Identification of tumor differentiation factor (TDF) in select CNS neurons

1 23

Brain Structure and Function ISSN 1863-2653 Brain Struct FunctDOI 10.1007/s00429-013-0571-1

Identification of tumor differentiationfactor (TDF) in select CNS neurons

Alisa G. Woods, Izabela Sokolowska,Katrin Deinhardt, Cristinel Sandu &Costel C. Darie

1 23

Your article is protected by copyright and

all rights are held exclusively by Springer-

Verlag Berlin Heidelberg. This e-offprint is

for personal use only and shall not be self-

archived in electronic repositories. If you wish

to self-archive your article, please use the

accepted manuscript version for posting on

your own website. You may further deposit

the accepted manuscript version in any

repository, provided it is only made publicly

available 12 months after official publication

or later and provided acknowledgement is

given to the original source of publication

and a link is inserted to the published article

on Springer's website. The link must be

accompanied by the following text: "The final

publication is available at link.springer.com”.

ORIGINAL ARTICLE

Identification of tumor differentiation factor (TDF) in select CNSneurons

Alisa G. Woods • Izabela Sokolowska •

Katrin Deinhardt • Cristinel Sandu •

Costel C. Darie

Received: 27 November 2012 / Accepted: 30 April 2013

� Springer-Verlag Berlin Heidelberg 2013

Abstract Identification of central nervous system (CNS)

molecules elucidates normal and pathological brain func-

tion. Tumor differentiation factor (TDF) is a recently-found

protein secreted by the pituitary into the blood. TDF mRNA

was detected in brain; not heart, placenta, lung, liver, skeletal

muscle, or pancreas. However, TDF has an unclear function.

It is not known whether TDF is expressed only by pituitary or

by other brain regions. It is also not known precisely where

TDF is expressed in the brain or which cells produce TDF.

Database searching revealed that this molecule shares no

homology with any known protein. Therefore, we investi-

gated the distribution of TDF in the rat brain using immu-

nohistochemistry (IHC) and immunofluorescence (IF). TDF

protein was detected in pituitary and most other brain

regions. Double-staining for TDF and glial fibrillary acidic

protein (GFAP), an astrocyte marker, showed no co-locali-

zation. Double-staining for TDF with NeuN, a neuronal

marker, showed co-localization. Not all NeuN positive cells

were positive for TDF. Western blotting (WB) using NG108

neuroblastoma and GS9L astrocytoma cell lysate revealed

TDF immunoreactivity in cultured neuroblastoma, not

astrocytoma. These data suggest that TDF is localized in

neurons, not in astrocytes. This is the first report of any

cellular localization of TDF. TDF may have specific roles as

a pituitary-derived hormone and in the CNS, and appears to

be produced by distinct CNS neurons, not astroglia.

Keywords Hormone � Growth-factor � Neurons � CNS

Abbreviations

TDF Tumor differentiation factor

TDF-R TDF receptor

CNS Central nervous system

NG108-15 Neuroblastoma 9 glioma cell line

GS9L Astrocytoma cell line

GFAP Glial fibrillary acidic protein

NeuN Neuron-specific DNA-binding nuclear protein

which functions as a marker for neurons

IHC Immunohistochemistry

IF Immunofluorescence

WB Western blotting

Introduction

Novel proteins identified in brain can be helpful for

understanding normal, pathological, and developmental

processes in the CNS. Many proteins have distribution and

local production within the CNS, but may also have dis-

tinct functions outside of the CNS and may be distributed

A. G. Woods and I. Sokolowska contributed equally to this work.

A. G. Woods � I. Sokolowska � C. C. Darie (&)

Biochemistry and Proteomics Group, Department of Chemistry

and Biomolecular Science, Clarkson University, 8 Clarkson

Avenue, Potsdam, NY 13699-5810, USA

e-mail: [email protected]

A. G. Woods

Neuropsychology Clinic and Psychoeducation Services,

SUNY Plattsburgh, Plattsburgh, NY 12901, USA

K. Deinhardt

Centre for Biological Sciences, University of Southampton, Life

Sciences Building 85, Southampton SO17 1BJ, UK

K. Deinhardt

Institute for Life Sciences, University of Southampton, Life

Sciences Building 85, Southampton SO17 1BJ, UK

C. Sandu

Howard Hughes Medical Institute, The Rockefeller University,

1230 York Avenue, New York, NY 10065, USA

123

Brain Struct Funct

DOI 10.1007/s00429-013-0571-1

Author's personal copy

in the bloodstream. Specific proteins may have numerous

roles, including as growth factors within the CNS (medi-

ating neuroprotection, mitosis, process growth, and other

processes) as well as regulating hormonal systems (Dallner

et al. 2002; Friedrich et al. 2013; Guthrie et al. 1997;

Rezende et al. 2012; Roy et al. 2012; Wiese et al. 2012;

Woods et al. 1998, 1999).

One novel, understudied protein of interest is TDF. TDF

is a 108-amino acid polypeptide protein initially isolated

from a human pituitary cDNA library. It is found in human

whole brain extract and GH3 pituitary tumor cell line. TDF

has an incompletely established role. TDF and TDF-P1, a

20 amino acid peptide selected from the open reading

frame of TDF, can differentiate human breast and prostate

cancer cells. Specifically, TDF induces polarization and

formation of cell junctions and basement membrane, milk

protein synthesis, and E-cadherin overexpression. How-

ever, TDF has no known differentiation effect on fibro-

blasts, kidney, hepatoma, and leukemic lymphocytic cells.

