with intact cells, enabling detection of repair deficient
subpopulations, and providing an integrated measure of MGMT activity over the course of 24 hours in
cells, rather than a snapshot of MGMT activity at a single time point that the NR-1 probe
provides. The estimated dynamic range for FM-HCR in the set of cell lines considered here (62) is
larger than that of the NR-1 probe (4.3). Although the NR-1 probe failed to detect the very low
levels of MGMT activity in extracts from cell lines #4 and #5, the probe did distinguish
significant differences in activity among cell lines with higher MGMT activity (TK6+MGMT versus
#12, #14, and #16), whereas FM-HCR did not (S2 Table). However FM-HCR requires
transfection of live cells, and flow cytometric analysis, while the NR-1 probe can be used with cell
lysates, and the fluorescent signal can be measured using a plate reader. Furthermore, the
NR1 probe can be used to measure repair kinetics in real time under varied conditions, such as in
the presence of small molecule inhibitors. The NR-1 probe is also extremely cost-efficient
because it is inexpensive to synthesize and requires only cell lysates and a plate reader for
Clinical translation will require studies in primary patient samples to determine how well
these assays predict therapeutic outcomes assessed using standard measures such as the
9 / 13
Response Evaluation Criteria in Solid Tumors (RECIST). For cell-based assays, optimization
of tumor processing to maximize live cell content may be needed. Strategies for excluding
signal from tumor stroma should be considered, for example using cell surface markers or
Limitations of existing assays have usually resulted in binary classification of tumor MGMT
status, making it difficult to assess whether the distinction between tumors with low MGMT
and those that are completely lacking MGMT is clinically important. However, available data
are consistent with a continuous relationship between MGMT activity and temozolomide
]. Quantitative functional assays such as those presented herein will expand the
potential for future studies aimed at resolving this question. Since FM-HCR reporters have
been successfully transfected into both primary cells and cancer cells [
], and the NR-1 probe
is amenable to any cells from which lysates can be derived [
], both approaches could
potentially be used for studies of MGMT activity in cancerous and normal tissue.
Many cancers exhibit alterations in MGMT activity that may be exploited when treating
patients with alkylating chemotherapeutic agents, and MGMT activity in normal tissues may
predict inter-individual differences in alkylating agent toxicity. However, there remains a
critical need for accurate, functional assays that could be used to identify individuals with
MGMTdeficient cancers. The recently developed FM-HCR assays and fluorescent NR-1 probe both
overcome the problems associated with currently used indirect methods of measuring MGMT
activity, and merit consideration as alternatives for use in pre-clinical studies and in clinical
trials involving cancers where MGMT status may be associated with therapeutic outcomes.
S1 Fig. In vitro MGMT activity assay. A) Oligonucleotide digest assay for MGMT activity.
Repair of a PstI cleavage blocking O6MeG DNA lesion results in an 8 nt 32P labeled fragment
detectable by polyacrylamide gel electrophoresis. B) Polyacrylamide gel analysis of restriction
digest products following treatment of 4 pmol of oligonucleotide with 10?400 ?g of protein
extracts from 7 lymphoblastoid cell lines for 30 minutes at 37 ?C. C) Determination of linear
range and calculation of MGMT activity. Slopes calculated from the data in these plots are
reported in Fig 1, Panel A.
S2 Fig. Representative immunoblot and quantitation of MGMT protein levels in 7 cell
lines. Protein levels were below the limit of detection for Coriell #5, Coriell #4, and TK6.
S3 Fig. Gel electrophoretic analysis of methylation specific PCR products. Each lane shows
PCR products obtained from amplification of bisulfite converted genomic DNA from the
indicated cell lines with primers specific for unmethylated DNA (U), or methylated DNA (M).
S1 Table. MGMT assay data and statistics. Units are as follows: 32P Oligo, fmoles cleaved
oligonucleotide per microgram protein lysate; NR-1, Fluorescence signal, arbitrary units;
FM-HCR, % Reporter Expression; Western Blot, MGMT protein levels as a percentage of
actin protein levels; qPCR, MGMT transcript levels normalized to MGMT transcript levels in
10 / 13
S2 Table. Pairwise statistical comparisons. Cell lines that were significantly different from
one another for each MGMT assay are indicated with an asterisk ? ?; those that were not
significantly different are marked ?ns?.
