Bone marrow cells from patients with Shwachman‐Diamond syndrome abnormally express genes involved in ribosome biogenesis and RNA processing

Bone marrow cells from patients with Shwachman-Diamondsyndrome abnormally express genes involved in ribosomebiogenesis and RNA processing

Shwachman-Diamond Syndrome (SDS) is a multi-system

autosomal recessive disorder. Although rare, it is one of the

most common causes of inherited bone marrow failure

syndrome and exocrine pancreatic insufficiency (Shwachman

et al, 1964; Dror, 2005; Dror & Freedman, 1999; Dror et al,

2001). The major causes of increased morbidity and mortality

are related to cytopenia, leukaemia, particularly acute myeloid

leukaemia (AML), malabsorption and short stature (Smith

et al, 1996; Dror et al, 2002). The only curative therapy for the

bone marrow complications in SDS is haematopoietic stem cell

transplantation. However, for as yet unknown reasons,

regimen-related toxicity is currently high in these patients

(Dror, 2005).

SBDS is a novel gene (Lai et al, 2000). Homozygous

mutations in SBDS were found in most SDS patients

(Boocock et al, 2003). The 250 amino acid Sbds protein is

highly conserved and SBDS mRNA is ubiquitously expressed

(Boocock et al, 2003). The structure of human Sbds has not

been solved. However, the AF0491 homolog in archaebacteria

has three domains (Shammas et al, 2005). Structural analysis

of these domains suggested that the N-terminal domain,

which consists of b1-b2-a1-a2-b3-loop-b4-a3-a4-b5 strands,

is involved in protein–protein interactions. The middle

domain has a common winged-helix-turn-helix fold associ-

ated with DNA binding (Luscombe et al, 2000). The

C-terminal domain comprises a four stranded-sheet with

two helices on one side, which share structural homology

with the RNA recognition motif, (Newberry et al, 1999) but

can also bind DNA (Selenko et al, 2003) or proteins (Wu

et al, 2002). The functions of SBDS are not yet known.

However, the gene has been suggested to play a role in

ribosome biogenesis and RNA processing, (Koonin et al,

2001; Wu et al, 2002) chemotaxis (Wessels et al, 2006) and

cell survival (Dror & Freedman, 2001; Rujkijyanont et al,

2008). The gene seems to be essential and the loss of human

SBDS or its mouse or yeast orthologue causes a prominent

Piya Rujkijyanont,1,2 Sally-Lin Adams,1,2

Joseph Beyene3 and Yigal Dror1,2

1Cell Biology Program, Research Institute and2Division of Hematology/Oncology, Department of

Pediatrics, and 3Child Health and Evaluative

Sciences Program, Research Institute,

The Hospital for Sick Children, University of

Toronto, Toronto, Ontario, Canada

Received 26 October 2008; accepted for

publication 26 February 2009

Correspondence: Dr Yigal Dror, Division of

Hematology/Oncology, Department of

Pediatrics, The Hospital for Sick Children, 555

University Avenue, Toronto, Ontario M5G 1X8,

Canada. E-mail: [email protected]

Summary

Shwachman-Diamond Syndrome (SDS) is a multi-system genetic disorder

with bone marrow failure. SBDS, the gene associated with SDS, has been

postulated to play a role in ribosome biogenesis and RNA processing, but its

functions are still unknown. To study whether these pathways are interrupted

when Sbds protein is lost, we studied the expression of related genes in

patient SBDS)/) cells by an oligonucleotide microarray. We first analysed

ribosomal protein (RP) genes, which are normally co-regulated. In SDS, 27 of

the 85 RP genes were downregulated. Among the downregulated RP genes,

seven are known to be associated with the inhibition of apoptosis. RPS27L,

which mediates p53-dependent induction of apoptosis, was the only

upregulated RP gene. Interestingly, several genes involved in RP mRNA

transcription were downregulated without affecting the expression of genes

involved in mRNA degradation, suggesting that the downregulation of the

RP gene expression might be at the transcriptional level. Importantly we also

found dysregulation of multiple genes involved in rRNA transcription and

pre-rRNA processing. We conclude that SDS marrow cells exhibit major

dysregulation of RP, RNA processing and RNA transcription genes.

Keywords: Shwachman-Diamond syndrome, ribosomal proteins, gene

expression, microarray, bone marrow failure.

research paper

First published online 6 May 2009doi:10.1111/j.1365-2141.2009.07692.x ª 2009 Blackwell Publishing Ltd, British Journal of Haematology, 145, 806–815

decrease in cell growth and viability (Zhang et al, 2006;

Rujkijyanont et al, 2008).

Ribosome biogenesis is a complex process. One critical

pathway involves the transcription of pre-rRNA in the

nucleolus, processing of pre-rRNA and rRNA and transport

to the cytoplasm. Structural analysis (Wu et al, 2002),

chromosomal localization (Koonin et al, 2001) and synthetic

genetic array (Koonin et al, 2001; Savchenko et al, 2005) data

from non-human cell systems suggest a role of Sbds in RNA

processing or ribosome biogenesis. Also, the yeast homologue,

YLR022c, was found to be associated with the maturation of

60s ribosomal subunits and translational activation of ribo-

somes in yeast (Menne et al, 2007). Human Sbds concentrates

in the nucleolus during the G1 and G2 phases (Austin et al,

2005). Its nucleolar localization is dependent on active rRNA

transcription (Ganapathi et al, 2007).

Another critical process during ribosome biogenesis is the

synthesis of ribosomal proteins (RPs). RP genes are comprised

of 3–10 exons and contain a terminal oligopyrimidine tract

(TOP) in the 5¢ UTR (Yoshihama et al, 2002). Positive

regulators of the RP gene transcription by the yeast RNA

Polymerase II include the protein kinase TOR, protein kinase

A, Ifh1, Fhl1, Rpd3, Rap1, Abf1, Sep1 and Esa1. Negative

regulators include the proteins Yak1 and Crf1. Despite the

importance of RP genes in cell viability, little is known about

their RP transcriptional regulation and stability in higher

eukaryotes and humans. Meyuhas and Perry (1980) found an

equal abundance of the various RP mRNAs in mouse L cells . It

has been postulated that a coordinated expression of the RP

genes is required to ensure a roughly equimolar accumulation

of ribosomal proteins. This is probably achieved by ensuring

an equivalent loading of RNA polymerases on the RP gene

promoters (Hariharan et al, 1989) and a comparable strength

of these promoters in driving RP expression (Hariharan et al,

1989). Several cis-acting regulatory elements and transcription

factor binding sites of mammalian RP genes have been studied

(Hariharan & Perry, 1989; Hariharan et al, 1989; Meyuhas &

Klein, 1990; Colombo & Fried, 1992; Harris et al, 1992;

