The epigenetic factor Kmt2a/Mll1 regulates neural progenitor proliferation and neuronal and glial differentiation

The Epigenetic Factor Kmt2a/Mll1 Regulates NeuralProgenitor Proliferation and Neuronal and GlialDifferentiation

Yin-Cheng Huang,1,2 Hung-Yu Shih,3 Sheng-Jia Lin,3 Ching-Chi Chiu,4 Tsu-Lin Ma,3

Tu-Hsueh Yeh,2,4,5 Yi-Chuan Cheng3

1 Department of Neurosurgery, Chang Gung Memorial Hospital at Linkou Medical Center,Taoyuan, Taiwan

2 College of Medicine, Chang Gung University, Taoyuan, Taiwan

3 Graduate Institute of Biomedical Sciences, College of Medicine, Chang Gung University,Taoyuan, Taiwan

4 Neuroscience Research Center, Chang Gung Memorial Hospital at Linkou Medical Center, Taoyuan,Taiwan

5 Section of Movement Disorders, Department of Neurology, Chang Gung Memorial Hospital at LinkouMedical Center, Taoyuan, Taiwan

Received 5 June 2014; revised 14 September 2014; accepted 29 September 2014

ABSTRACT: Multiple epigenetic factors play a

critical role in cell proliferation and differentiation. How-

ever, their function in embryogenesis, especially in neural

development, is currently unclear. The Trithorax group

(TrxG) homolog KMT2A (MLL1) is an important epige-

netic regulator during development and has an especially

well-defined role in hematopoiesis. Translocation and

aberrant expression of KMT2A is often observed in many

tumors, indicating its proto-oncogenic character. Here,

we show that Kmt2a was essential for neural develop-

ment in zebrafish embryos. Disrupting the expression of

Kmt2a using morpholino antisense oligonucleotides and

a dominant-negative variant resulted in neurogenic phe-

notypes, including downregulated proliferation of neural

progenitors, premature differentiation of neurons, and

impaired gliogenesis. This study therefore revealed a

novel function of Kmt2a in cell proliferation and differ-

entiation, providing further insight into the function of

TrxG proteins in neural development and brain

tumors. VC 2014 Wiley Periodicals, Inc. Develop Neurobiol 00: 000–

000, 2014

Keywords: Kmt2a; neural progenitors; proliferation;

differentiation; zebrafish

INTRODUCTION

Epigenetic regulatory mechanisms play a critical role

in brain development and in the development of vari-

ous diseases and tumors. The Trithorax group (TrxG)

and Polycomb group (PcG) proteins are the major

chromatin modulators that activate or silence target

gene expression, and defects in these proteins cause

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

Correspondence to: Y.-C. Cheng ([email protected]) orT.-H. Yeh ([email protected]).

Yin-Cheng Huang and Hung-Yu Shih contributed equally tothis work.

Contract grant sponsor: Chang Gung Memorial Hospital; con-tract grant numbers: CMRPD3B0041, CMRPD3B0042.

Contract grant sponsor: National Science Council of Taiwan;contract grant number: 102-2311-B-182-002-MY3.� 2014 Wiley Periodicals, Inc.Published online 00 Month 2014 in Wiley Online Library(wileyonlinelibrary.com).DOI 10.1002/dneu.22235

1

homeotic transformations. The Lysine (K)-specific

methyltransferase 2A [KMT2A, also known as mixed

lineage leukemia 1 (MLL1)] has been identified as

the mammalian ortholog of Drosophila trithorax(trx) and belongs to the Trithorax group (TrxG) of

proteins (Djabali et al., 1992). The encoded protein

contains 9–10 zinc-finger motifs and a highly con-

served SET [Su(var), Enhancer of zeste, trx] domain.

KMT2A mediates chromatin modifications through

its histone H3 lysine 4 methyltransferase activity and

is known to directly regulate homeotic genes. Multi-

ple chromosomal translocations involving this gene

are the cause of certain acute lymphoid leukemia and

acute myeloid leukemia (Mohan et al., 2010).

The in vivo function of KMT2A has been analyzed

in invertebrates and vertebrates. Drosophila mutant of

trx showed homeotic transformation (Mozer and

Dawid, 1989). The Kmt2a homozygous knockout was

embryonically lethal, and most of the mice died before

embryonic day 12.5. By contrast, heterozygous knock-

out mice survived and exhibited growth retardation,

hematopoietic defects, and skeletal malformation (Yu

et al., 1995). The phenotypical differences between

the homozygous and the heterozygous littermates sug-

gested a dosage-sensitive regulation by the KMT2A

protein. Mice harboring a truncation of Kmt2a exhibit

various phenotypes, such as failure of preimplantation

(Ayton et al., 2001), fetal liver hematopoiesis (Yagi

et al., 1998), and acute leukemia (Dobson et al., 1999),

depending on the particular allele knocked out and the

tissues in, which Kmt2a was deleted. Furthermore,

knockdown of Kmt2a expression in zebrafish embryos

resulted in hematopoietic defects, which is a conserved

phenotype observed in mammals (Wan et al., 2011).

