Biochimica et Biophysica Acta 1800 (2010) 565573

Contents lists available at ScienceDirect

Biochimica et Biophysica Acta

j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / b b a g e n

Aberrant RNA splicing in RHD 7-9 exons of DEL individuals in Taiwan:

A mechanism study
Hsiang-Chun Liu a, Hock-Liew Eng f, Yu-Fen Yang b, Ya-Hui Wang e, Kuan-Tsou Lin g,
Hua-Lin Wu b,c,d, Tsun-Mei Lin a,b,h,i,
a
Institute of Basic Medical Sciences, National Cheng Kung University, Tainan, Taiwan
b
Department of Medical Laboratory Science and Biotechnology, National Cheng Kung University, Tainan, Taiwan
c
Biochemistry and Molecular Biology, National Cheng Kung University, Tainan, Taiwan
d
Cardiovascular Research Center, National Cheng Kung University, Tainan, Taiwan
e
Department of Pathology, National Cheng Kung University, Tainan, Taiwan
f
Department of Clinical Pathology, Chang Gung Memorial Hospital, Kaohsiung Medical Center, Chang Gung University College of Medicine, Kaohsiung, Taiwan
g
Kaohsiung Blood Center, Taiwan Blood Services Foundation, Kaohsiung, Taiwan
h
Department of Medical Research, E-DA Hospital /I-SHOU University, Kaohsiung, Taiwan
i
Department of Laboratory Medicine, E-DA Hospital /I-SHOU University, Kaohsiung, Taiwan

a r t i c l e i n f o a b s t r a c t

Article history: Background: The Rh blood D group provides a clinically important model of aberrant splicing with skipped
Received 4 September 2009 exons. Approximately 30% of serologically D-negative Chinese individuals have an intact RHD gene (DEL
Received in revised form 28 January 2010 phenotype) and induce allo-immunization in transfusions. The RHD1227GNA polymorphism occurs in >95%
Accepted 16 February 2010 DEL phenotype of Asian descent. The effects of RHD 1227A and a novel allele on exon 9 splicing were examined.
Available online 25 February 2010
Results: Amplied DEL RNA products revealed that 3 transcripts involved skipping of exons 8-9, exon 9, or exon
9 with an inserted 170-bp cryptic exon located between exons 7 and 8. A novel, single nucleotide
Keywords:
RHD gene
polymorphism was identied in the 7th intron, (IVS7) 923C>T, and present in all DEL patients. The odds ratio
Single nucleotide polymorphism of RHD1227G>A allele with DEL phenotype was 2711. Splicing analysis of transcripts from minigenes
DEL containing the 1227GNA allele, but not the (IVS7) 923C>T allele, demonstrated aberrant exon 9 skipping.
Splicing mutation Conclusions: A combined haplotype of 1227G>A and IVS7 923C>T alleles was apparent in >95% DEL Chinese
individuals. RHD1227A mutation signicantly increased aberrant mRNA splicing, producing a hybrid RHD
mRNA lacking exon 9. These results provide a molecular basis of the DEL phenotype in the Chinese population.
2010 Elsevier B.V. All rights reserved.

1. Introduction
Alternative RNA splicing yields related proteins with diverse ne
specicity that can differentially inuence one or more signal trans-
duction pathways and cellular functions [1,2]. The identication of
aberrant splicing as the cause of several diseases [3] has sparked interest
in splicing mutations. Splicing defects account for 15% of disease-
causing mutations in humans [4]. Splicing involves the precise identi-
cation and joining of exons and removal of introns by the spliceosome
[5]. The spliceosome contains ve snRNAs (U1, U2, and U46) and N200
polypeptides in a large multicomponent ribonucleoprotein complex [5].
The spliceosome recognizes conserved consensus sequences which
include splicing enhancers and silencers [6,7]. The spliceosome protein
SC35 binding to its motif can promote the splicing process, whereas the
SF2/ASF protein can repress it [8,9]. The spliceosome also recognizes the
5 (donor site) and 3 (acceptor site) splice sites that ank mammalian
Corresponding author. Department of Medical Research, E-DA Hospital /I-SHOU
University, Kaohsiung, 800, Taiwan. Tel.: + 886 7 6150011x2801; fax: + 886 7 7354727.
E-mail address: ltmei@mail.ncku.edu.tw (T.-M. Lin).
exons [2] . The last two nucleotides of an exon are also semi-conserved;
A often occurs in the 2 position (64%), while predominantly G is in
the 1 position (73%). Most defects are point mutations within 5 donor
and 3 acceptor splice sites located at the exonintron boundary [6,10].
Changes in the 1 position of an exon can reduce donor site recognition
by the spliceosome and yield aberrant splicing [6,10]. Mutations of
splicing consensus sequences can result in exon skipping [11] and/or
the activation of cryptic splice sites, thereby producing aberrant mRNAs.
The effect of the splicing mutation depends on its position in the mRNA
and the translational reading frame of the resultant mRNA.
As a clinically relevant model system for aberrant splicing, the
presence of disparate Rh antigens during transfusions and pregnancy
can induce allo-immunization. Rhesus (Rh) antigens are acylated red
blood cell (RBC) transmembrane proteins. The 30- to 32-kDa Rh
antigens, D and CcEe, are encoded by two highly homologous genes,
RHD and RHCE, respectively, which are separated by approximately
30,000 bp. Both genes contain 10 exons, span approximately 75 kb on
chromosome 1p3436, and are anked by Rhesus boxes [12]. The RhD
protein encoded by exons 110 contains 417 amino acid residues.
Multiple forms of RhD antigen are generated by combinatorial splicing

0304-4165/$ see front matter 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.bbagen.2010.02.006
566 H.-C. Liu et al. / Biochimica et Biophysica Acta 1800 (2010) 565573

Table 1
PCR primer sequences.

