Three Gene-Targeted Mouse Models of RNA Splicing
Factor RP Show Late-Onset RPE and Retinal Degeneration
John J. Graziotto,1,2Michael H. Farkas,1Kinga Bujakowska,3Bertrand M. Deramaudt,1
Qi Zhang,1Emeline F. Nandrot,3Chris F. Inglehearn,4Shomi S. Bhattacharya,3,5and
Eric A. Pierce1
PURPOSE. Mutations in genes that produce proteins involved in
mRNA splicing, including pre-mRNA processing factors 3, 8,
and 31 (PRPF3, 8, and 31), RP9, and SNRNP200 are common
causes of the late-onset inherited blinding disorder retinitis
pigmentosa (RP). It is not known how mutations in these
ubiquitously expressed genes lead to retina-specific disease. To
investigate the pathogenesis of the RNA splicing factor forms
of RP, the authors generated and characterized the retinal
phenotypes of Prpf3-T494M, Prpf8-H2309P knockin mice. The
retinal ultrastructure of Prpf31-knockout mice was also inves-
METHODS. The knockin mice have single codon alterations in
their endogenous Prpf3 and Prpf8 genes that mimic the most
common disease causing mutations in human PRPF3 and
PRPF8. The Prpf31-knockout mice mimic the null alleles that
result from the majority of mutations identified in PRPF31
patients. The retinal phenotypes of the gene targeted mice
were evaluated by electroretinography (ERG), light, and elec-
RESULTS. The RPE cells of heterozygous Prpf3?/T494Mand
Prpf8?/H2309Pknockin mice exhibited loss of the basal infold-
ings and vacuolization, with accumulation of amorphous de-
posits between the RPE and Bruch[b]’s membrane at age two
years. These changes were more severe in the homozygous
mice, and were associated with decreased rod function in the
Prpf3-T494M mice. Similar degenerative changes in the RPE
were detected in Prpf31?mice at one year of age.
CONCLUSIONS. The finding of similar degenerative changes in
RPE cells of all three mouse models suggests that the RPE may
be the primary cell type affected in the RNA splicing factor
forms of RP. The relatively late-onset phenotype observed in
these mice is consistent with the typical adult onset of disease
in patients with RP. (Invest Ophthalmol Vis Sci. 2011;52:
place in a large ribonucleoprotein complex, the spliceosome,
which in addition to the pre-mRNA substrate is made up of five
core small nuclear ribonucleoprotein complexes (snRNPs),
and a host of accessory proteins which assemble in a stepwise
fashion on each intron to carry out the splicing reaction (for
reviews see Refs. 1, 2). Mutations in five protein factors integral
to this process have been implicated in the inherited blinding
disorder retinitis pigmentosa (RP).3–7These factors are ubiqui-
tously expressed and are common to the U4/U6/U5 tri-snRNP
complex. The pre-mRNA processing factor 3 (PRPF3) protein is
associated with the U4/U6 snRNP complex, and is necessary
for the integrity of a spliceosomal precursor, the U4/U6/U5
tri-snRNP complex, without which splicing cannot occur.8,9
PRPF8 is a U5 snRNP and U4/U6/U5 tri-snRNP component and
is thought to be an assembly platform at the core of the
activated spliceosome (for review see Ref. 10). PRPF31 is also
a component of the U4/U6/U5 tri-snRNP and is involved in
maintaining the stability of the complex.11,12
SNRNP200 interact with PRPF3 and PRPF8, respectively, and
arealsocomponents of the
plex.7,13,14Due to the ubiquitous nature of RNA splicing, there
is much debate regarding the underlying disease mechanism by
which mutations in the RNA splicing factors lead to photore-
ceptor specific disease.
RP is the most common inherited form of blindness, with an
estimated prevalence of approximately 1:1000 to 4000 peo-
ple.15–18This translates into ?1.5 to 6 million individuals
affected with RP worldwide. RP patients typically present in
early adulthood with progressive night blindness and loss of
peripheral visual field due to the loss of rod photoreceptor
cells of the retina. The cone photoreceptors die after the rods,
leading to a further loss of daytime vision, which gradually
progresses to complete blindness in later life.19RP displays all
three modes of Mendelian inheritance, and most of the 37
genes implicated in non-syndromic RP are expressed specifi-
cally in either photoreceptor cells or the retina20(for a current
listing of retinal disease genes see RetNet: http://www.sph.
uth.tmc.edu/Retnet/). As a group, the five pre-mRNA process-
ing factor forms of dominant RP constitute a leading cause of
re-mRNA splicing is an essential step in the expression of
most eukaryotic transcripts. This ubiquitous process takes
From the1F.M. Kirby Center for Molecular Ophthalmology, Scheie
Eye Institute, University of Pennsylvania School of Medicine, Philadel-
phia, Pennsylvania; the2Department of Neuroscience, University of
Pennsylvania, Philadelphia, Pennsylvania;3INSERM UMR_S968, CNRS
UMR_7210, Universite ´ Pierre et Marie Curie-Paris 6, Centre de Recher-
che Institut de la Vision, Paris, France;4Section of Ophthalmology and
Neuroscience, Leeds Institute of Molecular Medicine, University of
Leeds, St. James’s University Hospital, Leeds, United Kingdom; and the
5Institute of Ophthalmology, University College London, London,
Supported by The Foundation Fighting Blindness, Research to
Prevent Blindness, the Zeigler Foundation, the F. M. Kirby Foundation,
the Mackall Foundation Trust, and the National Eye Institute (Grant
T32-EY-007035; JJG), Institut National de la Sante ´ et de la Recherche
Me ´dicale (INSERM), Universite ´ Pierre et Marie Curie-Paris 6, Centre
National de la Recherche Scientifique (CNRS), De ´partment de Paris,
Marie Curie Actions (European Reintegration Grant (PERG04-GA-2008-
231125); KB), Fondation Voir et Entendre and Fondation Bettencourt
Schueller (Young Investigator Grants; EFN), and Agence Nationale de
la Recherche (Chaire d[b]’Excellence; SSB).
Submitted for publication January 11, 2010; revised June 14, 2010;
accepted July 28, 2010.
Disclosure: J.J. Graziotto, None; M.H. Farkas, None; K.