Conceptualization: Zachary D. Nagel, Leona D. Samson.
Funding acquisition: Leona D. Samson.
Investigation: Zachary D. Nagel, Andrew A. Beharry, Patrizia Mazzucato, Gaspar J. Kitange.
Methodology: Zachary D. Nagel, Andrew A. Beharry, Leona D. Samson.
Resources: Andrew A. Beharry.
Supervision: Jann N. Sarkaria, Eric T. Kool, Leona D. Samson.
Writing ? original draft: Zachary D. Nagel, Leona D. Samson.
Writing ? review & editing: Zachary D. Nagel, Andrew A. Beharry, Gaspar J. Kitange, Jann N.
Sarkaria, Eric T. Kool, Leona D. Samson.
11 / 13
12 / 13
1. Fu D , Calvo JA , Samson LD . Balancing repair and tolerance of DNA damage caused by alkylating agents . Nat Rev Cancer . 2012 ; 12 ( 2 ): 104 - 20 . https://doi.org/10.1038/nrc3185 PMID: 22237395
2. Pegg AE . Repair of O-6-alkylguanine by alkyltransferases . Mutat Res-Rev Mutat Res . 2000 ; 462 ( 2- 3 ): 83 - 100 .
3. Pegg AE . Multifaceted Roles of Alkyltransferase and Related Proteins in DNA Repair, DNA Damage , Resistance to Chemotherapy, and Research Tools. Chem Res Toxicol . 2011 ; 24 ( 5 ): 618 - 39 . https://doi. org/10.1021/tx200031q PMID: 21466232
4. Stupp R , Mason WP , van den Bent MJ , Weller M , Fisher B , Taphoorn MJB , et al. Radiotherapy plus concomitant and adjuvant temozolomide for glioblastoma . New England Journal of Medicine . 2005 ; 352 ( 10 ): 987 - 96 . https://doi.org/10.1056/NEJMoa043330 PMID: 15758009
5. Rivera AL , Pelloski CE , Gilbert MR , Colman H , De La Cruz C , Sulman EP , et al. MGMT promoter methylation is predictive of response to radiotherapy and prognostic in the absence of adjuvant alkylating chemotherapy for glioblastoma . Neuro Oncol . 2010 ; 12 ( 2 ): 116 - 21 . Epub 2010/02/13. https://doi.org/ 10.1093/neuonc/nop020 PMID: 20150378 .
6. Myrnes B , Giercksky KE , Krokan H . Interindividual variation in the activity of O6-methyl guanine-DNA methyltransferase and uracil-DNA glycosylase in human organs . Carcinogenesis . 1983 ; 4 ( 12 ): 1565 - 8 . Epub 1983/12/01. PMID: 6652869 .
7. Vahakangas K , Trivers GE , Plummer S , Hayes RB , Krokan H , Rowe M , et al. O(6) -methylguanine-DNA methyltransferase and uracil DNA glycosylase in human broncho-alveolar lavage cells and peripheral blood mononuclear cells from tobacco smokers and non-smokers . Carcinogenesis . 1991 ; 12 ( 8 ): 1389 - 94 . Epub 1991/08/11. PMID: 1860159 .
8. Sagher D , Karrison T , Schwartz JL , Larson R , Meier P , Strauss B. Low O6 -alkylguanine DNA alkyltransferase activity in the peripheral blood lymphocytes of patients with therapy-related acute nonlymphocytic leukemia . Cancer Res . 1988 ; 48 ( 11 ): 3084 - 9 . Epub 1988/06/01. PMID: 2452685 .
9. Sabharwal A , Waters R , Danson S , Clamp A , Lorigan P , Thatcher N , et al. Predicting the myelotoxicity of chemotherapy: the use of pretreatment O6-methylguanine-DNA methyltransferase determination in peripheral blood mononuclear cells . Melanoma research . 2011 ; 21 ( 6 ): 502 - 8 . Epub 2009/06/30. https:// doi.org/10.1097/CMR.0b013e32832ccd58 PMID: 19561552 .
10. Rudiger HW , Schwartz U , Serrand E , Stief M , Krause T , Nowak D , et al. Reduced O6-methylguanine repair in fibroblast cultures from patients with lung cancer . Cancer Res . 1989 ; 49 ( 20 ): 5623 - 6 . Epub 1989/10/15. PMID: 2551492 .
