RESEARCH ARTICLE

Human Glioblastoma-associated
Microglia/Monocytes Express a Distinct
RNA Profile Compared to Human Control
and Murine Samples
Frank Szulzewsky,1,2 Sonali Arora,2 Lot de Witte,3 Thomas Ulas,4 Darko Markovic,1,5
Joachim L. Schultze,4,6 Eric C. Holland,2 Michael Synowitz,1,7 Susanne A. Wolf,1 and
Helmut Kettenmann1

Glioblastoma (GBM) is the most aggressive brain tumor in adults. It is strongly infiltrated by microglia and peripheral monocytes
that support tumor growth. In the present study we used RNA sequencing to compare the expression profile of CD11b1 human
glioblastoma-associated microglia/monocytes (hGAMs) to CD11b1 microglia isolated from non-tumor samples. Hierarchical clus-
tering and principal component analysis showed a clear separation of the two sample groups and we identified 334 significantly
regulated genes in hGAMs. In comparison to human control microglia hGAMs upregulated genes associated with mitotic cell
cycle, cell migration, cell adhesion, and extracellular matrix organization. We validated the expression of several genes associated
with extracellular matrix organization in samples of human control microglia, hGAMs, and the hGAMs-depleted fraction via
qPCR. The comparison to murine GAMs (mGAMs) showed that both cell populations share a significant fraction of upregulated
transcripts compared with their respective controls. These genes were mostly related to mitotic cell cycle. However, in contrast to
murine cells, human GAMs did not upregulate genes associated to immune activation. Comparison of human and murine GAMs
expression data to several data sets of in vitro-activated human macrophages and murine microglia showed that, in contrast to
mGAMs, hGAMs share a smaller overlap to these data sets in general and in particular to cells activated by proinflammatory stim-
ulation with LPS 1 INFc or TNFa. Our findings provide new insights into the biology of human glioblastoma-associated microglia/
monocytes and give detailed information about the validity of murine experimental models.

GLIA 2016;64:1416–1437
Key words: glioma, glioblastoma, microglia, macrophage, monocyte, RNA sequencing

Introduction only 12–15 months after diagnosis, despite surgical tumor

resection, radio-, and chemotherapy (Wen and Kesari, 2008).
G lioblastoma multiforme (GBM, grade IV glioma) is a
highly aggressive brain tumor and accounts for around
50% of all malignant brain neoplasms (Dolecek et al., 2012).
Reasons for treatment failure include the highly invasive
growth and the cellular and genetic inter- and intra-tumor
Patients suffering from GBM have an average survival time of heterogeneity of these tumors (Brennan et al., 2009; Gill

View this article online at wileyonlinelibrary.com. DOI: 10.1002/glia.23014

Published online June 20, 2016 in Wiley Online Library (wileyonlinelibrary.com). Received Mar 16, 2016, Accepted for publication May 12, 2016.

Address correspondence to: Cellular Neurosciences, Max-Delbrueck-Center for Molecular Medicine in the Helmholtz Society, Berlin, Germany, Robert Roessle Str.
10 13125, Berlin.
E-mail: kettenmann@mdc-berlin.de

From the 1Department of Cellular Neurosciences, Max-Delbrueck-Center for Molecular Medicine in the Helmholtz Society, Berlin, Germany; 2Department of Human
Biology, Fred Hutchinson Cancer Research Center, Seattle, WA, USA; 3Department of Psychiatry, Brain Center Rudolf Magnus, University Medical Center Utrecht,
Utrecht, the Netherlands; 4Department of Genomics and Immunoregulation, Life and Medical Sciences Institute University of Bonn, Bonn, Germany; 5Department
of Neurosurgery, Helios Clinics, Berlin, Germany; 6Platform for Single Cell Genomics and Epigenomics at the German Center for Neurodegenerative Diseases and
the University of Bonn, Bonn, Germany; 7Department of Neurosurgery, University Medical Center Schleswig-Holstein (UKSH), Kiel, Germany

Michael Synowitz, Susanne A. Wolf and Helmut Kettenmann contributed equally to this work.

Additional Supporting Information may be found in the online version of this article.

1416 V
C 2016 Wiley Periodicals, Inc.

et al., 2014; Kim et al., 2015; Meyer et al., 2015; Patel et al.,
2014; Phillips et al., 2006; Sottoriva et al., 2013).
Similar to the situation in other solid tumors, tissue-
resident macrophages—microglia in the brain—and infil-
trating monocytes are attracted toward GBM in large num-
bers and can amount up to 30% of the cells in the tumor
tissue (Charles et al., 2011; Kerber et al., 2008; Watters
et al., 2005). These glioblastoma-associated microglia/
monocytes (GAMs) are exposed to tumor-derived factors
and are programmed into a tumor specific phenotype (Hu
et al., 2015; Vinnakota et al., 2013). GAMs actively sup-
port tumor growth, e.g. by secreting matrix-degrading
enzymes, cytokines, and immunosuppressive factors (Car-
valho da Fonseca et al., 2014; Coniglio et al., 2012; Feng
et al., 2015; Hu et al., 2014; Li and Graeber, 2012; Mar-
kovic et al., 2009). In addition, recent studies using glioma
mouse models have shown that GAMs-directed therapeutic
approaches can be beneficial for disease intervention (Kees
et al., 2012; Lisi et al., 2014; Markovic et al., 2011; Pyon-
teck et al., 2013; Sarkar et al., 2014; Xu et al., 2014).
However, it is unclear if these findings also hold true for
human GAMs (hGAMs) and if approaches that were devel-
oped in mouse models can be successfully applied for the
treatment of human GBMs (Butowski et al., 2015). Thus, a
better understanding of hGAMs is crucial for the develop-
ment of future therapies.
The bipolar axis of in vitro macrophage activation, rang-
ing from proinflammatory activation after stimulation with
LPS or IFNc (previously termed M1 activation) to anti-
inflammatory activation after stimulation with IL4 (previously
termed M2 activation) has been used to classify GAMs in the
past. However, this simplistic model of macrophage activation
has been challenged by many studies and the M1/M2 model
of macrophage polarization has been replaced by a multi-
dimensional model of macrophage activation (Edin et al.,
2012; Ginhoux et al., 2015; Murray et al., 2014; Reinartz
et al., 2014; Xue et al., 2014). To define a glioma-associated
phenotype it seems more promising to compare GAMs with
the phenotypes of a broader activation spectrum than with
M1 and M2 polarized macrophages only. Several studies have
investigated the cytokine expression levels of GAMs isolated
from human and murine samples. Studies on murine GAMs
(mGAMs) have shown that these cells produce several anti-
inflammatory molecules (e.g., TGFb1, ARG1, and IL10),
and molecules supporting tissue remodeling and angiogenesis
(e.g., VEGF, MMP2, MMP9, and MT1-MMP), but in turn
also produce pro-inflammatory molecules (e.g., IFNc, TNFa,
IL1b, IL-6, and CXCL10) (Gabrusiewicz et al., 2011; Sica
et al., 2006; Umemura et al., 2008; Ye et al., 2012). In con-
trast, studies on hGAMs could not detect the expression of
August 2016
Szulzewsky et al.: Human GAMs RNA Sequencing
several proinflammatory factors, such as IFNc, TNFa, and
IL1b (Hattermann et al., 2014; Hussain et al., 2006a,b).
We have previously published a limited transcriptome
experiment based on microarrays that investigated the expres-
sion pattern of CD11b1 mGAMs isolated from experimental
GL261 mouse glioma compared with naive brain-resident
microglia and peripheral macrophages (Szulzewsky et al.,
2015). We could show that these cells exhibit an activation state
that only partially overlaps with the expression pattern of classi-
cal pro- or anti-inflammatory activated cultured macrophages.
In addition we identified several factors produced by mGAMs
that might support glioma progression. It is of high importance
to validate rodent models to determine relevant pathways, mol-
ecules and therapeutic targets for potential human treatment
strategies. So far a direct comparison of murine and human
expression profiles of CD11b1 microglial cells is lacking.
In this study, we isolated CD11b1 cells from human
