(The FASEB Journal. 2000;14:1595-1602.)
© 2000 FASEB
Molecular misreading in non-neuronal cells
FRED W. VAN LEEUWEN1,
ELLY M. HOL,
ROB W. H. HERMANUSSEN,
MARC A. F. SONNEMANS,
EWOUD MORAAL,
DAVID F. FISCHER,
DANA A. P. EVANS
,
KUM-FAI CHOOI*,
J. PETER H. BURBACH and
DAVID MURPHY*
Netherlands Institute for Brain Research, Research Group Molecular Misreading, Amsterdam, The Netherlands;
* Neuropeptide Laboratory, Institute of Molecular and Cell Biology, Singapore 0511, Republic of Singapore, and University of Bristol, Molecular Neuroendocrinology Research Group, Department of Medicine, Bristol Royal Infirmary, Bristol, U.K.; and
Section of Molecular Neurosciences, Rudolf Magnus Institute for Neurosciences, Department of Medical Pharmacology, Utrecht University, The Netherlands
1Correspondence: Research Group Molecular Misreading, Netherlands Institute for Brain Research, Meibergdreef 33, 1105 AZ Amsterdam, The Netherlands. E-mail: f.van.leeuwen{at}nih.knaw.nl
 |
ABSTRACT
|
|---|
+1 Frame-shifted proteins such as amyloid precursor
protein+1 and ubiquitin-B+1 have been
identified in the neuropathological hallmarks of Alzheimer’s disease.
These frameshifts are caused by dinucleotide deletions in GAGAG motifs
of messenger RNA encoded by genes that have maintained the unchanged
wild-type DNA sequence. This process is termed ‘molecular
misreading’. A key question is whether this process is confined to
neurons or whether it could also occur in non-neuronal cells. A
transgenic mouse line (MV-B) carrying multiple copies of a rat
vasopressin minigene as a reporter driven by the MMTV-LTR promotor was
used to screen non-neuronal tissues for molecular misreading by means
of detection of the rat vasopressin+1 protein and mutated
mRNA. Molecular misreading was demonstrated to occur in several organs
(e.g., epididymis and the parotid gland) where transgenic vasopressin
expression is abundant, but its penetrance is variable both between and
within tissues. This implies that non-neural tissues too, could be
affected by cellular derangements caused by molecular misreading.—van
Leeuwen, F. W., Hol, E. M., Hermanussen, R. W. H.,
Sonnemans, M. A. F., Moraal, E., Fischer, D. F., Evans,
D. A. P., Chooi, K.-F., Burbach, J. P. H., Murphy,
D. Molecular misreading in non-neuronal cells.
Key Words: RNA errors • dinucleotide deletions • frame-shifted proteins
|
INTRODUCTION
|
|---|
WE HAVE RECENTLY shown that dinucleotide deletions
(
GA, GU, and CU) occur in numerous neuronal transcripts (e.g.,
vasopressin (VP), amyloid precursor protein (APP), and ubiquitin B
(UBB), whereas their corresponding genes were not mutated (1
, 2)
. This inaccurate conversion of genomic information into
nonsense mRNA has been termed ‘molecular misreading’
(3)
. A dinucleotide deletion in the open reading frame of
an mRNA molecule leads to translation into a protein with a wild-type
(wt) NH2 terminus, but at the site of the RNA
mutation the reading frame is shifted. As a result, the carboxyl
terminus is translated in the +1 reading frame leading to the synthesis
of a so-called ‘+1 protein’, indicated for instance as
VP+1. By an unknown mechanism, molecular
misreading always occurs in or near simple repeats, such as GAGAG or
CUCU, and often generates premature translation termination codons
downstream of the dinucleotide deletion. Subsequently, we reported that
neurons in postmortem brains of individuals affected either by
Alzheimer’s disease or Down’s syndrome (but also of elderly
nondemented controls) display a high degree of molecular misreading.
The resulting mutated APP+1 and
UBB+1 proteins accumulate in the regions with the
neuropathological hallmarks—neuritic plaques, neuropil threads, and
tangles of Alzheimer’s disease—and may act in a dominant-negative
fashion in aging and especially in neurodegeneration (1)
.