Name of primer Nucleotide sequence (5 3) Position Amplied size (bp) Accession

RHD exon 17
RHD_1s CTAGCTAGCTAGATGAGCTCTAAGTAC 58 73 1080 NM_016124
RH8 ATTGCCGGCTCCGACGGTATC 1126 1106 NM_016124
RHD exon 710
RH7 AGCTCCATCATGGG TAC AA 1031 1050 294 NM_016124
RhD_3UTR CCGCTCGAGCGGTTAAAATCCAACAGCCAAATG 1312 1292 NM_016124
RHD intron 7 sequence
RHD_in7s(693) AAGTGAATTGGGC ACG ATAC 693 712 678 AB035194
RHD_in7as(1370) GTCCCTGCACTGTGATGA 1370 1353 AB035194
Minigene detection
Mini-PCR-PFa 5-CTGGCTAGCTAGTGGATT ACTCGCTCA 98,99598,981 AL031542
Mini-PCR-PRa AGACTCGAGCGGCAGTTAAGTCTCTCACT 72,31772,346 AL031542
Internal control
beta-1A GTCTTCCCCTCCATCGTG 161 178 255 NM_001101
beta-1 S GTCATCTTCTCGCGGTTG 432 415 NM_001101
a
Primers contain not only pseudo-exon sequence but also restriction enzyme and vector sequence to avoid targeting on endogenous genes.

involving exons 7, 8, and 9 [13]. Individuals are clinically classied as the detection of RHD 1227 and RHD IVS7 + 923 polymorphisms, as
D-positive or D-negative, based on serological and molecular analysis. well as the hybrid Rhesus box by SSPPCR. The study was approved by
Approximately 15% of Caucasians are D-negative, mostly due to the Ethics Committee of Cheng Kung University Hospital, and written
deletion of the upstream and downstream Rhesus boxes in RHD [14]. informed consent was obtained from each participant.
The D-negative blood type occurs in only 0.1% to 0.5% of the Asian Ten milliliters of whole blood was collected in EDTA tubes. Phenotypes
population [15]. However, over 30% of D-negative Han Chinese were identied by polyclonal anti-D antibody (Dominion Biologicals
weakly express D antigen, which is only detectable by adsorption and Limited, Dartmouth, Nova Scotia, Canada) and by adsorption and elution
elution tests and is called the DEL phenotype [1518]. The incidence of tests. Genomic DNA from the buffy coat fraction was extracted using the
DEL phenotype in the D Caucasian populations ranges from 0.18% Blood & Tissue Genomic DNA Extraction Miniprep kit (Viogene, Hsichih,
[14] to 5% [19]. RHD genetic variations and the molecular basis of Taiwan). RHD 1227A, 1227G, and the hybrid Rhesus box were analyzed
the DEL phenotype are hotly pursued topics in transfusion medicine by sequence-specic primerpolymerase chain reaction (SSPPCR) [26].
because of their clinical importance and the difculty in detecting DEL
mutants [20]. While the D-negative phenotype usually involves a 2.2. cDNA study of the RHD gene
large deletion spanning the Rhesus boxes, the DEL phenotype has a
grossly intact RHD gene [21]. Mutations in the RHD gene associated Total RNA was isolated from whole blood using a QIAamp RNA
with the DEL phenotype include (i) partial deletion of exon 9 in 21 Blood Mini kit (Qiagen, Stanford, CA). Total poly (A)+ RNA was reverse-
Taiwanese DEL donors [22]; (ii) two alleles, RHD intravening sequence transcribed into single-stranded cDNA by Moloney murine leukemia
(IVS)IVS1 + 1A and RHD nucleotide 1227A in three Japanese DEL donors virus reverse transcriptase using an oligo d(T)20 primer (Promega, Inc.
[23]; (iii) RHD 885 T (missense), RHD IVS3 + 1A alleles (splice-site Madison, WI). The cDNA was directly amplied by PCR using the RHD
mutation) [14]; (iv) RHD 1227A or RHD 1227GNA (K409K) [14,17,24 exon 17 and RHD exon 710 primer sets listed in Table 1. The PCR
26]; (v) two variants in intron 7 (6984CbG and 9901(A)18N19) [20]; products were separated on 2% agarose gel. PCR bands were isolated
(vi) intron 9 variant (3425(T)15N14) [20]; and (vii) hybrid rhesus and subcloned into a T&A cloning vector (Yeastern Biotech Co., Taipei,
boxes [18,20,26]. RHD nucleotide 1227 is located at position 1 of the Taiwan). Subclones were directly sequenced with the AmpliCycle
exon 9intron junction, and the 1227GNA (K409K) mutation is very Sequencing Kit (Perkin-Elmer, Branchburg, NY) and an ABI 310 sequencer
prevalent (N95%) in DEL phenotype of Asians [20,25]. Although the (Applied Biosystem, Foster City, CA). Alignment of DNA sequences and
mechanisms of weak D antigen expression in the DEL phenotype remain
elusive, aberrant mRNA splicing and intron retention have been
proposed, based on splicing mutations in several diseases [6,7,10,27,28]. Table 2
Thus, we hypothesized that aberrant mRNA splicing, exon en- Primer sequences for detection of RHD 1227, RHD IVS7 + 923 alleles and hybrid Rhesus
hancers and silencers, and intron retention may play major roles box.
in the DEL phenotype of Han Chinese population. To further dissect
Primer Nucleotide sequence (5 3) Position Accession
the potential mechanisms, we examined the RHD DEL DNA and cDNA
RHD 1227A
structure, and its splicing patterns, especially near its 3 end, in a
E9s-3A GATGACCAAGTTTTCTGGAAA 12651284 NM_016124
population of Chinese individuals. In9as-3 GTTCTGTCACCCGCATGTCAG 327307 AB035185
RHD 1227G
2. Materials and methods E9s-3G GATGACCAAGTTTTCTGGAAG 12651284 NM_016124
In9as-3 GTTCTGTCACCCGCATGTCAG 327307 AB035185
Hybrid Rhesus box
2.1. Patients, blood samples, and RHD 1227 allele determination RhBox-s TGAGCCTATAAAATCCAAAGCAAGTTAG 49604987 AJ252311
RhBox-as CCTTTTTTTGTTTGTTTTTGGCGGTGC 77347708 AJ252312
Thirty-eight unrelated individuals with DEL (n = 26), D-positive RHD IVS7 293C
(n = 10), and D-negative (ccdee, n = 2) phenotypes were initially In7(923C) AGTTATCTGCCCAAGGTCACC 902922 AB035194
In7as(1139) TGGAACCCATCACAGAGGCG 11571138 AB035194
recruited at the Kaohsiung Blood Center, Taiwan Blood Service
RHD IVS7 293T
Foundation. Unrelated healthy blood donors (220 males, 179 females; In7(923T) AGTTATCTGCCCAAGGTCACT 902922 AB035194
mean age = 32 years, range = 1860 years) of Han Chinese origin living In7as(1139) TGGAACCCATCACAGAGGCG 11571138 AB035194
in Taiwan were recruited as controls. An additional 94 unrelated, DEL Growth hormone
phenotype blood donors were provided by Kaohsiung Blood Center, IC-s GCCTTCCCAACCATTCCCTTA 55605580 NG_001334
IC-as TAGACGTTGCTGTCAGAGGC 61956176 NG_001334
Chinese Blood Service Foundation, and the samples were utilized for
 H.-C. Liu et al. / Biochimica et Biophysica Acta 1800 (2010) 565573 567