11. Bugni JM , Han J , Tsai MS , Hunter DJ , Samson LD . Genetic association and functional studies of major polymorphic variants of MGMT . DNA Repair (Amst) . 2007 ; 6 ( 8 ): 1116 - 26 . Epub 2007/06/16. https://doi. org/10.1016/j.dnarep. 2007 . 03 .023 PMID: 17569599 .
12. Zhang Z , Wang L , Wei S , Liu Z , Wang LE , Sturgis EM , et al. Polymorphisms of the DNA repair gene MGMT and risk and progression of head and neck cancer . DNA Repair (Amst) . 2010 ; 9 ( 5 ): 558 - 66 . Epub 2010/03/09. https://doi.org/10.1016/j.dnarep. 2010 . 02 .006 PMID: 20206583 .
13. Pegg AE , Fang QM , Loktionova NA . Human variants of O-6-alkylguanine-DNA alkyltransferase . DNA Repair . 2007 ; 6 ( 8 ): 1071 - 8 . https://doi.org/10.1016/j.dnarep. 2007 . 03 .012 PMID: 17482892
14. Liu L , Lee K , Schupp J , Koc ON , Gerson SL . Heterogeneity of O6-alkylguanine-DNA-alkyltransferase measured by flow cytometric analysis in blood and bone marrow mononuclear cells . Clinical cancer research: an official journal of the American Association for Cancer Research . 1998 ; 4 ( 2 ): 475 - 81 . Epub 1998/05/14. PMID: 9516939 .
15. Chen X , Zhang M , Gan H , Wang H , Lee JH , Fang D , et al. A novel enhancer regulates MGMT expression and promotes temozolomide resistance in glioblastoma . Nat Commun . 2018 ; 9 ( 1 ): 2949 . Epub 2018/07/29. https://doi.org/10.1038/s41467-018-05373-4 PMID: 30054476 .
16. Beharry AA , Nagel ZD , Samson LD , Kool ET . Fluorogenic Real-Time Reporters of DNA Repair by MGMT, a Clinical Predictor of Antitumor Drug Response . PLoS One . 2016 ; 11 ( 4 ): e0152684 . Epub 2016 /04/02. https://doi.org/10.1371/journal.pone.0152684 PMID: 27035132 .
17. Nagel ZD , Margulies CM , Chaim IA , McRee SK , Mazzucato P , Ahmad A , et al. Multiplexed DNA repair assays for multiple lesions and multiple doses via transcription inhibition and transcriptional mutagenesis . Proceedings of the National Academy of Sciences of the United States of America . 2014 ; 111 ( 18 ): E1823 - 32 . Epub 4/22/14. https://doi.org/10.1073/pnas.1401182111 PMID: 24757057
18. Nagel ZD , Kitange GJ , Gupta SK , Joughin BA , Chaim IA , Mazzucato P , et al. DNA repair capacity in multiple pathways predicts chemoresistance in glioblastoma multiforme . Cancer research . 2016 .
19. Thomas A , Tanaka M , Trepel J , Reinhold WC , Rajapakse VN , Pommier Y. Temozolomide in the Era of Precision Medicine . Cancer research . 2017 ; 77 ( 4 ): 823 - 6 . Epub 2017/02/06. https://doi.org/10.1158/ 0008 - 5472 .CAN- 16 -2983 PMID: 28159862 .
20. Skopek TR , Liber HL , Penman BW , Thilly WG . Isolation of a human lymphoblastoid line heterozygous at the thymidine kinase locus: Possibility for a rapid human cell mutation assay . Biochemical and Biophysical Research Communications . 1978 ; 84 ( 2 ): 411 - 6 . http://dx.doi.org/10.1016/ 0006 - 291X ( 78 ) 90185 - 7 . PMID: 214074
21. Hickman MJ , Samson LD . Apoptotic signaling in response to a single type of DNA lesion, O(6)-methylguanine . Mol Cell . 2004 ; 14 ( 1 ): 105 - 16 . Epub 2004/04/08. PMID: 15068807 .