Molecular misreading has so far only been described in neuronal tissues
(1
, 2)
. Since neurons are postmitotic, the question arises
whether molecular misreading also occurs in non-neuronal cell types, a
possibility that has important implications for the etiology of a
number of human pathologies. As the mechanism of molecular misreading
is as yet unknown, we chose the rat VP genomic sequence known to give
rise to molecular misreading (2
, 4)
. Introns are also
present, since these might be involved in molecular misreading, as was
demonstrated for pre-mRNA editing (5)
. Thus, a rat VP
minigene construct allows molecular misreading and detection of
transgene (tg) frameshift proteins with appropriate antibodies as a
marker. Mice were generated with this construct under the control of
the promoter-enhancer of the mouse mammary tumor virus (MMTV) long
terminal repeat (lines MV-A and MV-B), specifically expressing VP in
epithelial cells of secretory organs (6
7
8)
. Frame-shifted
and wt transcripts of the tg were visualized by in situ
hybridization and immunocytochemically with an antibody directed
against a peptide sequence (rat VP+1) that would
be generated as a result of a dinucleotide deletion in a GAGAG motif in
rat wt VP transcripts (9)
. The results clearly show that
molecular misreading is a general source of transcript errors and thus
may be widely involved in cellular derangements.
|
MATERIALS AND METHODS
|
|---|
Transgenic mice
The MMTV-VP transgene consists of the mouse mammary tumor virus
long terminal repeat (MMTV-LTR) linked to a 2.2 kb fragment of the rat
VP structural gene that contains the three VP exons and two introns,
flanked by 36 and 160 bp, respectively, of 5' and 3' UTR sequences
(Fig. 1A
). After injection of fertilized one-cell mouse eggs with
the MMTV-VP construct, two founder animals, MV-A and MV-B, were
identified by Southern analysis and expanded into lines by mating with
wt C57/BL10 x CBA/J mice. All animals examined were obligate tg
heterozygotes. Southern analysis revealed that 36 copies of the
transgene per haploid genome were present in the MV-B mice that had
been used in this study (for details about MV-A and MV-B, see ref
6
).
View larger version (19K):
[in this window]
[in a new window]
|
Figure 1. A) The transgenic construct. The MMTV-VP fusion gene
used for the generation of tg mice (MV-A and MV-B) and the
corresponding positions of probes used in Southern and Northern blot
hybridization. The three VP exons are shown (A through C; Gen Bank Acc.
No. X01637). The transgene consists of 1.4 kb MMTV-LTR joined to 2.2 kb
of the rat VP gene (exons A-C are indicated). Two transcriptional
initiation sites are present, one in the MMTV-LTR (long arrow) and the
other in the VP gene (short arrow). Abbreviations: H,
HindIII, P, Pst I. TJ: oligonucleotide
probe (35 mer) complementary to the MMTV sequence prior to the junction
of the tg (TJ). This probe anneals to the same VP transcripts as the
3'VP probes annealing to the last 48 bases of VP glycopeptide (see also
panel B; for details, see ref 6
).
Transgenic offspring was identified by Southern blotting using a random
primed  32P-labeled VP genomic fragment (1557 bp) as a
probe. B) Organization of the rat vasopressin (VP) gene
and prohormone. The VP gene is composed of three exons (A, B, and C,
separated by two introns) that encode the VP precursor
(9)
. The number of amino acids of the VP precursor is
mentioned. The VP, neurophysin (NP), and glycopeptide (GP) moieties are
cleaved by post-translational modifications at, respectively, a dibasic
(Lys-Arg) and monobasic (Arg) cleavage site (stippled area) by
the processing enzymes (e.g., PC1 and PC2). In the rat epididymis, PC1
and PC2 are weakly expressed but the processing of the rat VP precursor
in the mouse might be performed by other enzymes (10)
.
SP = signal peptide.  = GAGAG motif (first motif was chosen for
construction of oligonucleotide nucleotide (365) and for
hybridization of GA deleted VP; see Fig. 6
). For specificity of
antisera (black bar), see also ref 11
for JAN (recognizing
the VP precursor) and C3 final (recognizing AA 128–145), ref
12
for YL-3 (recognizing AA 7–17), and refs 11
, 13
for THR (recognizing AA 98–105). Rat NP (RNP) recognizes the
constant region of NP (amino acids 22–86; ref 14
).
|
|
Northern blotting
Total RNA was isolated from various organs of 7-wk-old sexually
mature male (MV-B) tg mice. Northern blots were controlled for equal
loading and transfer in two ways. 1) After transfer, the
ethidium bromide stained filter was viewed and photographed under
ultraviolet light, and 2) filters were routinely reprobed
with oligonucleotides corresponding with the housekeeping mRNAs
encoding either GAPDH or -tubulin (for details, see ref
6
and Fig. 2
).
Immunohistochemistry
After intraperitoneal injection of Avertin anesthetic and
intracardial perfusion fixation with saline (0.9%), followed by
Bouin’s solution, the brains, parotid glands, and reproductive tracts
(testis and epididymis) of six male MV-B mice were isolated. Three mice
were 7 wk old (tg1–3) and three were 60 wk old (tg 4–6). Six male wt
nullizygotes (littermates, wt1–3, 4–6 wt) served as controls.