Its presence in the pituitary and human blood serum sug-

gests that it is a hormone (Platica et al. 1992, 2004; So-

kolowska et al. 2012a, b, c). The size of TDF is consistent

with an anterior pituitary-derived hormone, not a much

smaller posterior pituitary-derived hormone (Caldwell and

Young 2006; Roy et al. 2012; Sokolowska et al. 2012a). It

is not clear that TDF has a CNS function; however, anterior

pituitary-derived hormones are produced outside of the

anterior pituitary—for example in the nervous, muscular,

immune, or reproductive systems (Harvey 2010).

The putative TDF receptor (TDF-R) was recently iden-

tified. The cellular localization of TDF-R has not been

confirmed in any tissue (Sokolowska et al. 2011, 2012a, b,

c). Glucose Regulated Protein-78 (GRP78) and Heat

Shock-70 (HSP70) proteins are potential TDF-R candidates

in breast cells. GRP78 and HSP70 are both present in the

mammalian nervous system and may be neuroprotective

(Goldenberg-Cohen et al. 2011; Kim et al. 2012). Their

presence in the CNS suggests that, if these proteins are

indeed receptors for TDF, TDF may have CNS activity.

Based on evidence that TDF is present in brain extract

and its cellular pattern is unknown, we initiated this study

to test the hypothesis that (1) TDF is produced by the

pituitary, as would be consistent for a blood-borne hor-

mone and (2) to further investigate whether TDF is present

in other brain regions. Therefore, we examined TDF pro-

tein in rat brain using IHC and IF. TDF protein was

detected in pituitary and most other brain regions. Double-

staining for detection of TDF and glial fibrillary acidic

protein (GFAP), an astrocyte marker, showed no co-

localization. Double-staining for TDF with NeuN, a neu-

ronal marker, showed co-localization. Not all NeuN posi-

tive cells were TDF-immunopositive. A subset of isolated

primary hippocampal neurons grown in cell culture were

also TDF-immunopositive. WB of NG108 neuroblastoma

and GS9L astrocytoma cell lysate revealed TDF immuno-

reactivity in cultured neuroblastoma, not astrocytoma.

These data suggest that TDF is localized in neurons, not

astrocytes. This is the first report showing TDF in any cell.

TDF may have specific roles as a pituitary-derived hor-

mone and in the CNS, where it appears to be produced by

distinct CNS neurons.

Materials and methods

Generation of the TDF antibody

Three peptides from the ORF of TDF protein were synthesized

and used as antigens for anti-TDF generation. Peptides were

TDF-P1 (NH2-RESQGTRVGQALSFLCKGTA-COOH),

TDF-P2 (NH2-LLLEGPDHQYSL-COOH), and TDF-P3

(NH2-HTNTIQTQNMKH-COOH). Two rabbit polyclonal

anti-TDF antibodies were produced: anti-TDF-P1 antibody

(anti-TDF-P1-Ab) and anti-TDF-P1P2P3 antibody (anti-

TDF-P1-Ab). Creative Biolabs generated the antibodies. For

both IHC and WB, we used only anti-TDF-P1-Ab. We used

anti-TDF-P1P2P3-Ab to confirm anti-TDF-P1-Ab results.

Histology

Frozen, paraformaldehyde-fixed rat brains in optimum

cutting temperature (OCT) compound (Sakura Finetek

USA, Inc., Dublin, OH, USA) were purchased (Charles

River Laboratories, Wilmington, MA, USA) and stored at

-80 �C until sectioned (five 3-month-old female brains

and three male brains). Sections were cut sagittally

(25 microns, -25 �C) via Microm HM 500 M cryostat

(Microm International GmBH, Walldorf, Germany) into

phosphate buffer (PB, pH 7.4) for IHC. Pituitary tissue was

dissected from the brain, isolated and frozen in OCT

compound, and shipped from Charles River Laboratories.

We sectioned pituitary tissue via cryostat separately from

the rest of the brain; however, tissue preparation was

otherwise identical for all tissue sections.

IHC

Sections were processed for IHC free-floating in PB. Sec-

tions were blocked 2 h at room temperature (blocking

buffer: 10 % normal donkey serum, 2 % BSA in PB con-

taining 0.3 % Triton-X). After washing, sections were

incubated overnight at 4 �C with primary antibodies in

blocking buffer: rabbit polyclonal anti-TDF-Ab, 1:200,

mouse monoclonal anti-GFAP, 1:100 (Sigma-Aldrich, St.

