Induction of translationally controlled tumor protein
(TCTP) by transcriptional and post-transcriptional
mechanisms
Irina Schmidt, Michael Fa
¨
hling, Benno Nafz, Angela Skalweit and Bernd-Joachim Thiele
Charite
´
, Universita
¨
tsmedizin Berlin, Institut fu
¨
r Vegetative Physiologie, Germany
Translationally controlled tumor protein (TCTP) is a
highly conserved 19 kDa GDP ⁄ GTP exchange factor
[1,2] that is involved in control of cell growth and pro-
liferation by regulating the activity of Rheb (Ras
homolog enriched in brain), a Ras superfamily GTPase
[3]. It has been identified in a broad spectrum of
eukaryotic organisms, from plants, yeasts, fungi, para-
sites, insects, fishes, and birds up to mammals, and is
involved in a wide range of cellular processes [4].
TCTP attracted interest mainly as an antiapoptotic
factor [5–7], for its role in tumorigenesis [8,9], and as a
tubulin-binding [10] or Ca
2+
-binding protein [11,12].
Moreover, it has other, extracellular, functions as a
histamine-releasing factor and cytokine [13,14].
TCTP synthesis is extensively regulated at the tran-
scriptional and post-transcriptional levels [4,15,16]. It is

tion of the interaction of RNA-binding proteins with UTRs by UV-cross-
linking. PMA, forskolin, dioxin, cobalt and nickel induced TCTP
expression in 24 h in both cell lines about 2.2–3.2-fold at the mRNA level
and 1.6–2.2-fold at the protein level. The highest induction rate, 4.5–5.0-
fold at the mRNA level and 3.5–4.0-fold at the protein level, was observed
with copper. TPT1 promoter assays showed transcriptional activation by
PMA, forskolin and dioxin (2.0–3.1-fold) and a 7.0–8.0-fold increase by
copper, whereas cobalt and nickel had no effect. Deletion analysis revealed
that copper-dependent transcriptional control was transmitted by a metal-
responsive element residing in the TPT1 promoter. Post-transcriptional
activation of TCTP expression was associated with the action of dioxin,
nickel, cobalt (1.8–2.3-fold) and copper (2.5–3.0-fold), whereas stimulation
of TCTP synthesis by copper was mediated by the TCTP mRNA 3¢-UTR
(3.2-fold) but not by the 5¢-UTR (0.5-fold). mRNA stabilization was found
to mediate these effects of cobalt and nickel. Post-transcriptional regulation
was associated with qualitative and quantitative changes in the binding of
specific RNA-binding proteins to UTRs.
Abbreviations
ARE, AU-rich element; CREB, cAMP-responsive element-binding protein; HIF, hypoxia inducible factor; MRE, metal-responsive element;
PMA, 4b-phorbol 12-myristate 13-acetate; RNA-BP, RNA-binding protein; TCTP, translationally controlled tumor protein; TPT1, gene coding
for human TCTP.
5416 FEBS Journal 274 (2007) 5416–5424 ª 2007 The Authors Journal compilation ª 2007 FEBS
brain, but at high levels in tissues undergoing active cell
division and in tumor cells [8,9,17]. A wide variety of
chemicals are able to induce TCTP synthesis, including
substances such as phorbol esters and lipopolysaccha-
rides [18], cytotoxic drugs [19], Ca
2+
[11], heavy metals
[20], and dioxins [21]. However, less is known about

assayed. For characterization of mRNA-related regula-
tive capacity, 5¢- and 3 ¢-UTRs of TCTP mRNA were
cloned upstream and downstream of the luciferase gene
containing a constitutive promoter. Figures 1 and 2
show TCTP mRNA and protein quantification data,
demonstrating the way in which TCTP expression
responded to the six selected substances. It is evi-
dent that PMA, forskolin, dioxin, cobalt and nickel
were able to induce TCTP in both cell lines at the
mRNA and protein levels by 2.2–3.2-fold (mRNA) and
1.6–2.2-fold (protein) in 24 h. The highest induction
rate was observed with copper, i.e. 4.5–5.0-fold
(mRNA) and 3.5–4.0-fold (protein).
Figure 3 depicts TPT1 promoter ⁄ luciferase reporter
gene assays. It is obvious that the transcriptional activ-
ity of the TPT1 promoter has only a partial similarity
to TCTP mRNA and TCTP protein expression. In
both cell lines, PMA, forskolin and dioxin stimulated
transcription about 2.0–3.1-fold. Only copper stimu-
lated the luciferase reporter gene at a high rate
between 7.0-fold and 8.0-fold, whereas nickel and
cobalt did not stimulate transcription; instead, they
reduced luciferase expression slightly. In Fig. 4, the
influence of potential TCTP stimulators on luciferase
expression is shown, with the luciferase gene put under
control of TCTP mRNA UTRs. A moderate post-
transcriptional stimulatory effect was seen with dioxin,
nickel and cobalt (1.8–2.3-fold), and again the highest
rate was seen with copper (2.5–3.0-fold).
The finding that copper ions are able to stimulate