Vibratome sections (50 µm) were made from the brain and kept in
Tris-NaCl buffer (0.05 M Tris, 0.9% NaCl, pH 7.4). Extraneuronal
tissues were dehydrated with alcohol, embedded in paraplast, and 8
µm-thick sections were hydrated and washed in Tris-NaCl. All sections
were incubated as described (1)
. The presence of the VP
precursor and its derivatives was determined with five different
antibodies, each directed against different epitopes of the VP
precursor (Fig. 1B
. Table 1
). For detection of VP, anti-glycopeptide (GP) antiserum (C3 final or
VP-GP), which recognizes the unique carboxyl terminus of the VP
precursor, was used preferentially. To assess the occurrence of the +1
frameshift mutation (i.e., VP+1), antibodies were
used raised against a synthetic peptide predicted by the
carboxyl-terminal amino acid sequence after either of the two GA
deletions in the GAGAG motifs of VP mRNA (i.e., AGRASPGAAA: Dan E; ref
2
; Fig. 1B
). The degree of coexistence of VP-GP
and VP+1 in the epididymis was determined by
counting the profiles of double-labeled cells in pairs of sections
stained for VP-GP and VP+1 of four MV-B mice. The
peroxidase-anti-peroxidase method was used with 3,3'-diaminobenzidine
as a chromogen and 0.2% nickel ammonium sulfate as an intensifying
agent. Preimmune C3 final and DanE antisera preabsorbed with the GP
22–39 and DanE, respectively, served as controls. Rat neurophysin
(RNP) antibody was absorbed with rat NP. Specificity controls were not
only carried out on tg and wt mice, but also on hypothalamic sections
of an aged Wistar rat, which served as a positive control for
VP+1 immunoreactivity (2)
.
View this table:
[in this window]
[in a new window]
|
Table 1. Transgenic expression: immunoreactivity of rat VP precursor, its
processed products, and rat VP+1 in the transgenic mouse
testis, epididymis, and parotid glanda
|
|
In situ hybridization
Adjacent paraffin sections (8 µm) of tg and wt male
reproductive tract were dewaxed, rehydrated, and deproteinated for 20
min with 0.2 M HCl, followed by a 15 min treatment with proteinase K.
Thereafter the sections were dehydrated, delipidated with
chloroform, rehydrated, and incubated for 20 min in 0.1 M
phosphate-buffered saline/0.1% v/v Triton X-100. The sections were
hybridized with either a VP antisense oligomer recognizing WT rat VP-GP
mRNA (i.e., 5'GCCCGTCCAGCTGCGTGGCGTTGCT3'), a VP antisense oligomer
recognizing the GA deletion at position 365 (365, i.e.,
5'GCTACTCGACGCACCGGCTC3') (2)
, or a VP sense oligomer. The
oligomers were labeled at the 3' end using terminal transferase
(Boehringer Mannheim, Mannheim, Germany) and
35S-dATP (NEN/Dupont, Boston, Mass.). Each
section was hybridized with 106 cpm labeled
oligomer/100 µl hybridization buffer, containing 30% formamide, 0.5
M NaCl, 5x Denhardt’s solution, 10 mM Tris-HCl (pH 7.5), 1 mM EDTA,
10% dextran sulfate, 50 mM dithiothreitol, 100 µg/ml herring sperm
DNA, and 500 µg/ml tRNA. The sections were incubated overnight at
42°C in a humidified chamber, followed by washing at 55°C for
1 x 45 min in 1x SSC and 3 x 45 min in 0.3x SSC. Sections
were dehydrated in graded ethanols containing 300 mM ammonium acetate,
and after drying were dipped in NTB-2 photoemulsion (Kodak). After 2.5
wk, the sections were developed, counterstained with hematoxylin,
dehydrated, and embedded in Entellan.
Controls for wt VP mRNA expression included labeling of hypothalamic
sections of the Wistar and homozygous Brattleboro rat. VP (365) mRNA
labeling was validated in hypothalamic sections of homozygous
Brattleboro rats (2)
. Using a VP sense probe, no labeling
was found in the tg and wt male reproductive tract or hypothalamic
sections of the Wistar rats and homozygous Brattleboro rats.
|
RESULTS
|
|---|
Using Northern analysis, tg VP RNA in MV-B mice was found in the
brain, the Harderian, submaxillary and parotid glands, the kidney, and
the male reproductive tract (testis, epididymis, vas deferens, seminal
vesicles, and prostate (Fig. 2)
. VP expression in the MV-A strain
(6)
and the expression of other MMTV-driven constructs
showed a similar pattern (15)
. No tg VP mRNA was found in
other organs (e.g., lung and intestine). In the brain, the tg VP RNA
was detected in the cerebral cortex, cerebellum, midbrain, and
hypothalamus. The expression of tg VP transcripts was modest in the
brain but very high in the testis, epididymis, prostate, and parotid
gland (Fig. 2)
.