Louis, MO, USA), and mouse monoclonal anti-NeuN,

1:100 (Chemicon/Millipore). Sections were rinsed and

Brain Struct Funct

123

Author's personal copy

incubated in Alexa Fluor 488 conjugated goat anti-rabbit

(1:500, Molecular Probes, Eugene, OR, USA) and Alexa-

Fluor 546 conjugated goat anti-mouse (1:500; Molecular

Probes, Eugene, OR, USA) in blocking buffer. After

washing, sections were mounted on gelatin-coated slides

with Prolong Gold anti-fade reagent with DAPI (Molecular

Probes). To remove background, some sections were also

treated with 10 mM citrate buffer pH 6.0 for 20 min at

95 �C. Controls included primary antibody omission and

pre-incubation of anti-TDF-P1-Ab or anti-TDF-P1P2P3

with antigen—TDF-P1. We also incubated anti-TDF-P1

antibodies with random peptide, Glu-1-Fibrinopeptide B

(GluFib, EGVNDNEEGFFSAR). Sections were visualized

via Nikon Eclipse TE200 inverted microscope equipped in

IPlab 4.0 and LSM 710 confocal microscope (Zeiss)

equipped with X, Y, and Z lasers. Anatomy was identified

via rat brain atlas (Kruger et al. 1995).

Cell culture

Cell lines (NG108-15 neuroblastoma 9 glioma and GS9L

astrocytoma), grown as described (Darie et al. 2011; So-

kolowska et al. 2012b, c; Spellman et al. 2008) for 24 h in

serum free medium, were then lysed, cleared, and used for

WB.

Neuronal culture

DIV78 rat hippocampal neurons were fixed in 4 % para-

formaldehyde and 20 % sucrose for 10 min, the fixative

was quenched with 50 mM NH4Cl in PBS for 5 min, and

cells were permeabilized with 0.1 % Triton-X 100 in PBS

for 2 min before blocking. Coverslips were blocked with

10 % normal goat serum, 2 % bovine serum albumin, and

0.25 % fish skin gelatin in Tris-buffered saline for 30 min,

incubated with anti-TDF diluted in blocking solution for

30 min and washed thrice with Tris-buffered saline,

0.25 % fish skin gelatin and incubated with Alexa Fluor

555-conjugated goat anti-rabbit antibody in blocking buffer

for 30 min, washed and mounted using Mowiol 488. Cells

were imaged using a LSM510 laser-scanning confocal

microscope equipped with a 40 9 Plan Neofluor NA1.3

DIC oil-immersion objective (Carl Zeiss Microimaging).

Images were processed using LSM510 software (Zeiss) and

ImageJ (NIH).

WB

WB was performed as described (Sokolowska et al. 2012b,

c). Primary antibodies were rabbit polyclonal anti-TDF-P1

and mouse monoclonal anti-a-tubulin (Sigma-Aldrich, St.

Louis, MO, USA). Secondary antibodies were goat anti-

rabbit IgG-HRP and goat anti-mouse IgG-HRP (Santa Cruz

Biotechnology, Santa Cruz, CA, USA). For antibody

specificity, anti-TDF-P1-Ab was pre-incubated with TDF-

P1 4 h, room temperature, before use.

Results

TDF lack of homology

TDF polypeptide has 108 amino acids and predicted

molecular mass of 12,704 Da. The TDF protein sequence

was found in six different databases (PRF, EMBL CDS,

IPI, TREMBLNEW, H_INV, and UniProtKB). Although

the protein is named (TDF), these sequences do not have

any names for the corresponding gene. BLAST (http://

blast.ncbi.nlm.nih.gov/Blast.cgi) search for TDF in all

databases did not find homology with another protein.

Searches of the Allen Brain (http://www.brain-map.org/)

atlas and Gensat (http://www.gensat.org) produced 0 hits.

Database search (www.expasy.ch) for post-translational

processing of TDF revealed a Histidine rich region, two

potential Protein kinase C phosphorylation sites, and two

potential myristoylation sites. No signal sequence or gly-

cosylation site was identified.

Anti-TDF-P1-Ab is specific against their native

antigens

We tested antibodies for specificity. Figure 1 shows

staining using anti-TDF-P1-Ab in cerebellum (Fig. 1a).

Pre-incubation of anti-TDF-P1-Ab with TDF-P1 followed

by IHC completely abolished and greatly reduced staining

in all brain regions examined (Fig. 1b). Pre-incubation of

anti-TDF-P1-Ab with GluFib peptide, a random peptide,

followed by IHC did not affect staining, which was iden-

tical to anti-TDF-P1-Ab alone (Fig. 1c). Staining using

anti-TDF-P1P2P3-Ab (Fig. 1d) produced an identical pat-

tern as anti-TDF-P1-Ab (Fig. 1a). IHC using an identical

protocol omitting anti-TDF-P1 primary antibody produced

no specific staining (Fig. 1e), as did omission only of

secondary antibody (Fig. 1f). Incubation of anti-TDF-P1-

Ab with TDF-P1 also abolished WB results (Fig. 1g).

These data suggest that anti-TDF-Abs are highly specific.

TDF is present in select neurons, with a specific

distribution in the brain

IHC using anti-TDF-P1-Ab resulted in consistent staining

patterns throughout brain for all eight brains. No differ-

ences in staining patterns occurred between female and

male brains. Table 1 shows the overall distribution of

staining throughout brain, including relative intensity of

staining in somata and in fibers. In pituitary TDF stained

Brain Struct Funct

123

Author's personal copy

cell bodies (Fig. 2a), however, double-staining with DAPI

revealed that not all cells were TDF-immunopositive

(Fig. 2a). In cerebellum, TDF staining occurred in the

Purkinje cells with robust staining in cell bodies (Fig. 2b).