FEBS Journal 274 (2007) 5416–5424 ª 2007 The Authors Journal compilation ª 2007 FEBS 5417
same type of luciferase reporter gene experiment as
described in Fig. 3. The deletion of the putative MRE
resulted in a nearly complete loss of the ability of the
TPT1 promoter to respond to copper ions. The resid-
ual luciferase activity dropped to about 10% as com-
pared to the construct with an intact MRE (Fig. 5A).
As demonstrated in Fig. 4, UTRs of mRNA are
targets mediating post-transcriptional regulation of
TCTP expression. To discriminate between 5¢- and
3¢-UTR-mediated actions, we created luciferase con-
structs that were separately under control of the
TCTP 5¢-or3¢ -UTR and performed copper-depen-
dent transfection ⁄ reporter gene assays (Fig. 5B). The
results show that the post-transcriptional copper
stimulation of TCTP expression was associated exclu-
sively with the 3¢-UTR (3.2-fold stimulation). In con-
trast, the 5¢-UTR had an inhibitory effect of about
50%, and the 5¢⁄3¢-UTR combination resulted in
intermediate values (about 2.5-fold stimulation).
Nickel and cobalt seem to stimulate TCTP expres-
sion, not by transcriptional activation but solely by
mRNA-targeted post-transcriptional processes. The
results shown in Fig. 1 demonstrate that the action of
cobalt and nickel led to an increase in mRNA concen-
tration, but not to activation of the promoter (Fig. 3).
This could imply an influence on mRNA stability. To
clarify this, we inhibited transcription with actinomy-
cin D and followed the kinetics of TCTP mRNA
decay (Fig. 6). Interestingly, inhibition of transcription

luciferase gene in the vector pGLbasic. Plasmids were transfected
into Calu-6 and Cos-7 cells in the presence or absence of TCTP-
inducing substances. For normalization of transfection efficiency,
cells were cotransfected with Renilla luciferase. Luciferase assays
were performed as described in Experimental procedures (n ¼ 6).
Fig. 4. Effect of TCTP mRNA 5¢- and 3¢-UTR on luciferase reporter
gene expression in the presence of TCTP-inducing substances.The
complete 5¢- and 3¢-UTR of human TCTP mRNA were cloned
upstream and downstream of the luciferase coding sequence into
the pGL3promoter vector containing a constitutive SV40 promoter,
thereby replacing luciferase UTRs. Cells were transfected in the
presence or absence of TCTP-inducing substances, and luciferase
activity was assayed as described in Fig. 3 (n ¼ 6).
TCTP induction I. Schmidt et al.
5418 FEBS Journal 274 (2007) 5416–5424 ª 2007 The Authors Journal compilation ª 2007 FEBS
RNA–protein complexes with the electromobility shift
assay; however, significant differences in the electro-
phoretic migration behavior of the complexes were
observed neither with control nor with effector-treated
cell extracts (not shown). To detect more subtle differ-
ences in the composition of RNA–protein complexes,
we performed UV-crosslinking experiments (Fig. 7).
The results can be summarized as follows: (a) the pat-
terns of protein binding to the 5¢-UTR or 3¢-UTR dif-
fer significantly, although some proteins seem to bind
to both; (b) the pattern of 5¢-UTR and 3¢-UTR protein
binding of PMA-treated cell extracts does not differ as
compared to the control; (c) dioxin-, nickel- and
cobalt-treated extracts show qualitative (appearance of
new bands and suppression of control bands) and