Bujakowska, None; B.M. Deramaudt, None; Q. Zhang, None;
E.F. Nandrot, None; C.F. Inglehearn, None; S.S. Bhattacharya,
None; E.A. Pierce, None
Corresponding author: Eric A. Pierce, F.M. Kirby Center for Molecu-
lar Ophthalmology, University of Pennsylvania School of Medicine, 422
Curie Boulevard, Philadelphia, PA 19104; email@example.com.
Biochemistry and Molecular Biology
Investigative Ophthalmology & Visual Science, January 2011, Vol. 52, No. 1
Copyright 2011 The Association for Research in Vision and Ophthalmology, Inc.
dominant RP, second only to mutations in rhodopsin.21,22For
most RP genes, the mechanisms by which the mutations lead
to disease are not understood, but since most of the disease
genes are photoreceptor specific, it follows naturally that a
photoreceptor-specific disease results. The pre-RNA splicing
factor forms of RP are different in that they are due to muta-
tions in ubiquitous proteins involved in RNA splicing in all cells
and tissues of the body, and yet still lead to retina-specific
disease, suggesting a novel pathway to retinal degeneration.
There are several potential explanations for the retinal de-
generation in splicing-factor RP. Splicing-factor RP could result
from haploinsufficiency of functional splicing factors. Sub-op-
timal levels of pre-mRNA processing factors could lead to
retinal damage because the retina is among the most biosyn-
thetically active tissues in the body. A second hypothesis is that
spliceosomes containing the mutant factors are competent to
splice most introns correctly, but when assembled on introns
that, even under normal circumstances, splice relatively ineffi-
ciently, the mutant proteins compromise the correct splicing
of those introns. The correlate to this hypothesis is that one or
more vital retinal transcripts contain these introns, which then
lead to RP when mis-spliced. The third hypothesis is that the
mutations result in a gain of function to these factors which is
toxic to the retina (for review see Ref. 23).
Our laboratories have produced mouse models to study the
effects of mutations in PRPF3, PRPF8 and PRPF31 on vision. A
recent report from one of our groups described Prpf3 knock-
out mice, in which heterozygous animals had no evident retinal
phenotype at two years of age, arguing against the first hypoth-
esis of haploinsufficiency.24In contrast, homozygosity of the
Prpf31-A216P knockin allele caused early embryonic lethality,
consistent with the idea that a majority of mutations in PRPF31
result in null alleles and the disease is caused by haploinsuffi-
ciency of this splicing factor.4,25–33Heterozygous Prpf31?/A216P
and Prpf31?/?mice did not develop photoreceptor dysfunction
at ages up to 18 months.33Here we describe generation and
characterization of Prpf3 knockin mice bearing the p.T494M
missense mutation found in RP18 patients, and Prpf8 knockin
mice with the p.H2309P missense mutation found in RP13.3,5
Additionally, we report ultrastructural characterization of the het-
erozygous Prpf31 knockout (Prpf31?/?) mouse model.33We
find that at aged time points, all three animal models display
degenerative changes in the retinal pigment epithelium (RPE).
The RPE degeneration is associated with decreased retinal func-
tion in the Prpf3-T494M mice.
MATERIALS AND METHODS
Animal research was performed under protocols approved by the
Institutional Animal Care and Use Committees at the University of
Pennsylvania and the Centre de Recherche Institut de la Vision, Uni-
versite ´ Pierre et Marie Curie-Paris, and conforms to the ARVO State-
ment for Use of Animals in Ophthalmic and Vision Research. Wild-Type
C57BL/6J mice and FLPe deleter mice were obtained from Jackson
Laboratories, (Bar Harbor, ME).34
Production of Targeting Vectors
Prpf3-T494M. We used homologous recombination to screen a
lambda phage mouse genomic library as in Zhang et al.35Using a floxed
tetracycline resistance cassette flanked by regions of homology to the
10thintron of the Prpf3 gene, we isolated five clones spanning differ-
ent regions of the C terminus of Prpf3. Of these clones, one contained
a 10 kb insert spanning exons 10–16. This region of Prpf3 is the most
highly conserved and contains exon 11, which is the location of the
p.T494M mutation found in RP18 patients. To make the knockin vector
(See Fig. 1) we transformed this clone into bacteria expressing cre-
recombinase to remove the tetracycline cassette.35We then used
oligonucleotide-directed mutagenesis to mutate the 11th exon, result-
ing in the p.T494M mutation (QuikChange, Stratagene, LaJolla, CA). A
silent base change which eliminates an XmnI digest site was also
included for rapid genotyping purposes. Next, another round of ho-
mologous recombination was performed to insert the positive selec-
tion cassette for targeting in ES cells using short regions of homology
to the 10th intron to create the final targeting vector for the knockin36
Prpf8-H2309P. We created the Prpf8-H2309P targeting vector
using a similar approach to that described above, starting with a BAC
clone which contained the Prpf8 gene. We again used recombineering
techniques to isolate a 16 kb portion of the Prpf8 gene containing
exons 30 through 42 (Fig. 1H). We added appropriate selection cas-
settes, and mutated exon 42 to introduce the p.H2309P substitution
Transfection of ES Cells, Verification of Targeting,
and Production of Mice
After sequence verification that the knock-in constructs contained the
desired mutation and that the rest of the exons were correct, the
linearized vectors were electroporated into 129SvEvTac mouse embry-
onic stem (ES) cells.37Embryonic stem cells were cultured in embry-
onic stem cell medium (Dulbecco[b]’s Modified Eagle[b]’s Medium
(DMEM; GIBCO; Gaithersburg, MD) with 15% fetal bovine serum (Hy-
clone Laboratories, Logan, UT), 1% non essential amino acids (GIBCO),
0.1 mM ?-mercaptoethanol (Sigma, St. Louis, MO), and 1250 U/mL
leukemia inhibitory factor (Chemicon International, Temecula, CA) on
a primary mouse embryo fibroblast monolayer (Chemicon). G418- and
ganciclovir-resistant clones were isolated and expanded for verification
of targeting by Southern blot analysis using probes in the 5? and 3?