Intense staining was observed in Purkinje cell dendrites.

Staining of apparent processes but only occasional cell

bodies occurred in cerebellar granule cells and staining of

processes appeared within cerebellar white matter

(Fig. 2b). In hippocampus, large cell bodies predominantly

within the hilus were TDF-immunopositive (Fig. 2c).

Processes emerging from these cell bodies and traversing

through the granule cell and molecular layers were also

TDF-immunopositive (Fig. 2c). In addition, occasional

large cell bodies located within the hippocampal granule

cell layer or molecular layers and their processes also

stained positively using TDF IHC (Fig. 2c, f). Notably, all

small cells of the granule cell layer were identifiable with

DAPI, but these were never TDF-immunopositive

(Fig. 2c). Occasional large cell bodies within hippocampal

CA3 and CA1 also stained using anti-TDF (not shown). In

caudoputamen, primarily processes, but some cell bodies

were TDF-immunopositive (Fig. 2d). Select neurons with

pyramidal cell morphology and processes stained in cere-

bral cortex (Fig. 2e).

TDF co-localizes with a neuronal marker,

not with an astroglial marker

To investigate whether TDF is localized in astrocytes or

neurons, we performed TDF-GFAP double-staining for

localization of TDF in astrocytes and a TDF-NeuN double-

staining for localization of TDF in neurons.

TDF-GFAP double IHC showed no co-localization of

these two proteins in any region examined (Fig. 3a–d). For

example, GFAP IHC did not correspond with TDF-im-

munopositivity in pituitary (Fig. 3a), cerebellum (Fig. 3b),

or hippocampus (Fig. 3c, d). In contrast, when we exam-

ined the TDF—NeuN double IHC, TDF antibody almost

always co-localized with NeuN antibody (Fig. 3e–h). The

only exception was in Purkinje cells of the cerebellum,

which did not stain for NeuN, as previously described

(Weyer and Schilling 2003), but did stain positively with

anti-TDFP1 (Fig. 3f).

+ TDF Ab- TDF-P1

+ TDF Ab+ TDF-P1

A B

C D

pclpcl

pclpcl

gclgcl

gcl

gcl

ml

ml

ml

ml

G

E F

gcl gcl

ml ml

pcl pcl

Fig. 1 TDF antibodies are specific. IHC of brain and pituitary using

anti-TDF-Ab (green) counter stained with DAPI (blue). a Represen-

tative cerebellar staining. Purkinje cell bodies and dendrites stain

largely abolished with TDFP1 pre-incubation (b), not GluFib pre-

incubation (c). Anti-TDF-P1P2P3-Ab demonstrates similar staining

(d). Omission of primary (e) or secondary (f) antibodies produces no

cerebellar staining. Scale bar 100 lM. g WB for anti-TDF-P1

specificity. NG108 neuroblastoma and TDF-P1 peptide were analyzed

by WB using anti-TDF-P1-Ab or anti-TDF-P1-Ab pre-incubated with

TDF-P1 peptide. After incubation with secondary antibody, the

reaction was visualized by ECL. TDF protein in NG108 cell lysate

appears as two bands (about 50 and 80 kDa), likely monomers and

dimers. TDF-P1 peptide was identified at 3 kDa, close to expected

molecular mass. Molecular mass markers shown in kDa

Brain Struct Funct

123

Author's personal copy

Not all NeuN-immunopositive neurons stained TDF

immunopositively

Not all NeuN-immunopositive neurons are positive for

TDF. For example, in striatal fundus (Fig. 3e) some TDF

and NeuN IHC co-localized; however, in some instances

only NeuN staining occurred. In cerebellum, NeuN staining

was most intense within the cell-dense granule cells

(Fig. 3f). Apparent processes and occasional small cell

bodies were TDF-immunopositive in this region (Fig. 3f).

In the hippocampus, we observed that the large TDF-im-

munopositive cells in the hilus are also NeuN positive

(Fig. 3g). However, in hippocampus, other NeuN-immu-

nopositive cell bodies, such as the small cell bodies within

the granule cell layer (not shown) and pyramidal cell layers

(Fig. 3g) were NeuN-immunopositive, not TDF-immuno-

positive. In cerebral cortex, cells with pyramidal cell

morphology stained using both anti-TDF and anti-NeuN;

however, other NeuN-immunopositive cells failed to also

stain TDF-immunopositive (Fig. 3h). We also isolated

hippocampal neurons and immunostained them with anti-

TDF-Ab (Fig. 4). Again, staining was visible in a subset of

neurons within the culture and was prominent in axons and

cell bodies, to a lesser degree along dendrites, and in some

cases was also found in the nucleus. Staining appeared to

be enriched in the neurites of primary hippocampal

neurons.

TDF is present in neuron-like NG108 neuroblastoma

cells, but not in astrocyte-like GS9L astrocytoma cells

We used WB to test for TDF protein in lysate from NG108

mouse neuroblastoma 9 rat glioma cell line as neuron

model system and GS9L astrocytoma cell line as an

astrocyte model system. As shown in Fig. 5a, TDF protein

was detected by anti-TDF-Ab in the lysate from NG108

neuroblastoma cells, but not from GS9L astrocytoma cells.