By immunocytochemistry using C3 final (VP-GP) antiserum in wt mice,
normal endogenous rat VP gene products were shown within cell bodies
and their neurites in the classic VP synthesizing areas such as the
supraoptic, suprachiasmatic, and paraventricular nuclei of the
hypothalamus (Fig. 3
; ref 16
). In agreement with the Northern blot (Fig. 2)
and
the radioimmunoassay data (6)
, ectopic expression of rat
VP-GP immunoreactivity was prominent in many tg MV-B mouse brain areas,
including cells of the thalamic nuclei (Fig. 3)
. No rat
VP+1 staining was found in either wt or tg mouse
brain, indicating the absence of molecular misreading of the rat VP
gene in these ectopic brain areas of VP tg expression. Preimmune and
preabsorbed DanE antisera failed to show any staining in these areas.
Wistar brain sections, serving as a positive control, showed
VP+1 immunoreactivity in solitary cells of the
supraoptic nucleus, as reported before (2)
.
View larger version (129K):
[in this window]
[in a new window]
|
Figure 3. Transversal vibratome section of the tg mouse brain
immunocytochemically stained with C3 final antiserum showing ectopic
expression in various thalamic nuclei (T) and endogenous VP expression
in the hypothalamus (suprachiasmatic (SCN), supraoptic (SON), and
paraventricular (PVN) nucleus). Note that endogenous VP fibers from the
SCN are present in the lateral habenula (LH) and the paraventricular
nucleus of the thalamus (arrowhead). Bar = 200 µm; asterisk = third ventricle, OC = optic chiasm.
|
|
The tg mice of the MV-B lineage displayed ectopic and specific
immunoreactivity for VP, NP, and GP in many principal cells lining the
caput of the epididymis (Table 1
; Fig. 4A
, B
, C
, D
, E
and Fig. 5
) and in the Leydig cells of the testis (Fig. 4G
, H
, I
). No staining was seen after incubation
with preimmune and the preabsorbed C3 final and DanE antiserum or with
preabsorbed RNP antiserum in the epididymis (cf. Fig. 4F
) and testis (Fig. 4H
) of tg mice. No
immunoreactivity for VP gene products was found in epididymis (Fig. 4F
) and testis (Fig. 4I
) of wt mice. These
data were complemented with in situ hybridization supporting
the outcome of the immunocytochemical results. In tg MV-B mice, the
principal cells of the epididymis (Fig. 6A
) and the Leydig cells (Fig. 6D
) were intensely labeled,
whereas in wt mice no VP labeling was seen (Fig. 6G
, H
, I
). Thus, the VP precursor is not
expressed in the wt mouse epididymis and testis.
View larger version (164K):
[in this window]
[in a new window]
|
Figure 4. Immunocytochemical staining of the vasopressin (VP) precursor (JAN,
A, YL-3, B) and its processed parts:
region of neurophysin (RNP, C), and the unique carboxyl
terminus (THR, D) and glycoprotein (C3 final,
E) in paraplast sections of a transgenic mouse
epididymis (A–F) and testis
(G–I; refs 17
,18
). F) A
nontransgenic littermate was stained with RNP. The same negative
staining was obtained with all other wt VP preimmune and preabsorbed
antibodies (cf. Fig. 1B
). The Leydig cells of the testis
(large arrowheads) of all transgenic mice were also positive for all
parts of the VP precursor (G, THR staining) and negative
with VP+1 (H, DanE) or preadsorbed C3 final
or RNP antiserum. In wt mouse, no reaction was present in the
epididymis and testis with various VP antisera or with VP+1
(I, DanE). Note that the Leydig cells in panels
G–I are present in the interstitium between the
seminiferous tubules; elongated spermatids can be seen in their lumina
(asterisks). A–F) Small arrowhead =
stereocilia at the apical site of principal cells bordering lumen (l)
of the epididymal duct. Magnification bar = 50 µm.
|
|
View larger version (110K):
[in this window]
[in a new window]
|
Figure 5. Partial coexistence of VP+1 in transgenic VP-expressing
cells. Consecutive paraplast sections of tg mouse caput epididymis
immunocytochemically stained with VP-GP (A, C) and
VP+1 (B, D) antiserum. Note that the
ectopically expressed VP precursor is selectively present in many
principal cells (with a basally located nucleus, large arrowhead in
panel D), but that VP+1 protein (B,
D) is present only in a subpopulation of these VP
precursor-containing cells. At a higher magnification (C,
D), coexistence of VP and VP+1 is shown (small
arrowheads). s = stereocilia at the apical site of principal cells
in the lumen (l) of the epididymal duct. Note in panels C,
D that the nuclei of the immunoreactive cells are situated
basally and extend to the basal lamina, identifying them as principal
cells. Apical cells are never VP-GP positive (17
, 18)
. Bar
in panels A, B = 50 µm; bar in panels C,
D = 20 µm.