Table 3
Primer sequences for construction of pseudo-vector.

Primer Nucleotide sequence (5 3) Position Amplied size (bp) Accession

Exon A
Dys-ex18-F CTAGCTAGCTAGTGGATTACTCGCTCA 98,99598,981 189 AL031542
Dys-ex18-R CCCAAGCTTGGGCACAGATAACAAAGCAC 98,82998,846 AL031542
Exon B
Dys-ex20-F CGGGATCCCGCTGGTTTGAAATCATTC 72,52972,514 234 AL031542
Dys-ex20-R CCGCTCGAGCGGCAGTTAAGTCTCTCACT 72,31772,346 AL031542
Exon 9 minigene
Mini-ex9-F GGGGTACCCCCGATAGAGTGAGACTC 4582 4597 554 AB035196
Mini-ex9-R CGGGATCCCGGTATCAGGGAAATGG 202 215 AB035185
Intron 7 minigene
Mini-in7-F GGGGTACCCCAAGTGAATTGGGCACGATAC 693 712 699 AB035194
Mini-in7-R CGGGATCCCGTCCCTGCACTGTGATGA 1370 1353 AB035194

protein sequence comparisons were performed using NCBI- BLAST and
BioEdit-Clustal W Multiple alignment computer software. The reading
frame of the RHD transcripts was predicted by NCBI ORF nder. Further
analysis of the ESE motif recognition score was performed using ESEnder
(http://rulal.cshl.edu/tools/ESE/), as described previously [6,8,9].
2.3. Direct sequencing of the PCR product of intron 7
Genomic DNA from 10 DEL and 10 D-positive blood samples were
PCR-amplied using RHD intron 7 sequence primer pairs in Table 1 to
further characterize unusual transcript variants of intron 7. The PCR
products of intron 7 of the RHD gene were puried from 2% agarose
gels (Gel/PCR DNA Fragments Extraction Kit, Geneaid Inc., Taipei,
Taiwan), and sequenced with the AmpliCycle Sequencing Kit.
2.4. SSPPCR for allele determination
2.5. In vivo splicing assay
To study the splicing pattern, we constructed a minigene vector that
contained two exons (exons A and B) and one intervening intron using
the multicloning site in the pcDNA4/V5-His B mammalian expression
vector (Invitrogen, Carlsbad, CA) (Fig. 2) [28]. Briey, 4 sets of test
sequences of the exon and its anking introns were amplied (Table 3),
digested with KpnI and BamHI and inserted into the analogously digested
multicloning site of a hybrid minigene. After sequence verication, these
hybrid minigenes were transfected into HeLa and K562 cells for a
splicing assay [29,30]. Cells were harvested 24 hours after transfection,
and total RNA was extracted using REzol C&T (PROtech Technologies,
Taiwan). Total RNA (3 g) was subjected to reverse transcription using
random hexamer primers. PCR products of the spliced product were
amplied with relevant primers (Table 3), analyzed by agarose gel
electrophoresis, and conrmed by direct sequencing.

Genomic DNA samples from 399 unrelated healthy donors of Han
Chinese origin and 94 unrelated DEL blood donors were examined
for the presence of the hybrid Rhesus box, RHD 1227 and a novel
polymorphism by SSPPCR using the primers in Table 2. Primers
for the growth hormone gene were used to generate a 629-bp PCR
fragment as an internal control.
2.6. Statistical analysis
The allele frequencies were calculated as the number of detected
alleles divided by the population examined. The HardyWeinberg equi-
librium utilized the 2 test to compare the observed and expected
frequencies of various genotypes. Odds ratios (ORs) for allelic frequency

Fig. 1. Elucidation of RHD transcript structures from D-positive, DEL, and true D-negative individuals with different RHD 1227 genotypes. (A) RTPCR amplication of the cDNA
fragments of RHD exon 17, exon 710, and -actin. The results obtained from different genotype samples are as indicated. Lane M shows the molecular mass standards of the 100-bp
ladder. (B) Sequencing results of the PCR products, and the linkages of exon 7 and exon 8, exon 8 and exon 9, and exon 9 and exon 10 are shown.
568 H.-C. Liu et al. / Biochimica et Biophysica Acta 1800 (2010) 565573