In NG108 cell lysate, TDF protein occurred as two bands

(80 and 50 kDa), most likely monomer and dimer, also

observed in other WB experiments. The apparent mass of

TDF (50 kDa) differs from its calculated mass (16 kDa),

suggesting that it is intensely post-translationally modified.

The intensity of TDF bands in WB varied in different

experiments. Anti-a-tubulin-Abs were used to evaluate

levels of protein loaded per gel lane on the same blot

membrane. These data suggest that TDF protein is pro-

duced in the neuron-like NG108 cells, not in GS9L astro-

cyte-like cells.

Discussion

This is the first cellular localization of TDF—a protein

lacking homology to other known proteins—in brain using

IHC. This is also the first cellular localization for TDF

protein demonstrated in any organ, confirming the RNA

data which show that TDF is produced by the pituitary

gland and in the brain (Platica et al. 2004). We found TDF

immunoreactivity in several regions of brain, in neurons,

and their processes. Specifically, cerebellar Purkinje

cells, large-soma neurons in the hilus, as well as sparse,

Table 1 Overall distribution of TDF staining throughout brain

Brain region Intensity of staining

Somata Fibers

Cerebral cortex

Layer I–III – –

Layer IV–V ??? ???

Layer VI – –

Hippocampal formation

Granule cell layer – –

Subgranular layer ??? ??

Polymorph layer (hilus) ? ??

Subiculum ?? ??

Entorhinal cortex ?? ?

Inner molecular layer ? ?

Outer molecular layer – –

CA1 pyramidal layer – –

CA3 pyramidal layer – –

CA1 stratum lacunosum ? ??

CA3 stratum lacunosum ?? ??

Thalamus

External medullary lamina – –

Posterior group of the thalamus ? ?

Medial geniculate nucleus ?? ??

Lateral posterior nucleus ? –

Lateral dorsal nucleus ? –

Ventrobasal nuclear complex ?? –

Basal ganglia

Caudoputamen ? ?

Striatum ?? ???

Globus pallidus ? –

Brainstem

Superior colliculus ?? ??

Inferior colliculus ? –

Cerebellum

Molecular layer – ???

Purkinje cell layer ??? ???

Granular cell layer ? ?

Deep cerebellar nuclei – –

Olfactory cortex ?? ??

Piriform cortex – ?

Pituitary ??? –

Intensity of staining: ???, strong; ??, moderate; ?, weak; –, none

Brain Struct Funct

123

Author's personal copy

large-soma cells in granule cell region and pyramidal cell

region of the hippocampus, individual cells in the striatal

fundus and caudoputamen as well as pituitary gland cells

and apparent pyramidal neurons in the cerebral cortex.

TDF immunoreactivity was never observed in association

with anti-GFAP-positive astrocytes, but was consistently

associated with a subset of anti-NeuN-positive neurons and

the processes emitting from these cells. In support of

neuronal, but not astroglial localization, we also observed

that TDF protein is present in cultured neuroblastoma, but

not astrocytoma.

We tested the specificity of our antibody using several

types of controls, including incubation with the TDFP1

peptide versus a random peptide, comparison with a dif-

ferent antibody generated against a different portion of

TDF, and omission of the primary and secondary anti-

bodies. These methods have been established as valid

controls for measuring antibody specificity (Burry 2011).

As additional evidence, incubation with citrate treatment

did not abolish or reduce TDF staining, and this is known

to reduce non-specific staining and, indeed, reduced non-

specific staining for our other antibodies. Citrate reduces

background and unmasks antigenic sites (Shi et al. 1993).

Our database searches indicate a lack of homology of

TDF to other known proteins. We have further determined

that previous localization of TDF in the nervous system has

not been performed. This indicates that this protein may

have distinct, unexplored roles in the CNS. It is of partic-

ular interest to note that TDF is present in neurons, not

astroglia, and is further found in neuronal processes.

Presence in the processes may indicate that TDF serves a

structural role, stabilizing the cytoskeleton. It should also

be noted that TDF protein appeared to be predominantly

located in dendritic processes. Dendritic proteins can play a

role in increasing synaptic strength or efficacy (Sutton and

Schuman 2006), suggesting that TDF may be important for

establishing synapses in specific neuronal systems.

The presence of TDF protein in neuronal processes

additionally brings up the question of whether or not this

protein is released from neurons or acts intrinsically. In the

present study, we found that neuroblastoma produces TDF,

but also found that TDF lacks a signal sequence for

secretion. This suggests the possibility that TDF acts as

only an autocrine factor within the same neuron that pro-

duces it, but is not released to affect neighboring cells. It

should be noted, however, that despite lacking a conven-

tional signal sequence, other unconventional secretion

mechanisms have been identified for other proteins

(Glebov and Walter 2012). In addition, we have looked at

both the putative TDF-R and for the presence of TDF

protein, in neuroblastoma, not neurons. Although these

cells are not neurons or astroglia, they do share some

A pit

cp

B C

hil

gcl

ml

D E ctx F

gcl

pcl

ml

gcl

V

Fig. 2 TDF is found in pituitary and in other brain regions. IHC of

brain and pituitary tissues using anti-TDF-Ab (green) counter stained

with DAPI (blue). a Staining in pituitary. Some DAPI? cells were

TDF-immunopositive, others were not. b Cerebellum. TDF immuno-

reactivity in Purkinje cell bodies and dendrites. TDF immunoreac-

tivity is only occasionally seen in the granule cell layer.