|
|
View larger version (210K):
[in this window]
[in a new window]
|
Figure 6. In situ hybridization of paraplast sections incubated
with the 5' VP-GP antisense (A, D, G, H), 365
(B, E, I), and 5' VP-GP sense probe (C,
F) in the epididymis (A–C, G) and testis
(D, F, H, I). All sections were counterstained
with hematoxylin. Note heavy labeling in the principal cells of the
transgenic mouse (A) and its absence in its wt
littermate (G). In two ependymal ducts
(B) principal cells display the frame-shifted VP
transcript (small arrowheads), but not in its sense control
(C, small arrowheads). In some ductuli 5' VP-GP mRNA was
(partially) absent (asterisks in panel A). As a control,
the sense probe revealed no staining in the epididymis
(C) and the Leydig cells of the testis
(F, large arrowheads). Comparison of the adjacent
sections of panels B, C shows the same
ductuli, heavily labeled in panel B but not in panel
C (asterisks). In the upper left corner of the
epididymis (A–C), part of the transition of the testis
(with spermatozoa) and epididymis displays only background labeling. In
the Leydig cells of the testis, transgenic VP was clearly expressed
(D, large arrowheads). However, no labeling was seen in
the Leydig cells with the 365 probe (E, large
arrowheads). In nontransgenic littermates, no labeling was apparent
with either probe in the epididymis (G) and testis
(H–I) in Leydig cells (large arrowheads). Magnification
bar = 50 µm.
|
|
VP+1 immunoreactivity coexisted with VP-GP
immunoreactivity (Fig. 5C, D
), whereas stringent in
situ hybridization with the 365 probe showed that
VP+1 mRNA was also expressed in the principal
cells of the epididymis (Fig. 6B
). This indicates that
molecular misreading of the VP tg occurs in a subpopulation (20–50%)
of tg VP-GP immunoreactive principal cells (Fig. 5A
, B
, C
, D
). The frameshift occurred in the same GAGAG
motif as reported before (ref 2
and Fig. 6B, E, I
). No difference with age in the fraction of
VP+1 positive/VP immunoreactive cells was
observed in the epididymis of 7- and 60-wk-old MV-B-mice (both
n=2). VP+1 immunoreactivity was absent
in the Leydig cells of the testis of tg and wt mice (Fig. 4H, I
). Specific in situ hybridization with the
365 probe confirmed this result at the transcript level (Fig. 6E, I
).
Positive controls for transcript labeling of VP and
VP+1 yielded labeling of cells in the
magnocellular neurons of the supraoptic nucleus in the Wistar and
homozygous Brattleboro rat, respectively, as reported before
(2)
. The VP sense probe yielded no positive cells in the
male reproductive tract (Fig. 6C, F
).
In the parotid gland of the MV-B mice numerous acinar cells showed
intense immunoreactivity for VP-GP and RNP (Fig. 1B
), which
was in agreement with the Northern blot (Fig. 2)
and with
radioimmunoassay data in the MV-A and MV-B mice (6)
. The
VP+1 staining was present in a subpopulation
(5–10%) of the VP positive acinar cells of the parotid gland (data
not shown). Using well-characterized VP antisera (Fig. 1B
),
no immunoreactivity for the VP precursor was found in the parotid gland
of wt mice. As in the epididymis, no reaction was seen after incubation
with preimmune and preabsorbed C3 final and DanE or with preabsorbed
anti-rat NP serum. Thus, molecular misreading of the VP tg occurs here
as well, although to a much lower degree than in the epididymis.
|
DISCUSSION
|
|---|
Modification of gene transcripts is possible not only by means of
programmed rearrangement of DNA for functional expression of genes, but
also at the RNA level in a variety of tissues, including neurons, by
alternative splicing and substitutional mRNA editing (19
, 20)
. We recently reported a new mechanism (termed molecular
misreading) in which, perhaps by unfaithful transcription of genomic
information, dinucleotide deletions are introduced into the mRNA of
neuronal genes (1
2
3)
.
In this report we have demonstrated molecular misreading of a rat VP tg
in several non-neuronal tissues after its ectopic expression in MV-B
mice. The high copy number of the rat VP tg construct (36 copies per
haploid genome; Fig. 1A
and Fig. 2
) results in
overexpression in numerous tissues where in wt mice no VP expression is
present (cf. Fig. 4F, I
and Fig. 6G
, H
, I
).
This construct is therefore an ideal substrate for molecular
misreading. The availability of a panel of VP antibodies allows the use
of the VP minigene as a reporter for molecular misreading (Table I)
.