GBM samples and non-tumor control samples and performed
a genome-wide RNA sequencing analysis. We compared the
hGAMs expression profile to our previous mGAMs data set
and in vitro-activated macrophages and microglia and could
show that, in contrast to mGAMs, hGAMs did not upregu-
late genes associated with immune activation.
Materials and Methods
Ethics Statement
Freshly resected patient samples were provided by the Department
for Neurosurgery Charite University Hospital and the Helios Clinics
(both Berlin, Germany). Handling and analysis of these tissues was
performed in accordance with the 1964 Helsinki declaration and its
later amendments and according to the rules and with the approval
of the Ethical Committee and with patient’s written consents (Ethics
Committee: “Ethikkommission der Charite—Universit€atsmedizin
Berlin,” application number: EA4/098/11).
Postmortem tissue was obtained from the Netherlands Brain
Bank (NBB; Amsterdam, the Netherlands, www.brainbank.nl).
Informed consent was obtained by the NBB for brain autopsy and
for the use of material and clinical data for research purposes, in
compliance with institutional and national ethical guidelines.
Cell Isolation
Human tissue samples were dissociated using the Neural Tissue Dis-
sociation Kit (Miltenyi Biotec, Bergisch-Gladbach, Germany) accord-
ing to the manufacturer’s instructions. To remove the myelin we
followed a protocol published elsewhere (Olah et al., 2012). In brief,
the brain cell suspension was mixed with a 22% Percoll (Geyer, Ren-
ningen, Germany) solution and a layer of cold PBS (Gibco-Invitro-
gen, Darmstadt, Germany) was added on top. Centrifugation at
950g with slow acceleration and without breaks created a gradient
that separated the cell pellet on the bottom of the tube from the
myelin which was carefully aspirated. Cells were collected, washed
once with PBS, and centrifuged again at 300g for 10 min. Subse-
quently, the cell pellet was resuspended in PBS, containing 0.5%
1417
FCS and 2 mM EDTA, and labeled with anti-CD11b microbeadsTM
(Miltenyi Biotec, 130-093-634). The MACS isolation was carried
out according to the manufacturer’s instructions and cells were sub-
sequently used for RNA isolation. MACS-purified cells were stained
with anti-human CD11b antibody conjugated to PE (eBioscience,
12-0118-41, San Diego, CA) and analyzed using a BD LSR Fortessa
(BD Bioscience, Franklin Lakes, NJ).
RNA Protocols
Total RNA was isolated using the RNeasy Plus Mini Kit (Qiagen,
Hilden, Germany). On-column DNase 1 (Qiagen) digestion was
performed and total RNA was eluted in RNase-free water. RNA
yield was measured using a Nanodrop 1000 (Nanodrop, Wilming-
ton, DE) spectrophotometer and quality was assessed using an Agi-
lent 2100 Bioanalyzer (Agilent, Santa Clara, CA). Samples were
stored at 2808C until further use. For qRT-PCR, first strand cDNA
synthesis of RNA was done using the Superscript II (Invitrogen)
reverse transcriptase according to the manufacturer’s instructions. For
mRNA transcription oligo-dT primers (Invitrogen, Darmstadt, Ger-
many) were used. cDNA was stored at 2208C until further
processing.
RNA Sequencing
Total RNA was converted into libraries of double stranded cDNA
molecules as a template for high throughput sequencing following
the manufacturer’s recommendations using the Illumina TruSeq
RNA Sample Preparation Kit v2. Shortly, mRNA was purified from
100 ng of total RNA using poly-T oligo-attached magnetic beads.
Fragmentation was carried out using divalent cations under elevated
temperature in Illumina proprietary fragmentation buffer. First
strand cDNA was synthesized using random oligonucleotides and
SuperScript II. Second strand cDNA synthesis was subsequently per-
formed using DNA Polymerase I and RNase H. Remaining over-
hangs were converted into blunt ends via exonuclease/polymerase
activities and enzymes were removed. After adenylation of 30 ends of
DNA fragments, Illumina PE adapter oligonucleotides were ligated
to prepare for hybridization. DNA fragments with ligated adapter
molecules were selectively enriched using Illumina PCR primer
PE1.0 and PE2.0 in a 15 cycle PCR reaction. Size-selection and
purification of cDNA fragments with preferentially 200 bp in length
was performed using SPRIBeads (Beckman-Coulter, Krefeld, Ger-
many). Size-distribution of cDNA libraries was measured using the
Agilent high sensitivity DNA assay on a Tapestation system (Agi-
lent). cDNA libraries were quantified using KAPA Library Quantifi-
cation Kits (Kapa Biosystems, London, UK). After cluster generation
on a cBot, a 75 bp single-read run was performed on a HiSeq1500.
Base calling and de-multipexing was performed using CASAVA ver-
sion 1.8.
Bioinformatic Analysis
RNAseq reads were aligned to UCSC hg19 using TopHat2 (v
2.0.12) (Kim et al., 2013), allowing for two mismatches. For each
sample, a count table for all genes was generated using R function
summarizeOverlaps from the R/Bioconductor (Gentleman et al.,
2004; R Development Core Team, 2011) package GenomicAlign-
1418
ments (v 1.7.3) using only the primary alignments of each read
(Lawrence et al., 2013). The count table was prefiltered to keep only
those genes for which each of the tumor samples had a count of
more than 3. To assess the sample similarity between samples
belonging to the same group, we rlog-transformed the raw count
table using R function rlog from the R/Bioconductor package
DESeq2 (v 1.11.8) (Love et al., 2014) and then used the R function
dist to calculate Euclidean distance between samples. We computed
the PCA and MDS for the above distances and visualized it using R
package ggplot2 (v 1.0.1) (Wickham, 2009). Differential expression
analysis between the tumor and the normal samples was performed
using the Bioconductor package DESeq2 (v 1.11.8) in R (version
3.1). A one-way anova was also performed using DESeq2 (v 1.11.8)
to assess the difference between the three groups. A gene was consid-
ered significant if the adjusted P value was <0.05 and the absolute
log2 fold change was more than1 for DESeq2. Sequence data was
deposited at Sequence Read Archive (SRA) under SRA accession
number SRP071750. Raw and FPKM counts per gene per sample
can be accessed at NCBI Gene Expression Omnibus under
GSE80338.
Gene Ontology (GO)-based enrichment tests were imple-
mented using GOseq (v 1.23.0) (Young et al., 2010), which corrects
for gene length bias. GO terms with corrected P values < 0.05 were
considered significantly enriched (q value < 0.05). Upregulated genes
from various lists were enriched for Reactome and KEGG Pathways
using R/Bioconductor package ReactomePA (v 1.15.4) and cluster-
Profiler (v 2.5.4), respectively (Croft et al., 2014; Kanehisa and
Goto, 2000; Kanehisa et al., 2016; Milacic et al., 2012; Yu and He,
2016; Yu et al., 2012).
Fisher Hypergeometric tests were implemented to compare
gene sets from other data sets with the DE genes from the human
GAMs RNASeq analysis to see if the DE genes in the set are more
common than the DE genes not in the set.
The functional analyses were generated through the use of
QIAGEN’s Ingenuity Pathway Analysis (IPA, QIAGEN Redwood
City, www.qiagen.com/ingenuity).
Transcription factor binding site enrichment analysis was done
using Webgestalt (Wang et al., 2013; Zhang et al., 2005), iRegulon
(Janky et al., 2014), and Cytoscape (Shannon et al., 2003).
Statistical analysis for qRT-PCRs was done with GraphPad
Prism 6.
Data sets for comparison were downloaded from NCBI Gene
Expression Omnibus or EBI ArrayExpress. Data from murine
GAMs (Szulzewsky et al., 2015): E-MTAB-2660. Data from in
vitro-stimulated human macrophages (Xue et al., 2014): GSE47189.
Data from in vitro-stimulated murine microglia: GSE77064. Data
from human lung cancer-associated immune cells (Durrans et al.,
2015): GSE68795.
qRT-PCR
Quantitative RT-PCR reactions were performed using the SYBR
Select Mastermix (Applied Biosystems, Foster City, CA) according to
the manufacturer’s instructions on a QuantStudio 7 Flex Real-Time
PCR System (Applied Biosystems).
Volume 64, No. 8

