This article appeared in a journal published by Elsevier. The attachedcopy is furnished to the author for internal non-commercial researchand education use, including for instruction at the authors institution

and sharing with colleagues.

Other uses, including reproduction and distribution, or selling orlicensing copies, or posting to personal, institutional or third party

websites are prohibited.

In most cases authors are permitted to post their version of thearticle (e.g. in Word or Tex form) to their personal website orinstitutional repository. Authors requiring further information

regarding Elsevier’s archiving and manuscript policies areencouraged to visit:

http://www.elsevier.com/authorsrights

Author's personal copy

Novel heparan sulfate assay by using automated high-throughput massspectrometry: Application to monitoring and screeningfor mucopolysaccharidoses

Tsutomu Shimada a,1, Joan Kelly b,1, William A. LaMarr b, Naomi van Vlies c, Eriko Yasuda a, Robert W. Mason a,William Mackenzie a, Francyne Kubaski a, Roberto Giugliani d, Yasutsugu Chinen e, Seiji Yamaguchi f,Yasuyuki Suzuki g, Kenji E. Orii h, Toshiyuki Fukao h, Tadao Orii h, Shunji Tomatsu a,h,⁎a Nemours/Alfred I duPont Hospital for Children, Wilmington, DE, USAb Agilent Technologies, Inc., Wakefield, MA, USAc Laboratory of Genetic Metabolic Disease, Academic Medical Center, University of Amsterdam, The Netherlandsd Medical Genetics Service, HCPA and Department of Genetics, UFRGS, Porto Alegre, Brazile Department of Pediatrics, Faculty of Medicine University of the Ryukyus, Okinawa, Japanf Department of Pediatrics, Shimane University, Izumo, Japang Medical Education Development Center, Gifu University, Gifu, Japanh Department of Pediatrics, Graduate School of Medicine, Gifu University, Gifu, Japan

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

Article history:Received 9 July 2014Accepted 9 July 2014Available online 21 July 2014

Keywords:Automated high-throughput mass spectrometryMucopolysaccharidosesHeparan sulfateBiomarkerNewborn screening

Mucopolysaccharidoses (MPS) are caused by deficiency of one of a group of specific lysosomal enzymes, resultingin excessive accumulation of glycosaminoglycans (GAGs). We previously developed GAG assay methods usingliquid chromatography tandemmass spectrometry (LC–MS/MS); however, it takes 4–5 min per sample for anal-ysis. For the large numbers of samples in a screening program, a more rapid process is desirable. The automatedhigh-throughput mass spectrometry (HT-MS/MS) system (RapidFire) integrates a solid phase extraction robot toconcentrate and desalt samples prior to direction into the MS/MS without chromatographic separation; therebyallowing each sample to beprocessedwithin 10 s (enabling screening ofmore thanonemillion samples per year).The aim of this study was to develop a higher throughput system to assay heparan sulfate (HS) using HT-MS/MS,and to compare its reproducibility, sensitivity and specificity with conventional LC–MS/MS.HS levels were measured in the blood (plasma and serum) from control subjects and patients with MPS II, III, or IVand in dried blood spots (DBS) from newborn controls and patients with MPS I, II, or III. Results obtained from HT-MS/MS showed 1) that there was a strong correlation of levels of disaccharides derived from HS in the blood, be-tween those calculated using conventional LC–MS/MS and HT-MS/MS, 2) that levels of HS in the blood were signif-icantly elevated in patients with MPS II and III, but not in MPS IVA, 3) that the level of HS in patients with a severeformofMPS IIwas higher than that in an attenuated form, 4) that reduction of bloodHS levelwas observed inMPS IIpatients treated with enzyme replacement therapy or hematopoietic stem cell transplantation, and 5) that levels ofHS in newborn DBS were elevated in patients with MPS I, II or III, compared to those of control newborns.In conclusion, HT-MS/MS provides much higher throughput than LC–MS/MS-based methods with similar sensi-tivity and specificity in an HS assay, indicating that HT-MS/MS may be feasible for diagnosis, monitoring, andnewborn screening of MPS.

© 2014 Elsevier Inc. All rights reserved.