Yoganathan et al, 1992; Chung & Perry, 1993; Genuario et al,

1993; Overman et al, 1993; Davies & Fried, 1995; Safrany &

Perry, 1995; Genuario & Perry, 1996; Curcic et al, 1997;

Antoine & Kiefer, 1998; Kirn-Safran et al, 2000). No regulatory

element common to all RP genes has been identified, although

certain elements were present in several of the genes that were

studied. None of the RP genes studied contained a canonical

TATA box in the )25 to )30 regions, but some have a ‘TATA-

like’ A/T-rich sequence, which might bind the general

transcription factor TBP. Some of the regulatory elements in

the RP promoter-proximal region contain binding sites for the

transcription factors Gabp, Sp1 and Yy1.

As stated above, SBDS is also postulated to play a role in RNA

processing. RNA processing is necessary to generate a mature

mRNA, tRNA and rRNA from their primary transcripts.

Processing of pre-mRNA involves multiple steps including

5¢end capping with 7-methylguanylate, polyadenylation of the

3¢end, and splicing. Processing of pre-rRNA involves methyl-

ation, pseudouridylation and cleavage into the three mature

rRNA: 18S, 5Æ8S and 28S. Processing of pre-tRNA includes the

removal of an extra segment at the 5¢end, the removal of an

intron in the anticodon loop, the replacement of two U residues

at the 3¢end by CCA, and the modification of some residues to

characteristic bases.

Given that data on the human SBDS and its homologues

suggest a role in ribosome biogenesis and RNA processing, we

hypothesized that defects in these fundamental processes in

Sbds-deficient cells cause prominent dysregulation of genes

involved in these pathways. Sbds loss may have an either direct

effect on the regulation of these genes or it may have an

indirect effect. To test our hypothesis, we performed a gene

expression analysis using an oligonucleotide microarray

approach and analysed genes involved in ribosome biogenesis

and RNA processing. We used cells from SDS patient bone

marrows as a model because this system is markedly affected in

the disease. The SDS bone marrow is usually hypocellular and

contains reduced marrow progenitors (Dror, 2005).

Patients and methods

Bone marrow aspiration from patients and control subjects

These studies were approved by the Hospital for Sick Children

Research Ethics Board and informed written consent was

obtained from patients and controls, or their legal guardians,

prior to sample collection. Patients were diagnosed with SDS

based on institutional clinical criteria (Dror & Freedman,

2002) which includes clear evidence for both haematological

and exocrine pancreatic dysfunction. In all patients, the

diagnosis was supplemented by positive SBDS gene mutation

analysis. The clinical details and genetic defects of the patients

have previously been reported (Rujkijyanont et al, 2007).

Bone marrow aspirate samples were collected from the

posterior superior iliac crest into a syringe containing

preservative-free heparin. In addition to nine SDS patients,

seven haematologically healthy donors of bone marrows for

transplantation served as controls. Cells were layered over

Ficoll-Paque PLUS and centrifuged for 20 min at 500 g as

previously described (Dror & Freedman, 1999). The light-

density cell fraction was collected, washed and used immedi-

ately.

Gene expression by oligonucleotide microarray

Ficoll-extracted marrow mononuclear cells from patients and

controls were analysed by oligonucleotide microarray using the

HG_U133_ Plus 2.0 GeneChip (Affymetrix Inc., Santa Clara,

CA, USA) as previously described (Rujkijyanont et al, 2007).

Briefly, 2Æ5 lg of total RNA from each sample was reverse

transcribed using T7-Oligo(dT) promoter primers. This was

followed by second-strand cDNA synthesis and in vitro

transcription using T7 RNA polymerase and biotinylated

Gene Expression in Bone Marrow Cells from Patients with SDS

ª 2009 Blackwell Publishing Ltd, British Journal of Haematology, 145, 806–815 807

nucleotides, resulting in biotin-labelled complementary RNA

(cRNA). The biotinylated targets were column-purified and

fragmented, and then 15 lg was used for hybridization with

the GeneChip. The hybridization, staining and destaining steps

were carried out using the Fluidics Station 400 (Affymetrix

Inc.). Scanning was carried out with the GeneChip� Scanner

3000 (Affymetrix Inc.). A hybridization image was obtained

and processed in the GeneChip Operating Software, and all

experiments were scaled to a target intensity of 150 in their

signal values. Gene Expression Signal, Detection Call, Signal

Log Ratio and Change Call were calculated using the GeneChip

Operating Software and the statistical algorithm published

previously by Affymetrix (Hubbell et al, 2002; Liu et al, 2002).

Oligonucleotide microarray data analysis was performed in

R Statistical Computing Language, Release 2.0.1 (Ihaka &

Gentleman, 1996). Signal quantification and normalization

were performed using Robust Multichip Analysis (RMA) in

the Bioconductor Affy package (http://www.bioconductor.org)

(Irizarry et al, 2003). The testing of genes for differential

expression was carried out using linear models and

Empirical Bayes Analysis (Limma package in http://www.

bioconductor.org) (Smyth, 2004). Genes were ranked based on

an empirical Bayes Log Odds of Differential Expression (B

statistics in Limma) and the level of significance was

determined using the False Discovery Rate (FDR) adjusted P

values. To screen for differentially expressed RP-related genes,

moderated T-statistic values were used, which enabled the

analysis of a relatively small number of replicates per group

where the gene-specific variance estimates can be unstable.

Moderated t-tests are superior to a fold change value because

the latter does not allow for the assessment of the significance

of observed differences in the presence of biological and

experimental variation.