Primers for target genes were designed to recognize all vali-
dated mRNA splice variants of the gene, relying on the RefSeq
sequences of the UCSC genome browser (http://genome.ucsc.edu/).
The primer sequences were generated using the Primer-Blast tool
(http://www.ncbi.nlm.nih.gov/tools/primer-blast/). Amplification of
unintended targets was excluded by BLAST search (http://blast.ncbi.
nlm.nih.gov/Blast.cgi). Calculation of unfavorable secondary struc-
tures and primer-dimer amplification was done with the IDT oligo
analyzer (http://eu.idtdna.com/analyzer/applications/oligoanalyzer/).
ACTB Fwd: GAGCTACGAGCTGCCTGAC; ACTB Rev:
GACTCCATGCCCAGGAAGG; ADAMDEC1 Fwd: ATCACT-
CAAAAGCCCAGAGAAAG; ADAMDEC1 Rev: TAGGTTCATCA-
CATCAAACACAAAG; ADAMTS2 Fwd: ATCACGCCATCTTCCT
CACA; ADAMTS2 Rev: GCCACCACAAACGCTGA;
AIF1 Fwd: CGAATGCTGGAGAAACTTGGA; AIF1 Rev:
GAAAGTCAGGGTAGCTGAACG; CTSL Fwd: GACATCCC-
TAAGCAGGAGAAGG; CTSL Rev: AATCCGTAGCCAACCACC
AG;
FN1 Fwd: ACAGAACTATGATGCCGACCA; FN1 Rev:
CACGACCATTCCCAACACAC;
GFAP Fwd: CGCACGCAGTATGAGGCAA; GFAP Rev:
GACTCCAGGTCGCAGGTCAA;
HARS Fwd: GGTGAAGAATGAGATGGTGGGA; HARS
Rev: GCCTGCTTGTTTTGGGATAGT;
IBSP Fwd: ATCCGAAGAAAATGGAGATGACAG;
IBSP Rev: GTGTTGTAGCAGAAAGTGTGGTATT;
RND3 Fwd: GATGCTGTGCTGATTTGCTTTGA; RND3
Rev: CGTCTGCCTGTGATTGGAGAG; SDC3 Fwd: GATG
ACTCCTTTCCCGATGATG; SDC3 Rev: TGTCTCAATGCCCG
ACTCC.
Results
Gene Expression Profile of Human
Glioblastoma-associated Microglia/Monocytes
To identify tumor-regulated transcripts in hGAMs, we used
MACS to isolate CD11b1 cells from human glioblastoma
samples (primary and recurrent tumors) and from non-tumor
brain samples (samples from epilepsy, hippocampal biopsies,
and postmortem cortex samples) (Supporting Information
Table S1). FACS analysis of sorted samples revealed a high
purity of CD11b1 sorted cells. We isolated total RNA from
these samples and performed RNA sequencing on an Illumina
HiSeq 1500. We validated the enrichment of CD11b1 cells
in our samples using qPCR analyzing the expression of Iba1
and GFAP in the CD11b positive and negative fractions and
found a high enrichment of Iba1 in the positive fractions
(Supporting Information Fig. S1).
During data processing we implemented a nonspecific
filter to select only genes that had at least three reads in each
hGAMs sample which resulted in 11,777 remaining genes.
Principal component analysis (PCA) and multidimensional
scaling (MDS) plots of the 16 samples clearly distinguished
the eight hGAMs samples from microglia-derived from post-
August 2016
Szulzewsky et al.: Human GAMs RNA Sequencing
mortem and hippocampal and epilepsy tissues. The eight
tumor-derived samples cluster together on the left side of the
plot, whereas the five postmortem nontumor samples cluster
together on the right side. The three non-tumor samples iso-
lated from two epilepsy patients and one patient suffering
from hippocampal sclerosis cluster between these two groups
but align toward the postmortem nontumor samples (Fig.
1A, Supporting Information Fig. S2). In addition, unsuper-
vised hierarchical clustering based on the variance of all
11,777 genes showed a clear separation of the eight hGAMs
samples from the eight control samples (Fig. 1B). The cluster-
ing indicated no separation of hGAMs isolated from primary
and recurrent glioblastoma samples.
We performed a one-way ANOVA to compare the gene
expression profiles of the eight hGAMs samples, the three
microglia samples isolated from epilepsy/hippocampal biop-
sies, and the five microglia samples isolated from postmortem
tissues (Fig. 1C, Supporting Information Fig. S3 and Sup-
porting Information dataset S1). Nearly 200 genes were sig-
nificantly upregulated in hGAMs when compared with the
epilepsy and hippocampal control samples, whereas only 12
genes were downregulated. When compared with the post-
mortem control samples 384 genes were upregulated and 180
genes were downregulated in hGAMs. However, we found no
genes being upregulated and only 15 genes downregulated in
the samples isolated from epilepsy and hippocampal tissues
when compared with the samples isolated from postmortem
tissues, indicating that these two control groups were very
similar.
As the two control group data sets clustered similarly in
the unsupervised hierarchical clustering and based on the
results of the ANOVA we combined both control groups and
compared the eight hGAMs samples to all eight control sam-
ples for all subsequent analysis. We identified 292 and 42
genes that were significantly (Padj  0.05)  two-fold up- or
downregulated, respectively (Fig. 1D,E; Tables 1 and 2; Sup-
porting Information Tables S2 and S3; Supporting Informa-
tion dataset S2).
hGAMs Upregulate Genes Associated with “Mitotic
Cell Cycle” and “Extracellular Matrix Organization”
We performed gene ontology (GO) analysis to functionally
link the 292 up- and 42 downregulated genes in hGAMs.
Terms related to “mitotic cell cycle”, “extracellular matrix
organization”, “organ development”, “cell adhesion”, “cell rec-
ognition”, “positive regulation of cell migration”/“leukocyte
migration”, “protein phosphorylation”, “collagen catabolic”/
“metabolic process” were significantly overrepresented in the
292 upregulated genes in hGAMs (Table 3, Supporting Infor-
mation dataset S3). Interestingly also the terms for “blood
vessel development”/“angiogenesis” and “response to
1419
FIGURE 1: Altered gene expression in hGAMs compared with human control microglia A: Principal component analysis (PCA) plot of all
16 CD11b1 MACS-sorted samples shows a clear separation of tumor-derived hGAMs and non-tumor-derived samples. B: Unsupervised
hierarchical clustering of the 16 samples based on the variances of all 11,777 genes. C: A one-way ANOVA was applied to calculate the
number of differentially expressed genes between hGAMs, control microglia isolated from post mortem samples, and control microglia
isolated from epilepsy and hippocampal samples. D: The number of differentially expressed genes between the eight hGAMs and all
eight control CD11b1 samples. E: Heatmap of the 292 up- and 42 downregulated differentially expressed genes between the eight
hGAMs and all eight control CD11b1 samples.