1. Introduction

Mucopolysaccharidoses (MPS) are a group of lysosomal storage dis-eases caused by deficiency of the lysosomal enzymes required for the

degradation of glycosaminoglycans (GAGs) such as chondroitin sulfate(CS), dermatan sulfate (DS), heparan sulfate (HS), keratan sulfate(KS), and hyaluronan. There are 11 known enzyme deficiencies,resulting in seven distinct forms of MPS with a collective incidence ofmore than 1 in 25,000 live births. Accumulation of GAGs causes progres-sive damage of multiple tissues including the brain, lung, heart, liver,kidney, and bone. Most clinical signs and symptoms for MPS patientsdo not appear immediately after birth; however, the subsequent onsetof clinical signs and symptoms progresseswith age. Although the symp-toms and severity of MPS vary with individual patient and subtype of

Molecular Genetics and Metabolism 113 (2014) 92–99

⁎ Corresponding author at: Skeletal Dysplasia Lab, Nemours Biomedical Research,Nemours/Alfred I. duPont Hospital for Children, 1600 Rockland Rd., Wilmington, DE,USA. 19899-0269. Fax: 302 651 6888.

E-mail address: stomatsu@nemours.org (S. Tomatsu).1 The authors state that the first two authors should be regarded as joint first authors.

http://dx.doi.org/10.1016/j.ymgme.2014.07.0081096-7192/© 2014 Elsevier Inc. All rights reserved.

Contents lists available at ScienceDirect

Molecular Genetics and Metabolism

j ourna l homepage: www.e lsev ie r .com/ locate /ymgme

Author's personal copy

MPS, the average life span in patients with a severe form is one to twodecades if untreated.

Currently, enzyme replacement therapy (ERT), hematopoietic stemcell transplantation (HSCT), substrate reduction therapy (SRT), andgene therapy are clinically in use or are being investigated in clinical tri-als for some types ofMPS patient. Starting these treatments at birth or ata very early stagewill provide themost significant impact on the clinicalcourse of the disease [1]. To provide better quality of life for the patients,early diagnosis and treatment are required. Since unique clinical fea-tures in most MPS patients are not apparent until they are around twoyears old and general physicians have little knowledge ofMPS,most pa-tients aremisdiagnosed or undiagnosed until irreversible damage to thebrain and/or bones has started. Thus, a novel, accurate, sensitive, eco-nomical and rapid diagnostic method applicable to newborn screening(NBS) for MPS is critical to improving therapeutic efficacy.

Currently, conventional screening methods for MPS are dye-spectrometric methods such as dimethylmethylene blue (DMB) [2–4]and Alcian Blue [5,6] which measure total GAGs from urine samples.When urine assays provide a positive result, the definitive diagnosis isdetermined by measuring the enzyme activities in lymphocytes or fi-broblasts. However, these current methods cannot be applied to bloodand/or tissue extractswithout prior protease, nuclease or hyaluronidasetreatment [7]. Total GAG concentration in urine does not reflect the se-verity of the neurological or skeletal signs and symptoms [4] and sub-stantial overlap of the total urine GAG level between age-matchedcontrols and MPS IV patients is observed [4], resulting in misdiagnosisof many patients. Thin layer chromatography, which is a semi-quantitative method for each GAG, is used to allow preliminary classifi-cation of patients among subgroups to reduce the number of enzymeassays necessary for final diagnosis [8,9]. There are several othermethods for measuring specific GAGs in the blood, including ELISA[10,11] and HPLC [12,13]. However, ELISA kits are not commerciallyavailable for all types of GAGs and they cannot measure subtype of HSor KS. HPLC is not suitable for mass screening since the assay is time-consuming with low sensitivity.

NBS is recognized as an essential, preventive public health pro-gram for early identification of diseases in newborns that can affecttheir long-term health. A suitable method for NBS must be inexpen-sive and should be performed using a dried blood spot (DBS) from aGuthrie card. At present there is no reliable NBS to cover all types ofMPSs. We and others have developed new methods to assay CS, DS,HS, and KS simultaneously in the blood, urine, and/or DBS samplesby using liquid chromatography tandem mass spectrometry (LC–MS/MS) [7,14–23]. The LC–MS/MS method not only shows sensitiv-ity and specificity for detecting all subtypes of MPS, but also moni-tors therapeutic efficacy in MPS patients and animal models [1,16,22,24,25]; however, since LC processing is time consuming, themain drawback of this method is throughput. This factor may limitits utility for assaying large numbers of samples.