c Hippocampus. Large cell bodies and processes stain in hilus,

traversing into molecular layer. Molecular layer cell bodies were also

TDF-immunopositive. d In caudoputamen, occasional cell bodies but

primarily processes are TDF-immunopositive. e Cortex, large cell

bodies with pyramidal cell morphology and processes are TDF-

immunopositive. Many DAPI? cells do not possess TDF immuno-

reactivity. f Within hippocampal granule cell layer, a single large

neuron and its processes stains. TDF immunoreactivity is primarily in

processes in this region. Scale bar 50 lm. pit pituitary, ml molecular

layer, pcl Purkinje cell layer, gcl granule cell layer, hil hilus, cpcaudoputamen, ctx cortex

Brain Struct Funct

123

Author's personal copy

characteristics with neurons and astrocytes and are fre-

quently used as model systems (Kenney-Herbert et al.

2011). They may, however, not share all characteristics

with neurons and astroglia.

It should be noted that in the pituitary gland, GFAP

labels a unique cell type known as the pituicyte in both the

anterior and posterior pituitary (Redecker and Fechner

1989). These cells are believed to be resident astrocytes of

the pituitary, but not hormone-secreting cells (Hatton

1988). We never observed GFAP and TDF co-localization

in the pituitary. NeuN staining in pituitary was non-exis-

tent, consistent with previous published observations (Wolf

et al. 1996). TDF is therefore produced in the pituitary also

by select cells and specifically not by pituicytes. Based on

the size of TDF and its similarity in size to an anterior

pituitary-produced hormone, we might therefore expect it

to be produced by this region of pituitary and by hormone-

secreting cells.

The role of TDF in the CNS is currently unknown;

however, clues may be gained based on the predominant

localization of TDF in specific neuronal populations, for

example by cerebellar Purkinje cells and hippocampal hilar

cells as well as other cells with the appearance of hippo-

campal interneurons. Notably, these cell populations pro-

duce the neurotransmitter GABA (Sastry et al. 1997),

suggesting that TDF may be predominantly produced by

GABAergic neurons. Subsets of these specific brain cells

seem to be more vulnerable in certain neurological con-

ditions. For example, Purkinje cell number is reduced in

autism (Bauman and Kemper 2005) and hippocampal hilar

neurons are reduced in temporal lobe epilepsy (Sloviter

1996). Based on its localization to these vulnerable cell

populations, it is conceivable that TDF may serve a neu-

roprotective role in neurons that are more susceptible to

injuries. Further studies in neurodegenerative disease

models could assess whether or not this factor is indeed

neuroprotective. Further investigation of specific cell types

producing TDF (such as in the hippocampus) may shed

light on the possible functional role of this molecule in the

brain.

Our previous work suggests the TDF receptor may be a

HSP. Since our studies identifying TDF-R were conducted

in cultured breast and prostate cells, the concept that TDF

may act on a HSP as its receptor in the CNS is hypothetical

at this point. It is possible that TDF acts on a HSP receptor

globally in the body or serves a distinct function via a

different cellular receptor in the CNS. It should be noted

that HSPs do have distinct roles in the CNS for buffering

against insults and providing neuroprotection. For example,

HSPs have been identified as putative biomarkers for CNS

ischemia and hypoxia. HSP expression increases in

response to these insults and mediates cellular protection

A B C

CA1

hil

gcl

D

stf E F

CA3

hil

G H ctx

pcl

ml

gcl

pit

pcl gcl

pyr

or

wm

pyr

Fig. 3 TDF is found in neurons but not in astroglia. Immuno-staining

of brain and pituitary using double-staining TDF and GFAP co-

localization (a–d) or of TDF and NeuN (e–h). a GFAP-immunopos-

itive (?) cells (red) do not co-localize with TDF-immunopositive (?)

cells (green) in pituitary (small arrowhead anti-TDF? alone, largearrowhead anti-GFAP? alone, a–d). GFAP-immunopositive astro-

cytes (red) do not co-localize with TDF-immunopositive cells (green)

in cerebellum (b), hippocampal dentate gyrus (c) or hippocampal

hilus (d). e TDF immunoreactive cell bodies co-localize with NeuN,

however, all anti-NeuN? neurons are not anti-TDF? (large arrow-heads NeuN alone, triangular arrowheads NeuN co-localized with

TDF, e–h). f In cerebellum, TDF alone appears in the Purkinje cell

layer (small arrowhead). Some small cells in the granule cell layer

and white matter are anti-TDF? and co-localize with NeuN. White

matter processes are also anti-TDF?. g Pyramidal cells in hippo-

campal CA3 are anti-NeuN? but not anti-TDF?. In surrounding

hilus, cells are anti-NeuN? and anti-TDF?. h Neurons with

pyramidal cell morphology are both anti-NeuN? and anti-TDF?,

processes for these cells have only TDF immunoreactivity. Some cells

are only anti-NeuN?. pit pituitary, ml molecular layer, pcl Purkinje

cell layer, gcl granule cell layer, or stratum oriens of hippocampus,

pyr pyramidal cells, CA3 Cornus Ammonus 3 of hippocampus, hilhilus, stf striatal fundus, wm white matter, ctx cortex. Scalebar 50 lM

Brain Struct Funct

123

Author's personal copy

(Hecker and McGarvey 2011). HSPs, including HSP70,

can be induced in both neurons and in glia following sei-

zures, hypothermia, and ischemia and inducing HSP

expression has been suggested as a therapeutic target for a

variety of neurodegenerative disorders (van Noort 2008).