wounding”/“wound healing” were significantly overrepre-
sented, functions that are typical features of other solid-
tumor-associated macrophages. In contrast, there were no sig-
nificantly overrepresented terms for the 42 downregulated
genes in hGAMs compared with the eight control samples.
We used Webgestalt to analyze the 292 upregulated
genes in hGAMs for known transcription factor binding sites
and found a highly significant enrichment for E2F family
members, as well as E2F1 and E2F4 in particular (Supporting
Information dataset S4). We performed an additional analysis
using iRegulon and found an enrichment for several E2F
family members, FOXM1, TFDP1, FOSL2, SINA3, and
JUND (Supporting Information data set S5). In accordance
with this, the expression of E2F1, E2F7, E2F8, and FOXM1
was significantly upregulated in hGAMs. The expression of
E2F4 was 1.4 fold downregulated. E2F transcription factors
are key regulators of cell cycle progression, and an upregula-
tion of these genes might facilitate the high abundance of
genes related to mitosis and cell cycle regulation/progression
in the upregulated genes.
We used ingenuity pathway analysis (IPA) to further
analyze our data set. In accordance with the GO analysis we

1420 Volume 64, No. 8

 Szulzewsky et al.: Human GAMs RNA Sequencing

TABLE 1: Highest Upregulated Genes in Human Glioblastoma-associated Microglia/Monocytes Compared with

Controls

Symbol refseq_nm FC P value Padj Tumor Normal

mean expr mean expr

IBSP NM_004967 446.60 5.55E-46 6.53E-42 10.28 3.59

ADAMDEC1 NM_001145271 88.07 2.33E-26 3.92E-23 8.04 3.30
CA12 NM_001218 87.82 1.04E-33 6.15E-30 7.02 2.68
DTL NM_001286229 60.87 7.23E-26 1.06E-22 5.60 2.22
CD109 NM_001159587 55.74 5.65E-27 1.66E-23 8.69 3.97
CCL18 NM_002988 53.91 1.64E-15 9.67E-13 5.71 2.05
CHI3L2 NM_001025197 51.43 2.19E-17 1.84E-14 7.41 3.16
MYBL2 NM_001278610 46.25 1.17E-27 4.60E-24 6.63 2.83
FN1 NM_001306129 44.76 1.60E-17 1.45E-14 13.46 8.39
FAM111B NM_001142703 43.96 6.25E-22 6.70E-19 5.46 2.34
TNC NM_002160 40.50 8.04E-13 3.16E-10 9.52 4.57
HAS2 NM_005328 37.10 6.12E-12 2.00E-09 4.18 1.57
CRISPLD1 NM_001286777 34.93 3.94E-11 1.16E-08 4.87 1.79
PKIB NM_001270393 30.50 1.90E-14 9.32E-12 6.92 3.35
CHL1 NM_001253387 30.45 8.05E-13 3.16E-10 6.50 2.94
E2F8 NM_001256371 28.44 3.97E-13 1.67E-10 4.18 1.96
DLGAP5 NM_001146015 26.21 1.10E-22 1.29E-19 5.17 2.47
RRM2 NM_001034 25.61 7.81E-27 1.84E-23 7.89 4.22
PLOD2 NM_000935 25.18 4.55E-20 4.46E-17 8.17 4.57
RGMA NM_001166283 24.49 4.71E-10 9.57E-08 6.49 3.03
CA9 NM_001216 24.17 7.98E-10 1.59E-07 3.60 1.63
MT1H NM_005951 23.98 3.25E-09 5.32E-07 4.86 1.95
SDC1 NM_001006946 22.58 5.74E-09 8.55E-07 3.39 1.53
SERPINA3 NM_001085 22.02 1.02E-07 1.09E-05 9.31 4.53
CDK1 NM_001130829 21.95 5.89E-11 1.61E-08 7.40 3.41

found a highly significant enrichment for the pathways “cell
cycle control of chromosomal replication” and “mitotic roles
of polo-like kinases” (Fig. 2).
Human and Mouse GAMs Share Transcripts
Associated with “Mitotic Cell Cycle” and
“Extracellular Matrix Organization”
We have previously shown that mGAMs isolated from experi-
mental GL261 mouse gliomas exhibit a gene expression pat-
tern that is different from na€ıve microglia or in vitro-activated
macrophages (Szulzewsky et al., 2015). Both human and
murine GAMs were isolated with the same method and
sorted using CD11b as a marker. We calculated the signifi-
cantly regulated genes (fold change  2, Padj  0.05), excluded
genes that are not present in human and found 940 genes
that were upregulated and 1005 genes that were
downregulated in mGAMs in comparison to na€ıve mouse
microglia (Supporting Information dataset S6).
To investigate similarities between human and murine
GAMs we compared both data sets for overlap using gene set
enrichment analysis. We found that the 292 upregulated genes
in hGAMs were significantly enriched in the 940 upregulated
genes in mGAMs (P 5 2.2 3 10216) and 111 genes (38% of
the upregulated genes in hGAMs) were upregulated in both

August 2016 1421

 TABLE 2: Highest Downregulated Genes in Human Glioblastoma-associated Microglia/Monocytes Compared with
Controls

Symbol refseq_nm FC P value Padj Tumor Normal

mean expr mean expr

MLXIPL NM_032951 210.37 4.50E-08 5.15E-06 8.14 11.07

AANAT NM_001088 29.89 3.88E-11 1.16E-08 5.05 7.52
BATF3 NM_018664 27.89 3.58E-15 1.91E-12 5.31 7.81
ADAMTSL2 NM_001145320 26.94 5.34E-06 0.000379 7.90 10.25
HIST2H2BF NM_001024599 26.73 5.35E-08 6.00E-06 6.06 8.15
IPCEF1 NM_001130699 26.58 0.000665 0.027181 7.67 10.14
SPNS2 NM_001124758 26.48 6.29E-05 0.003383 7.73 10.26
C1orf228 NM_001004307 26.27 4.70E-05 0.002634 5.69 7.56
HIST2H2BE NM_003528 26.26 6.08E-07 5.33E-05 8.14 10.09
JAG2 NM_002226 26.16 1.48E-05 0.000962 6.58 8.70
JAK3 NM_000215 26.12 0.000156 0.007645 9.51 11.37
MCF2L NM_001112732 26.00 1.12E-05 0.000747 9.15 11.49
HIST1H2BD NM_021063 26.00 0.000418 0.017522 5.80 7.52
DHRS9 NM_001142270 25.98 7.55E-05 0.003916 6.45 9.17
FAM149A NM_001006655 25.93 3.49E-05 0.002016 6.65 8.78
BHLHE41 NM_030762 25.88 2.12E-06 0.000167 10.53 12.71
TAL1 NM_001287347 25.83 0.000357 0.015235 8.36 10.86
SLC15A2 NM_001145998 25.83 0.001825 0.062672 7.44 9.63
LOC102724323 NR_120674 25.72 0.000498 0.020719 4.97 6.67
ZNF514 NM_032788 25.48 1.59E-05 0.001023 5.75 7.62
SRGAP2 NM_001042758 25.43 4.32E-10 9.08E-08 10.44 12.41
ST6GALNAC3 NM_001160011 25.23 5.35E-05 0.002914 5.99 7.79
LINC01268 NR_038863 25.19 0.001135 0.041238 5.85 7.60
RGL3 NM_001035223 25.11 0.000351 0.015055 6.68 8.40
GSTM2 NM_000848 25.03 0.000736 0.029592 8.27 10.06

hGAMs and mGAMs in comparison to the corresponding naive
control microglia (Fig. 3A,B, Supporting Information dataset
S7). We found no significant overlap between the
downregulated genes in hGAMs and mGAMs.
We performed a GO analysis with the 111 genes that
were upregulated in both human and murine GAMs. Most of
the significantly overrepresented terms were related to
“mitotic cell cycle”; however, although less abundant, the
terms “cell adhesion”, “response to stimulus”, and
“extracellular matrix organization” were also significantly
overrepresented (Fig. 3C,D, Supporting Information dataset
S8). This indicates that the upregulation of transcripts
involved in cell cycle control is a common feature of both
human and murine GAMs. Using Webgestalt we analyzed
these shared genes for known transcription factor binding
sites and found a highly significant enrichment for E2F fam-
ily members. In fact, both E2F1 and E2F8 were upregulated
in both human and murine GAMs.
In contrast, when looking at the 181 genes that were
significantly upregulated in human GAMs but not in murine
GAMs, we only found very few significantly overrepresented
GO terms. These terms were related to “system/organ devel-
opment”, “extracellular matrix organization”, “cell recog-
nition”, “cell motility”, and “collagen metabolic/catabolic
process”. However, these 181 genes were not significantly
enriched for terms related to “mitotic cell cycle” (Fig. 3C,

1422 Volume 64, No. 8

 TABLE 3: Significantly Enriched Gene Ontology Terms in the Upregulated Genes of Glioblastoma-associated Microglia/Monocytes

Category Term Over Enriched Genes in Gene symbols

August 2016
represented genes category
P value

GO:0000278 Mitotic 1.80E-20 73 854 CDKN3, NDC80, CENPE, CENPF, NES, UBE2C, CHEK1, CIT, ZWINT,
cell cycle CDCA5, E2F7, KIF18B, ESCO2, DHFR, E2F1, SKA1, SKA3, TPX2,
FOXM1, NCAPH, KIF4A, ASPM, BIRC5, ID4, KIF11, KIFC1, MAD2L1,
MCM2, MCM4, MYBL2, NEK2, ORC1, NUSAP1, GTSE1, GINS2,
NCAPG2, CDCA8, CEP55, FANCI, MCM10, PRKAR2B, CASC5, RRM1,
RRM2, CLSPN, CENPK, NCAPG, BUB1, BUB1B, TOP2A, TTK, TYMS,
WEE1, CENPM, E2F8, FZD3, CDT1, CDC45, NUF2, CCNB1, TICRR,
PRC1, PKMYT1, AURKB, CD28, KIF23, ESPL1, DLGAP5, CDK1,
MELK, GINS1, CDC6, KIF14
GO:0030198 Extracellular 5.45E-10 26 233 FBLN5, COL1A2, COL5A2, COL6A1, COL8A1, CTSL, AGT, EFEMP1,
matrix organization FN1, ANXA2, HAS2, TNC, IBSP, ITGA3, ITGA4, LAMC1, LGALS3,
LOX, MMP14, PLOD2, SDC1, SDC4, SPARC, TGFBI, ADAMTS2,
SDC3
GO:0048513 Organ development 1.92E-09 85 1717 GPNMB, CENPF, NES, CCR7, COL1A2, COL5A2, COL8A1, CD109,
E2F7, ESCO2, ACE, AEBP1, AGT, E2F1, APLNR, ECM1, EDNRB, AHR,
AK4, EFEMP1, PALLD, FOXM1, FLNB, CLEC5A, CADM1, FJX1,
ASPM, GCNT1, GPC1, ANXA2, HAS2, HMGB3, HOXB6, TNC, IBSP,
ID4, IL2RA, IL6, IL7R, AQP1, ITGA3, ITGA4, AREG, CCDC103,
LGALS1, LOX, SMAD1, KITLG, MKI67, MMP14, MT1G, NNMT,
NT5E, PLAU, PRRX1, ACP5, GNG12, ANKH, C8orf4, VANGL2,
PTPRZ1, RFX4, RRM1, SDC1, SDC4, SPARC, TGFBI, TGM2, TK1,
TOP2A, TYMS, VIM, E2F8, FZD3, C11orf63, KCNAB2, NRP1, APLN,
CCNB1, PAPSS2, CD28, ADAMTS2, CELSR1, MELK, KIF14
GO:0009611 Response to 4.88E-08 42 651 CENPE, CHL1, CLU, CCR7, CNR1, COL1A2, CD109, DHFR, DPYSL3,
wounding AGT, EDNRB, F3, FABP5, FN1, KIF4A, ANXA2, TNC, IL2RA, IL6,
AQP1, ITGA3, ITGA4, KIF11, KIFC1, LGALS1, LOX, NT5E, PLAU,
ACP5, PRKAR2B, VANGL2, CCL18, CCL20, SDC1, SDC4, SPARC,
WEE1, CCNB1, PAPSS2, CD28, KIF23, CELSR1
GO:0007155 Cell adhesion 4.65E-07 47 831 CDH2, GPNMB, FBLN5, CHL1, CCR7, COL6A1, COL8A1, PLEKHA7,
AGT, FN1, CADM1, GCNT1, TNFRSF21, ADAMDEC1, HAS2, TNC,
IBSP, IGFBP2, IL2RA, IL6, IL7R, ITGA3, ITGA4, RND3, LAMC1,
LGALS1, LGALS3, LY9, KITLG, MMP14, NRCAM, NT5E, PCDHGC3,
PLAU, EPDR1, ACKR3, CADM3, S100A10, SDC4, SIGLEC1, TGFBI,
TGM2, PPAP2B, CD28, CELSR1, SDC3, KIF14