surrounding region outside the targeting vector. The presence of the
T494M and H2309P mutations was verified by PCR followed by se-
quencing. Two correctly targeted clones for each gene were injected
into C57BL/6J blastocysts to produce chimeric mice at the Transgenic
and Chimeric Mouse Core Facility at the University of Pennsylvania.38
Highly chimeric founder mice for each targeted allele were back-
crossed with C57BL/6J mice. Agouti F1 progeny were screened for the
targeted Prpf3 and Prpf8 alleles by PCR and Southern blotting. We
obtained germline transmission from at least two founder mice for
both targeted alleles. The F1 Prpf3-Neo-T494M and Prpf8-Neo-H2309P
mice were crossed with FLPeR mice to remove the neomycin selection
cassettes.34Excised Prpf3-T494M and Prpf8-H2309P mice were back-
crossed with C57BL/6J mice to generate heterozygous F1 mice. Male
and female F1 Prpf3-T494M or Prpf8-H2309P mice were inter-crossed
to generate homozygous F2 mice. Expected Mendelian ratios were
obtained from these crosses, and the resultant mice were healthy and
fertile, with normal lifespan. The Prpf31?/?mice were generated as
Genotyping of Mice
Genotyping of Prpf3-T494M and Prpf8-H2309P knockin mice and
Prpf31 knockout mice was performed by Southern blotting and PCR.
For Prpf3 Southern blotting, 2 probes were used. Probe 1 was ampli-
fied from genomic DNA 5? to the targeting vector using primers:
forward 5?-CATTCTTAGCATTAAAGCAACAAAATC-3? and reverse 5?-
ACAGACCCAATCAGCTAAACCACCTATC-3?. It detects a wild-type SpeI
fragment of 8 kb and a targeted 6 kb fragment. Probe 2 was amplified
3? to the targeting vector using primers: forward 5?-GCCCATGACCCG-
TACGGTGTGTGTATC-3? and reverse 5?-TGCTTCACGTTCCCTGCAT-
GTTCCATCC-3?. It detects a 16.9 kb wild-type BstXI fragment versus a
19.5 kb targeted BstXI fragment. For Prpf8 Southern blotting Probe1
was amplified 5? to the vector using primers: forward 5?-CCCT-
TCAGCTCGTCCCTGGTTTGAG-3? and reverse 5?-GATCTAGAAAAT-
ATCCAACTCCACAACTTACTGC-3?. It detects a wild-type NotI frag-
ment of 17 kb and a 10 kb targeted fragment.
IOVS, January 2011, Vol. 52, No. 1
RNA Splicing Factor RP Mice 191
For genotyping by PCR, the following primer sets were used. For
Prpf3-T494M mice, amplification with forward primer 5?-CTCATTCAT-
TGGCATTAAAAATAAATAAACTCC-3? and reverse primer 5?-CGTTG-
GCCTCTTCATGCGCTCTGTCGTGAC-3?, followed by digestion of the
PCR products with Xmn1 was used. The presence of a band at 649 bp
signifies the mutant allele. Prpf8 mice were genotyped using primers
Prpf3-T494M Knockin Vector
S = SpeI site
B = BstXI site
G A G G T C C N C A G G C C C
G A G G T C C A C A G G C C C
G A G G T C C C C A G G C C C
Prpf8-H2309P Knockin Vector
and Prpf8-H2309P knockin mice. (A)
Exons 9 to 16 of the Prpf3 locus with
restriction sites indicated. (B) Knockin
vector with T494M mutation, and pos-
itive and negative selection cassettes.
(C) Targeted Prpf3-Neo-T494M locus.
(D) Final targeted Prpf3-T494M
knockin allele. (E) Southern blot
analysis of wild-type and Prpf3-Neo-
T494M shown in (C) using indicated
5? and 3? probes for SpeI and BstXI
digests, respectively. (F) Genomic se-
quence data from wild-type, het-
erozygous, and homozygous Prpf3-
T494M mice. (G) Exons 25 to 42 of
the Prpf8 locus. (H) Prpf8 Knockin
vector containing exons 30–42, the
H2309P mutation, and selection cas-
sette for ES targeting. (I) Targeted
Prpf8-Neo-T494M locus. (J) Final tar-
geted Prpf8-H2309P knockin allele.
(K) Southern blot analysis of targeted
Prpf8 knockin mice detects the
shorter NotI fragment in targeted
mice. (L) Sequence analysis of wild-
type and targeted Prpf8 genomic
DNA showing replacement of the en-
dogenous codon with the mutant
Targeting of Prpf3-T494M
192Graziotto et al.
IOVS, January 2011, Vol. 52, No. 1
located in the 3? UTR (forward primer 5?-CCAGACAAGCTGCTGACAT-
TCAGCAGTC-3? and reverse primer: 5?-TAGAGCTGTAAGTGGTCACT-
TCAGGC-3?). The mutant allele is 400 bp larger than the wild-type.
Prpf31?/?mice were identified by PCR using primers in exons 4 and
8 (forward primer: 5?-CTCCTGAGTACCGAGTCATTGTGGATGC; re-
verse primer: 5?-GTAGAAGAGAAGCCAGACAGGGTCTTGC). The mu-
tant allele was 897 bp shorter than the wild-type.
Four or more retinas of 4-week-old mice of the indicated genotype
were pooled and RNA isolated using reagent (Trizol; Invitrogen, Carls-
bad, CA). Fifteen to 20 micrograms of total RNA were loaded per lane
on a denaturing 0.8% agarose gel, transferred overnight to a nylon
membrane (Schleicher & Schuell, Keene, NH), cross-linked and stored
or hybridized according to standard protocols.39A 652-bp radiolabeled
probe against the mouse Prpf3 transcript was amplified from cDNA
using primers: forward 5?-CAGATGATGGAAGCAGCAACACGAC-3? and
reverse 5?-TTCTAGCAGCTTGTGAAATCTCT-3?. This probe spans ex-
ons 5 to 8 of the Prpf3 transcript. To detect the mouse Prpf8 tran-
script, a 1300-bp probe was amplified using primers: forward 5?-
GCTGCCGGATTATATGTCAGAGGAGAAGC-3? and reverse 5?-GTAC-
TTAAGCAGCTTCTGATAAGAGACTCG-3?. This probe spans exons 1 to
9 of the Prpf8 transcript. To assess total RNA per lane, a probe against
the housekeeping gene acidic ribosomal phosphoprotein P0 (36B4)
was used.40Probes were hybridized to membranes overnight at 65°C,
washed, and exposed to phosphor screens for detection. Phosphor
screens were scanned using a phosphorimager (Storm Phosphorim-
ager, GE Healthcare, Piscataway, NJ) and quantified using imaging
software (ImageQuant 5.2, GE Healthcare).