Fig. 1 (continued).
 H.-C. Liu et al. / Biochimica et Biophysica Acta 1800 (2010) 565573 569

were calculated using generalized estimated equation (GEE). Linkage Table 4

disequilibrium (LD) was calculated using Haploview (Broad Institute of Comparison of frequencies of RHD 1227A, Hybrid Rhesus box and RHD IVS7 + 923T in
Chinese individuals with the DEL phenotype and the general population.
MIT and Harvard, Boston, MA). All statistical analysis was performed
using SAS 9.0 software (SAS Institute Inc., Cary, NC), and the level of DEL General OR (95% CI) P value
statistical signicance was set at P b 0.05. population

N = 94 % N = 399 %
3. Results RHD 1227
A 108 97.3 10 1.3 2711 (73410,005)
3.1. Analysis of RHD mRNA transcripts in DEL individuals G 3 2.7 753 98.7 Ref b 0.0001
Hybrid Rhesus box
+ 77 41.0 35 4.4 15.1 (9.723.6)
To determine the location of DEL-associated mutations, the RHD - 111 59.0 763 95.6 Ref b 0.0001
mRNA of DEL individuals with different genotypes (A/A and A/) RHD IVS7 + 923
were analyzed by RTPCR and compared with the mRNA obtained from T 111 100 628 82.3
D-positive (G/G, G/, and G/A) and true D-negative (/) indivi- C 0 0 135 17.7
RHD 1227/RHD IVS7 + 923
duals. All genotypes of D-positive and DEL individuals generated an
A/T 108 97.3 10 1.3 2224 (6038214)
exon 17 product with an expected size of 1080 bp (Fig. 1A). In contrast, G/C 0 0 135 17.7
the exons 710 (294 bp) segments of the RHD transcript were absent G/T 3 2.70 618 81.0 Ref b 0.0001
from the amplied cDNA of DEL individuals but present in the healthy OR: odds ratio; CI: condence interval.
controls. Three splicing variants were detected (Fig. 1A) and their cDNA Percentage calculated from the actually existed number of RHD alleles.

fragments were cloned and sequenced. The two shorter cDNA fragments Statistical signicance (P b 0.05).
of 220 bp and 140 bp indicated that their respective transcripts had
skipped exons 9 and exons 89, respectively. Sequencing revealed that frequencies of RHD 1227A and the hybrid Rhesus box were signicantly
the RNA template for the 390-bp RTPCR band had retained a 170-bp elevated in DEL patients compared to normal individuals (P b 0.0001).
cryptic exon between exons 7 and 8 and had skipped exon 9 (Fig. 1B). Because all DEL patients exhibited the RHD IVS7 + 923T phenotype, its
ORF analysis of the splicing variants indicated that the transcript OR was not determinable. The OR for the combined RHD 1227A and
lacking exon 9 encoded a protein of 463 amino acid residues with a RHD IVS7 + 923T alleles were signicantly elevated in DEL individuals
new C-terminal extension, which began at codon 384. The transcript (P b 0.0001). The highest OR (OR = 2711) was the allelic frequency of
lacking exons 89 encoded a DEL protein of 396 amino acids and the RHD1227A with the DEL phenotype.
C-terminus had 39 novel amino acid residues. ORF analysis of the
transcript, which harbored the insertion of a 170-bp cryptic exon 3.4. Analysis of RHD 1227 and IVS7 + 923 polymorphisms
between exon 7 and exon 8 but no exon 9, displayed a missense
protein of 431 amino acid residues which included 74 novel residues We hypothesized that the two polymorphisms may alter splicing
at the C-terminal. The AAs at the C-terminus of the splicing variants patterns by modulating the strength of their adjacent splice sites. The
were different from the normal RHD protein encoded by exons 110. effect of RHD 1227 and IVS7 + 923 polymorphisms on the strength of
Further sequencing analysis of DEL genomic DNA revealed the the splice sites was calculated according to the Splice Site Prediction by
presence of the RHD 1227A allele in most DEL samples but no mutations Neural Network software (SSPNN, www.fruity.org/seq_tools/splice.
in the vicinity of intron 8 and intron 9 or in the 3 UTR encoding region html). In general, sequences that have a high score are considered
(data not shown). to be strong (the maximum is 1.0 and corresponds to the consensus
sequence). The 3 splice sites scores for RHD 1227G and 1227A exon 9
3.2. Analysis of the RHD intron 7 were 0.97 and 0.49, respectively. The apparent weakness of the exon 9
donor site with RHD 1227A compared with RHD 1227G contrasted
To further characterize the 390-bp PCR product, the RNA from 10 strikingly with its splicing behavior. The 5 splice site consensus
DEL and 10 D-positive blood samples were RTPCR-amplied using sequence derived from the highest frequency, namely viz. AG/GURAGU
intron 7 primers and subsequently sequenced. Sequencing of the resulting (where R = purine, and/or indicates the exonintron border), is com-
390-bp PCR band from the 10 DEL blood samples conrmed the presence plementary to the nucleotides at positions 2 to sn + 6 of U1 small
of a 170-bp cryptic exon in intron 7 and the absence of exon 9 (data not nuclear (sn) RNA. But no difference in the strength of the splice sites
shown). A novel single nucleotide polymorphism IVS7+923CNT was was detected using the novel mutation IVS7 + 923CNT in intron 7.
detected in all 10 DEL individuals, but not in D-positive blood samples. We next performed a search of normal and mutated RHD intron
The polymorphism IVS7+923CNT was initially demonstrated to have 7 sequences to identify binding motifs of known exonic splicing
a linkage disequilibrium with RHD 1227A in DEL individuals. enhancers (ESE). Binding motifs that predict functional ESEs for three
of the most well-characterized SR proteins (SF2/ASF, SC35 and SRp40)
3.3. Prevalence of RHD 1227A, hybrid Rhesus box, and RHD IVS7 + 923T were found in the 5 portion of the cryptic exon of intron 7 (Table 5).
Comparison of introns IVS7 + 923T and IVS7 + 923C indicated that the
The RHD 1227A and hybrid Rhesus box polymorphisms have been mutation modestly disrupted one of the potential SC35 binding site.
associated with the DEL phenotype. The frequencies of RHD IVS7 + 923T IVS7 + 923T also slightly strengthened the overlapping preexisting
variants as well as RHD 1227A and hybrid Rhesus box in the genomic SC35 and SRp40 score (SC35: 3.15 vs. 2.46; SRp40: 5.77 vs. 4.28) and
DNA from 94 unrelated DEL individuals were compared to 399 un- depressed the SF2/ASF score (data not shown). These data suggested
related healthy donors of Han Chinese origin by SSPPCR (Table 4). that IVS7 + 923T may exert a greater effect on splicing efciency than
The distribution of genotypes for the RHD IVS7+ 923 polymorphism the IVS7+ 923C allele.
conformed to the predictions of HardyWeinberg (2 = 6.108, P =
0.107). The frequencies of RHD IVS7+ 923C and RHD IVS7 + 923T were 3.5. In vivo splicing assay
17.7% (135/763) and 82.3% (628/763), respectively. To further evaluate
the impact of the RHD 1227A, hybrid Rhesus box, and RHD IVS7 + 923T The RHD 1227A and RHD IVS7 + 923T polymorphisms that occur in
variants of DEL, we calculated the ORs and 95% condence intervals almost all Chinese DEL individuals may alter RHD transcription, splicing,
(CIs) associated with the frequencies of RHD 1227A, hybrid Rhesus box, and translation. To examine the potential effect of RHD 1227A and
and RHD IVS7+ 923T (Table 4). The estimated ORs for both the allelic RHD IVS7 + 923T polymorphisms on the efciency of alternate splicing
570 H.-C. Liu et al. / Biochimica et Biophysica Acta 1800 (2010) 565573