1423
Szulzewsky et al.: Human GAMs RNA Sequencing
 Supporting Information datasets S9, S10, and S11). This

CDH2, COL1A2, COL8A1, E2F7, AGT, ECM1, AHR, F3, FOXM1, FN1,
indicates that most of the genes that were upregulated in

ANXA2, HAS2, IL6, AQP1, LOX, MMP14, NRCAM, ANGPTL4, PLAU,

hGAMs and related to “mitotic cell cycle” were also upregu-

LGALS3, KITLG, MMP14, PLAU, CCL20, SPARC, PPAP2B, NRP1

CCR7, AGT, F3, FN1, CXCL2, HAS2, IL6, AQP1, ITGA3, ITGA4,
lated in mGAMs isolated from experimental GL261 mouse

CATSPER1, CCR7, CNR1, CADM1, LGALS3, NRCAM, CADM3,

PRRX1, ACKR3, SPARC, TGFBI, SCG2, E2F8, PPAP2B, NRP1
gliomas and that a cell cycle activation is a common feature
of both mouse and human GAMs.
Next, we investigated the enrichment of Reactome and
KEGG pathways in genes that were upregulated in hGAMs
and mGAMs, as well as the genes that were shared between
human and murine GAMs and also genes that were only
upregulated in one species but not the other. Similar to the
results from the GO analysis, the most significant enriched
Reactome pathways in the upregulated genes in hGAMs were
related to “mitotic cell cycle” and “extracellular matrix organ-
ization” (Fig. 4A, Supporting Information dataset S12). The
most significantly enriched genes in the upregulated genes in
mGAMs were related as well to “mitotic cell cycle” and
“extracellular matrix organization”, but in contrast to the
COLEC12, NRP1

upregulated genes in hGAMs also to Reactome pathways

Gene symbols

related to “immune system” and “cytokine signaling” (Fig.

4B). In accordance to the results from the GO analysis, the
Reactome pathways related to “mitotic cell cycle” were also
highly enriched in the 111 genes shared between human and
murine GAMs. In turn, genes that were specific for either
hGAMs or mGAMs were not enriched in “mitotic cell
Genes in
category

cycle”-related pathways (Fig. 4C–E). We found a similar

result when looking at the KEGG pathways (Supporting
357

233
62

Information Fig. S4).

Enriched

Human GAMs Express Several Extracellular Matrix-

genes

associated Genes
27

18

The terms “extracellular matrix space/organization” was pres-

9

ent in the significantly overrepresented GO terms, as well as

represented

KEGG and Reactome, pathways of the significantly upregu-

1.42E-06

3.81E-05

4.88E-05
P value

lated genes in hGAMs and mGAMs. We used qRT-PCR to

Over

measure the expression of eight genes related to “extracellular

matrix space/organization” in the CD11b1 hGAMs samples
in comparison to the three epilepsy/hippocampal non-tumor
samples and the CD11b2 fractions of four matched tumors.
Positive regulation
of cell migration
Cell recognition

We could validate an overall higher expression of all investi-

development
Blood vessel

gated genes in hGAMs when compared with control micro-

glia. When comparing the expression in hGAMs versus the
CD11b2 tumor fraction we found that IBSP, SPP1, ADAM-
Term
TABLE 3: Continued

DEC1, ADAMTS2, SDC3, and CTSL were expressed at

higher levels in the CD11b1 fraction, whereas FN1 and
GO:0001568

GO:0008037

GO:0030335

RND3 showed no clear pattern of expression between the

CD11b1 or CD11b2 fraction different tumor samples (Fig.
Category

5). SPP1, ADAMDEC1, SDC3, RND3, and FN1 were upreg-

ulated in both hGAMs and mGAMs, whereas IBSP, CTSL
and ADAMTS2 were only upregulated in hGAMs.

1424 Volume 64, No. 8

 Szulzewsky et al.: Human GAMs RNA Sequencing

FIGURE 2: Human glioblastoma-associated microglia/monocytes upregulate cell cycle-associated genes Schematics of the pathways “cell
cycle control of chromosomal replication” and “mitotic roles of polo-like kinases”. Proteins highlighted in red are encoded by genes
that were significantly upregulated (Fold change  2, Padj £ 0.05) in hGAMs. Images were created through the use of IPA.

Human GAMs Lack Transcripts Associated with
Immune Activation
Gene ontology analysis revealed no significant overrepresenta-
tion of terms related to “immune activation/response” in the
upregulated genes in hGAMs. In contrast, these terms were
significantly overrepresented in the significantly upregulated
genes in mGAMs isolated from GL261 tumors (Fig. 3C). In
addition, we found no significant activation or inhibition of
general immune pathways in hGAMs using IPA (Supporting
Information Fig. S5), whereas, in mGAMs we found a signifi-
cant activation of the pathways “interferon signaling”,
“complement system”, and “pattern recognition receptors”.
We analyzed how genes that were upregulated in
mGAMs and were part of the GO term “immune response”
were regulated in hGAMs (Supporting Information dataset
S13). There was no correlation of the fold change of immune
genes between the hGAMs experiment and the mGAMs
experiment (Spearman’s correlation coefficient 5 0.00797),
indicating that these genes are regulated differently in human
versus mouse. Out of these 112 genes only 10 genes were sig-
nificantly >2-fold upregulated in hGAMs in comparison to
human control microglia. Eighteen additional genes showed a
non-significant >2-fold higher expression in hGAMs com-
pared with the human control microglia. In turn, six of the
112 genes were non-significantly lower expressed >2-fold in
hGAMs compared with human control microglia.
We used Webgestalt and iRegulon to search for pre-
dicted transcription factor binding sites in the 112 genes asso-
ciated with the GO term “immune response” and found
enrichment for the factors IRF1, IRF2, IRF7, and NFjB in
both analyses. From these factors only Irf7 was significantly
upregulated 18-fold in mGAMs in comparison to murine
control microglia, whereas Irf1, Irf2, Rela, Relb, Nfkb1, and
Nfkb2 were unregulated. In contrast, IRF7 was nonsignifi-
cantly 1.7-fold lower expressed in hGAMs in comparison to
human control microglia. Similar to mGAMs IRF1, IRF2,
RELA, RELB, NFKB1, and NFKB2 were also not regulated in
hGAMs.
We next compared the expression of different immune
markers between human and mouse GAMs (Table 4, Sup-
porting Information Table S4). Several genes were similarly
regulated in both species; however hGAMs displayed a lower

August 2016 1425

FIGURE 3: Comparison of human and mouse glioblastoma-associated microglia/monocytes A: Overlap between upregulated and down-
regulated genes in human and murine GAMs. B: 111 upregulated genes are shared between human and murine GAMs. Genes high-
lighted in red are associated with “mitotic cell cycle” or related GO terms. C: The 111 shared genes are mostly related to “mitotic cell
cycle” or related terms. The font size of specific terms corresponds to the abundance and significance of the overrepresented terms. D:
Significantly overrepresented GO terms in genes specific for hGAMs and mGAMs and genes that are shared between both species.

expression of several proinflammatory marker genes. Although
still highly expressed, IL1B expression was not significantly
upregulated in hGAMs. In contrast, Il1b expression was 2.46
fold upregulated in mGAMs. In addition, although almost
three-fold upregulated in mGAMs, TNF expression was not
significantly regulated in hGAMs. In addition, although
upregulated in mGAMs, IFNG and NOS2 (coding for iNOS)
expression were both not regulated in hGAMs and only
expressed at very low levels. However, also several marker
genes associated with anti-inflammatory immune activation
showed a different regulation in hGAMs when compared
with mGAMs. CD163 expression was strongly upregulated in
hGAMs, whereas the expression of TGFB3, CCL22, CCL17,
and CCL1 expression was downregulated. Furthermore, in
hGAMs ARG2 expression was not regulated (Arg2 expression
was highly upregulated in mGAMs), and although signifi-
cantly upregulated, both ARG1 and MMP13 were expressed
only at very low levels. This highlights the fundamental dif-
ference between GAMs isolated from human GBM samples
and from experimental GL261 mouse gliomas, as the latter
cell type possess features that are related to “immune
activation” whereas hGAMs lack those features.
Comparison of Human and Murine GAMs to
In Vitro-Stimulated Human Macrophages
To further investigate the immune activation states of these
cells we next compared the expression pattern of hGAMs and
mGAMs to in vitro-activated human macrophages and
murine microglia. We have previously shown that mouse
GAMs share only a part of their up- and downregulated

1426 Volume 64, No. 8

 Szulzewsky et al.: Human GAMs RNA Sequencing

FIGURE 4: Significantly enriched Reactome pathways in upregulated genes in hGAMs and mGAMs and in genes that are shared between
both species Significantly enriched pathways for genes A: upregulated in hGAMs versus human control microglia B: upregulated in
mGAMs versus murine control microglia C: genes that were upregulated in hGAMs but not upregulated in mGAMs D: genes upregu-
lated in mGAMs but not upregulated in hGAMs E: genes upregulated in both hGAMs and mGAMs. Listed pathways were selected to
avoid redundant listing.

genes with in vitro-activated mouse macrophages (Szulzewsky
et al., 2015). In addition, we have previously compared the
gene expression patterns of human macrophages after stimula-
tion with several factors in vitro (Xue et al., 2014). We could
show that in addition to the conventional pro- and anti-
inflammatory activation with LPS 1 IFNc (in the old classifi-
cation assigned as M1) and IL4 (in the old classification
assigned as M2), respectively, macrophages can also acquire
activation states diverging from this bipolar axis when stimu-
lated with free fatty acids, high density lipoprotein (HDL), or
combinations of stimuli associated with chronic inflamma-
tion. Using 28 different substances we could identify nine
major activation classes. We selected one stimulation from
each of the nine major classes and calculated the up- and
downregulated genes in comparison to un-stimulated control
macrophages for these data sets (fold change  2, Padj  0.05).
Subsequently we used gene set enrichment analysis to calcu-
late the significant overlap between each of the data sets,
hGAMs, and mGAMs. We selected data sets for treatment
with GC (dexamethasone), IL4, LPS 1 IFNc, oleic acid,
PGE2 (prostaglandin E2), TNFa, TNFa 1 PGE2, TPP
(TNFa 1 PGE2 1 TLR2-ligand), and ultra-pure LPS 1
immune complexes (upLPS 1 IC).
We first compared the upregulated genes in hGAMs to
the upregulated genes in the different in vitro-stimulated mac-
rophage sets. We found that the 292 upregulated genes in

August 2016 1427

FIGURE 5: qRT-PCR validation of transcripts associated with extracellular matrix organization. We measured the expression of eight
genes that were upregulated in hGAMs and associated with extracellular matrix organization by qRT-PCR in CD11b1 control microglia
(n53), CD11b1 hGAMs (n58) and the CD11b2 depleted fraction of corresponding GBM samples (n54). Connected dots represent
paired samples from the same tumor. Significances were calculated using Kruskal–Wallis test followed by Dunn’s test for multiple
comparisons.