The automated high-throughput mass spectrometry (HT-MS/MS)system (RapidFire; Agilent Technologies) eliminates the chromato-graphic process, enabling sample-to-sample cycle times to be reducedto seconds, thereby removing the bottleneck of throughputwhilemain-taining the quality and reliability of standard LC–MS/MS read-outs. Thesample is aspirated to amatrix for concentration anddesalting, followedby direct injection into MS/MS without chromatographic separation.Each sample is processed within 10 s, yielding much faster throughputthan traditional LC–MS/MS based methods. A single 384 well plate canbe read in less than 40 min, indicating that this HT-MS/MS system cananalyze over one million samples annually. Moreover, HT-MS/MS hasbeen shown to provide sensitivity and specificity equivalent to a scintil-lation proximity assay [26], and an LC–MS/MS assay for some applica-tions [19,27]. The speed and efficiency of HT-MS/MS can allow forexperiments that would otherwise be deemed untenable under normalcircumstances. The HT-MS/MS system has been validated as suitable formany drug discovery programs [26,28–34], and ADME (Absorption,

Distribution,Metabolism and Excretion) based applications [27].We re-cently reported that HT-MS/MS could measure HS levels in controlhuman blood much faster than LC–MS/MS, and showed a strong corre-lation with levels measured by LC–MS/MS [19].

In this studywe have evaluated a higher throughput system to assayHS levels in the blood and DBS from newborn patients with MPS byusing HT-MS/MS, and have compared its reproducibility, sensitivity,and specificity with conventional LC–MS/MS.

2. Materials and methods

2.1. Materials

2.1.1. Standards and enzymesTo digest polymer HS to disaccharides, the enzyme heparitinasewas

obtained from Seikagaku Corporation (Tokyo, Japan). Chondrosine wasused as an internal standard (IS), and unsaturated disaccharides,[ΔDiHS-0S, 2-acetamido-2-deoxy-4-O-(4-deoxy-a-L-threo-hex-4-enopyranosyluronic acid)-D-glucose; ΔDiHS-NS, 2-deoxy-2-sulfamino-4-O-(4-deoxy-a-L-threo-hex-4-enopyranosyluronic acid)-D-glucose]were used for generation of standard curves. Stock solutions ofΔDiHS-0S (100 μg/ml), ΔDiHS-NS (100 μg/ml), and IS (5 mg/ml)were prepared separately in ddH2O. Standard working solutions ofΔDiHS-0S and ΔDiHS-NS, (15.625, 31.25, 62.5, 125, 250, 500, and1000 ng/ml), each mixed with IS solution (5 μg/ml), were prepared.

2.1.2. Blood and DBS samplesWe collected 95 blood (plasma or serum) samples from MPS pa-

tients as follows: 55 MPS II patients (age 2.0–35.0 years, mean11.4 years; 45 severe, 10 attenuated), 18 MPS III patients (age 4.0–27.0 years, mean 12.5 years; 6 IIIA, 12 IIIB), and 22 MPS IV patients(age 3.4–56.0 years, mean 17.2 years; 17 IVA, 5 IVB). We obtainedthe samples from patients with MPS II who received ERT (16) or HSCT(9), and one patient with MPS IIIB who received HSCT. All patientswere diagnosed as having below 5% of normal enzymatic activity.We also collected 133 blood (plasma or serum) samples from normalcontrols. The age of control samples ranged between 0 and 80 years(mean 5.2 years). In previous LC–MS/MS experiments, we confirmedthat specificity and sensitivity of each GAG value were comparableusing plasma or serum [7,14–17,19]. Diagnosis for MPS patientswas confirmed by enzyme assay at each institute providing samples.The clinical severity in patients with MPS II was assessed by the pres-ence (severe) or absence (attenuated) of central nervous system(CNS) involvement. Clinical phenotypes in all patients with MPS IIIand MPS IVA were defined as severe, and all patients with MPS IVBwere defined as attenuated in terms of growth compared to MPS IVApatients.

We also collected DBS samples from 22 anonymous control new-borns and 12 newborns with MPS (6 MPS I, 1 MPS II, and 5 MPS III).These MPS patients had been diagnosed enzymatically and clinically,and DBS samples obtained at birth were provided by the newbornscreening center after diagnosis. These control and patient sampleswere assayed in a blinded manner.

Written informed consentwas obtained for each patient prior to ini-tiation of the study, and the study was approved by the institutional re-view board of each institute providing samples for this study.