HSPs are also potentially protective against several neu-

rodegenerative disorders characterized by aberrant protein

folding, such as Alzheimer’s disease, Parkinson’s disease,

and amyotrophic lateral sclerosis (Chen and Brown 2007).

The role of TDF in the CNS is highly speculative at this

point and needs to be further elucidated. This initial

localization of an unexplored factor to select neurons in the

CNS adds an intriguing additional factor to the list of those

that are present in and that affect the nervous system.

Comprehensive characterization of TDF protein and

mRNA localization in the mammalian nervous system and

further inquiry into the potential role of TDF in the normal

and injured CNS are needed and warrants further

investigation.

Fig. 4 TDF is found in cultured

hippocampal neurons. Shown

are two examples of mature

primary hippocampal cultures

stained with anti-TDF-Abs.

Bottom panel shows a higher

magnification of the boxed

region. Scale bar 50 lm

kDa130

95

72

43

kDa

56

α-tubulin

TDF

Fig. 5 TDF is found in NG108 neuroblastoma but not in GS9L

astrocytoma cell lysate. WB for TDF protein in NG108 neuroblas-

toma and GS9L astrocytoma cell lysate. Two bands appear (monomer

and dimer). The blot was stripped and re-probed with anti-a-tubulin.

Molecular mass markers indicated in kDa

Brain Struct Funct

123

Author's personal copy

Acknowledgments We thank Dr. Kenneth Wallace (Clarkson

University) for advice on immunohistochemistry, as well as Dr. Ivan

Soltesz (UC Irvine) and Dr. Moses Chao (NYU) for helpful discus-

sions. We also thank Dr. Thomas A. Neubert (Skirball Institute, New

York University, New York, NY) and Ms. Jill Pflugheber (ST.

Lawrence University, Canton, NY) for providing the NG108 neuro-

blastoma and GS9L astrocytoma cell lines. This work was supported

in part by U.S. Army research office (DURIP grant #W911NF-11-1-

0304).

Conflict of interest The authors have no conflicts of interest to

disclose.

References

Bauman ML, Kemper TL (2005) Neuroanatomic observations of the

brain in autism: a review and future directions. Int J Dev

Neurosci 23(2–3):183–187

Burry RW (2011) Controls for immunocytochemistry: an update.

J Histochem Cytochem 59(1):6–12

Caldwell HK Young WS 3rd (2006) Oxytocin and vasopressin:

genetics and behavioral implications (1st ed, vol 40). Springer,

Berlin pp 573–607

Chen S, Brown IR (2007) Neuronal expression of constitutive heat

shock proteins: implications for neurodegenerative diseases. Cell

Stress Chaperones 12(1):51–58

Dallner C, Woods AG, Deller T, Kirsch M, Hofmann HD (2002)

CNTF and CNTF receptor alpha are constitutively expressed by

astrocytes in the mouse brain. Glia 37(4):374–378

Darie CC, Deinhardt K, Zhang G, Cardasis HS, Chao MV, Neubert

TA (2011) Identifying transient protein–protein interactions in

EphB2 signaling by blue native PAGE and mass spectrometry.

Proteomics 11(23):4514–4528

Friedrich N, Nauck M, Schipf S, Volzke H, Brabant G, Wallaschofski

H (2013) Cross-sectional and longitudinal associations between

insulin-like growth factor I and the metabolic syndrome: a

general population study in German Adults. Diabetes Metab Res

Rev (epub ahead of print)

Glebov K, Walter J (2012) Statins in unconventional secretion of

insulin-degrading enzyme and degradation of the amyloid-beta

peptide. Neurodegener Dis 10(1–4):309–312

Goldenberg-Cohen N, Raiter A, Gaydar V, Dratviman-Storobinsky O,

Goldstein T, Weizman A et al. (2011) Peptide-binding GRP78

protects neurons from hypoxia-induced apoptosis. Apoptosis.

doi:10.1007/s10495-011-0678-x

Guthrie KM, Woods AG, Nguyen T, Gall CM (1997) Astroglial

ciliary neurotrophic factor mRNA expression is increased in

fields of axonal sprouting in deafferented hippocampus. J Comp

Neurol 386(1):137–148

Harvey S (2010) Extrapituitary growth hormone. Endocrine 38(3):

335–359

Hatton GI (1988) Pituicytes, glia and control of terminal secretion.