1428 Volume 64, No. 8

August 2016
TABLE 4: Comparison of the Expression of Several Immune Genes in Human and Mouse Glioblastoma-associated Microglia/Monocytes in Comparison
to the Respective Control Microglia

Proinflammatory markers Anti-inflammatory markers

Gene symbol Expression Regulation Regulation Gene Expression Regulation Regulation

hGAMs hGAMs mGAMs symbol hGAMs hGAMs mGAMs
Receptors and signal transduction
TLR2 12.00 21.06 1.61 CD163 14.77 6.20 25.16
CD86 10.89 1.09 21.12 CD204 12.56 3.40 13.55
TLR4 10.86 21.07 21.33 STAT3 11.68 21.04 1.20
STAT1 10.80 2.02 4.19 CD206 11.59 1.72 1.54
IL1R1 8.61 1.69 1.31 IL1R2 10.43 1.39 14.00
CD80 5.68 2.26 2.96 TLR1 10.05 21.36 1.87
TLR8 8.59 21.23 9.24
Secreted factors and cytokines
IL1B 13.11 1.34 2.46 IL1RN 11.41 4.37 13.16
TNF 9.22 21.31 21.98 IL10 9.81 2.84 3.50
IL18 9.05 21.53 1.14 IL6 8.07 5.69 1.43
IL15 5.83 1.45 21.38 ARG2 7.90 1.05 18.70
IL23a 4.28 1.36 1.41 TGFB3 4.80 21.23 9.83
IFNG 0.91 1.13 4.65 ARG1 3.1 3.27 56.99
NOS2 0.78 1.14 14.21
Chemokines
CCL3 12.77 2.57 21.22 CCL22 4.67 22.09 1.61
CCL4 12.08 3.36 21.91 CCL17 2.62 23.58 1.89
CCL2 10.14 3.83 1 CCL1 0.73 21.75 3.05
CCL5 7.21 4.70 9.41
CCL8 6.67 1.98 19.41
CXCL9 4.75 4.39 51.10

1429
Szulzewsky et al.: Human GAMs RNA Sequencing
hGAMs were most significantly enriched for genes upregu-
lated in macrophages stimulated with upLPS 1 IC (P 5 4.6
3 1029), oleic acid (P 5 4.4 3 1026), TNFa (P 5 1.2 3
1024), and dexamethasone (P 5 1.5 3 1024). A less signifi-
cant enrichment was found for the upregulated genes of mac-
rophages stimulated with TPP (P 5 1.8 3 1023), PGE2
(P 5 4.9 3 1023), LPS 1 IFNc (P 5 0.01), and TNFa 1
PGE2 (P 5 0.019) (Fig. 6A, Supporting Information Fig.
S6). Interestingly, the upregulated hGAMs genes were not sig-
nificantly enriched in the upregulated genes of macrophages
stimulated with IL4. Rather, the upregulated genes in hGAMs
highly significantly overlapped with the downregulated genes
in macrophages stimulated with IL4 (P 5 3 3 1025), indicat-
ing an inverse regulation between hGAMs and macrophages
treated with IL4. This is in contrast to previous studies
reporting that GAMs are activated towards an alternative acti-
vation phenotype that resembles the state of macrophages
treated with IL4 (previously termed M2 activation). In addi-
tion to IL4, the upregulated genes in hGAMs were signifi-
cantly overlapping with the downregulated genes of
macrophages stimulated with TPP (P 5 4.4 3 1024),
TNFa 1 PGE2 (P 5 3.2 3 1023), TNFa (P 5 0.025), and
LPS 1 IFNc (P 5 0.038). In addition, the downregulated
genes of macrophages treated with oleic acid (P 5 6.3 3
1023) constituted the only data set that was significantly
enriched when comparing the downregulated genes
of hGAMs with these data sets (Supporting Information
Fig. S7A).
In contrast, when comparing the upregulated genes in
mGAMs to the in vitro-activated human macrophages we
found a significant overlap to all nine conditions, with the
most significant overlaps to the upregulated genes of macro-
phages stimulated with TPP, LPS 1 INFc and TNFa 1 PGE2
(all P < 2.2 3 10216, Fig. 6B). In addition we found a highly
significant overlap to the upregulated genes of macrophages
stimulated with oleic acid (P 5 4.6 3 10212), TNFa
(P 5 6.1 3 10212), and upLPS 1 IC (P 5 1.8 3 1027). This
highlights a higher association of mGAMs with proinflamma-
tory immune activation which is lacking in hGAMs.
Although less strong, we also found a significant overlap to
macrophages stimulated with IL4 (P 5 0.015). We also found
a highly significant overlap when comparing the upregulated
genes in mGAMs to the downregulated genes in macrophages
stimulated with TPP (P 5 1.4 3 1028), upLPS 1 IC
(P 5 5.7 3 1025), TNFa 1 PGE2 (P 5 2.3 3 1024), and
LPS 1 IFNc (P 5 4.9 3 1024). When comparing the down-
regulated genes in mGAMs to the in vitro data sets we found
a significant enrichment for the upregulated genes of macro-
phages treated with dexamethasone (P 5 0.038; Supporting
Information Fig. S7B).
1430
Comparison of Human and Murine GAMs to Murine
Microglia Stimulated with LPS, IL4, and TGFb
Since—at least in murine experimental gliomas—in addition
to invading monocytes and monocyte-derived macrophages
also brain-resident microglia constitute a major cellular frac-
tion of GAMs we used data sets from murine adult microglia
that were stimulated with LPS, IL4, and TGFb. We calcu-
lated the significantly regulated genes (fold change  2,
P  0.05) in comparison to un-stimulated control microglia
and performed gene set enrichment analysis to calculate the
significant overlap between these data sets, hGAMs, and
mGAMs.
When comparing the upregulated genes in hGAMs we
only found a significant overlap to the upregulated genes for
microglia stimulated with TGFb (P 5 7.4 3 1024; Fig.
6C,E). This emphasizes the lack of immune activation in
hGAMs. We found no significant overlap between the down-
regulated genes in hGAMs and any of the in vitro data sets
(Supporting Information Fig. S7C).
The mGAMs were highly overlapping with the
upregulated genes in microglia stimulated with LPS (P < 2.2
3 10216) and the up- and downregulated genes in microglia
stimulated with IL4 (P 5 8.2 3 1026 and 8.8 3 1024,
respectively; Fig. 6D,F). In addition, we found a highly sig-
nificant overlap between the downregulated genes in mGAMs
and the downregulated genes in microglia stimulated with
LPS (P 5 6.5 3 10212) and IL4 (P 5 1.89 3 1026; Support-
ing Information Fig. S7D).
Interestingly, we found a completely inverse overlap
between mGAMs and microglia stimulated with TGFb. There
was no significant overlap between the upregulated genes or
the downregulated genes of both cell types, however, we
found a highly significant overlap between the upregulated
genes in mGAMs and the downregulated genes in TGFb-
stimulated microglia (P 5 2.6 3 10213) as well as between
the downregulated genes in mGAMs and the upregulated
genes in TGFb-treated microglia (1.9 3 1026).
The genes that were related to the GO term “mitotic
cell cycle” and were upregulated in the human samples were
not found to be regulated in the cultured macrophage or
microglial cells stimulated under different conditions (Sup-
porting Information Fig. S7E).
Human GAMs and Immune Cells Isolated from Lung
Tumors Share Gene Expression Patterns
To investigate similarities in gene expression between hGAMs
and tumor-associated monocytes/macrophages (TAMs) in
other solid cancers we used a published gene list of significantly
regulated genes from CD11b1CD331CXCR22 immature
monocytic myeloid cells and CD11b1CD332CXCR21 neu-
trophils that were isolated from human non-small cell lung
Volume 64, No. 8
FIGURE 6: Comparison of human and murine glioblastoma-associated microglia/monocytes to in vitro-stimulated macrophages and
microglia A: gene set enrichment analysis was used to compare the significantly regulated genes in hGAMs to data sets of in vitro stimu-
lated human macrophages. Macrophages were stimulated with either dexamethasone (GC), IL4, LPS 1 IFNc, Oleic Acid (OA), PGE2,
TNFa, TNFa 1 PGE2, TPP, ultra-pure LPS 1 immune complexes. B: Upregulated genes in mGAMs were compared with the same data
sets. C: Gene set enrichment analysis was used to compare the upregulated genes in hGAMs to murine microglia stimulated with LPS,
IL4, or TGFb. D: Comparison of upregulated genes of mGAMs with murine microglia stimulated with either TGFb, LPS, IL4. E: 66 out of
the 292 upregulated genes in hGAMs (22.6%) are also regulated in any of the significantly overlapping in vitro-activation data sets. F:
343 out of the 940 upregulated genes in mGAMs (36.5%) are also upregulated in any of the significantly overlapping in vitro-activation
data sets.
cancer (NSCLC) samples by flow cytometry (Durrans et al.
2015). Similar to hGAMs these cells express genes such as
SPP1, CCL7, GPNMB, FN1, MMP14, SPARC, and several col-
lagens, indicating that there are shared features that are similar
in TAMs from different types of cancer, and may reflect tumor-
specific responses of these cells.
We performed a gene set enrichment analysis and found
that the 292 upregulated genes in hGAMs were significantly
enriched in the gene sets of upregulated genes of both imma-
ture monocytic myeloid cells (P 5 4.95 3 10211) and neutro-
phils (P 5 0.00062) (Supporting Information Fig. S8). This
indicates that the regulation of certain gene expression pattern
in innate immune cells is shared between different solid
cancers.
Discussion
Glioblastoma-associated microglia/monocytes have been
shown to support tumor growth via the release of factors that
stimulate angiogenesis, invasion, or suppression of immunity
(Carvalho da Fonseca et al., 2014; Coniglio et al., 2012;
Feng et al., 2015; Hu et al., 2014; Li and Graeber, 2012; Lisi
et al., 2014; Markovic et al., 2009; Xu et al., 2014). Disrupt-
ing the interaction between glioma cells and GAMs has been
shown to be beneficial for survival in animal models (Kees
et al., 2012; Markovic et al., 2011; Pyonteck et al., 2013;
Sarkar et al., 2014; Xu et al., 2014). However, it is not clear
how well these concepts can be applied to humans. As an
example, in a recent clinical trial a therapeutic approach
directed against GAMs could not recapitulate the positive
results obtained in murine models (Butowski et al., 2015;
Pyonteck et al., 2013). The analysis of several factors released
by human and murine GAMs has made it clear that these
cells possess different features in both species (Gabrusiewicz
et al., 2011; Hattermann et al., 2014; Hussain et al.,
2006a,b; Sica et al., 2006; Umemura et al., 2008; Ye et al.,
2012).
There are two major challenges of studying hGAMs.
Because of the lack of naive control samples, control tissues
either originate from postmortem tissues or from diseased,
non-tumor patients (most prominently epilepsy patients). The
second challenge is the current lack of reliable surface markers
to distinguish between brain-resident microglia and infiltrat-
ing myeloid cells in humans. Surface marker combinations
exist to distinguish these cells in mice: CD11b1CD45low for
brain-resident microglia and CD11b1CD45high for invading
myeloid cells (Dunay et al., 2008; Muller et al., 2015). These
two distinct populations are not visible in the FACS analysis
of immune cells in human brains. There are several recent
attempts to find specific markers for microglia versus mono-
cytes/macrophages with mixed results (Butovsky et al., 2014;
Hickman et al., 2013; Pong et al., 2013). As a consequence,
1432
we used the sorting technique based on CD11b1 MACS to
be able to compare our murine and human data. It is worth-
while to notice here that with this approach we investigate all
myeloid cells including microglia, monocytes, macrophages,
dendritic cells, and neutrophils.
Our data show that a large percentage of the upregu-
lated genes in hGAMs are associated with functions related to
“mitotic cell cycle”. As mentioned before, GAMs are com-
prised of brain-intrinsic microglia, but also of immune cells
that invade from the periphery—among those monocytic
cells, tissue macrophages, and granulocytic cells. It is likely
that invading monocytic cells have a higher proliferative
capacity as intrinsic microglia. It is however conceivable that
also microglia are proliferating within the GBM tissue, as
recent findings in mice have shown that—yet unknown—
brain-resident microglia precursor cells can proliferate and
reconstitute the microglia population within the brain with-
out influx of peripheral cells (Bruttger et al., 2015; Elmore
et al., 2014). By analyzing enriched transcription factor bind-
ing sites of the upregulated cell cycle-associated genes we
found that E2F family members might be involved in the
regulation of these genes. A recent study could show that
E2f3 expression in murine lung cancer-associated macro-
phages is positively correlated with metastasis (Trikha et al.,
2015). E2f3 expression did not influence the survival or pro-
liferation of lung cancer-associated macrophages, but it posi-
tively influenced cytoskeleton rearrangements, cell migration,
and cell adhesion, all of which are features that were also
enriched in the upregulated genes in hGAMs. The expression
of E2F3 was not changed in hGAMs; however, the expression
of E2F7, E2F8, and E2F1 was significantly upregulated.
Human GAMs also upregulate and are the main
expressers of several genes that are associated with
“extracellular matrix organization”. Extracellular matrix
remodeling is thought to play an important role in tumor
promotion, in part by promoting stem-ness of glioma stem
cells, mediating enhanced tumor cell migration and invasion,
e.g. through degradation of the extracellular matrix, but also
by stimulating tumor cell growth and angiogenesis, as extra-
cellular matrix-bound growth factors are released during
matrix degradation. Our group has previously demonstrated
the effect of MMP14 and MMP9 on tumor promotion in a
glioma mouse model (Hu et al., 2014; Markovic et al., 2009,
2011). MMP14 is significantly upregulated in hGAMs and
mGAMs, whereas MMP9 is significantly downregulated in
both hGAMs and mGAMs. In addition, several other extra-
cellular matrix proteins are under investigation for their
tumor promoting role in different cancers, among them sev-
eral that were upregulated in hGAMs (Kruger et al., 2014;
Pasini et al., 2014; Pietras et al., 2014; Sangaletti et al., 2008;
Volume 64, No. 8