Quantitative real-time polymerase chain reaction (PCR)

Representative RP genes (RPL22 and RPL23), which were found

to be differentially expressed by the oligonucleotide microarray,

were also analysed by real-time PCR using SYBR Green chemistry

and Applied Biosystems’ 7500 Real Time PCR System. Briefly,

bone marrow mononuclear cell samples from SDS (SBDS)/))

patients and healthy control subjects were analysed. The primers

used for amplification of RPL22 were: forward primer:

5¢-GAATCATGGATGCTGCCAATT-3¢ and reverse primer

5¢-CCCAGCTTTTCCGTTCACTTT-3¢. The primers used for

amplification of RPL23 were: forward primer: 5¢-AACACAG-

GAGCCAAAAACCTGTATA-3¢ and reverse primer 5¢-GCTCT-

GGTTTGCCTTTCTTGAC-3¢. The primers used for ACTB

control gene were: forward primer 5¢-AGCCTCGCCTTTGC-

CGA-3¢, and reverse primer 5¢-CTGGTGCCTGGGGCG-3¢. To

quantify the target gene expression level in the real time PCR

experiments, thetargetgenecDNAlevelswerenormalizedagainst

ACTB and expressed as DCT. The difference between patient and

control samples was analysed by the standard t-test. In all of these

experiments, a P value of less than 0Æ05 was considered to be

significant (Fig 1A,B). In addition to RPL22 and RPL23, several

leukaemia-related genes that were deregulated in SDS and

confirmed by quantitative real-time PCR were previously

reported (Rujkijyanont et al, 2007).

Results

RP gene expression

We conducted an oligonucleotide microarray analysis of SDS

marrow mononuclear cells. Since the purification of enough

RNA from a single cell lineage at one maturation stage in SDS

is impossible, we used the whole population of marrow

mononuclear cells.

We have previously shown that all SDS cell lineages

(myeloid, erythroid and lymphoid) are quantitatively and

qualitatively affected. The average percentages of the various

bone marrow cell lineages, myeloid/erythroid/lymphoid/other

cells, were in the normal range: 46Æ4% (SD ±6Æ8)/31Æ1% (SD ±

7Æ3)/19Æ5% (SD ±5Æ5) and 3Æ0% (SD ± 1Æ3) respectively. Also,

no patient had maturation arrest of the myeloid or erythroid

series. Some genes with either mild to moderate expression

changes, or transient or lineage-restricted expression might

escape detection by this approach. However, genes with

significant differential expression between patients and

controls, particularly those that are dysregulated in all cells,

will be recognized.

(B)

*

0·2

0·4

0·6

0·8

1

1·2

Normal(N = 10)

SDS(N = 8)

RQ

(2–d

dC

t )

0·2

0·4

0·6

0·8

1·0

1·2

RQ

(2–d

dC

t )

(A)

*

Normal(N = 7)

SDS(N = 8)

Fig 1. Real time PCR analysis of RPL22 and RPL23 in SDS patients and normal subjects. Total RNA from patients and healthy control subjects were

analysed by quantitative real-time PCR using SYBR Green chemistry. The differences between the groups for both genes were significant with a P

value of 0Æ04 (*). (A) RPL22 data (B) RPL23 data.

P. Rujkijyanont et al

808 ª 2009 Blackwell Publishing Ltd, British Journal of Haematology, 145, 806–815

Using the Gene Set Analysis (Efron & Tibshirani, 2007), we

found that translation-related gene expression was the most

prominently deregulated gene cluster among 12 other path-

ways that are known to be defective in various bone marrow

failure conditions including genes for hematopoietic growth

factor receptors, hematopoietic signal transduction factors,

hematopoietic transcription factors, cell proliferation, cell

cycle, chemotaxis/adhesion, apoptosis, DNA repair, ribosome

biogenesis, translation, RNA transcription and telomere

maintenance. Of the 38 500 well-characterized human genes

on the HG_133_Plus2.0 GeneChip, we analysed 284 known

ribosomal protein and RNA processing-related genes. Inter-

estingly, the expression pattern of the RP mRNAs was not

identical for all of the genes (Fig 2A,B; Table SI). Among 85 RP

genes obtained from the Ribosomal Protein Gene Database

(http://ribosome.med.miyazaki-u.ac.jp), we identified 27 genes

that were prominently downregulated in patient samples

compared to controls, with T values of less than )2Æ3, and

the additional 13 genes that were mildly downregulated, with T

values between )1Æ9 and )2Æ3. Multiple RP genes involved in

cell growth and survival, including RPS9, RPS20, RPL6, RPL15,

RPL22, RPL23 and RPL29, had decreased expression. By

quantitative real-time PCR the RPL22 and RPL23 genes were

downregulated in the SDS marrow mononuclear cells com-

pared to controls (P < 0Æ05) (Fig 1A,B). Interestingly, only one

of the RP genes, RPS27L, was found to be upregulated with a

significant T value of 2Æ65. RPS27L is a novel ribosomal protein

gene whose expression was found to be induced in a p53-

dependent manner in a genome-wide chip-profiling study and

in multiple cancer cell models. While induced, RPS27L

promotes chemotherapy-induced apoptosis. (He & Sun,

2007). It is noteworthy that the pattern of deregulation was

not the same in all the gene clusters. For example, the growth

factor gene cluster showed upregulation compared to the

controls (Data not shown).

rRNA processing genes

Among the 48 analysed rRNA-related genes, 11 genes were

prominently downregulated with a T value of less than )2Æ3,

and the additional three genes were mildly downregulated,

with T values between )1Æ9 and )2Æ3. The downregulated

N80

P20

9

P60 P

9

P20

N78

N47

N49

P37

P88

P21

P14

N33

N45

P20

6

N42

RPS15RPS21RPS28RPS23KIAA0999RPS16LOC388720RPS4Y1RPS27L RPS9RPS26RPS8RPS25RPS7RPS19RPS5RPS20RPS3RPS29RPS2RPS18LOC439992RPS15A RPS24RPS11 RPS4XRPS6LOC440055RPS13RPS14LOC402057RPS10

–2 –1 0 1 2

(A)

N47

P20

N49

N78

N80

P37

P60

P14

P20

9

P20

6

P9

P88

P21

N33

N45

N42

C15orf15RPL26L1RPL31hCG_2040224RPL37RPL13RPL28RPL17RPL30RPL3LRPL37ARPL39LRPL10LEVI1

RPL36RPL15RPL41RPL10RPL27RPL3RPL32RPL9RPL21RPL23ARPLP1RPL27ARPL13AhCG_26523RPLP0-likeRPL29RPL10ARPL35ARPL12RPL5RPL35RPL23RPL8UBA52LOC388474RPL6RPL18RPL14RPL4RPL11RPLP2RPL34RPL24RPL38RPL19RPL36A

(B)