These studies demonstrated the important role of

KMT2A in hematopoiesis at both physiological and

pathological levels.

The enriched and ubiquitous expression of

KMT2A suggests its role in regulating cellular proc-

esses in tissues in addition to the hematopoietic sys-

tem. Previously, a conditional KMT2A knockout

study showed impaired neuronal differentiation in

postnatal mouse brain, demonstrating an essential

role of KMT2A in neurogenesis (Lim et al., 2009).

More recently, two studies revealed that KMT2A is

expressed in hypoxic conditions (Heddleston et al.,

2012) and is required for the growth of glioblastoma

stem cells (Gallo et al., 2013), suggesting that

KMT2A is associated with glial-derived tumors and

may have a potential role in gliogenesis. However,

the role of KMT2A in brain malignancies still

requires further characterization. The prenatal func-

tion of KMT2A in the developing nervous system

remains especially unclear, and its study may provide

valuable information toward elucidating the role of

KMT2A in brain tumorigenesis.

In this study, we interfered with the expression of

Kmt2a in zebrafish embryos to study the endogenous

role during neural development. Because the study of

mice suggested that KMT2A acts in a dosage-sensitive

manner and exhibit lethal actions in homozygous knock-

out mice (Yu et al., 1995), we knocked down without

completely abolishing the endogenous Kmt2a expression

using a kmt2a antisense morpholino. In addition, we

used a dominant-negative kmt2a variant to confirm the

results of the knockdown experiment further. The results

revealed that the embryos with Kmt2a deficiency exhib-

ited decreased neural progenitor cell proliferation, prema-

ture differentiation of neurons, and defected gliogenesis.

Our data show a novel function for Kmt2a in the regula-

tion of cellular processes, providing further insights into

the diverse roles of Kmt2a in neural development and a

possible mechanism in brain tumor formation.

MATERIALS AND METHODS

Ethics Statement

All experiments were performed in strict accordance with

standard guidelines for zebrafish work and approved by the

Institutional Animal Care and Use Committee of Chang

Gung University (IACUC approval number: CGU12–039).

Fish Maintenance and Mutants

T€u (wild type) and Tg(gfap:egfp) zebrafish embryos were

purchased from the Zebrafish International Resource Center

(Oregon) and were raised, maintained, and paired under

standard conditions. The embryos were staged according to

the number of somites, hours postfertilization, and days

postfertilization (Kimmel et al., 1995).

Generation of Constructs

kmt2aN84 was polymerase chain reaction (PCR)-amplified

with high fidelity Pfu and with primers (F: 50-GGATCCGCCGCCACCATGGCGCACAGCTGTCGGT

GGC-30and R: 50-GGATCCCTATTATTCCTCCTC

CCCGCTACTGGAGCC-30) according to the GenBank

sequence (accession number: XM_005157583.1). The Kmt2amorpholino binding sequence was inserted upstream of an

enhanced green fluorescent protein (eGFP) reporter in the

pCS21 vector to create 50kmt2a-EGFP construct to evaluate

the specificity and efficiency of morpholino.

RNA and Morpholino Injection

Capped RNA encoding the full coding sequence of

kmt2aN84 was prepared as described previously (Chung

et al., 2011). Antisense morpholino oligonucleotides were

2 Huang et al.

Developmental Neurobiology

purchased from Gene Tools, LLC (Oregon). A morpholino

against kmt2a (TGCTGAGATCGCTCGTTCGGGGCTA)

that corresponds to 272 to 248 to the translation start site

was used. Basic Local Alignment Search Tool (Blast) anal-

ysis revealed homology of less than 20 bp identity for

kmt2a morpholino to other genomic sequences, none of

which corresponded to 50 UTR or exon–intron splicing site

of predicted or characterized genes, suggesting that the

morpholino would be specific for kmt2a. As a control

experiment, a morpholino designed with a random nucleo-

tide sequence not found in the zebrafish genome (50-CCTCTTACCTCAGTTACAATTTATA-30; Gene Tools)

or a morpholino with five bases mismatch to kmt2a mor-

pholino (50-TAcCCCCcAAgGAGCcATCTCAcCA-30; mis-

matched bases are indicated by small letters) was injected

in equal amounts to the kmt2a morpholino. All injections

were performed at the one- to two-cell stage and cRNAs or

morpholinos were introduced into blastomeres.