Remarkably, molecular misreading often occurs in tissues with a high tg
VP mRNA expression (Fig. 2)
, such as in the epididymis and parotid
gland, as revealed by the VP+1 staining in a
subpopulation of principal epididymal cells and acinar cells,
respectively. Both cell types express the rat VP tg protein as well
(Fig. 4A
, B
, C
, D
, E
, Fig. 6A
). These results are in
agreement with our previous data, suggesting that the detection of
molecular misreading is facilitated by enhanced transcriptional
activity, as shown before by the VP and APP genes in homozygous
Brattleboro rats (2
, 14)
and Down syndrome patients,
respectively (1)
. Despite the widespread but modest
ectopic expression of rat VP in the mouse brain, we did not observe any
immunoreactivity for VP+1. This may be related to
the low expression of tg VP mRNA in the brain (Fig. 2a
). On
the other hand, the data show a variable penetrance of
VP+1 immunoreactivity. In the Leydig cells of the
testis, tg VP expression is high (Fig. 2a
and Fig. 6D
) but VP+1 is not detectable, as
shown by immunocytochemistry (Fig. 4H
) and in
situ hybridization (Fig. 6E
). This indicates
that either the mechanisms of molecular misreading or the subsequent
degradation of the nonsense transcripts and the +1 proteins and their
clearance is tissue specific (Fig. 7
). Thus, in some tissues molecular misreading may be suppressed entirely
(e.g., Leydig cells in testis; Fig. 4G, H
and Fig. 6D, E
), almost entirely (acinar cells of the parotid gland),
or partially (principal epididymal cells; Fig. 5A
, B
, C
, D
, Fig. 6A, B
). The present finding that several tissues accumulate
aberrant proteins, possibly by a lack of RNA or protein quality
control, implies a common denominator. It has indeed been shown that
RNA surveillance is a candidate mechanism for detecting aberrant
transcripts (22
23
24)
. This system, which monitors and
reduces the number of aberrant transcripts by nonsense-mediated RNA
decay (NMD), might differ in its efficiency in various species and
genetic background, tissues, and within one cell population
(3)
. This could explain the variable penetrance of
molecular misreading reported in the present study.
View larger version (13K):
[in this window]
[in a new window]
|
Figure 7. Quality control of mRNA and proteins. Schematic representation of
consequences of molecular misreading. Introduction of RNA errors leads
to mutant transcripts. However, if these are not effectively monitored
by mRNA surveillance and degraded by nonsense-mediated decay (NMD),
they will result in proteins with an aberrant, often truncated carboxyl
terminus. If these mutant proteins are not effectively degraded by a
failing ubiquitin/proteasome system (21)
or cleared, they
will accumulate and could interfere with the function of wt proteins.
|
|
Contrary to alternative splicing and mRNA editing, both of which are
functional (25)
, molecular misreading is not likely to
have a physiological role. The introduction of premature termination
signals might be deleterious, i.e., NMD serving as an mRNA surveillance
mechanism to eliminate mRNAs encoding truncated polypeptides that could
act in a dominant-negative fashion and interfere with the function of
wt proteins (26)
. Thus, if molecular misreading becomes
manifest, it probably results in aberrant dysfunctional proteins, the
accumulation of which might be involved in cellular derangements during
disease or aging (Fig. 7)
. This has recently been clearly illustrated
not only in Alzheimer’s disease and Down syndrome, where
APP+1 and UBB+1 proteins
were detected in the neuropathological hallmarks (plaques, tangles and
neuropil threads), but also in aged nondemented controls in the initial
stages of neuropathology (1)
. The absence of obvious
problems in reproduction [in both males and females of the MV-A
and MV-B tg lines (6)
while showing transcript mutations]
indicates that molecular misreading of the VP tg in the male
reproductive organs and the resulting expression of
VP+1 protein do not result from a failing
reproductive system.
The mechanism of molecular misreading, which introduces errors
preferentially in or adjacent to GAGAG or CUCU motifs in mammalian
tissue and results in ‘dirty’ transcripts (27)
, is not
clear at present (for example, by slippage or stuttering; ref
3
). This process may occur either co- or
post-transcriptionally. These GA and CU dinucleotide repeats could
randomly occur in virtually every transcript (1
, 2)
. It is
interesting to realize that selection against frameshift mutations has
been shown to occur preferentially in primate exons and not so much in
introns (28)
. The present study shows that molecular
misreading also occurs in non-neuronal tissue. Therefore, a large
variety of gene transcripts—for instance, those involved in or
associated with pathological processes in a large array of age-related
diseases, such as cancer and diabetes mellitus—may undergo a similar
modification and are the subject of our present studies.
|
ACKNOWLEDGMENTS
|
|---|
We thank Drs. D. F. Swaab, J. Verhaagen, J. M.