RPL7

Fig 2. Expression of small ribosomal protein genes (A) and large ribosomal protein genes (B) in SDS mononuclear marrow cells compared to cells

from healthy controls. The green colour represents downregulation, and the red colour represents upregulation. The gene symbols are indicated on

the right, and the patients’ numbers are indicated on the bottom of the figure.



genes are involved in multiple steps of rRNA processing

especially rRNA modification including FBL for rRNA meth-

ylation (Yanagida et al, 2004) and DKC1 for rRNA pseudo-

uridylation (He et al, 2002). There was only one gene, EMG1,

which was found to be mildly upregulated (T = 2Æ00) (Fig 3A

& Table SII).

mRNA processing genes

Among the 123 genes involved in mRNA processing obtained

from GenMAPP (http://www.genmapp.org/HTML_MAPPs/

Human/MAPPIndex_Hs_Contributed.htm), 45 genes were

prominently downregulated with a T value of less than )2Æ3,

and the additional 11 genes were mildly downregulated, with T

values between )1Æ9 and )2Æ3 (Figs 3B and 4A,B Table SIII).

These genes are involved in mRNA capping and processing of

intron-containing pre-mRNAs, including the formation of

pre-mRNPs (ribonucleoproteins), mRNA splicing (major

pathway) and mRNA 3¢-end processing. Interestingly, only

three genes involved in mRNA processing were upregulated.

PRMT1 and SPOP were prominently upregulated (with a T

value of more than 2Æ3) whereas CLK3 was slightly upregulated

(with a T value between 1Æ9 and 2Æ3).

Genes involved in general transcription regulation andnon-RP TOP genes

Among the 21 genes involved in general transcription regu-

lation (Fig 4C & Table SIV), multiple genes were found to be

significantly downregulated with a T value of less than )2Æ3,

including those involved in RP gene transcription (e.g. GABPA

and YY1), RNA polymerases (e.g. POLR1B, POLR1D and

POLR3E) and one of the related transcription inhibitors

(DCP1A). The non-RP TOP genes (EEF1A1, EEF1B2 and

EEF2), which are involved in the binding of aminoacyl-tRNAs

to 80S ribosomes, the enzymatic delivery of aminoacyl tRNAs

to the ribosome and the translocation step in protein synthesis

respectively, were also downregulated with T values ranging

from )2Æ06 to )3Æ32 (Table SV).

Discussion

Studies on yeast (Menne et al, 2007) and human cells

(Ganapathi et al, 2007) suggest that Sbds may be involved in

RNA processing and ribosome biogenesis. RNA processing is

complex and involves multiple steps and a myriad of proteins.

Some of these proteins might process more than one type of

N80

N49

N78

P20

N47

N33

N45

N42

P20

9

P20

6

P88

P21

P60

P14

P37 P9

LOC284801EXOSC1C14orf145SH3BP4ERRFI1MRPL20DHX35SCARF2RPL5RPS11RPL23ARPS18NOP5/NOP58FBLDEDD2NOL5ADKC1NOLC1EIF4A3DIMT1LEMG1POP4GTF3ATFB2MFTSJ1UTP11LEXOSC8EXOSC9NOLA1RRP9TFB1MEXOSC2GEMIN4DIS3BOP1PDCD11EXOSC7EXOSC3DDX56NAGKEXOSC10MKI67IPEXOSC5FTSJ2EXOSC4

P20

N80

N49

N78

N47

P14

N45

N33

N42

P60P

9

P20

9

P20

6

P37

P88

P21

NCBP1RNMTHNRPRHNRPMRP11-78J21.1 METTL3RNGTTHNRPCNCBP2HNRPH1HNRPL SUPT5HPOLR2A HNRPA2B1 HNRPDHNRPH2HNRPA3 HNRPUHNRPA1 HNRPAB HNRPK

–2 –1 0 1 2

(A) (B)

Fig 3. Expression of rRNA-related genes (A) and expression of genes involved in mRNA capping, mRNA methylation and formation of pre-mRNPs

(B) in SDS marrow mononuclear cells compared to cells from healthy controls. The green colour represents downregulation, and the red colour

represents upregulation. The gene symbols are indicated on the right, and the patients’ numbers are indicated on the bottom of the figure.



RNA species. Ribosome biogenesis involves the synthesis and

assembly of rRNA with RPs, and requires additional factors,

such as RNA Polymerase I and II ribonucleoproteins and non-

ribosomal proteins. Because of the fundamental functions of

RNA, the alteration of the levels or function of any of these

proteins may lead to serious cellular phenotypes. The results of

our study suggest that Sbds loss abrogates various steps in

ribosome biogenesis and RNA processing, which may explain

the severe consequences of mutations in SBDS on multiple

tissues in patients with SDS.

Little is known about the components of ribosome biogen-

esis that are affected in Sbds-deficient cells. Using oligonu-

cleotide microarray technology, we found that SDS marrow

cells are characterized by the aberrant expression of multiple

genes encoding ribosomal proteins belonging to the large (L,

60S) and small (S, 40S) ribosomal subunits. It is noteworthy

P20

9

P20

6

P20

N47

N49

N80

N78

N33

N45

N42

P14

P88

P21P

9

P60

P37

PWCR1BRUNOL4TMED10SFRS1SRP54PRPF18CSTF1CSTF3PRKG1PTBP2SRPK2DNAJC8SMC1ASFRS8U2AF1PAPOLADDX20PABPN1CPSF2PSKH1DHX8SFRS9DHX15SNRPGYBX1SNRPD2SFRS2SFRS7SNRPESFRS5SFRS4FUSPRPF8SFRS3DDX1SF3B2CDC40SRPK1SNRPB2SNRPA1SF3B1SF3A1SNRPBPTBP1SRRM1SFRS6SFRS10SNRPD3SF3B5SF3A2CLP1FUSIP1DHX16CLK2PRPF4BSNRPD1SF3A3PRPF4CSTF2U1SNRNPBPDHX38CD2BP2SFRS12CPSF1SNRPALSM2NUDT21SF3B4RNPS1U2AF2CLK3PRPF6WDR57CPSF4PPM1GSNRP70PCBP2CPSF3PHF5ARBM5DHX9SFRS14SFRS16

(A) (B)