Histological Analysis

Digoxigenin-UTP labeled riboprobes were synthesized

according to the manufacturer’s instructions (Roche) and insitu hybridizations were performed as described previously

(Cheng et al., 2013). The color reaction was carried out using

the NBT/BCIP substrate (Roche). For immunohistochemistry,

the embryos were blocked in 5% goat serum and incubated

with mouse anti-HuC/HuD monoclonal 16A11 antibody (1/

500 dilution, Invitrogen), rabbit phospho-histone H3 antibody

(1:500, Millipore), or rabbit monoclonal antiactive caspase-3

(1:200, Abcam). Fluorochrome-conjugated antibodies Alexa

Fluor 488 goat anti-mouse (Invitrogen) was used to detect the

primary antibodies. Embryos were mounted with Vectashield

mounting medium (Vector Laboratories).

Quantitative Analysis

For quantitative real-time PCR (qPCR), embryos were

homogenized in TRIzol reagent (Invitrogen), and total

RNA was extracted using a standard method. cDNA was

synthesized from total RNA with random hexamer priming

using RevertAid First Strand cDNA Synthesis Kit (Fermen-

tas). qPCR was performed on an ABI StepOneTM Real-

Time PCR System (Applied Biosystems) with SYBR green

fluorescent label (Fermentas). Primers for sox2 (F: 50-CG

GAAAATGGCACAGGAGAA-30; R: 50-GTAATCCGGGT

GTTCCTTCATG-30), neurogenin1 (F: 50-CGCACACGGA

TGATGAAGACTCGCG-30; R: 50-CGGTTCTTCTTCAC-

GACGTGCACAGTGG-30), slc1a3 (F: 50-GTAACGGGGA

GACGCGTCTGCAGCG-30; R: 50-GATTATTCCCACGA

TGACGGCGGCG-30), gata1 (F: 50-ACACAGTCCAGTT

CGCCAAGT-30; R: 50-TGGAGAGGTGTTTTTGGGAA

A-30), and gapdh (F: 50-ACCCGTGCTGCTTTCTTGAC-30;R: 50-GACCAGTTTGCCGCCTTCT-30) were used. Gene

expression levels were normalized to gapdh and assessed

using the comparative CT (40 cycles) according to the manu-

facturer’s instructions (Applied Biosystems).

For Western blot analysis, embryos were homogenized

in sodium dodecyl sulfate (SDS) lysis buffer. Sixty micro-

grams was loaded on a 12% SDS polyacrylamide gel, trans-

ferred to a polyvinylidene difluoride (PVDF) membrane

and detected with anti-GFP antibody (1:1000, Invitrogen)

or tubulin (1:5000, Sigma). After washes, membranes were

incubated with goat anti-Mouse horseradish peroxidase

(HRP)-conjugated secondary Ab (Chemicon) and devel-

oped with ECL (Millipore). Band intensities were quanti-

fied using Multi Gaugre analysis software.

Statistical analysis was performed using Student’s t-test

in Microsoft ExcelVR 2007. The significance level was set at

p< 0.05. All reactions were performed in triplicate for each

sample.

RESULTS

kmt2a Morpholinos and the Dominant-Negative kmt2a Variant Effectively Inter-fere Kmt2a Expression

The sequence and expression of kmt2a have been

reported previously (Robinson et al., 2011; Wan

et al., 2011). We performed expression analysis with

more focus on the developing nervous system and

found that kmt2a was ubiquitously expressed in the

entire embryo from four-cell stage but later restricted

in the brain from 48-hours postfertilization (hpf), as

previously described [Fig. 1(A)].

To study the role of Kmt2a in the developing nerv-

ous system, we used a morpholino knockdown

approach to interfere with the endogenous expression

of Kmt2a. Two morpholinos were used in the previ-

ous study to knockdown Kmt2a expression, but these

morpholinos induced off-target cell apoptosis (Wan

et al., 2011). Therefore, we synthesized another anti-

sense morpholino that targeted the alternative transla-

tion start site of kmt2a mRNA to block protein

production. This morpholino showed no off-target

effects, as described later. BLAST analysis detected

<20 bp identity between the kmt2a morpholino and

other genomic sequences, none of which corre-

sponded to the 50-UTR or exon–intron splicing site of

the predicted or characterized genes, suggesting that

the morpholino may act specifically on kmt2a. The

specificity of the kmt2a morpholino was further con-

firmed by coinjection of a reporter construct that

fused with the kmt2a morpholino binding sequences

upstream of an eGFP reporter. Specific knockdown

of the translation of this construct (evident by eGFP

loss) was observed on coinjection with the kmt2amorpholinos but not with the 5-mismatch control