Retinas were solubilized by sonication in LDS sample buffer (Invitro-
gen) and 100 micrograms of reduced protein were separated in each
lane of 3–8% Tris-Acetate polyacrylamide gel (NuPage; Invitrogen).
Proteins were transferred electrophoretically to polyvinylidene difluo-
ride (PVDF) membrane (Invitrogen), and blocked in 10% nonfat dry
milk solution for 1 hour at room temperature. Primary antibodies
against Prpf3 protein were a generous gift from James Hu.41Alkaline
phosphatase conjugated anti-rabbit secondary antibodies (Vector Lab-
oratories, Burlingame, CA) were used in conjunction with ECF reagent
(Amersham Biosciences, Piscataway, NJ) and blots were scanned with
an imager (Storm Phosphorimager or Typhoon Variable Mode Imager;
GE Healthcare). Band intensities were quantified (ImageQuant 5.2
software; GE Healthcare).
Electroretinography was performed as previously described.42Briefly,
full-field ERG[b]’s were recorded in a Ganzfeld on dark-adapted, anes-
thetized mice taking care to maintain 37°C body temperature at all
times. Pupils were dilated with 1% tropicamide. Retinal responses
were detected with platinum electrodes embedded in contact lenses
contacting the cornea and recorded using custom software.
Light and Electron Microscopy
Preparation of retinas for light and electron microscopy was performed
as previously described.24,43For histologic analysis of the retina ani-
mals were perfused with 4% paraformaldehyde in phosphate buffered
saline (PBS; Electron Microscopy Sciences, Hatfield, PA), and eyecups
were processed for cryosectioning. Ten-micron sections were cut,
mounted onto slides (Superfrost Plus, Fisher Scientific, Pittsburgh, PA),
and stained with alkaline toluidine blue for light microscopy. For
co-localization studies, antibody Y12 (ab3138; Abcam, Cambridge, MA)
was used for detection of snRNPs. For the measurement of ONL
thickness, a section from 12 o[b]’clock, through the optic nerve head,
and 6 o[b]’clock was used and measurements were taken at 10 inter-
vals in both directions toward the periphery starting at the optic nerve
head. For electron microscopy, perfused eyecups were transferred to
2% paraformaldehyde ? 2% glutaraldehyde in 0.2 M sodium cacody-
layte buffer (pH 7.4) and further processed for plastic sectioning in
EMbed812 (Electron Microscopy Sciences). One micrometer sections
were then cut and stained with alkaline toluidine blue for light micros-
copy and 60 to 80 nM ultrathin sections were stained with lead citrate
/uranyl acetate and examined using a transmission electron micro-
scope (FEI Tecnai, Hillsboro, OR).
Generation of Prpf3-T494M and Prpf8-H2309P Mice.
We generated Prpf3-T494M and Prpf8-H2309P knockin mice
using standard protocols as described in Materials and Methods
(Fig. 1). After removal of the selection cassettes via crosses
with FLPe deleter mice, heterozygous and homozygous
knockin mice were generated for the experiments described.
Both lines of homozygous knockin mice were healthy, and
fertile. The generation of the Prpf31?/?mice has been de-
Expression of Prpf3 and Prpf8. Northern blot analysis
and qRT-PCR analyses showed that the sizes and levels of the
mutant Prpf3 and Prpf8 transcripts produced in the retinas of
the homozygous Prpf3-T494M and Prpf8-H2309P mice were
similar to that observed in controls (Figs. 2A, 2B, 2C). Similarly,
Western blot analysis showed that the level of Prpf3 protein in
the retinas of homozygous knockin mice was also normal (Fig.
2D). No suitable antibodies for mouse Prpf8 were available to
perform this analysis in the Prpf8-H2309P line.
We asked whether the presence of the T494M mutation in
the Prpf3 protein alters its subcellular localization, as has been
reported following over-expression of mutant protein in cul-
tured cells.44Prpf3 was detected in photoreceptor inner seg-
ments, the outer nuclear layer, and in the inner nuclear layer of
the retina, consistent with its ubiquitous nature. This localiza-
tion was unchanged in the retinas of homozygous knockin
mice (Fig. 2E). Furthermore, we performed a co-localization
study with monoclonal antibody Y1245which recognizes Sm
proteins in snRNPs, to determine whether the mutation alters
the inclusion of Prpf3 into nuclear speckles, which are sites of
snRNP storage, assembly and modification (for review see Ref.
46). We found an interesting staining pattern in photoreceptor
nuclei, for both Prpf3 and the Y12 antibody. The staining
concentrates to nuclear speckles in the periphery of the nu-
cleus in photoreceptors, but in non-photoreceptor cells dis-
plays the more typical pattern of speckles throughout the
nuclei. In both cases we found that in wild-type versus ho-
mozygous knockin retina, the overlap of the two signals was
unchanged, indicating that the mutant Prpf3 is able to enter the
nucleus and incorporate into snRNPs (Fig. 2F).