N80

P20

9

P20

6

P20

N49

N78

N47

N33

N42

P14

N45

P60

P37P9

P88

P21

PNLDC1SPOP

DICER1CNOT7

NONO

LSM7SFPQ

RBM39XRN2

CNOT8DCP2

CUGBP1

NXF1DCP1B

PRPF3DCPS

PARN RBM17

DCP1A

TOE1

–2 –1 0 1 2 P9

P88

P21

N49

N47

N78

N80

N33

P20

N42

P20

9

P20

6

P37

N45

P60

P14

POLR2BPOLR3EMGC13098GABPA POLR1BPOLR1A POLR3DEEF2EEF1B2PABPC1 POLR1DPOLR2GRASA4YY1EEF1A1POLR2IPOLR2EPOLR2A POLR3KPOLR2CSP1POLR2KGABPB2POLR2F POLR2H

(C)

Fig 4. Expression of genes involved in mRNA Splicing (A), genes involved in mRNA degradation (B) and genes involved in RNA transcription, non-

RP TOP genes (C) in SDS marrow mononuclear cells compared to cells from healthy controls. The green colour represents downregulation, and the

red colour represents upregulation. The gene symbols are indicated on the right, and the patients’ numbers are indicated on the bottom of the figure.



that the majority of the dysregulated genes were downregu-

lated with only one gene showing upregulation.

An important question is to determine whether the dys-

regulation of the genes mentioned above are related to the

clinical SDS phenotype, in particular the reduced cell mass in

the bone marrow and the pancreas. The Sbds-deficient marrow

mononuclear cells in our study abnormally expressed RPS9,

RPS20, RPL6, RPL15, RPL22, RPL23 and RPL29. Due to the

prominent dysregulation in specific RP genes that have

fundamental cellular functions, it is unlikely that these

abnormalities do not play a mechanistic role in the cellular

or clinical phenotype of SDS. It is intriguing that not all of the

RP genes were downregulated. One possible mechanism is that

the collective effect of all the dysregulated RP genes is only on

translation insufficiency. Alternatively, an attractive link

between the SDS genotype and phenotype could be through

the disruption of the extraribosomal functions of each of the

dysregulated RPs or through the disruption of a combination

of ribosomal and extraribosomal functions of these RPs.

Some of the RP genes that were downregulated in SDS have

extra-ribosomal functions. For example, RPS9 is thought to

have a protective role in oxidative injury of neural cells. The

decreased expression of RPS9 in Neuro-2A cells causes a

significant decrease in cell viability with H2O2 treatment (Kim

et al, 2003). The expression of RPS20 was found to be

downregulated during the induction of apoptosis by using

dexamethasone and radiation in the human leukemic cell line

CEM C7 (Goldstone & Lavin, 1993). RPL6 and RPL23 have

recently been found to have a role in regulating multidrug

resistance in gastric cancer cells by suppressing drug-induced

apoptosis. RPL6 is able to upregulate Bcl-2 and downregulate

Bax (Du et al, 2005), and RPL23 can upregulate Bcl-2 and the

Bcl-2/Bax ratio in cells (Shi et al, 2004). Interestingly, RPL15

was also significantly upregulated in gastric cancer cells, and

the inhibition of its expression by siRNA suppressed the

growth of SGC7901 gastric cancer cells (Wang et al, 2006).

RPL22 was recently found to have an important role in

thymocyte development by regulating the activation of p53 in

a cell-lineage-restricted manner. RPL22 deficiency selectively

blocked the development of ab-lineage T cells by inducing p53

expression (Anderson et al, 2007). RPL29 was reported to play

a role in cell proliferation, migration and differentiation and is

involved in the modulation of the apoptotic response to

anticancer drugs. Using siRNA methods, a 50% reduction of

RPL29 expression caused an increase in the percentage of

apoptotic cells in the colon cancer cell lines, HT-29 and HCT-

116 (Liu et al, 2004).

Interestingly, RPS27L was the only RP gene found to be

upregulated in SDS marrow cells. Several ribosomal proteins

can regulate p53 and apoptosis through various mechanisms,

including inhibition of Mdm2, enhancing p53 translation and

enhancing p53 translocation to the mitochondria (Zhang et al,

2003; Takagi et al, 2005; Yoo et al, 2005). However, only

Rps27l has been identified as a direct target of p53. RPS27L has

been shown to contribute to p53-induced apoptosis upon

induction by p53 (He & Sun, 2007). p53 binds directly to the

consensus-binding site in the first intron of the RPS27L gene

and transactivates its expression. Furthermore, siRNA silencing

of RPS27L inhibits apoptosis induced by etoposide (He & Sun,

2007). Therefore, the increased expression of RPS27L and the

downregulation of at least some of the RP genes discussed

above might be related to the accelerated apoptosis observed in

Sbds-deficient cells (Dror & Freedman, 2001; Rujkijyanont

et al, 2008).

It is noteworthy that many genes involved in the transcrip-

tion of RP genes (e.g. GABPA and YY1) were downregulated

without the dysregulation of genes involved in mRNA

degradation. This suggests that the downregulation of the RP

genes in Sbds-deficient cells takes place at the transcriptional

level. The GABPA complex regulates the transcription of

mitochondrial enzyme subunits, including cytochrome C

oxidase and mitochondrial transcription factor A (Virbasius

& Scarpulla, 1994). Yy1 is an ubiquitous, zinc-finger-contain-

ing, transcription factor which is a member of the Polycomb

Group protein family. It affects genes involved in normal

biological processes, such as embryogenesis, differentiation,

replication and cellular proliferation, via its ability to initiate,

activate, or repress transcription (Gordon et al, 2006).

In addition to observing the dysregulation of RP genes in

SDS cells, we also observed the dysregulation of genes involved

in rRNA transcription and processing. In our study, the rRNA

transcription factor, Mki67ip, was shown to be downregulated.

Mki67ip is an RNA-binding nucleolar protein that interacts

with the fork-head-associated domain of the proliferation

marker protein, Ki-67. MKI67IP is widely expressed in adult

mouse tissues and is upregulated in denervated hind limb

muscle (Takagi et al, 2001). It was originally reported that the

FHA domain of this protein constitutes a region which is

conserved in a subset of fork-head-type transcription factors

(Hofmann & Bucher, 1995). This sequence profile has been

reported for a variety of proteins with diverse functions

involving transcription, DNA repair and cell cycle progression.

In addition, the pre-rRNA processing gene, FBL, was

observed to be downregulated in SDS cells. FBL encodes

fibrillarin, which is a component of a nucleolar small nuclear

ribonucleoprotein (snRNP) particle which participates in

pre-rRNA methylation.