patterns,
Table 5 the wild-type and mutant exons were subjected to in vitro RHD gene is approximately 17.6%32% [16,17,21,25,26]. The
Exonic splicing enhancer motifs predicted by searching intron 7 sequence with the ESEnder program.

SR protein type SRprotein binding motifs found in IVS7 + 923C sequence SRprotein binding motifs found in IVS7 + 923T sequence Intron 7 nucleotide positions
(intron 7, nucleotides 913993) (intron 7, nucleotides 913993) spanning the predicted ESE motifs

SF2/ASF CACCCGG CACTCGG 920926

AGCAGGT AGCAGGT 954960
ATGAGGA ATGAGGA 962968
TGGACGT TGGACGT 980986
SC35 GGTCACCC GGTCACTC 917924
GTCACCCG 918925
GGCTGGAA GGCTGGAA 925932
GGAACCTG GGAACCTG 929936
GGCTGAAG GGCTGAAG 948955
SRp40 TCACCCG TCACTCG 919925

splicing analysis using hybrid minigenes (Fig. 2). Transfection of the
control (RHD 1227G) minigene produced a full-length hybrid mRNA in
both HeLa and K562 cells (Fig. 3A). The 285-bp RTPCR products of the
transcripts from the (RHD 1227G) minigene contained exons A, 9, and B
in both HeLa and K562 cells (Fig. 3A). In contrast, transfection of the
RHD 1227A mutant exon 9 minigene produced a shorter mRNA species.
RTPCR and sequencing revealed that the smaller PCR product (211 bp)
lacked exon 9 (Figs. 3A and B). These data indicated that RHD 1227A
variant altered the splicing pattern of the hybrid minigenes.
Transfection of the minigenes carrying the intron IVS7 + 923C or
intron IVS7 + 923T variant into HeLa or K562 cells produced aberrant
splicing transcripts, detected by RTPCR (Fig. 4A). The 381-bp RTPCR
products contained exon A, intron 7, and exon B, based on sequencing
(Fig. 4B). Sequencing of the 211-bp RTPCR product showed that the
transcript contained exon A linked to exon B with no intron 7 (Fig. 4C).
The minigenes carrying the intron IVS7 + 923C or the intron IVS7 + 923T
variant produced a similar ratio of transcripts, indicating that the IVS7 +
923C or IVS7 + 923T variants did not modulate splicing in this system.
4. Discussion
The DEL phenotype is very rare (0.17%0.29%) in D-negative
Caucasian populations [14,19,31]. However, the incidence of DEL
phenotype in D-negative Asian individuals who carry a grossly intact
RHD1227A allele is very common in DEL Asians (73%100%)
compared with DEL Caucasians (1.5%33.3%) [16,17,19,21,31]. Based
on these reports, we suggest that the very low frequency of RHD1227A
allele in the Caucasian population contributes to their low incidence of
DEL phenotype.
The DEL phenotype exhibits the weakest expression of D antigens,
and additional mutations are being identied [19,20,32]. Aberrant
splicing mutations involving exon skipping, cryptic splice site usage,
and intron retention has led to numerous genetic diseases [1,3,10,33].
Similarly, our analyses of the RHD transcripts indicated that transcript
variants involved exon 9 skipping and possibly intron 7 retention. The
aberrantly spliced variants encoded RHD proteins with various
lengths of novel amino acids at the C-terminal. The exon 9 skipping
transcript expressed a protein containing 79 new amino acids at its C-
terminus. The transcript with a deletion of exon 9, and an insertion of
a 170-bp cryptic exon between exon 7 and exon 8 encoded a RHD
protein with 58 novel AA residues at its C-terminal. Our results
indicated that the GNA transition at RHD 1227 induced these splicing
variants in the C-terminal domain of RhD protein. Each splicing
variant resulted in 39, 74, or 78 novel amino acids in the C-terminus.
These novel C-terminal amino acids may alter the interactions of RhD
with both RhAG and ankyrin, which play key roles in the transport and
assembly of the Rh membrane complex to RBC membrane [3436].
The loss of a normal C-terminal portion of the RhD protein contributes