Retinal Function. Electroretinography was used to evalu-
ate the retinal function of the Prpf3 and Prpf8 knockin mice.42
We measured scotopic a-waves and b-waves to analyze the rod
photoreceptors and bipolar cells, respectively, as well as the
photopic b-waves to assess cone function. We compared wild-
type, heterozygous and homozygous knockin mice at various
time points up to 24 months of age. At time points up to 18
months, no significant differences were found between the
wild-type and knockin animals (data not shown). However, at
the oldest time point of 24 months, the maximal rod a-wave of
Prpf3 heterozygous and homozygous knockin mice was signif-
icantly decreased (Fig. 3A). Although the rod b-waves were
decreased as well, this decrease did not reach statistical signif-
icance. No significant differences were found in the cone
b-wave at this time point (Figs. 3B, 3C). No significant differ-
ences were found for Prpf8 knockin mice compared to Prpf8
IOVS, January 2011, Vol. 52, No. 1
RNA Splicing Factor RP Mice 193
Retinal Morphology and Ultrastructure. Light micros-
copy showed that there was no major photoreceptor degenera-
tion at any age tested in Prpf3-T494M (Fig. 4A) or Prpf8-H2309P
(Fig. 4B) mice. Similar results were also reported for 1-year-old
Prpf31?/?mice.33However, ultrastructural analyses did reveal
significant differences between the wild-type and knockin ani-
mals in the retinal pigment epithelium (RPE) for the Prpf3 and
Prpf8 mutant mouse lines at two years of age, and the Prpf31
mice at one year of age (Fig. 5). The wild type RPE appears
normal, with long apical microvilli, interdigitating with the pho-
toreceptor outer segments, and visible basal infoldings on the
basal side of the RPE cells (Fig. 5). In contrast, the RPE cells of
the heterozygous Prpf3-T494M mice and Prpf8-H2309P mice
exhibit loss of the basal infoldings and accumulation of amor-
phous deposits between the RPE and Bruch[b]’s membrane
(Figs. 5A, 5B). There are also vacuoles in the RPE cells of the
mutant mice. These changes are more severe in the homozy-
gous Prpf3-T494M and Prpf8-H2309P mice. Photoreceptor
cells in the knockin mice appear to be normal in histologic and
ultrastructural analyses. One-year-old Prpf31?/?mice also
demonstrate degenerative changes in the RPE, with loss of the
basal infoldings and accumulation of amorphous deposits be-
tween the RPE and Bruch’s membrane (Fig. 5C). Analysis
showed that all the mutant mice develop loss of basal infold-
ings and accumulation of deposits between the RPE and
Bruch’s membrane (n ? 3 per mutant line), compared to none
of the control mice (n ? 3 per mutant line).
The results presented above provide several important insights
into the pathogenesis of RNA splicing factor RP. First, all three
mouse models manifest degenerative changes in the RPE. The
finding of a similar phenotype in all three models suggests that
the RPE is the primary retinal cell type affected in these forms
of RNA splicing factor RP, which in turn leads to photorecep-
tor degeneration over the long term. The RPE changes are
associated with decreased photoreceptor function in the
Prpf3-T494M mice, although overt photoreceptor degenera-
tion was not observed in any of the models. The later-onset
phenotype in these three mouse models is consistent with the
adult-onset phenotype and vision loss observed in many pa-
tients with RNA splicing factor forms of RP.7,47,48These three
qRT-PCR of Prpf3 and Prpf8 normalized to β-actin
Prpf8 knockin transcripts and associ-
ated proteins. (A) Northern blot anal-
ysis of RNA from four wild-type and
four homozygous Prpf3-T494M reti-
nas. (B) Northern blot analysis of
four wild-type and four homozygous
green qRT-PCR for Prpf3 and Prpf8
from wild-type versus knockin retina
(n ? 3). (D) Western blotting of
Prpf3 from wild-type, heterozygous,
and homozygous knockin retina. (E)
Frozen sections of wild-type and
Prpf3-T494M/T494M retinas stained
with antibodies to Prpf3 (red). The
Prpf3 protein is located in the inner
segment, outer nuclear layer, and in-
ner nuclear layer in both wild-type
and homozygous knockin animals.
(F) Confocal microscopy of frozen
sections co-stained with Prpf3 (red)
and antibody Y12 (green). Prpf3 lo-
calizes to nuclear speckles detected
by Y12 in the periphery of photore-
ceptor nuclei (ONL panels) and
throughout the nuclei of the inner
nuclear layer (INL panels). The loca-
tion of Prpf3 is the same in wild-type
and Prpf3-T494M/T494M animals. IS,
inner segment; ONL, outer nuclear
layer; INL, inner nuclear layer.
Expression of Prpf3 and
194Graziotto et al.
IOVS, January 2011, Vol. 52, No. 1
mouse models provide a platform for future comparative stud-
ies to elucidate the mechanism of pathogenesis of the RNA
splicing factor forms of RP.
An interesting finding from this work is that all the Prpf
mutant mice develop loss of RPE basal infoldings and sub-RPE
deposits. The deposits share features with the basal deposits
associated with macular degeneration, including membranous
debris and vesicular structures.49Both the heterozygous and
homozygous mice accumulate these deposits, in agreement
with the dominant inheritance of the human conditions. Ultra-
structural analysis of the retina from one patient with RNA
splicing factor RP has been reported. In this sample, from a
patient with a mutation in PRPF8, the RPE was clearly abnor-
mal with loss of basal infoldings, but since the patient had
end-stage RP it is difficult to determine whether the RPE ab-
normalities were a cause or consequence of photoreceptor
loss.50,51While other forms of RP are not known to develop
sub-RPE deposits, the RPE and photoreceptors have an intimate
relationship and there are several types of retinal degeneration
caused by mutations in genes required for RPE function. Some
examples are LCA caused by mutations in LRAT and RPE65 and
RP caused by MERTK and RGR.52–55Further, RPE phagocytic
dysfunction in ?5 integrin-deficient mice leads to photorecep-
tor dysfunction, and accumulation of lipofuscin in RPE cells.56
The RPE is also a major part of the blood-retinal barrier, and
therefore regulates the extracellular environment of photore-
ceptors. One recent study reported that mice which lack the
RPE monocarboxylic acid transporter 3 have an altered pH in
the sub-retinal space, and that this leads to altered photorecep-
tor function.57One possible explanation for the decreased
a-wave ERG observed in the older Prpf3-T494M mice is that the
RPE defects observed lead to a similar alteration in the extra-
cellular milieu of photoreceptor outer segments. In short, the
RPE and the retina are integrally related, and problems in one
can have consequences for the other. Additional studies of the
effects of mutations in the three RNA splicing factors on RPE
function and its relationship to photoreceptor degeneration
The homozygous Prpf3-T494M and Prpf8-H2309P knockin
mouse lines are viable, demonstrating that the Prpf3-T494M
and Prpf8-H2309P mutations do not create null alleles, or the
animals would have died embryonically, as was recently re-
ported for Prpf3 knockout animals, Prpf8 knockout animals,
and Prpf31 knockin and knockout animals (Deramaudt BM, et
al. IOVS 2005;46:ARVO E-Abstract 5263).24,33The use of gene-
targeted knockin mice for these studies provided several im-
portant advantages over other methods that have been used to
study the RNA splicing factor forms of RP to date,44,58includ-
ing the ability to study the Prpf3-T494M protein and the Prpf8-
H2309P protein in the absence of wild-type Prpf3 or Prpf8 in
vivo, which cannot be readily accomplished in cell culture. We
have observed that in neither case do the expression levels of
the mutant Prpf3 or Prpf8 transcripts change, nor does the size
of the transcript, ruling out the possibility of the mutation
altering a splice signal within the Prpf3 or Prpf8 transcripts
themselves. Furthermore, the levels and location of Prpf3-
T494M protein within the retina are normal, and no nuclear
24-month-old mice. Significant decreases were found in Prpf3-Wt/T494M
(n ? 12, P ? 0.0149 via two-tailed Student’s t-test) and Prpf3-T494M/
T494M (n ? 11, P ? 0.0135)) compared to wild-type (n ? 9). No
significant changes were found in Prpf8 mice (Wt/Wt, n ? 12, Prpf8-Wt/
H2309P, n ? 6, Prpf8-H2309P/H2309P, n ? 12). (B) Scotopic b-wave
amplitudes as measured from baseline to b-wave peak were decreased in
Prpf3 knockin mice relative to wild-type, but this change did not reach
significance. (C) No significant differences were found in photopic b-
wave responses. Error bars indicate SEM.