Serres et al., 2014). This further highlights the important role
of hGAMs in glioblastoma.
We could show that there is a significant overlap between
the upregulated genes of murine and human CD11b1 GAMs.
This overlap includes genes related to “extracellular matrix
organization” and “cell adhesion”, but was primarily restricted
to genes related to “mitotic cell cycle regulation”. While several
genes related to “immune activation” were upregulated in
mGAMs this was not the case for human GAMs. In line with
this, IFNG and NOS2 were unchanged and expressed at only
very low levels in hGAMs and TNF was expressed at medium
levels, but not upregulated. The lack of genes related to
“immune response/activation” in whole tumor tissues of high
grade versus low grade glioma has previously been reported
(Mieczkowski et al., 2015). We have previously compared the
expression profiles of CD11b1 mGAMs to in vitro-activated
murine macrophages and could show that mGAMs express
markers associated with both pro- and anti-inflammatory acti-
vation (Szulzewsky et al., 2015). Here we compared the differ-
entially expressed genes of CD11b1 hGAMs and mGAMs to
those of in vitro-stimulated human macrophages and murine
microglia. We have recently demonstrated that macrophages
can acquire activation states that extend beyond the bipolar axis
of pro- and anti-inflammatory activation when stimulated with
free fatty acids, high density lipoprotein (HDL), or combina-
tions of stimuli associated with chronic inflammation (Xue
et al., 2014). In general we found a lower overlap between the
in vitro data sets to hGAMs than mGAMs. The comparison to
these in vitro data sets showed that hGAMs share a smaller over-
lap to cells treated in vitro with pro-inflammatory stimuli, such
as LPS 1 IFNc and TNFa, than mGAMs. This underlines the
results from the GO analysis which did not detect a significant
enrichment of terms related to “immune activation” in hGAMs,
whereas in mGAMs a high abundance of these terms was
detected. It remains to be shown if there is a homogeneously
activated GAMs population throughout the tumor or—which
seems more likely—if GAMs activation depends on the location
within or the stage of the tumor.
The lack of the GO terms related to “immune
response” in hGAMs versus control GAMs needs to be fur-
ther investigated. In contrast to murine samples, which allow
the use of truly na€ıve and age-matched samples, this is not
possible for the human setting. The brain tissue we used for
the isolation of our human control samples originated either
from postmortem tissues of aged patients (average donor age
of 80.6 years, however only one out of five donors showed
signs of cognitive dysfunction) or from epilepsy and hippo-
campal sclerosis tissue (average donor age of 37 years). In
contrast, the average donor age for the glioblastoma samples
was 59 years. Thus, a possible explanation for the lack of
immune activation in hGAMs may be an already present acti-
August 2016
Szulzewsky et al.: Human GAMs RNA Sequencing
vation in non-tumor control microglia which was isolated
from epilepsy and hippocampal tissues, as well as from post-
mortem cortex tissues. The genes that were related to
“immune response” were not expressed at high levels in con-
trol cells; however a direct comparison of the absolute gene
expression between control microglia from human and
mouse tissues is difficult due to the differences of microar-
ray and RNA sequencing. A recent study has found that,
based on the analysis of several immune related genes using
Nanostring technology hGAMs share features of unactivated
M0 macrophages (Gabrusiewicz et al., 2016), which is in
line with the absence of immune activation that we
detected. Moreover, the immune response of mGAMs in the
GL261 model could be explained by the nature of the
model itself. GL261 cells originate from C3H mice and are
not a syngeneic transplantation model and the immune
response of mGAMs isolated from these tumors may repre-
sent a mild version of graft versus host response (Akbasak
et al., 1991; Ausman et al., 1970; Seligman and Shear,
1939). In addition, the relatively short tumor latency of
around 3 weeks might be too short for complete reprogram-
ming GAMs. It remains to be shown if different glioma
models, such as genetically engineered mouse models or
xenograft models mimic the phenotype of hGAMs more
closely. This might be especially interesting in the light of
an ongoing discussion about the reliability of animal models
to study human immune activation (Mestas and Hughes,
2004; Reynolds and Haniffa, 2015; Shay et al., 2013).
Despite these differences, the GL261 model is suitable
to investigate distinct aspects that are shared between hGAMs
and mGAMs. Our findings provide new insights into the
biology of human glioblastoma-associated microglia/mono-
cytes and give detailed information about the validity of
murine experimental models. These data will also serve as a
guide to select potential therapeutic targets which are relevant
in the human context.
Acknowledgment
Grant sponsor: Deutsche Forschungsgemeinschaft; Grant
numbers: KE 329/30-1, SY 144/4-1, SZ 350/1-1; Grant
sponsor: NeuroCure. Grant sponsor: NIH Grant; Grant
number: U01CA160882-01A1; Grant sponsor: Deutsche For-
schungsgemeinschaft; Grant number: SFB645/SFB704
The authors thank Hans van Mierlo and Manja Litjens
from the Department of Translational Neuroscience at the
UMC Utrecht.
References
Akbasak A, Oldfield EH, Saris SC. 1991. Expression and modulation of major
histocompatibility antigens on murine primary brain tumor in vitro.
J Neurosurg 75:922–929.
1433
Ausman JI, Shapiro WR, Rall DP. 1970. Studies on the chemotherapy of
experimental brain tumors: Development of an experimental model. Cancer
Res 30:2394–2400.
Brennan C, Momota H, Hambardzumyan D, Ozawa T, Tandon A, Pedraza A,
Holland E. 2009. Glioblastoma subclasses can be defined by activity among
signal transduction pathways and associated genomic alterations. PloS One
4:e7752.
Bruttger J, Karram K, Wortge S, Regen T, Marini F, Hoppmann N, Klein M,
Blank T, Yona S, Wolf Y, Mack M, Pinteaux E, Muller W, Zipp F, Binder H,
Bopp T, Prinz M, Jung S, Waisman A. 2015. Genetic cell ablation reveals
clusters of local self-renewing microglia in the mammalian central nervous
system. Immunity 43:92–106.
Butovsky O, Jedrychowski MP, Moore CS, Cialic R, Lanser AJ, Gabriely G,
Koeglsperger T, Dake B, Wu PM, Doykan CE, Fanek Z, Liu L, Chen Z,
Rothstein JD, Ransohoff RM, Gygi SP, Antel JP, Weiner HL. 2014. Identifica-
tion of a unique TGF-beta-dependent molecular and functional signature in
microglia. Nat Neurosci 17:131–143.
Butowski N, Colman H, De Groot JF, Omuro AM, Nayak L, Wen PY,
Cloughesy TF, Marimuthu A, Haidar S, Perry A, Huse J, Phillips J, West BL,
Nolop KB, Hsu HH, Ligon KL, Molinaro AM, Prados M. 2015. Orally adminis-
tered colony stimulating factor 1 receptor inhibitor PLX3397 in recurrent glio-
blastoma: An ivy foundation early phase clinical trials consortium phase II
study. Neuro Oncol 17:131–143.
Carvalho da Fonseca AC, Wang H, Fan H, Chen X, Zhang I, Zhang L, Lima
FR, Badie B. 2014. Increased expression of stress inducible protein 1 in
glioma-associated microglia/macrophages. J Neuroimmunol 274:71–77.
Charles NA, Holland EC, Gilbertson R, Glass R, Kettenmann H. 2011. The
brain tumor microenvironment. Glia 59:1169–1180.
Coniglio SJ, Eugenin E, Dobrenis K, Stanley ER, West BL, Symons MH, Segall
JE. 2012. Microglial stimulation of glioblastoma invasion involves epidermal
growth factor receptor (EGFR) and colony stimulating factor 1 receptor (CSF-
1R) signaling. Mol Med 18:519–527.
Croft D, Mundo AF, Haw R, Milacic M, Weiser J, Wu G, Caudy M, Garapati
P, Gillespie M, Kamdar MR, Jassal B, Jupe S, Matthews L, May B, Palatnik S,
Rothfels K, Shamovsky V, Song H, Williams M, Birney E, Hermjakob H, Stein
L, D’Eustachio P. 2014. The reactome pathway knowledgebase. Nucleic Acids
Res 42:D472–D477.
Dolecek TA, Propp JM, Stroup NE, Kruchko C. 2012. CBTRUS statistical
report: Primary brain and central nervous system tumors diagnosed in the
United States in 2005–2009. Neuro Oncol 14:v1–49.
Dunay IR, Damatta RA, Fux B, Presti R, Greco S, Colonna M, Sibley LD. 2008.
Gr1(1) inflammatory monocytes are required for mucosal resistance to the
pathogen Toxoplasma gondii. Immunity 29:306–317.
Durrans A, Gao D, Gupta R, Fischer KR, Choi H, El Rayes T, Ryu S, Nasar A,
Spinelli CF, Andrews W, Elemento O, Nolan D, Stiles B, Rafii S, Narula N,
Davuluri R, Altorki NK, Mittal V. 2015. Identification of reprogrammed mye-
loid cell transcriptomes in NSCLC. PLoS One 10:e0129123.
Edin S, Wikberg ML, Dahlin AM, Rutegard J, Oberg A, Oldenborg PA,
Palmqvist R. 2012. The distribution of macrophages with a M1 or M2 pheno-
type in relation to prognosis and the molecular characteristics of colorectal
cancer. PLoS One 7:e47045.
Elmore MR, Najafi AR, Koike MA, Dagher NN, Spangenberg EE, Rice RA,
Kitazawa M, Matusow B, Nguyen H, West BL, Green KN. 2014. Colony-stimu-
lating factor 1 receptor signaling is necessary for microglia viability, unmask-
ing a microglia progenitor cell in the adult brain. Neuron 82:380–397.
Feng X, Szulzewsky F, Yerevanian A, Chen Z, Heinzmann D, Rasmussen RD,
Alvarez-Garcia V, Kim Y, Wang B, Tamagno I, Zhou H, Li X, Kettenmann H,
Ransohoff RM, Hambardzumyan D. 2015. Loss of CX3CR1 increases accumu-
lation of inflammatory monocytes and promotes gliomagenesis. Oncotarget
6:15077–15094.
Gabrusiewicz K, Ellert-Miklaszewska A, Lipko M, Sielska M, Frankowska M,
Kaminska B. 2011. Characteristics of the alternative phenotype of microglia/
macrophages and its modulation in experimental gliomas. PloS One 6:
e23902.
1434
Gabrusiewicz K, Rodriguez B, Wei J, Hashimoto Y, Healy LM, Maiti SN,
Thomas G, Zhou S, Wang Q, Elakkad A, Liebelt BD, Yaghi NK, Ezhilarasan R,
Huang N, Weinberg JS, Prabhu SS, Rao G, Sawaya R, Langford LA, Bruner
JM, Fuller GN, Bar-Or A, Li W, Colen RR, Curran MA, Bhat KP, Antel JP,
Cooper LJ, Sulman EP, Heimberger AB. 2016. Glioblastoma-infiltrated innate
immune cells resemble M0 macrophage phenotype. JCI Insight 1:2016;1. pii:
e85841.
Gentleman RC, Carey VJ, Bates DM, Bolstad B, Dettling M, Dudoit S, Ellis
B, Gautier L, Ge Y, Gentry J, Hornik K, Hothorn T, Huber W, Iacus S,
Irizarry R, Leisch F, Li C, Maechler M, Rossini AJ, Sawitzki G, Smith C,
Smyth G, Tierney L, Yang JY, Zhang J. 2004. Bioconductor: Open software
development for computational biology and bioinformatics. Genome Biol
5:R80.
Gill BJ, Pisapia DJ, Malone HR, Goldstein H, Lei L, Sonabend A, Yun J,
Samanamud J, Sims JS, Banu M, Dovas A, Teich AF, Sheth SA, McKhann
GM, Sisti MB, Bruce JN, Sims PA, Canoll P. 2014. MRI-localized biopsies
reveal subtype-specific differences in molecular and cellular composition at
the margins of glioblastoma. Proc Natl Acad Sci USA 111:12550–12555.
Ginhoux F, Schultze JL, Murray PJ, Ochando J, Biswas SK. 2015. New
insights into the multidimensional concept of macrophage ontogeny, activa-
tion and function. Nat Immunol 17:34–40.
Hattermann K, Sebens S, Helm O, Schmitt AD, Mentlein R, Mehdorn HM,
Held-Feindt J. 2014. Chemokine expression profile of freshly isolated human
glioblastoma-associated macrophages/microglia. Oncol Rep 32:270–276.
Hickman SE, Kingery ND, Ohsumi TK, Borowsky ML, Wang LC, Means TK, El
Khoury J. 2013. The microglial sensome revealed by direct RNA sequencing.
Nat Neurosci 16:1896–1905.
Hu F, a Dzaye OD, Hahn A, Yu Y, Scavetta RJ, Dittmar G, Kaczmarek AK,
Dunning KR, Ricciardelli C, Rinnenthal JL, Heppner FL, Lehnardt S, Synowitz
M, Wolf SA, Kettenmann H. 2015. Glioma-derived versican promotes tumor
expansion via glioma-associated microglial/macrophages Toll-like receptor 2
signaling. Neuro Oncol 17:200–210.
Hu F, Ku MC, Markovic D, a Dzaye OD, Lehnardt S, Synowitz M, Wolf SA,
Kettenmann H. 2014. Glioma-associated microglial MMP9 expression is
upregulated by TLR2 signaling and sensitive to minocycline. Int J Cancer
135:2569–2578.
Hussain SF, Yang D, Suki D, Aldape K, Grimm E, Heimberger AB. 2006a. The
role of human glioma-infiltrating microglia/macrophages in mediating antitu-
mor immune responses. Neuro Oncol 8:261–279.
Hussain SF, Yang D, Suki D, Grimm E, Heimberger AB. 2006b. Innate immune
functions of microglia isolated from human glioma patients. J Transl Med 4:15.
Janky R, Verfaillie A, Imrichova H, Van de Sande B, Standaert L, Christiaens
V, Hulselmans G, Herten K, Naval Sanchez M, Potier D, Svetlichnyy D,
Kalender Atak Z, Fiers M, Marine JC, Aerts S. 2014. iRegulon: From a gene
list to a gene regulatory network using large motif and track collections.
PLoS Comput Biol 10:e1003731.
Kanehisa M, Goto S. 2000. KEGG: Kyoto encyclopedia of genes and
genomes. Nucleic Acids Res 28:27–30.
Kanehisa M, Sato Y, Kawashima M, Furumichi M, Tanabe M. 2016. KEGG as
a reference resource for gene and protein annotation. Nucleic Acids Res 44:
D457–D462.
Kees T, Lohr J, Noack J, Mora R, Gdynia G, Todt G, Ernst A, Radlwimmer B,
Falk CS, Herold-Mende C, Regnier-Vigouroux A. 2012. Microglia isolated
from patients with glioma gain antitumor activities on poly (I:C) stimulation.
Neuro-Oncology 14:64–78.
Kerber M, Reiss Y, Wickersheim A, Jugold M, Kiessling F, Heil M, Tchaikovski
V, Waltenberger J, Shibuya M, Plate KH, Machein MR. 2008. Flt-1 signaling
in macrophages promotes glioma growth in vivo. Cancer Res 68:7342–7351.
Kim D, Pertea G, Trapnell C, Pimentel H, Kelley R, Salzberg SL. 2013.
TopHat2: Accurate alignment of transcriptomes in the presence of insertions,
deletions and gene fusions. Genome Biol 14:R36.
Kim H, Zheng S, Amini SS, Virk SM, Mikkelsen T, Brat DJ, Grimsby J,
Sougnez C, Muller F, Hu J, Sloan AE, Cohen ML, Van Meir EG, Scarpace L,
Laird PW, Weinstein JN, Lander ES, Gabriel S, Getz G, Meyerson M, Chin L,
Volume 64, No. 8