2.2. Methods

2.2.1. Sample preparationBlood specimens and standards were prepared as described previ-

ously [18,19]. First, 10 μl of each serum or plasma sample and 90 μl of50 mM Tris–hydrochloric acid buffer (pH 7.0) were placed in wells ofAcroPrep™ Advance 96-Well Filter Plates that have UltrafiltrationOmega 10 K membrane filters (PALL corporation, NY, USA). The filterplates were placed on the receiver and centrifuged at 2000 g for

93T. Shimada et al. / Molecular Genetics and Metabolism 113 (2014) 92–99

Author's personal copy

15min to remove free disaccharides. The membrane plates were trans-ferred to a fresh receiver plate. Standardswere added to unusedwells ofthe filter plate. Then, 20 μl of IS solution (500 ng/ml), 70 μl of 50 mMTris–hydrochloric acid buffer (pH 7.0), and 10 μl of heparitinase solu-tion (2mUper sample) were added to each filter well. The plate was in-cubated in a water bath at 37 °C for 15 h and centrifuged at 2000 g for15 min. The receiver plate containing disaccharides was storedat−20 °C until analysis by HT-MS/MS or LC–MS/MS. Disks of DBS sam-ples (3 mm in diameter) were punched out and incubated in 0.1% bo-vine serum albumin for 15 min on Omega 10 K membrane filters.After centrifugation at 2000 g for 15 min, the mixture of IS, 50 mMTris–hydrochloric acid buffer (pH 7.0), and heparitinase solution as de-scribed above was added to each filter well, and processed as per theblood samples.

2.2.2. Automated high-throughput mass spectrometry (HT-MS/MS) system(RapidFire)

Samples stored at−20 °C were allowed to thaw at ambient tem-perature. General operation of the HT-MS/MS has been describedpreviously [35]. In brief, 10 μl of sample was aspirated directly frommicrotiter plates and loaded onto the RapidFire microscale solid-phase extraction cartridge D (graphitic carbon) to remove saltswith HPLC-grade water in a 2.5-sec wash cycle. Analytes were thencoeluted into the mass spectrometer in a 3.5-s elution cycle using25% acetonitrile, 25% acetone containing 5 mM ammonium acetate.ΔDiHS-0S, ΔDiHS-NS and IS were measured using a Sciex API 4000QQQ mass spectrometer (Applied Biosystems, Concord, Ontario,Canada) in negative ESI mode following multiple-reaction monitor-ing (MRM) transitions at 378.2/174.3, 416.0/137.7 and 354.3/193.1,respectively. Sample data were integrated using the RapidFire Inte-grator software (Agilent Technologies, Inc: Santa Clara, CA) resultingin individual AUC (area under the curve) values for each of the threeanalytes in each sample. Each sample was prepared and analyzed induplicate on separate days. HS levels were measured and averaged.

2.2.3. LC–MS/MSThe chromatographic system consisted of the 1260 Infinity LC

(Agilent Technologies, Palo Alto, CA, USA) and a Hypercarb column(2.0 mm i.d. 50 mm, 5 μm, Thermo Electron, USA). The 6460 TripleQuad mass spectrometer (Agilent Technologies) was operated inthe negative ion detection mode. The procedure was described pre-viously [18]. The MRM of each disaccharide was the same as thatused for HT-MS/MS analysis.

2.2.4. Method validationIntra-day precision evaluated as coefficient of variation (CV) was

determined by replicate analyses (n = 5) of three different controlblood samples. Inter-day precision was determined by replicateanalyses (n = 5) of three different control blood samples on threeseparate days.

The selectivity of the assay was investigated by processing andanalyzing five independent samples by the method describedabove without enzymatic digestion. Calibration curves were con-structed by plotting the peak area ratio of the analytes to IS againstthe concentration of the analytes. Each calibration curve consistedof seven calibration points. HS values obtained by HT-MS/MS werecompared with those by LC–MS/MS.

2.2.5. Data analysisThe data were analyzed by ANOVA and Tukey's test. Analysis

was performed using SPSS for Windows (version 22.0, IBM, Chicago,IL, USA).

3. Results

3.1. MRM of HT-MS/MS

Typical MRM of HT-MS/MS for ΔDiHS-0S standards with serial dilu-tions and samples is shown in Fig. 1. Similar MRMswere obtained for ISand ΔDiHS-NS. The HT-MS/MS system allowed us tomeasure 8–9 sam-ples per min.