Retinal Function. (A) Maximum scotopic a-wave responses in
Thick plastic retinal sections
stained with toluidine blue. (A) 24-
month-old wild-type, Prpf3-Wt/
T494M, and Prpf3-T494M/T494M
mice. (B) 23-month-old wild-type,
H2309P/H2309P mice. Normal
ONL thickness and photoreceptor
outer segment length indicates no
major loss of photoreceptors at
these time points.
IOVS, January 2011, Vol. 52, No. 1
RNA Splicing Factor RP Mice195
aggregates were seen, in contrast to a recent study in which
aggregation of mutant PRPF3 was seen after over-expression of
the protein.44These findings are also consistent with the hy-
pothesis that mutations in these genes produce disease via a
dominant mechanism (dominant-negative or gain-of-function),
rather than haploinsufficiency.
The location of the Prpf3 and Sm proteins in the periphery
of the nuclei of photoreceptor cells is distinct from the typical
staining pattern for splicing proteins, which localize to speck-
les and Cajal bodies throughout the nucleus but exclude nu-
cleoli. The location of Prpf3 and Sm proteins in the mouse is
still speckled but peripheral, consistent with recent findings
that nocturnal mammal photoreceptor nuclei have a unique
inverted pattern with a heterochromatin center surrounded by
euchromatin, nascent transcripts as well as the splicing ma-
chinery.59However, the mutant Prpf3-T494M location does
not differ from that of the wild-type, implying no nuclear
import defects or defects incorporating into the snRNPs.
The findings described here provide a new model for the
pathogenesis of RNA splicing RP. Although the characteristic
loss of photoreceptors is the defining feature of RP, our mouse
models suggest that photoreceptor dysfunction in RNA splic-
ing factor RP may arise secondary to an unhealthy RPE. Given
these findings, we hypothesize that production of an aberrantly
spliced transcript or group of transcripts in RPE and/or photo-
receptor cells is responsible for the retinal degeneration phe-
notype observed in patients with RNA splicing factor RP. We
believe analyses of the transcriptomes of the RPE and retinas of
the Prpf3-T494M, Prpf8-H2309P, and Prpf31-knockout mice
will help identify the pathogenic splice alterations responsible
for retinal disease. This is a worthwhile endeavor, because
identification of the altered transcripts that cause retinal de-
generation will open the path to the development of therapies
for these blinding disorders. Indeed, there are now several
promising examples of the use of antisense oligonucleotides to
correct splicing errors caused by mutations in vivo, including a
recent report of efficacy in a small clinical trial.60,61
The authors thank Qin Liu (F.M. Kirby Center) and Christina Zeitz
(Institut de la Vision) for research advice; and Marian Humphries and
Peter Humphries (Trinity College, Dublin) for their help with creating
Prpf31?/?line and animal husbandry.
1. Jurica MS, Moore MJ. Pre-mRNA splicing: awash in a sea of pro-
teins. Mol Cell. 2003;12:5–14.
2. Wahl MC, Will CL, Luhrmann R. The spliceosome: design princi-
ples of a dynamic RNP machine. Cell. 2009;136:701–718.
tron microscopy of retina and RPE
from (A) two-year-old Prpf3-T494M
and littermate control mice, (B) two-
year-old Prpf8-H2309P and littermate
control mice, and (C) one-year-old
Prpf31 knockout and control mice.
In all three mouse models, loss of basal
infoldings and accumulation of amor-
phous material between Bruch’s mem-
brane and the RPE is evident (arrows).
These changes are shown in more
detail in the insets at the bottom of
each image, with the deposits out-
lined. Vacuoles are also present in
the RPE of the mutant mice. These
changes were not detected in the
control mice. All images were taken
at the same magnification (?1100).
Scale bars: 2 ?m. BI, basal infoldings;
BrM, Bruch’s membrane; OS, outer
segments; RPE, retinal pigment epi-
RPE Ultrastructure. Elec-
196 Graziotto et al.
IOVS, January 2011, Vol. 52, No. 1
3. Chakarova CF, Hims MM, Bolz H, et al. Mutations in HPRP3, a third
member of pre-mRNA splicing factor genes, implicated in autosomal
dominant retinitis pigmentosa. Hum Mol Genet. 2002;11:87–92.
4. Vithana EN, Abu-Safieh L, Allen MJ, et al. A human homolog of
yeast pre-mRNA splicing gene, PRP31, underlies autosomal domi-
nant retinitis pigmentosa on chromosome 19q13.4 (RP11). Mol
5. McKie AB, McHale JC, Keen TJ, et al. Mutations in the pre-mRNA
splicing factor gene PRPC8 in autosomal dominant retinitis pig-
mentosa (RP13). Hum Mol Genet. 2001;10:1555–1562.
6. Maita H, Kitaura H, Keen TJ, Inglehearn CF, Ariga H, Iguchi-Ariga
SM. PAP-1, the mutated gene underlying the RP9 form of dominant
retinitis pigmentosa, is a splicing factor. Exp Cell Res. 2004;300:
7. Zhao C, Bellur DL, Lu S, et al. Autosomal-dominant retinitis pig-
mentosa caused by a mutation in SNRNP200, a gene required for
unwinding of U4/U6 snRNAs. Am J Hum Genet. 2009;85:617–627.