Barnholtz-Sloan JS, Verhaak RG. 2015. Whole-genome and multisector
exome sequencing of primary and post-treatment glioblastoma reveals pat-
terns of tumor evolution. Genome Res 25:316–327.
Kruger TE, Miller AH, Godwin AK, Wang J. 2014. Bone sialoprotein and
osteopontin in bone metastasis of osteotropic cancers. Crit Rev Oncol Hema-
tol 89:330–341.
Lawrence M, Huber W, Pages H, Aboyoun P, Carlson M, Gentleman R,
Morgan MT, Carey VJ. 2013. Software for computing and annotating
genomic ranges. PLoS Comput Biol 9:e1003118.
Li W, Graeber MB. 2012. The molecular profile of microglia under the influ-
ence of glioma. Neuro-Oncology 14:958–978.
Lisi L, Laudati E, Navarra P, Dello Russo C. 2014. The mTOR kinase inhibitors
polarize glioma-activated microglia to express a M1 phenotype.
J Neuroinflamm 11:125.
Love MI, Huber W, Anders S. 2014. Moderated estimation of fold change
and dispersion for RNA-seq data with DESeq2. Genome Biol 15:550.
Markovic DS, Vinnakota K, Chirasani S, Synowitz M, Raguet H, Stock K, Sliwa
M, Lehmann S, Kalin R, van Rooijen N, Holmbeck K, Heppner FL, Kiwit J,
Matyash V, Lehnardt S, Kaminska B, Glass R, Kettenmann H. 2009. Gliomas
induce and exploit microglial MT1-MMP expression for tumor expansion.
Proc Natl Acad Sci USA 106:12530–12535.
Markovic DS, Vinnakota K, van Rooijen N, Kiwit J, Synowitz M, Glass R,
Kettenmann H. 2011. Minocycline reduces glioma expansion and invasion by
attenuating microglial MT1-MMP expression. Brain Behav Immun 25:624–
628.
Mestas J, Hughes CC. 2004. Of mice and not men: Differences between
mouse and human immunology. J Immunol 172:2731–2738.
Meyer M, Reimand J, Lan X, Head R, Zhu X, Kushida M, Bayani J, Pressey
JC, Lionel AC, Clarke ID, Cusimano M, Squire JA, Scherer SW, Bernstein M,
Woodin MA, Bader GD, Dirks PB. 2015. Single cell-derived clonal analysis of
human glioblastoma links functional and genomic heterogeneity. Proc Natl
Acad Sci USA 112:851–856.
Mieczkowski J, Kocyk M, Nauman P, Gabrusiewicz K, Sielska M, Przanowski
P, Maleszewska M, Rajan WD, Pszczolkowska D, Tykocki T, Grajkowska W,
Kotulska K, Roszkowski M, Kostkiewicz B, Kaminska B. 2015. Down-regulation
of IKKbeta expression in glioma-infiltrating microglia/macrophages is associ-
ated with defective inflammatory/immune gene responses in glioblastoma.
Oncotarget 6:33077–33090.
Milacic M, Haw R, Rothfels K, Wu G, Croft D, Hermjakob H, D’Eustachio P,
Stein L. 2012. Annotating cancer variants and anti-cancer therapeutics in
reactome. Cancers (Basel) 4:118021211.
Muller A, Brandenburg S, Turkowski K, Muller S, Vajkoczy P. 2015. Resident
microglia, and not peripheral macrophages, are the main source of brain
tumor mononuclear cells. Int J Cancer 137:278–288.
Murray PJ, Allen JE, Biswas SK, Fisher EA, Gilroy DW, Goerdt S, Gordon S,
Hamilton JA, Ivashkiv LB, Lawrence T, Locati M, Mantovani A, Martinez FO,
Mege JL, Mosser DM, Natoli G, Saeij JP, Schultze JL, Shirey KA, Sica A,
Suttles J, Udalova I, van Ginderachter JA, Vogel SN, Wynn TA. 2014. Macro-
phage activation and polarization: nomenclature and experimental guidelines.
Immunity 41:14–20.
Olah M, Raj D, Brouwer N, De Haas AH, Eggen BJ, Den Dunnen WF, Biber
KP, Boddeke HW. 2012. An optimized protocol for the acute isolation of
human microglia from autopsy brain samples. Glia 60:96–111.
Pasini FS, Zilberstein B, Snitcovsky I, Roela RA, Mangone FR, Ribeiro U, Jr.,
Nonogaki S, Brito GC, Callegari GD, Cecconello I, Alves VA, Eluf-Neto J,
Chammas R, Federico MH. 2014. A gene expression profile related to
immune dampening in the tumor microenvironment is associated with poor
prognosis in gastric adenocarcinoma. J Gastroenterol 49:1453–1466.
Patel AP, Tirosh I, Trombetta JJ, Shalek AK, Gillespie SM, Wakimoto H,
Cahill DP, Nahed BV, Curry WT, Martuza RL, Louis DN, Rozenblatt-Rosen O,
Suva ML, Regev A, Bernstein BE. 2014. Single-cell RNA-seq highlights intratu-
moral heterogeneity in primary glioblastoma. Science 344:1396–1401.
Phillips HS, Kharbanda S, Chen R, Forrest WF, Soriano RH, Wu TD, Misra A,
Nigro JM, Colman H, Soroceanu L, Williams PM, Modrusan Z, Feuerstein BG,
August 2016
Szulzewsky et al.: Human GAMs RNA Sequencing
Aldape K. 2006. Molecular subclasses of high-grade glioma predict progno-
sis, delineate a pattern of disease progression, and resemble stages in neuro-
genesis. Cancer Cell 9:157–173.
Pietras A, Katz AM, Ekstrom EJ, Wee B, Halliday JJ, Pitter KL, Werbeck JL,
Amankulor NM, Huse JT, Holland EC. 2014. Osteopontin-CD44 signaling in
the glioma perivascular niche enhances cancer stem cell phenotypes and pro-
motes aggressive tumor growth. Cell Stem Cell 14:357–369.
Pong WW, Walker J, Wylie T, Magrini V, Luo J, Emnett RJ, Choi J, Cooper
ML, Griffith M, Griffith OL, Rubin JB, Fuller GN, Piwnica-Worms D, Feng X,
Hambardzumyan D, DiPersio JF, Mardis ER, Gutmann DH. 2013. F11R is a
novel monocyte prognostic biomarker for malignant glioma. PLoS One 8:
e77571.
Pyonteck SM, Akkari L, Schuhmacher AJ, Bowman RL, Sevenich L, Quail DF,
Olson OC, Quick ML, Huse JT, Teijeiro V, Setty M, Leslie CS, Oei Y, Pedraza
A, Zhang J, Brennan CW, Sutton JC, Holland EC, Daniel D, Joyce JA. 2013.
CSF-1R inhibition alters macrophage polarization and blocks glioma progres-
sion. Nat Med 19:1264–1272.
R Development Core Team. 2011. R: A language and environment for statis-
tical computing. ISBN: 3-900051-07-0 http://www.R-project.org/.
Reinartz S, Schumann T, Finkernagel F, Wortmann A, Jansen JM, Meissner
W, Krause M, Schworer AM, Wagner U, Muller-Brusselbach S, Muller R. 2014.
Mixed-polarization phenotype of ascites-associated macrophages in human
ovarian carcinoma: Correlation of CD163 expression, cytokine levels and early
relapse. Int J Cancer 134:32–42.
Reynolds G, Haniffa M. 2015. Human and mouse mononuclear phagocyte
networks: A tale of two species? Front Immunol 6:330.
Sangaletti S, Di Carlo E, Gariboldi S, Miotti S, Cappetti B, Parenza M, Rumio
C, Brekken RA, Chiodoni C, Colombo MP. 2008. Macrophage-derived SPARC
bridges tumor cell-extracellular matrix interactions toward metastasis. Cancer
Res 68:9050–9059.
Sarkar S, Doring A, Zemp FJ, Silva C, Lun X, Wang X, Kelly J, Hader W,
Hamilton M, Mercier P, Dunn JF, Kinniburgh D, van Rooijen N, Robbins S,
Forsyth P, Cairncross G, Weiss S, Yong VW. 2014. Therapeutic activation of
macrophages and microglia to suppress brain tumor-initiating cells. Nat Neu-
rosci 17:46–55.
Seligman AM, Shear MJ. 1939. Studies in Carcinogenesis: Experimental pro-
duction of brain tumors in mice with methylcholanthrene. Am J Cancer 37:
364–395.
Serres E, Debarbieux F, Stanchi F, Maggiorella L, Grall D, Turchi L, Burel-
Vandenbos F, Figarella-Branger D, Virolle T, Rougon G, Van Obberghen-
Schilling E. 2014. Fibronectin expression in glioblastomas promotes cell
cohesion, collective invasion of basement membrane in vitro and orthotopic
tumor growth in mice. Oncogene 33:3451–3462.
Shannon P, Markiel A, Ozier O, Baliga NS, Wang JT, Ramage D, Amin N,
Schwikowski B, Ideker T. 2003. Cytoscape: A software environment for integrated
models of biomolecular interaction networks. Genome Res 13:2498–2504.
Shay T, Jojic V, Zuk O, Rothamel K, Puyraimond-Zemmour D, Feng T,
Wakamatsu E, Benoist C, Koller D, Regev A, ImmGen C. 2013. Conservation
and divergence in the transcriptional programs of the human and mouse
immune systems. Proc Natl Acad Sci USA 110:2946–2951.
Sica A, Schioppa T, Mantovani A, Allavena P. 2006. Tumour-associated
macrophages are a distinct M2 polarised population promoting tumour
progression: Potential targets of anti-cancer therapy. Eur J Cancer 42:717–727.
Sottoriva A, Spiteri I, Piccirillo SG, Touloumis A, Collins VP, Marioni JC,
Curtis C, Watts C, Tavare S. 2013. Intratumor heterogeneity in human glio-
blastoma reflects cancer evolutionary dynamics. Proc Natl Acad Sci USA 110:
4009–4014.
Szulzewsky F, Pelz A, Feng X, Synowitz M, Markovic D, Langmann T, Holtman
IR, Wang X, Eggen BJ, Boddeke HW, Hambardzumyan D, Wolf SA,
Kettenmann H. 2015. Glioma-associated microglia/macrophages display an
expression profile different from M1 and M2 polarization and highly express
Gpnmb and Spp1. PLoS One 10:e0116644.
Trikha P, Sharma N, Pena C, Reyes A, Pecot T, Khurshid S, Rawahneh M,
Moffitt J, Stephens JA, Fernandez SA, Ostrowski MC, Leone G. 2015. E2f3 in
1435
tumor macrophages promotes lung metastasis. Oncogene 10.1038/
onc.2015.429:[Epub ahead of print].
Umemura N, Saio M, Suwa T, Kitoh Y, Bai J, Nonaka K, Ouyang GF, Okada
M, Balazs M, Adany R, Shibata T, Takami T. 2008. Tumor-infiltrating
myeloid-derived suppressor cells are pleiotropic-inflamed monocytes/macro-
phages that bear M1- and M2-type characteristics. J Leukocyte Biol 83:
1136–1144.
Vinnakota K, Hu F, Ku MC, Georgieva PB, Szulzewsky F, Pohlmann A,
Waiczies S, Waiczies H, Niendorf T, Lehnardt S, Hanisch UK, Synowitz M,
Markovic D, Wolf SA, Glass R, Kettenmann H. 2013. Toll-like receptor 2
mediates microglia/brain macrophage MT1-MMP expression and glioma
expansion. Neuro-Oncology 15:1457–1468.
Wang J, Duncan D, Shi Z, Zhang B. 2013. WEB-based GEne SeT AnaLysis
toolkit (WebGestalt): Update 2013. Nucleic Acids Res 41:W77–W83.
Watters JJ, Schartner JM, Badie B. 2005. Microglia function in brain tumors.
J Neurosci Res 81:447–455.
Wen PY, Kesari S. 2008. Malignant gliomas in adults. N Engl J Med 359:492–507.
Wickham H. 2009. ggplot2: Elegant graphics for data analysis. New York:
Springer.
Xu S, Wei J, Wang F, Kong LY, Ling XY, Nduom E, Gabrusiewicz K, Doucette
T, Yang Y, Yaghi NK, Fajt V, Levine JM, Qiao W, Li XG, Lang FF, Rao G,
1436
Fuller GN, Calin GA, Heimberger AB. 2014. Effect of miR-142-3p on the M2
macrophage and therapeutic efficacy against murine glioblastoma. J Natl
Cancer Inst 106:106. pii: dju162 doi:10.1093/jnci/dju162.
Xue J, Schmidt SV, Sander J, Draffehn A, Krebs W, Quester I, De Nardo D,
Gohel TD, Emde M, Schmidleithner L, Ganesan H, Nino-Castro A, Mallmann
MR, Labzin L, Theis H, Kraut M, Beyer M, Latz E, Freeman TC, Ulas T,
Schultze JL. 2014. Transcriptome-based network analysis reveals a spectrum
model of human macrophage activation. Immunity 40:274–288.
Ye XZ, Xu SL, Xin YH, Yu SC, Ping YF, Chen L, Xiao HL, Wang B, Yi L, Wang QL,
Jiang XF, Yang L, Zhang P, Qian C, Cui YH, Zhang X, Bian XW. 2012. Tumor-asso-
ciated microglia/macrophages enhance the invasion of glioma stem-like cells via
TGF-beta1 signaling pathway. J Immunol 189:444–453.
Young MD, Wakefield MJ, Smyth GK, Oshlack A. 2010. Gene ontology analysis
for RNA-seq: Accounting for selection bias. Genome Biol 11:R14.
Yu G, He QY. 2016. ReactomePA: An R/Bioconductor package for reactome
pathway analysis and visualization. Mol Biosyst 12:477–479.
Yu G, Wang LG, Han Y, He QY. 2012. clusterProfiler: An R package for com-
paring biological themes among gene clusters. Omics 16:284–287.
Zhang B, Kirov S, Snoddy J. 2005. WebGestalt: An integrated system for
exploring gene sets in various biological contexts. Nucleic Acids Res 33:
W741–W748.
Volume 64, No. 8