Interestingly, similar to the trend with the RP genes, many

genes which play a role in mRNA processing were downreg-

ulated. These genes included those involved in mRNA capping,

formation of pre-mRNPs, mRNA splicing (spliceosomal A, B

and E complexes) and mRNA 3¢-end processing. The data

above shows multiple abnormalities in various steps of mRNA

processing. The broad downregulatory effect on RNA process-

ing genes in Sbds-deficient cells suggests a common regulatory

mechanism of these genes, which is directly or indirectly

affected by Sbds.

In summary, SDS cells lead to the abnormal expression of

multiple genes involved in ribosome biogenesis, and rRNA and

mRNA processing. Future studies will focus on the link



between Sbds-deficiency and these abnormalities, as well as on

the relationship between these abnormalities and the SDS

phenotype.

Acknowledgements

This study was supported by grants from the Canadian

Institute of Health Research MOP57720, Shwachman-Dia-

mond Syndrome International and Anemia Institute for

Research and Education. P. Rujkijyanont is holding a training

award from the Royal Thai Army.

References

Anderson, S.J., Lauritsen, J.P., Hartman, M.G., Foushee, A.M., Lefeb-

vre, J.M., Shinton, S.A., Gerhardt, B., Hardy, R.R., Oravecz, T. &

Wiest, D.L. (2007) Ablation of ribosomal protein L22 selectively

impairs alphabeta T cell development by activation of a p53-

dependent checkpoint. Immunity, 26, 759–772.

Antoine, M. & Kiefer, P. (1998) Functional characterization of tran-

scriptional regulatory elements in the upstream region and intron 1

of the human S6 ribosomal protein gene. Biochemical Journal, 336

(Pt 2), 327–335.

Austin, K.M., Leary, R.J. & Shimamura, A. (2005) The Shwachman-

Diamond SBDS protein localizes to the nucleolus. Blood, 106, 1253–

1258.

Boocock, G.R., Morrison, J.A., Popovic, M., Richards, N., Ellis, L.,

Durie, P.R. & Rommens, J.M. (2003) Mutations in SBDS are asso-

ciated with Shwachman-Diamond syndrome. Nature Genetics, 33,

97–101.

Chung, S. & Perry, R.P. (1993) The importance of downstream delta-

factor binding elements for the activity of the rpL32 promoter.

Nucleic Acids Research, 21, 3301–3308.

Colombo, P. & Fried, M. (1992) Functional elements of the ribo-

somal protein L7a (rpL7a) gene promoter region and their

conservation between mammals and birds. Nucleic Acids Research,

20, 3367–3373.

Curcic, D., Glibetic, M., Larson, D.E. & Sells, B.H. (1997) GA-binding

protein is involved in altered expression of ribosomal protein L32

gene. Journal of Cellular Biochemistry, 65, 287–307.

Davies, B. & Fried, M. (1995) The L19 ribosomal protein gene

(RPL19): gene organization, chromosomal mapping, and novel

promoter region. Genomics, 25, 372–380.

Dror, Y. (2005) Shwachman-Diamond syndrome. Pediatric Blood &

Cancer, 45, 892–901.

Dror, Y. & Freedman, M.H. (1999) Shwachman-Diamond syndrome:

An inherited preleukemic bone marrow failure disorder with aber-

rant hematopoietic progenitors and faulty marrow microenviron-

ment. Blood, 94, 3048–3054.

Dror, Y. & Freedman, M.H. (2001) Shwachman-Diamond syndrome

marrow cells show abnormally increased apoptosis mediated

through the Fas pathway. Blood, 97, 3011–3016.

Dror, Y. & Freedman, M.H. (2002) Shwachman-diamond syndrome.

British Journal of Haematology, 118, 701–713.

Dror, Y., Ginzberg, H., Dalal, I., Cherepanov, V., Downey, G., Durie,

P., Roifman, C.M. & Freedman, M.H. (2001) Immune function in

patients with Shwachman-Diamond syndrome. British Journal of

Haematology, 114, 712–717.

Dror, Y., Durie, P., Ginzberg, H., Herman, R., Banerjee, A., Cham-

pagne, M., Shannon, K., Malkin, D. & Freedman, M.H. (2002)

Clonal evolution in marrows of patients with Shwachman-Diamond

syndrome: a prospective 5-year follow-up study. Experimental

Hematology, 30, 659–669.

Du, J., Shi, Y., Pan, Y., Jin, X., Liu, C., Liu, N., Han, Q., Lu, Y., Qiao, T.

& Fan, D. (2005) Regulation of multidrug resistance by ribosomal

protein l6 in gastric cancer cells. Cancer Biology & Therapy, 4, 242–

247.

Efron, B. & Tibshirani, R. (2007) On testing the significance of sets of

genes. Annals of Applied Statistics, 1, 107–129.

Ganapathi, K.A., Austin, K.M., Lee, C.S., Dias, A., Malsch, M.M., Reed,

R. & Shimamura, A. (2007) The human Shwachman-Diamond

syndrome protein, SBDS, associates with ribosomal RNA. Blood,

110, 1458–1465.

Genuario, R.R. & Perry, R.P. (1996) The GA-binding protein can serve

as both an activator and repressor of ribosomal protein gene tran-

scription. Journal of Biological Chemistry, 271, 4388–4395.

Genuario, R.R., Kelley, D.E. & Perry, R.P. (1993) Comparative utili-

zation of transcription factor GABP by the promoters of ribosomal

protein genes rpL30 and rpL32. Gene Expression, 3, 279–288.

Goldstone, S.D. & Lavin, M.F. (1993) Isolation of a cDNA clone,

encoding the ribosomal protein S20, downregulated during the

onset of apoptosis in a human leukaemic cell line. Biochemical and

Biophysical Research Communications, 196, 619–623.

Gordon, S., Akopyan, G., Garban, H. & Bonavida, B. (2006) Tran-

scription factor YY1: structure, function, and therapeutic implica-

tions in cancer biology. Oncogene, 25, 1125–1142.

Hariharan, N. & Perry, R.P. (1989) A characterization of the elements

comprising the promoter of the mouse ribosomal protein gene

RPS16. Nucleic Acids Research, 17, 5323–5337.