(HIF), thereby simulating hypoxic conditions [23–25].
Our experiments demonstrate that cobalt and nickel do
not interfere with TPT1 transcription. Moreover, no
HIF-responsive element can be detected in the TPT1
promoter. On the other hand, as shown with erythro-
poietin or tyrosine hydroxylase, improved expression
by hypoxia is partially controlled by UTRs of the
appropriate mRNAs in interaction with RNA-BPs
[26,27]. In the case of TCTP mRNA, this view is sup-
ported by our RNA–protein interaction studies
(Fig. 7). Nickel as well as cobalt treatment changes the
pattern of protein binding to the 5¢-UTR and 3¢-UTR.
This indicates that RNA-BPs may contribute to
post-transcriptional control of TCTP mRNA stability.
Interestingly, the crosslinking patterns of cobalt- and
A
B
Fig. 5. Transcriptional and post-transcriptional control of TCTP
expression by copper. (A) Copper-dependent luciferase expression
in Cos-7 cells after transfection with luciferase vector pGL3 pro-
moter modified with the TPT1 promoter. The SV40 promoter of
pGL3 has been replaced by the TPT1 promoter () 312 ⁄ )1 ) con-
taining an MRE at position ) 81 ⁄ ) 68, or using a construct in which
the MRE was deleted (– MRE). Copper-related reporter gene
expression was dependent on the presence of the MRE (+ MRE)
in the TPT1 promoter. (B) Copper-dependent luciferase expression
after transfection with vector pGL3 promoter modified with TCTP
UTRs. The Luc UTRs have been replaced by TCTP mRNA 5¢- and
3¢-UTRs (5¢⁄3¢), by TCTP mRNA 5¢-UTR (5¢), or by TCTP mRNA 3¢-
UTR. The TCTP 3¢-UTR but not the 5¢-UTR stimulated reporter gene

response. The crosslinking pattern obtained with
dioxin-treated cell extracts differs significantly from
controls and from experiments with heavy metal-trea-
ted cells (Fig. 7), suggesting the participation of indi-
vidual RNA-BPs in this response. Potential candidate
proteins and mechanistic details have yet to be investi-
gated.
The TCTP-inducing capability of copper was known
from experiments with earthworms inhabiting copper-
contaminated soils [20]. No data, however, exist from
mammals or from mammalian cells in culture. In Calu-
6 and Cos-7 cells, copper turned out to be a potent
inducing agent, comparable to the situation in earth-
worms. From the mechanistic point of view, it is of
interest that the action of copper involves both a tran-
scriptional and a post-transcriptional effect, although
the strong promoter action exceeds the influence of the
UTRs by a factor of 3 in reporter gene assays. Proto-
types of metal-inducible genes, which play an important
role in metal ion homeostasis and which are effectively
induced by transitional metals such as copper, cadmium
or zinc, are the metallothionein genes. Metal-inducible
metallothionein transcription is regulated primarily
through the interaction between MREs and metal tran-
scription factor-1 [28,29]. The MRE consensus sequence
is described as a core motif with the sequence
TGC(G ⁄ A)CNC flanked by a short GC-rich domain
[30]. This sequence shows a high degree of homology
A
B

copper response at the mRNA level is associated with
the 3¢-UTR. With regard to the mRNA sequence, it has
been known since the first cloning of human TCTP
mRNA that three AU-rich elements (AREs) of the
AUUUA type reside in the TCTP 3¢-UTR [33]. In
general, AREs are associated with mRNA stabiliza-
tion ⁄ destabilization, depending on ARE type and on
the interaction with different ARE-binding proteins
such as AUF1 or HuR [34]. At present, we can only
speculate that these AREs could be targets for TCTP
mRNA stabilization, a process possibly triggered by
copper ions. The molecular details of the nature of the
cis-elements and trans-acting factors involved (RNA-
BPs or micro RNAs) and, in particular, the signaling
pathway related to the interaction with copper ions
remain to be investigated.
Recently, the renal transcriptional response of TCTP
to another heavy metal, uranium, was investigated.
Uranyl ions induced TCTP at the mRNA level about
two-fold and at the protein level at least five-fold [35].
This suggests a transcriptional as well as a post-tran-
scriptional mechanism in the stress response to ura-
nium, as for copper.
Experimental procedures
Cell culture and RNA ⁄ protein isolation
Human fetal lung carcinoma Calu-6 cells were maintained
in MEM ⁄ Earle’s (Biochrom AG, Berlin, Germany) con-
taining 10% heat-inactivated fetal bovine serum,
1 · MEM nonessential amino acids, 1 · MEM sodium
pyruvate, 0.065% sodium hydrogen carbonate, 200 l m l-