22. Fry RC , Svensson JP , Valiathan C , Wang E , Hogan BJ , Bhattacharya S , et al. Genomic predictors of interindividual differences in response to DNA damaging agents . Genes Dev . 2008 ; 22 ( 19 ): 2621 - 6 . https://doi.org/10.1101/gad.1688508 PMID: 18805990
23. Pearson SJ , Ferguson J , Santibanez-Koref M , Margison GP . Inhibition of O(6)-methylguanine-DNA methyltransferase by an alkyltransferase-like protein from Escherichia coli . Nucleic Acids Res . 2005 ; 33 ( 12 ): 3837 - 44 . https://doi.org/10.1093/nar/gki696 PMID: 16027108 .
24. Noonan EM , Shah D , Yaffe MB , Lauffenburger DA , Samson LD. O(6)-Methylguanine DNA lesions induce an intra-S-phase arrest from which cells exit into apoptosis governed by early and late multipathway signaling network activation. Integrative biology: quantitative biosciences from nano to macro . 2012 ; 4 ( 10 ): 1237 - 55 . https://doi.org/10.1039/c2ib20091k PMID: 22892544 .
25. Kitange GJ , Carlson BL , Schroeder MA , Grogan PT , Lamont JD , Decker PA , et al. Induction of MGMT expression is associated with temozolomide resistance in glioblastoma xenografts . Neuro-Oncology . 2009 ; 11 ( 3 ): 281 - 91 . https://doi.org/10.1215/ 15228517 -2008-090 PMID: 18952979
26. Collins FS , Brooks LD , Chakravarti A. A DNA polymorphism discovery resource for research on human genetic variation . Genome Res . 1998 ; 8 ( 12 ): 1229 - 31 . PMID: 9872978
27. Nagel ZD , Kitange GJ , Gupta SK , Joughin BA , Chaim IA , Mazzucato P , et al. DNA Repair Capacity in Multiple Pathways Predicts Chemoresistance in Glioblastoma Multiforme. Cancer Res . 2017 ; 77 ( 1 ): 198 - 206 . Epub 2016/10/30. https://doi.org/10.1158/ 0008 - 5472 .CAN- 16 -1151 PMID: 27793847 .
28. Zaidi NH , Liu L , Gerson SL . Quantitative immunohistochemical estimates of O6-alkylguanine-DNA alkyltransferase expression in normal and malignant human colon. Clinical cancer research: an official journal of the American Association for Cancer Research . 1996 ; 2 ( 3 ): 577 - 84 . Epub 1996/03/01. PMID: 9816206 .
29. Kitange GJ , Mladek AC , Carlson BL , Schroeder MA , Pokorny JL , Cen L , et al. Inhibition of histone deacetylation potentiates the evolution of acquired temozolomide resistance linked to MGMT upregulation in glioblastoma xenografts . Clinical cancer research: an official journal of the American Association for Cancer Research . 2012 ; 18 ( 15 ): 4070 - 9 . Epub 2012/06/08. https://doi.org/10.1158/ 1078 - 0432 .ccr- 12 - 0560 PMID: 22675172 .
30. Mendelsohn J. Personalizing Oncology: Perspectives and Prospects . Journal of Clinical Oncology . 2013 ; 31 ( 15 ): 1904 - 11 . https://doi.org/10.1200/JCO. 2012 . 45 .3605 PMID: 23589547
31. Hanahan D , Weinberg RA . Hallmarks of Cancer: The Next Generation . Cell . 2011 ; 144 ( 5 ): 646 - 74 . https://doi.org/10.1016/j.cell. 2011 . 02 .013 PMID: 21376230
32. Hegi ME , Diserens A , Gorlia T , Hamou M , de Tribolet N , Weller M , et al. MGMT gene silencing and benefit from temozolomide in glioblastoma . New England Journal of Medicine . 2005 ; 352 ( 10 ): 997 - 1003 . https://doi.org/10.1056/NEJMoa043331 PMID: 15758010
33. Inno A , Fanetti G , Di Bartolomeo M , Gori S , Maggi C , Cirillo M , et al. Role of MGMT as biomarker in colorectal cancer . World Journal of Clinical Cases: WJCC . 2014 ; 2 ( 12 ): 835 - 9 . https://doi.org/10.12998/ wjcc.v2. i12 .835 PMID: 25516857 .
34. Brandwein JM , Kassis J , Leber B , Hogge D , Howson-Jan K , Minden MD , et al. Phase II study of targeted therapy with temozolomide in acute myeloid leukaemia and high-risk myelodysplastic syndrome patients pre-screened for low O(6) -methylguanine DNA methyltransferase expression . British journal of haematology . 2014 ; 167 ( 5 ): 664 - 70 . Epub 2014/08/28. https://doi.org/10.1111/bjh.13094 PMID: 25160658 .