Fig. 2. Schematic representation of the hybrid minigenes. Each minigene vector was constructed to encode two cassette exons (A and B) anking an intervening sequence containing
a multicloning site. The PCR-amplied exon 9 and intron 7-encompassing regions were digested with KpnI and BamHI and inserted into the multicloning site. Each minigene vector
contained a cytomegalovirus (CMV) enhancerpromoter and a polyadenylation signal (BGH) for complete synthesis of mRNA. Box and lines indicate the exon and its anking
introns. The primers used in the RTPCR assay are represented by arrows.
 H.-C. Liu et al. / Biochimica et Biophysica Acta 1800 (2010) 565573 571

Fig. 3. Hybrid minigene containing RHD exon 9 splicing assay. Hybrid minigenes containing RHD exon 9 variants were transfected into K562 and HeLa cells. The transcribed mRNA
was amplied by reverse transcriptasepolymerase chain reaction (RTPCR). (A) RTPCR products electrophoresed on agarose gel are shown. A schematic depiction of the RTPCR
products is shown on the right. (B) Sequence of the 285-bp RTPCR product from RHD1227G minigene transcript: depicted the junction between exons A9 and exons 9B.
(C) Sequence of the 211-bp RTPCR product from RHD1227A minigene transcript: depicted the junction between exon A and B with no detection of exon 9. Nucleotide sequences at
the junctions between exons are shown.

to an Rhnull phenotype [36] and may be the major cause for the
perturbation of D protein presentation in the DEL erythrocytes.
We and others have demonstrated that most Chinese DEL indi-
viduals have a single nucleotide substitution 1227(GNA) located at the
splice donor site of RHD exon 9 [15,17,20,24,26] . Analogous to previous
mutations in 2 to +6 nucleotides for introns which disrupt base
pairing between U1 and the 5 splice site [3], the 1227GNA substitution
destroys the conserved sequence of a splice donor site, in agreement
with previous studies [20]. The destroyed donor splicing consensus
converts the exon 9 region into a pseudoexon and results in the skipping
of exon 9 during RHD mRNA processing. D-positive individuals with the
1227A/G genotype produced both normal and aberrant transcripts; and
the aberrant transcripts are probably linked to weak D expression,
which underlies the DEL phenotype. Taken together with previous
studies [20,26], the RHD1227 GNA allele reduced recognition by the
spliceosome and stimulated aberrant splicing of mRNA precursors in
vivo. However, the RHD1227 GNA allele was not observed in 3 of 108 DEL
individuals, which suggests that RHD 1227A allele is not essential,
in agreement with the mutational analysis of Chinese, Korean, and
European DEL patients [14,20,25]. Thus, additional mechanisms must
also be involved.
Further sequencing analysis of intron 7 identied a novel single
nucleotide polymorphism (RHD IVS7 + 923CNT) in all DEL individuals
tested. Since our distinct primers that amplied the 3 end gene were
located in exon 7 and the 3-UTR, our DEL mutant analyses were able
to detect both putative deletions of exon 9 and exons 89 as well as
an mRNA containing a 170-bp cryptic exon between 7 and 8. Because
RHD IVS7 + 923CNT allele is fairly common in the general population,
its association with DEL phenotype was not detected earlier by clas-
sical linkage disequilibrium studies. One possible mechanism of RHD
IVS7 + 923CNT would be the activation of cryptic exons within RHD
intron 7 by intron mutations that either create or strengthen splice
sites or create a branch site [7,30,37,38]. Although analysis suggested
that the RHD IVS7 + 923T allele affected splicing enhancer and
silencer on RHD gene, the predicted strength of the splice site was
similar to RHD IVS7 + 923C, as calculated with the SSPNN software.
572 H.-C. Liu et al. / Biochimica et Biophysica Acta 1800 (2010) 565573

Fig. 4. Hybrid minigene containing RHD intron 7 splicing assay. Hybrid minigenes containing RHD intron 7 variants (RHD IVS7 + 923C and RHD IVS7 + 923T) were tested by in vivo
splicing assay. The resultant mRNA was amplied by RTPCR and sequenced. (A) RTPCR products electrophoresed on a polyacrylamide gel. A schematic depiction of the RTPCR
products is shown on the right. (B) The 381-bp RTPCR product consisted of exons A, intron 7, and exon B. (C) The shorter RTPCR product did not contain intron 7, as exon A joined
directly to exon B. Nucleotide sequences at the junctions between exons are shown.