Hariharan, N., Kelley, D.E. & Perry, R.P. (1989) Equipotent mouse

ribosomal protein promoters have a similar architecture that

includes internal sequence elements. Genes and Development, 3,

1789–1800.

Harris, S.A., Dudov, K.P. & Bowman, L.H. (1992) Comparison of the

mouse L32 ribosomal protein promoter elements in mouse myo-

blasts, fibers, and L cells. Journal of Cellular Biochemistry, 50, 178–

189.

He, H. & Sun, Y. (2007) Ribosomal protein S27L is a direct p53 target

that regulates apoptosis. Oncogene, 26, 2707–2716.

He, J., Navarrete, S., Jasinski, M., Vulliamy, T., Dokal, I., Bessler, M. &

Mason, P.J. (2002) Targeted disruption of Dkc1, the gene mutated

in X-linked dyskeratosis congenita, causes embryonic lethality in

mice. Oncogene, 21, 7740–7744.

Hofmann, K. & Bucher, P. (1995) The FHA domain: a putative nuclear

signalling domain found in protein kinases and transcription fac-

tors. Trends in Biochemical Sciences, 20, 347–349.

Hubbell, E., Liu, W.M. & Mei, R. (2002) Robust estimators for

expression analysis. Bioinformatics, 18, 1585–1592.

Ihaka, R. & Gentleman, R. (1996) A language for data analysis and

graphics. Journal of Computational Graphical and Statistics, 5, 299–

314.

Irizarry, R.A., Hobbs, B., Collin, F., Beazer-Barclay, Y.D., Antonellis,

K.J., Scherf, U. & Speed, T.P. (2003) Exploration, normalization,

and summaries of high density oligonucleotide array probe level

data. Biostatistics, 4, 249–264.

Kim, S.Y., Lee, M.Y., Cho, K.C., Choi, Y.S., Choi, J.S., Sung, K.W.,

Kwon, O.J., Kim, H.S., Kim, I.K. & Jeong, S.W. (2003) Alterations in



mRNA expression of ribosomal protein S9 in hydrogen peroxide-

treated neurotumor cells and in rat hippocampus after transient

ischemia. Neurochemical Research, 28, 925–931.

Kirn-Safran, C.B., Dayal, S., Martin-DeLeon, P.A. & Carson, D.D.

(2000) Cloning, expression, and chromosome mapping of the

murine Hip/Rpl29 gene. Genomics, 68, 210–219.

Koonin, E.V., Wolf, Y.I. & Aravind, L. (2001) Prediction of the

archaeal exosome and its connections with the proteasome and the

translation and transcription machineries by a comparative-genomic

approach. Genome Research, 11, 240–252.

Lai, C.H., Chou, C.Y., Ch’ang, L.Y., Liu, C.S. & Lin, W. (2000) Iden-

tification of novel human genes evolutionarily conserved in

Caenorhabditis elegans by comparative proteomics. Genome Re-

search, 10, 703–713.

Liu, W.M., Mei, R., Di, X., Ryder, T.B., Hubbell, E., Dee, S., Webster,

T.A., Harrington, C.A., Ho, M.H., Baid, J. & Smeekens, S.P. (2002)

Analysis of high density expression microarrays with signed-rank

call algorithms. Bioinformatics, 18, 1593–1599.

Liu, J.J., Zhang, J., Ramanan, S., Julian, J., Carson, D.D. & Hooi, S.C.

(2004) Heparin/heparan sulfate interacting protein plays a role in

apoptosis induced by anticancer drugs. Carcinogenesis, 25, 873–879.

Luscombe, N.M., Austin, S.E., Berman, H.M. & Thornton, J.M. (2000)

An overview of the structures of protein-DNA complexes. Genome

Biology, 1, REVIEWS001.

Menne, T.F., Goyenechea, B., Sanchez-Puig, N., Wong, C.C., Tonkin,

L.M., Ancliff, P.J., Brost, R.L., Costanzo, M., Boone, C. & Warren,

A.J. (2007) The Shwachman-Bodian-Diamond syndrome protein

mediates translational activation of ribosomes in yeast. Nature

Genetics, 39, 486–495.

Meyuhas, O. & Klein, A. (1990) The mouse ribosomal protein L7 gene.

Its primary structure and functional analysis of the promoter region.

Journal of Biological Chemistry, 265, 11465–11473.

Meyuhas, O. & Perry, R.P. (1980) Construction and identification of

cDNA clones for mouse ribosomal proteins: application for the

study of r-protein gene expression. Gene, 10, 113–129.

Newberry, E.P., Latifi, T. & Towler, D.A. (1999) The RRM domain of

MINT, a novel Msx2 binding protein, recognizes and regulates the

rat osteocalcin promoter. Biochemistry, 38, 10678–10690.

Overman, P.F., Rhoads, D.D., Tasheva, E.S., Pyle, M.M. & Roufa, D.J.

(1993) Multiple regulatory elements ensure accurate transcription of

a human ribosomal protein gene. Somatic Cell and Molecular

Genetics, 19, 347–362.

Rujkijyanont, P., Beyene, J., Wei, K., Khan, F. & Dror, Y. (2007)

Leukaemia-related gene expression in bone marrow cells from

patients with the preleukaemic disorder Shwachman-Diamond

syndrome. British Journal of Haematology, 137, 537–544.

Rujkijyanont, P., Watanabe, K., Ambekar, C., Wang, H., Schimmer, A.,

Beyene, J. & Dror, Y. (2008) SBDS-deficient cells undergo accelerated

apoptosis through the Fas-pathway. Haematologica, 93, 363–371.

Safrany, G. & Perry, R.P. (1995) The relative contributions of various

transcription factors to the overall promoter strength of the mouse

ribosomal protein L30 gene. European Journal of Biochemistry, 230,

1066–1072.

Savchenko, A., Krogan, N., Cort, J.R., Evdokimova, E., Lew, J.M., Yee,

A.A., Sanchez-Pulido, L., Andrade, M.A., Bochkarev, A., Watson,

J.D., Kennedy, M.A., Greenblatt, J., Hughes, T., Arrowsmith, C.H.,

Rommens, J.M. & Edwards, A.M. (2005) The Shwachman-Bodian-

Diamond syndrome protein family is involved in RNA metabolism.

Journal of Biological Chemistry, 280, 19213–19220.

Selenko, P., Gregorovic, G., Sprangers, R., Stier, G., Rhani, Z., Kramer,

A. & Sattler, M. (2003) Structural basis for the molecular recogni-

tion between human splicing factors U2AF65 and SF1/mBBP.