morpholino [Supporting Information Fig. 1(A)]. A

previous study showed that Kmt2a is essential for

hematopoiesis in zebrafish embryos (Wan et al.,

2011). Downregulation of the hematopoietic marker

Kmt2a Regulates Neurodevelopment 3

Developmental Neurobiology

Figure 1 kmt2a is expressed in the developing nervous system and the expression is required for

neural development. (A) kmt2a expression was detected by in situ hybridization in the developing

nervous system during zebrafish embryogenesis. The embryo stages are shown in the bottom left

corner of each panel. All panels are lateral view, with anterior to the left except 0.75 hpf, 5.3 hpf,

and 8 hpf. Ubiquitous expression of kmt2a was detected from 0.75 hpf to 16 hpf. Strong expression

was detected in the entire nervous system at 24 hpf and later became restricted to specific brain

areas, persisting until the final stage that was analyzed (48 hpf). di, diencephalon; fb, forebrain; hb,

hindbrain; mb, midbrain; sp, spinal cord; t, tectum; hpf, hours postfertilization. (B) Interference

with Kmt2a expression using kmt2a morpholino or kmt2aN84 resulted in brain malformation (arrow-

heads). Note, however, that the midbrain-hindbrain boundaries (white arrowheads) and somitic

boundaries are unaffected. Scale bars: 200 lm.

gata1 as previously reported [Supporting Information

Fig. 1(B,C)] indicated the effectiveness of the kmt2amorpholino.

The specificity of the morpholino could not be con-

firmed by rescue experiments using kmt2a cRNA due

to the extra-long coding sequence (12,703 bp) that

made it difficult to amplify the cDNA, which also

could not be transcribed in vitro. Therefore, we con-

structed a deletion variant that contained only the 84-

amino acid N terminus of kmt2a (kmt2aN84). This N-

terminal minipeptide contains a highly conserved

Menin-binding motif and has been demonstrated to act

as a dominant negative form that disrupts normal func-

tioning of kmt2a during hematopoiesis (Wan et al.,

2011). Injection of kmt2aN84 downregulated gata1expression in a manner identical to that observed in

kmt2a knockdown embryos [Supporting Information

Fig. 1(B,C) and Fig. 1(B)]. Therefore, kmt2aN84 injec-

tion confirmed the specificity of the kmt2a morpholino

and was used in each experiment as described later.

Kmt2a Depletion is Sufficient to ReduceNeural Progenitor Proliferation

Embryos injected with kmt2a morpholino or

kmt2aN84 were first analyzed at 24 hpf for morpho-

logical defects. The injection of kmt2a morpholino or

kmt2aN84 produced an identical phenotype exhibiting

brain malformation, particularly thickening and

abnormally folded structures of the neural tube, indi-

cating that Kmt2a is required for neural development

[Fig. 1(B)]. A previous study showed aberrant seg-

mental boundaries of spinal ganglia and somites in a

Kmt2a deficient mouse (Yu et al., 1998). In contrast,

Figure 2 Defective Kmt2a expression decreased the proliferation of neural progenitors. (A) Insitu hybridization of 75%-epiboly embryos showed that sox2 expression was downregulated by the

kmt2a morpholino or kmt2aN84 cRNA, as confirmed by qPCR analysis on the right. (B) Embryos

were flat-mounted and double-labeled with sox2 and phospho-histone H3 antibody. The bottom

panels are representative of the enlargement regions of the upper panels, as indicated. sox2-express-

ing cells were pseudo-colored with fluorescent red and counterstained with phospho-histone H3

antibody for immunohistochemistry (fluorescent green) to locate proliferating neural progenitor

cells (arrowheads). Injection of the kmt2a morpholino or kmt2aN84 decreased proliferation of neural

progenitors. The proportions of phospho-histone H3- and sox2-positive cells among the total sox2-

positive cells were quantified, as shown on the right. Note that in the Kmt2a-deficient embryos, no

significant deviation was observed in the proportions of phosphohistone H3-positive and sox2-nega-

tive cells in the sox2-negative cells counted in adjacent surface ectoderm.

Kmt2a Regulates Neurodevelopment 5

Developmental Neurobiology

the brain and somitic boundaries appeared unaffected

by the injection of kmt2a morpholino or kmt2aN84

[Fig. 1(B)].

During development, the heterogeneous neurons

and glial cells are derived from neural stem cells and

progenitor cells originating from the neuroectoderm

in spatial- and temporal-related fashions. Therefore,

we first analyzed the effects of Kmt2a knockdown

using the neural progenitor marker sox2 during neural

induction. The result of whole-mount in situ hybrid-

ization showed significantly decreased sox2 expres-

sion [Fig. 2(A)]. Quantitative real-time PCR (qPCR)

analysis confirmed a 2-fold reduction in sox2 [Fig.