The minigenes carrying intron IVS7 + 923C vs IVS7 + 923T produced
similar ratios of the unspliced to spliced transcripts containing the
cryptic exon. In contrast to activation of cryptic exons by splice site
mutations in several diseases [6], the effect of this IVS7 + 923T allele
alone on forming new cryptic splice sites could not account for the
aberrant mRNA splicing. The retaining of a cryptic exon 7 was caused
by unknown mechanisms distinct from the ICS7 + 923T polymor-
phism in this in vivo splicing assay.
Presently, we demonstrated that RHD 1227A is strongly associated
with DEL (OR = 2711). The IVS7 + 923T allele was observed in 100% of
DEL patients and was prevalent (82.3%) in the general population. Based
on the analysis of Kralovicova et al. [6] on the combined splice site
mutations on aberrant splicing, the putative mechanism of RHD 1227A
splice site mutation would augment exon 9 or 89 skipping, which is
prevalent in the DEL phenotype. Similar to other human diseases [6],
these results strongly indicate that the RHD 1227A allele provides
the molecular defect that perpetuates the DEL phenotype in N95%
Chinese individuals. In addition, the results of pairwise LD analysis
suggest that DEL is connected with the simultaneous occurrence of both
RHD polymorphisms, IVS7 + 923CNT and 1227GNA. Therefore, further
studies to identify other polymorphisms in intron 7 of the RHD 1227A
allele that may increase retention of the cryptic exon are warranted.
In summary, genetic polymorphisms within the RHD genetic locus
contribute to the development of DEL. This study demonstrates that
the haplotype dened by RHD 1227A shows the strongest associa-
tion with DEL phenotype. The high prevalence of RHD 1227A in DEL
cohort and the OR of 2711 suggest a causal relationship. These
ndings warrant further study and provide a basis for probing the
mechanisms. Another possible interpretation is that the RHD1227A
and the IVS7 + 923T may be in linkage disequilibrium with uniden-
tied polymorphisms(s) that are the true variants causing the DEL
phenotype. These ndings suggest that the RHD 1227A mutation
 H.-C. Liu et al. / Biochimica et Biophysica Acta 1800 (2010) 565573
contributes to the molecular mechanisms underlying the defect in D
expression in Chinese individuals with the DEL phenotype.
Acknowledgments
This work was supported by the National Science Council [NSC 93
2314-B-006-120], Chang Gung University and Memorial Hospital
[CMRPG8055], and the MOE program for Promoting Academic
Excellent of University from Ministry of Education, Taiwan, Republic
of China [91-B-FA09-2-4].
The authors thank Ms. Ya-Chun Wang and Ms. Jung-Chin Chen
for providing technical assistance.
References
[1] T.A. Cooper, L. Wan, G. Dreyfuss, RNA and disease, Cell 136 (2009) 777793.
[2] M.C. Wahl, C.L. Will, R. Luhrmann, The spliceosome: design principles of a dynamic
RNP machine, Cell 136 (2009) 701718.
[3] J. Tazi, N. Bakkour, S. Stamm, Alternative splicing and disease, Biochim. Biophys.
Acta 1792 (2009) 1426.
[4] P.D. Stenson, E.V. Ball, M. Mort, A.D. Phillips, J.A. Shiel, N.S. Thomas, S. Abeysinghe,
M. Krawczak, D.N. Cooper, Human Gene Mutation Database (HGMD): 2003
update, Hum. Mutat. 21 (2003) 577581.
[5] C. Wachtel, J.L. Manley, Splicing of mRNA precursors: the role of RNAs and
proteins in catalysis, Mol. Biosyst. 5 (2009) 311316.
[6] J. Kralovicova, I. Vorechovsky, Global control of aberrant splice-site activation by
auxiliary splicing sequences: evidence for a gradient in exon and intron denition,
Nucleic Acids Res. 35 (2007) 63996413.
[7] N.J. Sakabe, S.J. de Souza, Sequence features responsible for intron retention in
human, BMC Genomics 8 (2007) 59.
[8] J. Wang, P.J. Smith, A.R. Krainer, M.Q. Zhang, Distribution of SR protein exonic
splicing enhancer motifs in human protein-coding genes, Nucleic Acids Res. 33
(2005) 50535062.
[9] H.X. Liu, S.L. Chew, L. Cartegni, M.Q. Zhang, A.R. Krainer, Exonic splicing enhancer
motif recognized by human SC35 under splicing conditions, Mol. Cell. Biol. 20
(2000) 10631071.
[10] E. Buratti, M. Chivers, J. Kralovicova, M. Romano, M. Baralle, A.R. Krainer, I.
Vorechovsky, Aberrant 5 splice sites in human disease genes: mutation pattern,
nucleotide structure and comparison of computational tools that predict their
utilization, Nucleic Acids Res. 35 (2007) 42504263.
[11] N. Kanai, F. Yanai, S. Hirose, K. Nibu, K. Izuhara, T. Tani, T. Kubota, A. Mitsudome, A
G to A transition at the last nucleotide of exon 6 of the gamma c gene (868GNA)
may result in either a splice or missense mutation in patients with X-linked severe
combined immunodeciency, Hum. Genet. 104 (1999) 3642.
[12] I. Mouro, Y. Colin, B. Cherif-Zahar, J.P. Cartron, C. Le Van Kim, Molecular genetic
basis of the human Rhesus blood group system, Nat. Genet. 5 (1993) 6265.
[13] E. Kajii, F. Umenishi, T. Omi, S. Ikemoto, Intricate combinatorial patterns of exon
splicing generate multiple Rh-related isoforms in human erythroid cells, Hum.
Genet. 95 (1995) 657665.
[14] T. Wagner, G.F. Kormoczi, C. Buchta, M. Vadon, G. Lanzer, W.R. Mayr, T.J. Legler,