35. Hong Q , Chen X , Ye H , Zhou A , Gao Y , Jiang D , et al. Association between the methylation status of the MGMT promoter in bone marrow specimens and chemotherapy outcomes of patients with acute myeloid leukemia . Oncology letters . 2016 ; 11 ( 4 ): 2851 - 6 . Epub 2016/04/14. https://doi.org/10.3892/ol. 2016 .4317 PMID: 27073563 .
36. Jiang X , Reardon DA , Desjardins A , Vredenburgh JJ , Quinn JA , Austin AD , et al. O6 -methylguanineDNA methyltransferase (MGMT) immunohistochemistry as a predictor of resistance to temozolomide in primary CNS lymphoma . J Neurooncol . 2013 ; 114 ( 1 ): 135 - 40 . Epub 2013/05/21. https://doi.org/10. 1007/s11060-013-1162-y PMID: 23686298 .
37. Fumagalli C , Pruneri G , Possanzini P , Manzotti M , Barile M , Feroce I , et al. Methylation of O6-methylguanine-DNA methyltransferase (MGMT) promoter gene in triple-negative breast cancer patients . Breast cancer research and treatment. 2012 ; 134 ( 1 ): 131 - 7 . Epub 2012/01/10. https://doi.org/10.1007/ s10549-011-1945-9 PMID: 22228432 .
38. Cros J , Hentic O , Rebours V , Zappa M , Gille N , Theou-Anton N , et al. MGMT expression predicts response to temozolomide in pancreatic neuroendocrine tumors . Endocrine-related cancer . 2016 ; 23 ( 8 ): 625 - 33 . Epub 2016/06/30. https://doi.org/10.1530/ERC-16-0117 PMID: 27353036 .
39. Hassel JC , Sucker A , Edler L , Kurzen H , Moll I , Stresemann C , et al. MGMT gene promoter methylation correlates with tolerance of temozolomide treatment in melanoma but not with clinical outcome . Br J Cancer . 2010 ; 103 ( 6 ): 820 - 6 . Epub 2010/08/26. https://doi.org/10.1038/sj.bjc.6605796 PMID: 20736948 .
40. Esteller M , Herman JG . Generating mutations but providing chemosensitivity: the role of O6-methylguanine DNA methyltransferase in human cancer . Oncogene . 2004 ; 23 ( 1 ): 1 - 8 . Epub 2004/01/09. https:// doi.org/10.1038/sj.onc.1207316 PMID: 14712205 .
41. Preusser M , Janzer RC , Felsberg J , Reifenberger G , Hamou MF , Diserens AC , et al. Anti-O6-methylguanine-methyltransferase (MGMT) immunohistochemistry in glioblastoma multiforme: Observer variability and lack of association with patient survival impede its use as clinical biomarker . Brain Pathology . 2008 ; 18 ( 4 ): 520 - 32 . https://doi.org/10.1111/j.1750- 3639 . 2008 . 00153 . x PMID : 18400046
42. Kreth S , Thon N , Eigenbrod S , Lutz J , Ledderose C , Egensperger R , et al. O-methylguanine-DNA methyltransferase (MGMT) mRNA expression predicts outcome in malignant glioma independent of MGMT promoter methylation . PLoS One . 2011 ; 6 ( 2 ): e17156 . Epub 2011 /03/03. https://doi.org/10. 1371/journal.pone.0017156 PMID: 21365007 .
43. Khalil S , Fabbri E , Santangelo A , Bezzerri V , Cantu C , Di Gennaro G , et al. miRNA array screening reveals cooperative MGMT-regulation between miR-181d-5p and miR-409-3p in glioblastoma . Oncotarget . 2016 ; 7 ( 19 ): 28195 - 206 . Epub 2016/04/09. https://doi.org/10.18632/oncotarget.8618 PMID: 27057640 .
44. Liu L , Xu-Welliver M , Kanugula S , Pegg AE . Inactivation and degradation of O(6)-alkylguanine-DNA alkyltransferase after reaction with nitric oxide . Cancer Res . 2002 ; 62 ( 11 ): 3037 - 43 . Epub 2002/05/31. PMID: 12036910 .