Molecular Cell, 11, 965–976.

Shammas, C., Menne, T.F., Hilcenko, C., Michell, S.R., Goyenechea, B.,

Boocock, G.R., Durie, P.R., Rommens, J.M. & Warren, A.J. (2005)

Structural and mutational analysis of the SBDS protein family.

Insight into the leukemia-associated Shwachman-Diamond

Syndrome. Journal of Biological Chemistry, 280, 19221–19229.

Shi, Y., Zhai, H., Wang, X., Han, Z., Liu, C., Lan, M., Du, J., Guo, C.,

Zhang, Y., Wu, K. & Fan, D. (2004) Ribosomal proteins S13 and

L23 promote multidrug resistance in gastric cancer cells by sup-

pressing drug-induced apoptosis. Experimental Cell Research, 296,

337–346.

Shwachman, H., Diamond, L.K., Oski, F.A. & Khaw, K.T. (1964) The

Syndrome of Pancreatic Insufficiency and Bone Marrow Dysfunc-

tion. Journal of Pediatrics, 65, 645–663.

Smith, O.P., Hann, I.M., Chessells, J.M., Reeves, B.R. & Milla, P.

(1996) Haematological abnormalities in Shwachman-Diamond

syndrome. British Journal of Haematology, 94, 279–284.

Smyth, G.K. (2004) Linear models and empirical bayes methods for

assessing differential expression in microarray experiments. Statis-

tical Applications in Genetics and Molecular Biology, 3, Article3.

Takagi, M., Sueishi, M., Saiwaki, T., Kametaka, A. & Yoneda, Y. (2001)

A novel nucleolar protein, NIFK, interacts with the forkhead asso-

ciated domain of Ki-67 antigen in mitosis. Journal of Biological

Chemistry, 276, 25386–25391.

Takagi, M., Absalon, M.J., McLure, K.G. & Kastan, M.B. (2005) Reg-

ulation of p53 translation and induction after DNA damage by

ribosomal protein L26 and nucleolin. Cell, 123, 49–63.

Virbasius, J.V. & Scarpulla, R.C. (1994) Activation of the human

mitochondrial transcription factor A gene by nuclear respiratory

factors: a potential regulatory link between nuclear and mitochon-

drial gene expression in organelle biogenesis. Proceedings of the

National Academy of Sciences of the United States of America, 91,

1309–1313.

Wang, H., Zhao, L.N., Li, K.Z., Ling, R., Li, X.J. & Wang, L. (2006)

Overexpression of ribosomal protein L15 is associated with cell

proliferation in gastric cancer. BMC Cancer, 6, 91.

Wessels, D., Srikantha, T., Yi, S., Kuhl, S., Aravind, L. & Soll, D.R.

(2006) The Shwachman-Bodian-Diamond syndrome gene encodes

an RNA-binding protein that localizes to the pseudopod of Dicty-

ostelium amoebae during chemotaxis. Journal of Cell Science, 119,

370–379.

Wu, L.F., Hughes, T.R., Davierwala, A.P., Robinson, M.D., Stoughton,

R. & Altschuler, S.J. (2002) Large-scale prediction of Saccharomyces

cerevisiae gene function using overlapping transcriptional clusters.

Nature Genetics, 31, 255–265.

Yanagida, M., Hayano, T., Yamauchi, Y., Shinkawa, T., Natsume, T.,

Isobe, T. & Takahashi, N. (2004) Human fibrillarin forms a sub-

complex with splicing factor 2-associated p32, protein arginine

methyltransferases, and tubulins alpha 3 and beta 1 that is inde-

pendent of its association with preribosomal ribonucleoprotein

complexes. Journal of Biological Chemistry, 279, 1607–1614.

Yoganathan, T., Bhat, N.K. & Sells, B.H. (1992) A positive regulator of

the ribosomal protein gene, beta factor, belongs to the ETS onco-

protein family. Biochemical Journal, 287 (Pt 2), 349–353.

Yoo, Y.A., Kim, M.J., Park, J.K., Chung, Y.M., Lee, J.H., Chi, S.G., Kim,

J.S. & Yoo, Y.D. (2005) Mitochondrial ribosomal protein L41



suppresses cell growth in association with p53 and p27Kip1.

Molecular and Cellular Biology, 25, 6603–6616.

Yoshihama, M., Uechi, T., Asakawa, S., Kawasaki, K., Kato, S., Higa, S.,

Maeda, N., Minoshima, S., Tanaka, T., Shimizu, N. & Kenmochi, N.

(2002) The human ribosomal protein genes: sequencing and com-

parative analysis of 73 genes. Genome Research, 12, 379–390.

Zhang, Y., Wolf, G.W., Bhat, K., Jin, A., Allio, T., Burkhart, W.A. &

Xiong, Y. (2003) Ribosomal protein L11 negatively regulates onco-

protein MDM2 and mediates a p53-dependent ribosomal-stress

checkpoint pathway. Molecular and Cellular Biology, 23, 8902–8912.

Zhang, S., Shi, M., Hui, C.C. & Rommens, J.M. (2006) Loss of the

mouse ortholog of the shwachman-diamond syndrome gene (Sbds)

results in early embryonic lethality. Molecular and Cellular Biology,

26, 6656–6663.

Supporting information

Additional Supporting Information may be found in the

online version of this article:

Table SI. Ribosomal protein-related gene expression in

Shwachman-Diamond syndrome bone marrow mononuclear

cells.

Table SII. rRNA-related gene expression in Shwachman-

Diamond syndrome bone marrow mononuclear cells.

Table SIII. mRNA processing gene expression in Shwach-

man-Diamond syndrome bone marrow mononuclear cells.

Table SIV. Expression of genes involved in global tran-

scription regulation in Shwachman-Diamond syndrome bone

marrow mononuclear cells.

Table SV. Expression of non-ribosomal protein 5¢ terminal

oligopyrimidine genes (Non RP TOP genes) in Shwachman-

Diamond syndrome bone marrow mononuclear cells.

Please note: Wiley-Blackwell are not responsible for the

content or functionality of any supporting materials supplied

by the authors. Any queries (other than missing material)

should be directed to the corresponding author for the article.



Bone marrow cells from patients with Shwachman‐Diamond syndrome abnormally express genes involved in ribosome biogenesis and RNA processing

Documents