2(A)]. kmt2aN84 injection also downregulated the

expression of sox2 by 4-fold [Fig. 2(A)]. We ques-

tioned whether the reduced number of neural progen-

itors in Kmt2a deficient embryos resulted from

induction of apoptosis or inhibition of proliferation

and further analyzed cell proliferation using a phos-

phohistone H3 antibody and counterstaining with

sox2 to localize the proliferating neural progenitors.

The result revealed a decreased number of proliferat-

ing neural progenitors in embryos injected with the

kmt2a morpholino or kmt2aN84, an observation that

was further confirmed by a count of the proliferating

cells in the sox2-positive and sox2-negative popula-

tions [Fig. 2(B)]. This result suggests that Kmt2a is

required for the proliferation of neural progenitors.

To investigate whether the loss of sox2-positive

neural progenitors was due to cell death, apoptotic

neural precursors were analyzed by looking for the

presence of proteolytic activation of the effector

caspase-3 by immunohistochemistry on Kmt2a defi-

cient embryos. Following in situ hybridization with

sox2 RNA probe, embryos were subjected to immu-

nohistochemical staining to determine whether apo-

ptosis was localized to neural progenitors (Fig. 3).

Injection of kmt2a morpholino or kmt2aN84 did not

cause a significant alteration in the number of apopto-

tic neural progenitors, suggesting that altered Kmt2a

expression had no effect on cell survival (Fig. 3).

This result also confirmed that the effect of the kmt2amorpholino was due to specific Kmt2a knockdown

but not off-target cell apoptosis.

Impaired Kmt2a Expression CausesPrecocious Neuronal Differentiation

After the formation of neural progenitors, the next

step in neurogenesis is the specification of neuronal

precursors within the neurogenic region, a process

regulated by proneural genes. We examined the role

of Kmt2 in neuronal precursors using the proneural

marker neurogenin1. The results showed a 1.9 to 2.9-

fold decrease in neurogenin1 expression in embryos

injected with kmt2a morpholino or kmt2aN84 ana-

lyzed at the bug stage [Fig. 4(A,B)]. Although this

effect could be explained by the decreased neural

progenitors as described earlier and, therefore,

Figure 3 kmt2a morpholino or kmt2aN84 cRNA injection had no effect on neural progenitor apo-

ptosis. Apoptotic neural progenitor cells were labeled for sox2 (fluorescent red) and activated

caspase-3 antibody (fluorescent green). As shown in E, apoptotic neural progenitor cells were quan-

tified by counting the proportions of activated caspase-3- and sox2-positive (or -negative) cells

among the total sox2-positive (or -negative) cells. *p< 0.05; **p< 0.01; n.s., not significant. Scale

bar: 100 lm, applies to all panels.

6 Huang et al.

Developmental Neurobiology

consequently caused the loss of neuronal precursors,

immunohistochemistry analysis using the postmitotic

neuronal marker HuC/D antibody revealed significant

upregulation of HuC/D in Kmt2a deficient embryos at

24 hpf [Fig. 4(C)]. Ectopic HuC/D-positive cells were

also observed in several regions where normal neuro-

nal differentiation would not occur [arrows in Fig.

4(C)]. This effect was confirmed by Western blot anal-

ysis showing a 3- to 4-fold increase in HuC/D expres-

sion in kmt2a deficient embryos [Fig. 4(D)].

Concurrently with the increased HuC/D expression,

the expression of neurogenin1 was significantly down-

regulated at 24 hpf [Fig. 4(E,F)]. These results sug-

gested that injection of the kmt2a morpholino or

kmt2aN84 could cause excessive neurogenesis. We

thus performed a time course analysis to investigate

whether impaired Kmt2a expression could elicit pre-

mature differentiation of neurons. The result showed

that the number of HuC/D-positive neurons was unal-

tered in Kmt2a deficient embryos in comparison to

control embryos at 14 hpf. However, HuC/D-positive

neurons started to be significantly upregulated in

Kmt2a deficient embryos from 16 hpf and onwards

(Fig. 5). This result indicated that neurons were pre-

maturely differentiated from neurogenein1-positive

precursors into HuC/D-positive differentiating neurons

and that this effect is separable from the decreased

proliferation of neural progenitors. The well-

organized, ladder-like arrays of neurons in the devel-

oping hindbrain were unaffected by the kmt2a mor-

pholino or kmt2aN84 cRNA injection (Figs. 4(C) and

5), suggesting that Kmt2a deficiency does not affect

Figure 4 Disrupted Kmt2a expression causes aberrant formation of neuronal precursors and

mature neurons. (A) At the bud stage, the expression level of neurogenin1 was significantly

decreased in embryos injected with the kmt2a morpholino or kmt2aN84 cRNA injected embryos in

comparison to the controls. (B) qPCR analysis confirmed the results obtained by in situ hybridiza-

tion in A. (C and E) The upper panels are enlargements of the hindbrain region, anterior to the top;

and the bottom panels present lateral views of the enlargements of the 3–9-somite levels of the spi-

nal cord. (C) HuC/D-expressing post-mitotic neurons increased massively in kmt2a morpholino or

kmt2aN84 cRNA injected embryos, as shown by immunohistochemical analysis with anti-HuC/D

antibody at 24 hpf. Note the ectopic HuC/D expression in the ventricular zone (arrowheads). (D)

Levels of HuC/D expression were confirmed by Western blot analysis and were quantified. (E) At

24 hpf, neurogenin1 expression was reduced significantly by the kmt2a morpholino and kmt2aN84.