J Exp Biol 139:67–79

Hecker JG, McGarvey M (2011) Heat shock proteins as biomarkers

for the rapid detection of brain and spinal cord ischemia: a

review and comparison to other methods of detection in thoracic

aneurysm repair. Cell Stress Chaperones 16(2):119–131

Kenney-Herbert EM, Ball SL, Al-Mayhani TM, Watts C (2011)

Glioblastoma cell lines derived under serum-free conditions can

be used as an in vitro model system to evaluate therapeutic

response. Cancer Lett 305(1):50–57

Kim N, Kim JY, Yenari MA (2012) Anti-inflammatory properties and

pharmacological induction of Hsp70 after brain injury. Inflam-

mopharmacology. doi:10.1007/s10787-011-0115-3

Kruger X, Saporta Y, Swanson Z (1995). The cell and fiber

architecture illustrated in three planes with stereotaxic coordi-

nates Photographic atlas of the rat brain (1st edn) Cambridge

University Press, Cambridge

Platica M, Chen HZ, Ciurea D, Gil J, Mandeli J, Hollander VP (1992)

Pituitary extract causes aggregation and differentiation of rat

mammary tumor MTW9/Pl cells. Endocrinology 131(6):

2573–2580

Platica M, Ivan E, Holland JF, Ionescu A, Chen S, Mandeli J et al

(2004) A pituitary gene encodes a protein that produces

differentiation of breast and prostate cancer cells. Proc Natl

Acad Sci USA 101(6):1560–1565

Redecker P, Fechner J (1989) Immunohistochemical study of cells

positive for glial fibrillary acidic protein (GFAP) in the human

pituitary gland, with special reference to folliculo-stellate cells.

Histochemistry 91(3):227–234

Rezende LF, Santos GJ, Santos-Silva JC, Carneiro EM, Boschero AC

(2012) Ciliary neurotrophic factor (CNTF) protects non-obese

Swiss mice against type 2 diabetes by increasing beta cell mass

and reducing insulin clearance. Diabetologia 55(5):1495–1504

Roy U, Sokolowska I, Woods AG, Darie CC (2012) Structural

investigation of tumor differentiation factor (TDF). Biotechnol

Appl Biochem 59:445–450

Sastry BR, Morishita W, Yip S, Shew T (1997) GABA-ergic

transmission in deep cerebellar nuclei. Prog Neurobiol 53(2):

259–271

Shi SR, Chaiwun B, Young L, Cote RJ, Taylor CR (1993) Antigen

retrieval technique utilizing citrate buffer or urea solution for

immunohistochemical demonstration of androgen receptor in

formalin-fixed paraffin sections. J Histochem Cytochem 41(11):

1599–1604

Sloviter RS (1996) Hippocampal pathology and pathophysiology in

temporal lobe epilepsy. Neurologia 11(Suppl 4):29–32

Sokolowska I, Woods AG, Wagner J, Dorler J, Wormwood K, Thome

J et al (2011) Mass spectrometry for proteomics-based investi-

gation of oxidative stress and heat shock proteins. In: Andreescu

S, Hepel M (eds) Oxidative stress: diagnostics, prevention, and

therapy. American Chemical Society, Washington

Sokolowska I, Woods AG, Gawinowicz MA, Roy U, Darie CC

(2012a) Characterization of tumor differentiation factor (TDF)

and its receptor (TDF-R). Cell Mol Life Sci (epub ahead of print)

Sokolowska I, Woods AG, Gawinowicz MA, Roy U, Darie CC

(2012b) Identification of a potential tumor differentiation factor

receptor candidate in prostate cancer cells. FEBS J 279(14):

2579–2594

Sokolowska I, Woods AG, Gawinowicz MA, Roy U, Darie CC

(2012c) Identification of potential tumor differentiation factor

(TDF) Receptor from steroid-responsive and steroid-resistant

breast cancer cells. J Biol Chem 287(3):1719–1733

Spellman DS, Deinhardt K, Darie CC, Chao MV, Neubert TA (2008)

Stable isotopic labeling by amino acids in cultured primary

neurons: application to brain-derived neurotrophic factor-depen-

dent phosphotyrosine-associated signaling. Mol Cell Proteomics

7(6):1067–1076

Sutton MA, Schuman EM (2006) Dendritic protein synthesis,

synaptic plasticity, and memory. Cell 127(1):49–58

van Noort JM (2008) Stress proteins in CNS inflammation. J Pathol

214(2):267–275

Weyer A, Schilling K (2003) Developmental and cell type-specific

expression of the neuronal marker NeuN in the murine

cerebellum. J Neurosci Res 73(3):400–409

Brain Struct Funct

123

Author's personal copy

Wiese S, Karus M, Faissner A (2012) Astrocytes as a source for

extracellular matrix molecules and cytokines. Front Pharmacol

3:120

Wolf HK, Buslei R, Schmidt-Kastner R, Schmidt-Kastner PK, Pietsch

T, Wiestler OD et al (1996) NeuN: a useful neuronal marker for

diagnostic histopathology. J Histochem Cytochem 44(10):

1167–1171

Woods AG, Guthrie KM, Kurlawalla MA, Gall CM (1998) Deaffer-

entation-induced increases in hippocampal insulin-like growth

factor-1 messenger RNA expression are severely attenuated in

middle aged and aged rats. Neuroscience 83(3):663–668

Woods AG, Poulsen FR, Gall CM (1999) Dexamethasone selectively

suppresses microglial trophic responses to hippocampal deaffer-

entation. Neuroscience 91(4):1277–1289

Brain Struct Funct

123

Author's personal copy