Anti-D immunization by DEL red blood cells, Transfusion 45 (2005) 520526.
[15] Q. Xu, M.G. Grootkerk-Tax, P.A. Maaskant-van Wijk, C.E. van der Schoot, Systemic
analysis and zygosity determination of the RHD gene in a D-negative Chinese Han
population reveals a novel D-negative RHD gene, Vox Sang. 88 (2005) 3540.
[16] C.P. Shao, J.H. Maas, Y.Q. Su, M. Kohler, T.J. Legler, Molecular background of Rh
D-positive, D-negative, D(el) and weak D phenotypes in Chinese, Vox Sang. 83
(2002) 156161.
[17] J.C. Chen, T.M. Lin, Y.L. Chen, Y.H. Wang, Y.T. Jin, C.T. Yue, RHD 1227A is an
important genetic marker for RhD(el) individuals, Am. J. Clin. Pathol. 122 (2004)
193198.
573
[18] J.Y. Kim, S.Y. Kim, C.A. Kim, G.S. Yon, S.S. Park, Molecular characterization of
D Korean persons: development of a diagnostic strategy, Transfusion 45 (2005)
345352.
[19] H. Polin, M. Danzer, W. Gaszner, D. Broda, M. St-Louis, J. Proll, K. Hofer, C.
Gabriel, Identication of RHD alleles with the potential of anti-D immunization
among seemingly D blood donors in Upper Austria, Transfusion 49 (2009)
676681.
[20] Q. Li, L. Hou, Z.H. Guo, L.Y. Ye, D.Q. Yue, Z.Y. Zhu, Molecular basis of the RHD gene
in blood donors with DEL phenotypes in Shanghai, Vox Sang 97 (2009) 139146.
[21] H. Okuda, M. Kawano, S. Iwamoto, M. Tanaka, T. Seno, Y. Okubo, E. Kajii, The RHD
gene is highly detectable in RhD-negative Japanese donors, J. Clin. Invest. 100
(1997) 373379.
[22] J.G. Chang, J.C. Wang, T.Y. Yang, K.W. Tsan, M.C. Shih, C.T. Peng, C.H. Tsai, Human
RhDel is caused by a deletion of 1,013 bp between introns 8 and 9 including exon
9 of RHD gene, Blood 92 (1998) 26022604.
[23] B.K. Singleton, C.A. Green, N.D. Avent, P.G. Martin, E. Smart, A. Daka, E.G. Narter-
Olaga, L.M. Hawthorne, G. Daniels, The presence of an RHD pseudogene con-
taining a 37 base pair duplication and a nonsense mutation in Africans with the Rh
D-negative blood group phenotype, Blood 95 (2000) 1218.
[24] C.P. Shao, W. Xiong, Y.Y. Zhou, Multiple isoforms excluding normal RhD mRNA
detected in Rh blood group Del phenotype with RHD 1227A allele, Transfus.
Apher. Sci. 34 (2006) 145152.
[25] C.F. Sun, J.P. Liu, D.P. Chen, W.T. Wang, T.T. Yang, Use of real time PCR for rapid
detection of Del phenotype in Taiwan, Ann. Clin. Lab. Sci. 38 (2008) 258263.
[26] Y.F. Yang, Y.H. Wang, J.C. Chen, H.L. Eng, T.M. Lin, Prevalence of RHD 1227A and
hybrid Rhesus box in the general Chinese population, Transl. Res. 149 (2007) 3136.
[27] P. Agudelo-Florez, C.C. Prando-Andrade, J.A. Lopez, B.T. Costa-Carvalho, A.
Quezada, F.J. Espinosa, M.A. de Souza Paiva, P. Roxo Jr., A. Grumach, C.A. Jacob,
M.M. Carneiro-Sampaio, P.E. Newburger, A. Condino-Neto, Chronic granuloma-
tous disease in Latin American patients: clinical spectrum and molecular genetics,
Pediatr. Blood Cancer 46 (2006) 243252.
[28] H.T. Thi Tran, Y. Takeshima, A. Surono, M. Yagi, H. Wada, M. Matsuo, A G-to-A
transition at the fth position of intron-32 of the dystrophin gene inactivates a
splice-donor site both in vivo and in vitro, Mol. Genet. Metab. 85 (2005) 213219.
[29] V.K. Tran, Y. Takeshima, Z. Zhang, M. Yagi, A. Nishiyama, Y. Habara, M. Matsuo,
Splicing analysis disclosed a determinant single nucleotide for exon skipping
caused by a novel intraexonic four-nucleotide deletion in the dystrophin gene,
J. Med. Genet. 43 (2006) 924930.
[30] J.S. Smythe, N.D. Avent, P.A. Judson, S.F. Parsons, P.G. Martin, D.J. Anstee, Expression of
RHD and RHCE gene products using retroviral transduction of K562 cells establishes
the molecular basis of Rh blood group antigens, Blood 87 (1996) 29682973.
[31] F.F. Wagner, A. Frohmajer, W.A. Flegel, RHD positive haplotypes in D negative
Europeans, BMC Genet. 2 (2001) 10.
[32] G.F. Kormoczi, C. Gassner, C.P. Shao, M. Uchikawa, T.J. Legler, A comprehensive
analysis of DEL types: partial DEL individuals are prone to anti-D alloimmunization,
Transfusion 45 (2005) 15611567.
[33] D. Baralle, M. Baralle, Splicing in action: assessing disease causing sequence
changes, J. Med. Genet. 42 (2005) 737748.
[34] S.A. Eyers, K. Ridgwell, W.J. Mawby, M.J. Tanner, Topology and organization of
human Rh (rhesus) blood group-related polypeptides, J. Biol. Chem. 269 (1994)
64176423.
[35] C.H. Huang, The human Rh50 glycoprotein gene. Structural organization and
associated splicing defect resulting in Rh(null) disease, J. Biol. Chem. 273 (1998)
22072213.
[36] V. Nicolas, C. Le Van Kim, P. Gane, C. Birkenmeier, J.P. Cartron, Y. Colin, I. Mouro-
Chanteloup, Rh-RhAG/ankyrin-R, a new interaction site between the membrane
bilayer and the red cell skeleton, is impaired by Rh(null)-associated mutation,
J. Biol. Chem. 278 (2003) 2552625533.
[37] E. Zuccato, E. Buratti, C. Stuani, F.E. Baralle, F. Pagani, An intronic polypyrimidine-
rich element downstream of the donor site modulates cystic brosis transmem-
brane conductance regulator exon 9 alternative splicing, J. Biol. Chem. 279 (2004)
1698016988.
[38] M. Raponi, M. Upadhyaya, D. Baralle, Functional splicing assay shows a pathogenic
intronic mutation in neurobromatosis type 1 (NF1) due to intronic sequence
exonization, Hum. Mutat. 27 (2006) 294295.