(F) qPCR analysis further confirmed the decreased expression of neurogenein1 at 24 hpf in E.

*p< 0.05; **p< 0.01.

Kmt2a Regulates Neurodevelopment 7

Developmental Neurobiology

neural migration and patterning. In addition, we exam-

ined whether Kmt2a deficiency affected the proliferat-

ing neurons using phosphohistone H3 antibody and

counterstaining with neurogenin1. The results demon-

strated that neuronal proliferation remained intact in

embryos injected with kmt2a morpholino or kmt2aN84

(Supporting Information Fig. 2), indicating that Kmt2a

was not required for neuronal proliferation.

Kmt2a is Required for Gliogenesis

In addition to producing neurons, neural progenitors

generate glial derivatives such as astrocytes and oli-

godendrocytes. We accordingly analyzed the role of

Kmt2a in gliogenesis. Injection of the kmt2a morpho-

lino or kmt2aN84 cRNA downregulated the expression

of the early glial marker slc1a3a (Glast in mammals)

Figure 5 Kmt2a-deficient embryos exhibit upregulated HuC/D expression. Significant upregula-

tion of HuC/D signals, as analyzed by immunohistochemistry using an anti-HuC/D antibody, can

be detected at 16 hpf and 18 hpf in embryos injected with either the kmt2a morpholino or kmt2aN84

cRNA. Quantitative data are presented as the mean 6 standard deviation normalized to the number

of controls. *p< 0.05; **p< 0.01; n.s., not significant. Scale bar: 100 lm, applies to all panels.

8 Huang et al.

Developmental Neurobiology

(Storck et al., 1992; Shibata et al., 1997; Hsieh et al.,

2013) and decreasing the expression by 2.0- to 3.3-

fold according to qPCR analysis (Fig. 6). These data

suggested that Kmt2a is required for gliogenesis. To

test whether the effects of slc1a3a were due to irregu-

lar glial differentiation, we evaluated the expression

of the mature glial cell markers using mag for

myelinated glial cells (Quarles, 2007), oilg1 for oli-

godendrocytes and Gfap for mature astrocytes.

Altered Kmt2a expression decreased the expression

of all three markers for glial (Fig. 6) as confirmed by

cell count, qPCR, and Western blot analysis (Fig. 6

and Supporting Information Fig. 3). Overall, these

results showed that Kmt2a is required for gliogenesis.

An explanation for the loss of glial cells in the

Kmt2a-deficient embryos is that Kmt2a deficiency

resulted in insufficient neural progenitors and conse-

quently led to loss of the glial derivatives. Another

possibility is that Kmt2a directly regulates gliogene-

sis, independent from its role in the generation of

neural progenitors. Alternatively, Kmt2a might regu-

late the fate determination between neurons and glia,

and thus, Kmt2a deficiency might cause biased dif-

ferentiation of neurons and downregulation of the

glial markers.

DISCUSSION

Recent studies have revealed the importance of

KMT2A in the epigenetic regulation of cell self-

renewal and differentiation in hematopoiesis.

KMT2A (or Kmt2a in zebrafish) deficient

Figure 6 Disrupted Kmt2a expression causes decreased glial precursors and derivatives. The

expression of slc1a3a, oilg1, and mag were analyzed by in situ hybridization (A), and Gfap was

examined in Tg(gfap:egfp) transgenic embryos (C). Injection of the kmt2a morpholino or kmt2aN84

cRNA downregulated the expression of all these glial markers. (B) qPCR analysis confirmed the

results obtained by in situ hybridization in A. (D) Levels of GFP expression in C were confirmed

by Western blot analysis and were quantified. Quantitative data are presented as the means 6 stan-

dard deviations. *p< 0.05; **p< 0.01; n.s., not significant.