8. Lauber J, Plessel G, Prehn S, et al. The human U4/U6 snRNP
contains 60 and 90kD proteins that are structurally homologous to
the yeast splicing factors Prp4p and Prp3p. RNA. 1997;3:926–941.
9. Anthony JG, Weidenhammer EM, Woolford JL, Jr. The yeast Prp3
protein is a U4/U6 snRNP protein necessary for integrity of the
U4/U6 snRNP and the U4/U6.U5 tri-snRNP. RNA. 1997;3:1143–
10. Grainger RJ, Beggs JD. Prp8 protein: at the heart of the spliceo-
some. RNA. 2005;11:533–557.
11. Makarova OV, Makarov EM, Liu S, Vornlocher HP, Luhrmann R.
Protein 61K, encoded by a gene (PRPF31) linked to autosomal
dominant retinitis pigmentosa, is required for U4/U6*U5 tri-snRNP
formation and pre-mRNA splicing. EMBO J. 2002;21:1148–1157.
12. Schaffert N, Hossbach M, Heintzmann R, Achsel T, Luhrmann R.
RNAi knockdown of hPrp31 leads to an accumulation of U4/U6
di-snRNPs in Cajal bodies. EMBO J. 2004;23:3000–3009.
13. Maita H, Kitaura H, Ariga H, Iguchi-Ariga SM. Association of PAP-1 and
Prp3p, the products of causative genes of dominant retinitis pigmen-
tosa, in the tri-snRNP complex. Exp Cell Res. 2005;302:61–68.
14. van Nues RW, Beggs JD. Functional contacts with a range of
splicing proteins suggest a central role for Brr2p in the dynamic
control of the order of events in spliceosomes of Saccharomyces
cerevisiae. Genetics. 2001;157:1451–1467.
15. Bunker CH, Berson EL, Bromley WC, Hayes RP, Roderick TH.
Prevalence of retinitis pigmentosa in Maine. Am J Ophthalmol.
16. Xu L, Hu L, Ma K, Li J, Jonas JB. Prevalence of retinitis pigmentosa
in urban and rural adult Chinese: The Beijing Eye Study. Eur J
17. Grondahl J. Estimation of prognosis and prevalence of retinitis
pigmentosa and Usher syndrome in Norway. Clin Genet. 1987;31:
18. Haim M. Epidemiology of retinitis pigmentosa in Denmark. Acta
Ophthalmol Scand Suppl. 2002:1–34.
19. Berson EL. Retinitis pigmentosa. The Friedenwald Lecture. Invest
Ophthalmol Vis Sci. 1993;34:1659–1676.
20. Pierce EA. Pathways to photoreceptor cell death in inherited
retinal degenerations. Bioessays. 2001;23:605–618.
21. Sullivan LS, Bowne SJ, Birch DG, et al. Prevalence of disease-
causing mutations in families with autosomal dominant retinitis
pigmentosa: a screen of known genes in 200 families. Invest
Ophthalmol Vis Sci. 2006;47:3052–3064.
22. Hartong DT, Berson EL, Dryja TP. Retinitis pigmentosa. Lancet.
23. Mordes D, Luo X, Kar A, et al. Pre-mRNA splicing and retinitis
pigmentosa. Mol Vis. 2006;12:1259–1271.
24. Graziotto JJ, Inglehearn CF, Pack MA, Pierce EA. Decreased levels
of the RNA splicing factor Prpf3 in mice and zebrafish do not cause
photoreceptor degeneration. Invest Ophthalmol Vis Sci. 2008;49:
25. Wang L, Ribaudo M, Zhao K, et al. Novel deletion in the pre-mRNA
splicing gene PRPF31 causes autosomal dominant retinitis pigmen-
tosa in a large Chinese family. Am J Med Genet A. 2003;121A:235–
26. Xia K, Zheng D, Pan Q, et al. A novel PRPF31 splice-site mutation
in a Chinese family with autosomal dominant retinitis pigmentosa.
Mol Vis. 2004;10:361–365.
27. Sato H, Wada Y, Itabashi T, Nakamura M, Kawamura M, Tamai M.
Mutations in the pre-mRNA splicing gene, PRPF31, in Japanese
families with autosomal dominant retinitis pigmentosa. Am J Oph-
28. Abu-Safieh L, Vithana EN, Mantel I, et al. A large deletion in the
adRP gene PRPF31: evidence that haploinsufficiency is the cause
of disease. Mol Vis. 2006;12:384–388.
29. Rivolta C, McGee TL, Rio FT, Jensen RV, Berson EL, Dryja TP.
Variation in retinitis pigmentosa-11 (PRPF31 or RP11) gene expres-
sion between symptomatic and asymptomatic patients with dom-
inant RP11 mutations. Hum Mutat. 2006;27:644–653.
30. Sullivan LS, Bowne SJ, Seaman CR, et al. Genomic rearrangements
of the PRPF31 gene account for 2.5% of autosomal dominant
retinitis pigmentosa. Invest Ophthalmol Vis Sci. 2006;47:4579–
31. Waseem NH, Vaclavik V, Webster A, Jenkins SA, Bird AC, Bhatta-
charya SS. Mutations in the gene coding for the pre-mRNA splicing
factor, PRPF31, in patients with autosomal dominant retinitis pig-
mentosa. Invest Ophthalmol Vis Sci. 2007;48:1330–1334.
32. Rio FT, Wade NM, Ransijn A, Berson EL, Beckmann JS, Rivolta C.
Premature termination codons in PRPF31 cause retinitis pigmen-
tosa via haploinsufficiency due to nonsense-mediated mRNA de-
cay. J Clin Invest. 2008;118:1519–1531.
33. Bujakowska KM, Maubaret C, Chakarova CF, et al. Study of gene-
targeted mouse models of splicing factor gene Prpf31 implicated
in human autosomal dominant retinitis pigmentosa (RP). Invest
Ophthalmol Vis Sci. 2009;50:5927–5933.
34. Rodriguez CI, Buchholz F, Galloway J, et al. High-efficiency deleter
mice show that FLPe is an alternative to Cre-loxP. Nat Genet.