Ruijter (University of Amsterdam), and M. A. Corner for critically
reading the manuscript and Ms. O. Pach and Ms. W. T. P.
Verweij for secretarial support. We are grateful to Dr. A. G.
Robinson (UCLA) and Dr. J. Verbalis (Georgetown University, Washington,
D.C.) for providing the anti-NP and YL-3 antisera and rat NP, to J.T.M.
Vreeburg (Erasmus University, Rotterdam) for helpful discussions, and
to Dr. N.G. Seidah (Clinical Research Institute, Montreal, Canada) for
helpful discussions on the enzymatic processing of ectopically
expressed VP. GP 22–39 was supplied by Dr. J. van Nispen, Organon Oss,
The Netherlands. Supported by ‘Het Vernieuwingsfonds’ of the Royal
Netherlands Academy of Arts and Sciences, Matty Brand Foundation, Jan
Dekker and Ludgardine Bouwman Foundation, Internationale Stichting
Alzheimer Onderzoek, Hersenstichting Nederland Human Frontier Science
Program Organization (HFSP-RG 0148/1999B), and NWO, Priority Program
Memory Processes and Dementia (GPD). D.M. was generously supported by
the Wellcome Trust.
Received for publication September 7, 1999.
Revision received February 7, 2000.
|
REFERENCES
|
|---|
-
Van Leeuwen, F. W., De Kleijn, D. P. V., Van den Hurk, H. H., Neubauer, A., Sonnemans, M. A. F., Sluijs, J. A., Köycü, S., Ramdjielal, R. D. J., Salehi, A., Martens, G. J. M., Grosveld, F. G., Burbach, J. P. H., Hol, E. M. (1998) Frameshift mutants of ß-amyloid precursor protein and ubiquitin-B in Alzheimer’s and Down patients. Science 279,242-247[Abstract/Free Full Text]
-
Evans, D. A. P., Van der Kleij, A. A. M., Sonnemans, M. A. F., Burbach, J. P. H., Van Leeuwen, F. W. (1994) Frameshift mutations at two hotspots in vasopressin transcripts in post-mitotic neurons. Proc. Natl. Acad. Sci. USA 91,6059-6063[Abstract/Free Full Text]
-
Van Leeuwen, F. W., Burbach, J. P. H., Hol, E. M. (1998) Mutations in RNA: a first example of molecular misreading in Alzheimer’s disease. Trends Neurosci 21,331-335[Medline]
-
Finch, C. E., Goodman, M. F. (1997) Relevance of ‘adaptive’ mutations arising in nondividing cells of microorganisms to age-related changes in mutant phenotypes of neurons. Trends Neurosci 20,501-507[Medline]
-
Herb, A., Higuchi, M., Sprengel, R., Seeburg, P. H. (1996) Q/R site editing in kainate receptor GluR5 and GluR6 pre-mRNAs requires distant intronic sequences. Proc. Natl. Acad. Sci. USA 93,1875-1880[Abstract/Free Full Text]
-
Chooi, K. F., Carter, D. A., Biswas, S., Lightman, S. L., Ho, M.-Y., Murphy, D. (1994) Ectopic vasopressin expression in MMTV-Wnt-1 transgenic mice modifies mammary tumor differentiation and pathology. Cancer Res 54,6434-6440[Abstract/Free Full Text]
-
Clermont, Y., Flannery, J. (1970) Mitotic activity in the epithelium of the epididymis in young and old adult rats. J. Biol. Reprod. 3,283-292
-
Denny, P. C., Ball, W. D., Redman, R. S. (1997) Salivary glands: a paradigm for diversity of gland development. Crit. Rev. Oral Biol. Med. 8,51-75[Abstract/Free Full Text]
-
Ivell, R., Richter, D. (1984) Structure and comparison of the oxytocin and vasopressin genes from rat. Proc. Natl. Acad. Sci. USA 81,2006-2010[Abstract/Free Full Text]
-
Seidah, N. G., Chrétien, M., Day, R. (1994) The family of subtilisin/kexin like pro-protein and pro-hormone convertases: divergent or shared functions. Biochimie (Paris) 76,197-209[Medline]
-
Sonnemans, M. A. F., Evans, D. A. P., Burbach, J. P. H., Van Leeuwen, F. W. (1996) Immunocytochemical evidence for the presence of vasopressin in intermediate sized neurosecretory granules of solitary neurohypophyseal terminals in the homozygous Brattleboro rat. Neuroscience 72,225-231[Medline]
-
Verbalis, J. G., Hoffman, G. E., Rosenbaum, L. C., Nilaver, G., Loh, Y. P. (1991) Generation and characterization of an antiserum directed against neurohypophyseal prohormones. J. Neuroendocrinol. 3,267-272