Kmt2a Regulates Neurodevelopment 9

Developmental Neurobiology

experiments have been performed in vivo using

knockout mice (Yu et al., 1995) and morpholino

knockdown zebrafish (Wan et al., 2011), which

revealed the essential role of KMT2A (Kmt2a) in

embryonic hematopoiesis; however, no previous

study has described the role of KMT2A (Kmt2a) defi-

ciency in causing any abnormal development in the

nervous systems. Conditional knockout of Kmt2a in

the neural stem cells of the subventricular zone of the

postnatal nervous system demonstrated that although

the neural stem cells survive, proliferate, and effi-

ciently differentiate into glial lineages, neuronal dif-

ferentiation is severely impaired (Lim et al., 2009).

By contrast, other studies have suggested a potential

role of KMT2A in gliogenesis by stating that

KMT2A is required for the growth of the glioblas-

toma cells (Heddleston et al., 2012; Gallo et al.,

2013). Therefore, the role of KMT2A in the nervous

system, particularly in cell progression including pro-

liferation, survival, and differentiation, remains

unclear. This study evaluated the physiological role

of the Kmt2a protein in the development of the nerv-

ous system by disrupting the Kmt2a expression using

the kmt2a morpholino and the dominant-negative

kmt2a variant kmt2aN84. Disrupted endogenous

expression of Kmt2a in the zebrafish embryos

decreased the proliferation of neural progenitor cells,

which resemble neural stem cells in many ways but

undergo a limited number of replication cycles invivo, indicating that Kmt2a is essential for the prolif-

eration of neural progenitors or stem cells and might

thus regulate the proliferation of neural progenitor

and brain tumor stem cells through a similar mecha-

nism. Moreover, the Kmt2a-deficient embryos exhib-

ited premature differentiation of neurons and

defective gliogenesis, which indicated that Kmt2a is

essential for the inhibition of neuronal differentiation

as well as appropriate glial differentiation; this fur-

ther strengthens the explanations regarding the onco-

genic role of Kmt2a in brain tumor formation.

The molecular regulation of KMT2A (Kmt2a) in

neural progenitor proliferation, neurogenesis, or glio-

genesis is not clear. The best established target of

KMT2A are the Homeobox genes (HOX genes),

which are fundamental for segmental identity during

early development (Krivtsov and Armstrong, 2007).

We found that hoxa9a, one of the well-characterized

genes downstream of Kmt2a, was downregulated in

Kmt2a-deficient embryos (unpublished data), sug-

gesting that Kmt2a also positively regulates hoxa9a.

However, hoxa9a is not expressed in neural progeni-

tor populations, suggesting that Kmt2a-mediated neu-

ral progenitor proliferation is regulated by other

molecules. Some potential candidates are the compo-

nents in the Notch signaling pathway. Research in

Drosophila and mammals suggests that the Notch

pathway is regulated by TrxG proteins such as the

histone demethylase UTX and the chromatin remod-

eler BRM (Schuettengruber et al., 2011). In particu-

lar, the Drosophila homologue of Kmt2a, trithorax,

was shown to collaborate with Notch in gene activa-

tion (Bejarano and Milan, 2009). Notch-mediated

signaling plays a fundamental role in a variety of

neural developmental processes and in the pathoge-

nesis of several human cancers (Roy et al., 2007). In

the developing nervous system, Notch signaling posi-

tively regulates the maintenance of neural progenitors

and later governs the decision between neuronal and

glial lineages. Deficient Notch signaling resulted in

decreased proliferation of neural progenitors, prema-

ture neuronal differentiation, and defected gliogene-

sis (Chung et al., 2011; Cheng et al., 2013), which

highly resembles the phenotypes observed in Kmt2a-

deficient embryos. Therefore, it is worthwhile to fur-

ther investigate the regulatory mechanism between

KMT2A (Kmt2a) and the Notch signaling pathway in

neural development and tumorigenesis.

The authors thank David Wilkinson for neurogenin1and Paul Scotting for the sox2 constructs used in making

the riboprobes. The authors are also grateful to the Taiwan

Zebrafish Core facility at ZeTH and the Zebrafish Core in

Academia Sinica for providing fish.

REFERENCES

Ayton P, Sneddon SF, Palmer DB, Rosewell IR, Owen MJ,

Young B, Presley R, et al. 2001. Truncation of the Mll gene

in exon 5 by gene targeting leads to early preimplantation

lethality of homozygous embryos. Genesis 30:201–212.

Bejarano F, Milan M. 2009. Genetic and epigenetic mecha-

nisms regulating hedgehog expression in the Drosophila

wing. Dev Biol 327:508–515.

Cheng YC, Scotting PJ, Hsu LS, Lin SJ, Shih HY, Hsieh

FY, Wu HL, et al. 2013. Zebrafish rgs4 is essential for

motility and axonogenesis mediated by Akt signaling.

Cell Mol Life Sci 70:935–950.

Chung PC, Lin WS, Scotting PJ, Hsieh FY, Wu HL, Cheng

YC. 2011. Zebrafish Her8a is activated by Su(H)-depend-

ent notch signaling and is essential for the inhibition of

neurogenesis. PLoS One 6:e19394.