35. Zhang P, Li MZ, Elledge SJ. Towards genetic genome projects:
genomic library screening and gene-targeting vector construction
in a single step. Nat Genet. 2002;30:31–39.
36. Lee EC, Yu D, Martinez de Velasco J, et al. A highly efficient
Escherichia coli-based chromosome engineering system adapted
for recombinogenic targeting and subcloning of BAC DNA.
37. Matise MP, Auerbach W, Joyner AL. Production of targeted embry-
onic stem cell lines. In: Joyner AL, ed. Gene Targeting: A Practical
Approach, 2nd ed. Oxford, UK: Oxford University Press; 2000:
38. Nagy A, Gertsenstein M, Vintersten K, Behringer RR. Manipulat-
ing the Mouse Embryo, 3rd ed. Plainview, NY: Cold Spring Harbor
Laboratory Press; 2003.
39. Sambrook J, Fritsch EF, Maniatis T. Molecular Cloning: A Labora-
tory Manual. Plainview, NY: Cold Spring Harbor Laboratory Press;
40. Laborda J. 36B4 cDNA used as an estradiol-independent mRNA
control is the cDNA for human acidic ribosomal phosphoprotein
PO. Nuc Acid Res. 1991;19:3998.
41. Gonzalez-Santos JM, Wang A, Jones J, Ushida C, Liu J, Hu J. Central
region of the human splicing factor Hprp3p interacts with
Hprp4p. J Biol Chem. 2002;277:23764–23772.
42. Lyubarsky AL, Falsini B, Pennesi ME, Valentini P, Pugh EN, Jr. UV-
and midwave-sensitive cone-driven retinal responses of the mouse:
a possible phenotype for coexpression of cone photopigments.
J Neurosci. 1999;19:442–455.
43. Liu Q, Lyubarsky A, Skalet JH, Pugh EN, Jr, Pierce EA. RP1 is
required for the correct stacking of outer segment discs. Invest
Ophthalmol Vis Sci. 2003;44:4171–4183.
44. Comitato A, Spampanato C, Chakarova C, Sanges D, Bhattacharya
SS, Marigo V. Mutations in splicing factor PRPF3, causing retinal
degeneration, form detrimental aggregates in photoreceptor cells.
Hum Mol Genet. 2007;16:1699–1707.
45. Lerner EA, Lerner MR, Janeway CA, Jr, Steitz JA. Monoclonal
antibodies to nucleic acid-containing cellular constituents: probes
for molecular biology and autoimmune disease. Proc Natl Acad Sci
U S A. 1981;78:2737–2741.
46. Lamond AI, Spector DL. Nuclear speckles: a model for nuclear
organelles. Nat Rev Mol Cell Biol. 2003;4:605–612.
IOVS, January 2011, Vol. 52, No. 1
RNA Splicing Factor RP Mice 197
47. Towns KV, Kipioti A, Long V, et al. Prognosis for splicing factor Download full-text
PRPF8 retinitis pigmentosa, novel mutations and correlation be-
tween human and yeast phenotypes. Hum Mutat. 2010.
48. Vaclavik V, Gaillard MC, Tiab L, Schorderet DF, Munier FL. Variable
phenotypic expressivity in a Swiss family with autosomal domi-
nant retinitis pigmentosa due to a T494M mutation in the PRPF3
gene. Mol Vis. 2010;16:467–475.
49. Curcio CA, Millican CL. Basal linear deposit and large drusen are
specific for early age-related maculopathy. Arch Ophthalmol.
50. Szamier RB, Berson EL. Retinal ultrastructure in advanced retinitis
pigmentosa. Invest Ophthalmol Vis Sci. 1977;16:947–962.
51. To K, Adamian M, Berson EL. Histologic study of retinitis pigmen-
tosa due to a mutation in the RP13 gene (PRPC8): comparison with
rhodopsin Pro23His, Cys110Arg, and Glu181Lys. Am J Ophthal-
52. den Hollander AI, Roepman R, Koenekoop RK, Cremers FP. Leber
congenital amaurosis: genes, proteins and disease mechanisms.
Prog Retin Eye Res. 2008;27:391–419.
53. Morimura H, Saindelle-Ribeaudeau F, Berson EL, Dryja TP. Mutations
in RGR, encoding a light-sensitive opsin homologue, in patients with
retinitis pigmentosa [Letter]. Nature Genet. 1999;23:393–394.
54. Gal A, Li Y, Thompson DA, et al. Mutations in MERTK, the human
orthologue of the RCS rat retinal dystrophy gene, cause retinitis
pigmentosa. Nat Genet. 2000;26:270–271.
55. Nandrot E, Dufour EM, Provost AC, et al. Homozygous deletion in
the coding sequence of the c-mer gene in RCS rats unravels general
mechanisms of physiological cell adhesion and apoptosis. Neuro-
biol Dis. 2000;7:586–599.
56. Nandrot EF, Kim Y, Brodie SE, Huang X, Sheppard D, Finnemann
SC. Loss of synchronized retinal phagocytosis and age-related
blindness in mice lacking alphavbeta5 integrin. J Exp Med. 2004;
57. Daniele LL, Sauer B, Gallagher SM, Pugh EN, Jr, Philp NJ. Altered
visual function in monocarboxylate transporter 3 (Slc16a8)
knockout mice. Am J Physiol Cell Physiol. 2008;295:C451–
58. Yuan L, Kawada M, Havlioglu N, Tang H, Wu JY. Mutations in
PRPF31 inhibit pre-mRNA splicing of rhodopsin gene and cause
apoptosis of retinal cells. J Neurosci. 2005;25:748–757.
59. Solovei I, Kreysing M, Lanctot C, et al. Nuclear architecture of rod
photoreceptor cells adapts to vision in mammalian evolution. Cell.
60. Aartsma-Rus A, Janson AA, van Ommen GJ, van Deutekom JC.
Antisense-induced exon skipping for duplications in Duchenne
muscular dystrophy. BMC Med Genet. 2007;8:43.
61. van Deutekom JC, Janson AA, Ginjaar IB, et al. Local dystrophin
restoration with antisense oligonucleotide PRO051. N Engl J Med.
198 Graziotto et al.
IOVS, January 2011, Vol. 52, No. 1