(VWR International, Dresden,
Germany), 300 lm CoCl
2
(Sigma) or 300 lm NiCl
2
(Sigma), for the times indicated. For inhibition of transcrip-
tion, actinomycin D (MoBiTech, Goettingen, Germany)
was used at a concentration of 10 lgÆmL
)1
.
For RNA and protein isolation, cells were washed with
ice-cold NaCl ⁄ P
i
. RNA was prepared using RNA-Bee
(Biozol Diagnostica Vertrieb GmbH, Eching, Germany), in
accordance with the manufacturer’s protocol. Protein
extracts (10 000 g supernatants, S10, Sigma 3K15 centrifuge
with 12154-H rotor, Sigma Laborzentrifugen GmbH,
Fig. 7. Interaction of RNA-BPs with TCTP mRNA 5¢- and 3¢-UTR
analyzed by UV-crosslinking. The complete 5¢-UTR (134 nucleotides)
and 3¢-UTR1 (257 nucleotides) of the main TCTP mRNA were tran-
scribed in vitro in the presence of [
32
P]UTP and incubated with
cytoplasmic cell extracts (S10) of Calu-6 cells (control) or cells, in
which TCTP was induced for 24 h by PMA, dioxin, NiCl
2
, or CoCl
2
.

Dreieich, Germany) programmed with individual software
(luminoscan rii; R. Mrowka, Institute of Physiology,
Charite
´
, Berlin). The transfection with the Renilla lucifer-
ase served as control.
Molecular cloning procedures
For analysis of the TCTP promoter in luciferase reporter
gene assays, the promoterless luciferase vector pGL3basic
was used. Nucleotides ) 1 ⁄ ) 312 of the human TPT1 pro-
moter were amplified from a genomic clone [22] and
inserted into the multiple cloning region (KpnI ⁄ NphEI site)
of pGL3basic.
For analysis of TCTP UTRs, the luciferase vector
pGL3 promoter (Promega; constitutive SV40 promoter)
was modified. Therefore, the vector-specific 5¢- and
3¢-UTRs of luciferase mRNA were replaced by human
TCTP mRNA UTRs. The UTRs (5¢-UTR, 134 nucleo-
tides, and 3¢-UTR, 257 nucleotides) [16] were amplified by
PCR, and restriction sites were added by primer extension.
The 5¢-UTR of TCTP mRNA was cloned using the
pGL3p vector-specific HindIII and NcoI restriction sites,
and the 3¢-UTR using the XbaI and BamHI restriction
sites. The quality of processed vectors was confirmed by
sequencing. The resulting vector constructs expressed a
constitutively transcribed luciferase transcript with or with-
out specific TCTP UTRs.
The ) 73 ⁄ ) 80 nucleotide-deleted human TPT1 promoter
) 1 ⁄ ) 312 construct in pGL3p was obtained by PCR for
assessment of the influence of the MRE region on luciferase

reprobed with anti-b-actin (Chemicon) or anti-(glyceralde-
hyde-3-phosphate dehydrogenase) (Acris Antibodies
GmbH, Hiddenhausen, Germany) antibodies, with a sec-
ondary antibody (anti-mouse; Santa Cruz) to detect rela-
tive b-actin and glyceraldehyde-3-phosphate dehydrogenase
levels as loading control.
UV-crosslinking
UV-crosslinking experiments were performed as described
previously [36]. Briefly, the following protocol was used.
In vitro transcripts representing the 5¢-or3¢-UTR of TCTP
mRNA (cloned in TOPO II vector, transcription with T7
or SP6 polymerase) were radioactively labeled using
[
32
P]UTP[aP] (800 CiÆmmol
)1
; MP Biomedicals GmbH,
Heidelberg, Germany). For UV-crosslinking experiments 1–
2ng of [
32
P]UTP[aP]-labeled in vitro transcripts represent-
ing 200 000 c.p.m. were incubated with 40–60 lg of cyto-
solic protein extract for 15 min in a total volume of 10 lL
at room temperature in 10 mm Hepes, pH 7.2, 3 mm
MgCl
2
, 5% glycerol, 1 mm dithiothreitol, 150 mm KCl, and
2UÆlL
)1
RNaseOUT (Invitrogen, Karlsruhe, Germany), in