Djabali M, Selleri L, Parry P, Bower M, Young BD, Evans

GA. 1992. A trithorax-like gene is interrupted by chromo-

some 11q23 translocations in acute leukaemias. Nat

Genet 2:113–118.

Dobson CL, Warren AJ, Pannell R, Forster A, Lavenir I,

Corral J, Smith AJ, et al. 1999. The mll-AF9 gene fusion

in mice controls myeloproliferation and specifies acute

myeloid leukaemogenesis. EMBO J 18:3564–3574.

10 Huang et al.

Developmental Neurobiology

Gallo M, Ho J, Coutinho FJ, Vanner R, Lee L, Head R,

Ling EK, et al. 2013. A tumorigenic MLL-homeobox net-

work in human glioblastoma stem cells. Cancer Res 73:

417–427.

Heddleston JM, Wu Q, Rivera M, Minhas S, Lathia JD,

Sloan AE, Iliopoulos O, et al. 2012. Hypoxia-induced

mixed-lineage leukemia 1 regulates glioma stem cell

tumorigenic potential. Cell Death Differ 19:428–439.

Hsieh FY, Ma TL, Shih HY, Lin SJ, Huang CW, Wang

HY, Cheng YC. 2013. Dner inhibits neural progenitor

proliferation and induces neuronal and glial differentia-

tion in zebrafish. Dev Biol 375:1–12.

Kimmel CB, Ballard WW, Kimmel SR, Ullmann B,

Schilling TF. 1995. Stages of embryonic development of

the zebrafish. Dev Dyn 203:253–310.

Krivtsov AV, Armstrong SA. 2007. MLL translocations,

histone modifications and leukaemia stem-cell develop-

ment. Nat Rev Cancer 7:823–833.

Lim DA, Huang YC, Swigut T, Mirick AL, Garcia-

Verdugo JM, Wysocka J, Ernst P, et al. 2009. Chromatin

remodelling factor Mll1 is essential for neurogenesis

from postnatal neural stem cells. Nature 458:529–533.

Mohan M, Lin C, Guest E, Shilatifard A. 2010. Licensed to

elongate: A molecular mechanism for MLL-based leu-

kaemogenesis. Nat Rev Cancer 10:721–728.

Mozer BA, Dawid IB. 1989. Cloning and molecular charac-

terization of the trithorax locus of Drosophila mela-

nogaster. Proc Natl Acad Sci USA 86:3738–3742.

Quarles RH. 2007. Myelin-associated glycoprotein (MAG):

Past, present and beyond. J Neurochem 100:1431–1448.

Robinson BW, Germano G, Song Y, Abrams J, Scott M,

Guariento I, Tiso N, et al. 2011. mll ortholog containing

functional domains of human MLL is expressed through-

out the zebrafish lifespan and in haematopoietic tissues.

Br J Haematol 152:307–321.

Roy M, Pear WS, Aster JC. 2007. The multifaceted role of

notch in cancer. Curr Opin Genet Dev 17:52–59.

Schuettengruber B, Martinez AM, Iovino N, Cavalli G.

2011. Trithorax group proteins: Switching genes on and

keeping them active. Nat Rev Mol Cell Biol 12:799–814.

Shibata T, Yamada K, Watanabe M, Ikenaka K, Wada K,

Tanaka K, Inoue Y. 1997. Glutamate transporter GLAST

is expressed in the radial glia-astrocyte lineage of devel-

oping mouse spinal cord. J Neurosci 17:9212–9219.

Storck T, Schulte S, Hofmann K, Stoffel W. 1992. Struc-

ture, expression, and functional analysis of a Na(1)-

dependent glutamate/aspartate transporter from rat brain.

Proc Natl Acad Sci USA 89:10955–10959.

Wan X, Hu B, Liu JX, Feng X, Xiao W. 2011. Zebrafish

mll gene is essential for hematopoiesis. J Biol Chem 286:

33345–33357.

Yagi H, Deguchi K, Aono A, Tani Y, Kishimoto T, Komori

T. 1998. Growth disturbance in fetal liver hematopoiesis

of Mll-mutant mice. Blood 92:108–117.

Yu BD, Hanson RD, Hess JL, Horning SE, Korsmeyer SJ.

1998. MLL, a mammalian trithorax-group gene, func-

tions as a transcriptional maintenance factor in morpho-

genesis. Proc Natl Acad Sci USA 95:10632–10636.

Yu BD, Hess JL, Horning SE, Brown GA, Korsmeyer SJ.

1995. Altered Hox expression and segmental identity in

Mll-mutant mice. Nature 378:505–508.

Kmt2a Regulates Neurodevelopment 11

Developmental Neurobiology