-
Van Leeuwen, F. W., Evans, D. A. P., Meloen, R., Sonnemans, M. A. F. (1994) Differential neurophysin immunoreactivities in solitary magnocellular neurons of the homozygous Brattleboro rat indicate an altered neurophysin moiety. Brain Res 635,328-330[Medline]
-
Van Leeuwen, F., Van der Beek, E., Seger, M., Burbach, P., Ivell, R. (1989) Age-related development of a heterozygous phenotype in solitary neurons of the homozygous Brattleboro rat. Proc. Natl. Acad. Sci. USA 86,6417-6420[Abstract/Free Full Text]
-
Korpelainen, E. I., Karkkainen, M. J., Tenhunen, A., Lakso, M., Rauvala, H., Vierula, M., Parvinen, M., Alitalo, K. (1998) Overexpression of VEGF in testis and epididymis causes infertility in transgenic mice: evidence for nonendothelial targets for VEGF. J. Cell Biol. 14,1705-1712
-
De Vries, G. J., Buijs, R. M., Van Leeuwen, F. W., Caffé, A. R., Swaab, D. F. (1985) The vasopressinergic innervation of the brain in normal and castrated rats. J. Comp. Neurol. 233,236-254[Medline]
-
Hamilton, D. W. (1975) Structure and function of the epithelium lining the ductuli efferentes, ductus epididymis and ductus deferens in the rat. Hamilton, D. W. Greep, R. O. eds. Handbook of Physiology, Sec. 7, Vol. 5, Male Reproductive System ,259-301 American Physiological Society Washington, D.C..
-
Abe, K., Takano, H., Ito, T. (1983) Ultrastructure of the mouse epididymal duct with special reference to the regional differences of the principal cells. Arch. Histol. Jpn. 46,51-68[Medline]
-
Schatz, D. G., Oettinger, M. A., Schlissel, M. S. (1992) V(D)J recombination: molecular biology and regulation. Annu. Rev. Immunol. 10,359-383[Medline]
-
Newman, A. J. (1994) Pre-mRNA splicing. Curr. Opin. Genet. Dev. 4,298-304[Medline]
-
Schwartz, A. L., Ciechanover, A. (1999) The ubiquitin-proteasome pathway and pathogenesis of human diseases. Annu. Rev. Med. 50,57-74[Medline]
-
Applequist, S. A., Selg, M., Raman, C., Jäck, H.-M. (1997) Cloning and characterization of HUPF1, a human homolog of the Saccharomyces cerevisiae nonsense mRNA-reducing UPF1 protein. Nucleic Acids Res 25,814-821[Abstract/Free Full Text]
-
Hentze, M. W., Kulozik, A. E. (1999) A perfect message: RNA surveillance and nonsense-mediated decay. Cell 96,307-310[Medline]
-
Frischmeyer, P. A., Dietz, H. C. (1999) Nonsense-mediated mRNA decay in health and disease. Hum. Mol. Genet. 8,1893-1900[Abstract/Free Full Text]
-
Benne, R. (1996) RNA editing: how a message is changed. Curr. Opin. Genet. Dev. 6,221-231[Medline]
-
Cali, R. M., Anderson, P. (1998) mRNA surveillance mitigates genetic dominance in Caenorhabditis elegans. Mol. Gen. Genet. 260,176-184[Medline]
-
Bridges, B. A. (1999) Dirty transcripts from clean DNA. Science 284,62-63[Free Full Text]
-
Metzgar, D., Bytof, J., Wills, C. (2000) Selection against frameshift mutations limits microsatellite expansion in coding DNA. Genome Res. 10,72-80[Abstract/Free Full Text]
This article has been cited by other articles:
|
|
|
|
 
F. W. van Leeuwen, P. van Tijn, M.A.F. Sonnemans, B. Hobo, D. M.A. Mann, C. Van Broeckhoven, S. Kumar-Singh, P. Cras, G. Leuba, A. Savioz, et al.
Frameshift proteins in autosomal dominant forms of Alzheimer disease and other tauopathies
Neurology,
January 24, 2006;
66(1_suppl_1):
S86 - S92.
[Abstract]
[Full Text]
[PDF]
|
|
|
TOP
ABSTRACT
INTRODUCTION
-32P 3'VP (a). The arrowhead
indicates the positions of the endogenous VP RNA in the hypothalamic
total RNA sample. The 3'VP probe, an oligonucleotide complementary to
bases encoding the last 16 amino acids of the VP-GP sequence detects
the endogenous VP mRNA in the hypothalamus (arrowhead) as well as the
tg mRNA. Subsequent reprobing with the oligonucleotide (TJ) probe
directed at the MMTV sequence prior to the junction of the transgene
(b; see also Fig. 1A