tor on the translation elongation factor eEF1A. Proc
Natl Acad Sci USA 100, 13892–13897.
3 Hsu YC, Chern JJ, Cai Y, Liu M & Choi KW (2007)
Drosophila TCTP is essential for growth and prolifera-
tion through regulation of dRheb GTPase. Nature 445,
785–788.
4 Bommer UA & Thiele BJ (2004) The translationally
controlled tumour protein (TCTP). Int J Biochem Cell
Biol 36, 379–385.
5 Li F, Zhang D & Fujise K (2001) Characterization of
fortilin, a novel antiapoptotic protein. J Biol Chem 276,
47542–47549.
6 Liu H, Peng HW, Cheng YS, Yuan HS & Yang-Yen HF
(2005) Stabilization and enhancement of the antiapop-
totic activity of Mcl-1 by TCTP. Mol Cell Biol 25, 3117–
3126.
7 Rinnerthaler M, Jarolim S, Heeren G, Palle E, Perju S,
Klinger H, Bogengruber E, Madeo F, Braun RJ, Brei-
tenbach-Koller L et al. (2006) MMI1 (YKLO56c,
TMA19), the yeast orthologue of the translationally
controlled tumor protein (TCTP) has apoptotic func-
tions and interacts with both microtubules and mito-
chondria. Biochim Biophys Acta 1757, 631–638.
8 Chung S, Kim M, Choi W, Chung J & Lee K (2000)
Expression of translationally controlled tumor protein
mRNA in human colon cancer. Cancer Lett 156, 185–
190.
9 Tuynder M, Susini L, Prieur S, Besse S, Fiucci G,
Amson R & Telerman A (2003) Biological models and
genes of tumor reversion: cellular programming through

478–496.
16 Thiele H, Berger M, Lenzner C, Ku
¨
hn H & Thiele BJ
(1998) The promoter structure and complete sequence
of the rabbit gene coding for the translationally con-
trolled tumor protein (TCTP) P23. Eur J Biochem 257,
62–68.
17 Thiele H, Berger M & Thiele BJ (2000) Expression of
the gene and processed pseudogenes coding for the
human and rabbit translationally controlled tumour
protein (TCTP). Eur J Biochem 267, 5473–5481.
18 Walsh BJ, Gooley AA, Williams KL & Breit SN
(1995) Identification of macrophage activation
association proteins by two-dimensional gel electro-
phoresis and microsequencing. J Leukocyte Biol 57,
507–512.
19 Sinha P, Kohl S, Hu
¨
tter G, Kern M, Ko
¨
ttgen E,
Dietel M, Lage H, Schno
¨
lzer M & Schadendorf D
(2000) Identification of novel proteins associated with
the development of chemoresistance in malignant
melanoma using two-dimensional electrophoresis.
Electrophoresis 21, 3048–3057.
20 Sturzenbaum SR, Kille P & Morgan AJ (1998) Identifi-

26 Paulding WR & Czyzyk-Krzeska MF (1999) Regulation
of tyrosine hydroxylase mRNA stability by protein-
binding, pyrimidine-rich sequence in the 3¢-untranslated
region. J Biol Chem 274, 2532–2538.
27 Czyzyk-Krzeska MF & Bendixen AC (1999) Identifica-
tion of poly(C) binding protein in the complex associ-
ated with 3¢ untranslated region of erythropoietin
messenger RNA. Blood 93, 2111–2120.
28 Radtke F, Heuchel R, Georgiev O, Hergersberg M,
Gariglio M, Dembic Z & Schaffner W (1993) Cloned
transcription factor MTF-1 activates the mouse metallo-
thionein I promoter. EMBO J 12, 1355–1362.
29 Matti MD & Freedman JH (2004) Copper-inducible
transcription: regulation by metal- and oxidative stress-
responsive pathways. Am J Physiol Cell Physiol 286,
C293–C301.
30 Giedroc DP, Chen X & Apuy JL (2001) Metal response
element (MRE)-binding transcription factor-1 (MTF-1):
structure, function, and regulation. Antioxid Redox Sig-
nal 3, 577–596.
31 Varela-Nallar L, Toledo EM, Larrondo LF, Cabral ALB,
Martins VR & Inestrosa NC (2005) Induction of cellu-
lar prion protein gene expression by copper in neurons.
Am J Physiol Cell Physiol 290, C271–C281.
32 Sadhu C & Gedamu L (1989) Metal-specific posttran-
scriptional control of human metallothionein genes. Mol
Cell Biol 9, 5738–5741.
33 Gross B, Gaestel M, Bo
¨
hm H & Bielka H (1989) cDNA