3.2. Precision

Results of intra- and inter-assay precision for ΔDiHS-0S and ΔDiHS-NS in control blood samples are as follows. The intra-assay precisionvalues/coefficient of variation (CV) determined from the analysis ofΔDiHS-0S and ΔDiHS-NS for control blood samples are less than 14.7and 14.9%, respectively. The inter-assay precision values/CVs for thesedisaccharides in control blood samples are less than 13.9 and 13.7%,respectively. These results demonstrate the reproducibility and preci-sion of the method.

3.3. Correlation between HT-MS/MS and LC–MS/MS

All blood samples were also measured with the LC–MS/MS systemthat separates the disaccharides with a chromatographic system. Therewere strong correlations betweenHT-MS/MS and LC–MS/MS inmeasure-ments of both ΔDiHS-0S (r2 = 0.9409) and ΔDiHS-NS (r2 = 0.889), andregression coefficients for ΔDiHS-0S and ΔDiHS-NS were 1.2 and 1.0,respectively (Fig. 2)

3.4. Measurements of GAGs in the blood

3.4.1. ControlThe levels ofΔDiHS-0S andΔDiHS-NSwere the highest in newborns

and decreased until three years of age. After three years of age, the levelsof these two HS remained stable (Fig. 3).

3.4.2. Mucopolysaccharidosis type II (MPS II)Levels of bloodΔDiHS-0S andΔDiHS-NS in untreated patientswith a

severe form of MPS II (n = 24) were significantly higher, comparedwith those in control groups (Table 1, Fig. 3). The increase in untreatedpatients with an attenuated form of MPS II (n= 9) did not reach statis-tical significance with the small sample size. The level of ΔDiHS-0S inuntreated patients with a severe form of MPS II was higher than thatin patients with an attenuated form. The level of ΔDiHS-NS was alsoelevated in the untreated patients with the severe form of MPS II com-pared to those with the attenuated form (Table 1).

The levels ofΔDiHS-0S in the blood fromHSCT-treated patients witha severe form of MPS II 8.9 years (average) after treatment were signif-icantly reduced, compared with untreated patients (Table 1). Levels ofΔDiHS-0S in the blood from these HSCT treated patients were also sig-nificantly lower than those in the blood from ERT-treated patients(Table 1). Seven patients with the severe form of MPS II providedblood specimens before and an average of two years after startingERT. Levels of ΔDiHS-0S and ΔDiHS-NS were significantly decreasedafter ERT treatment (Fig. 4).

3.4.3. Mucopolysaccharidosis type III (MPS III)Untreated patients with MPS IIIA and MPS IIIB (n = 6 and n = 11,

respectively) had significant elevation of levels of ΔDiHS-0S andΔDiHS-NS from the blood compared with those in control groups(Table 2; Fig. 3). A 15 year-oldMPS IIIB female, who had undergone suc-cessful HSCT at 1 year and 10 months, had blood levels ofΔDiHS-0S andΔDiHS-NS in the normal range (Table 2). This patient could speak withlimited words, recognize her parents, and walk by herself at the age of15 years. Thus, her cognitive and physical function was much better,compared with that in untreated age-matched MPS IIIB patients,

94 T. Shimada et al. / Molecular Genetics and Metabolism 113 (2014) 92–99

Author's personal copy

indicating that the biochemical correction by HSCT resulted in an im-proved clinical outcome.

3.4.4. Mucopolysaccharidosis type IV (MPS IV)Blood HS level in untreated and ERT-treated patients with MPS IVA

(n=11and n=6, respectively) showed the same levels as normal con-trols. Untreated patients with MPS IVB also had the same range of HScompared with normal controls (Table 3).

3.5. Measurements of GAGs in newborn DBS

The levels of ΔDiHS-0S and ΔDiHS-NS extracted from DBS of new-borns later diagnosedwithMPS I andMPS III were significantly elevatedcompared to control newborn samples. The levels of ΔDiHS-0S andΔDiHS-NS from DBS of a newborn later diagnosed with MPS II were 3and 1.5 times higher than those from DBS from control newborns(Table 4). For all three forms of MPS, ΔDiHS-0S distinguished patientsfrom controls better than ΔDiHS-NS.

4. Discussion

We have demonstrated 1) that the RapidFire platform is capable ofdetecting disaccharides of ΔDiHS-0S and ΔDiHS-NS in the blood withthe same reproducibility, sensitivity, and specificity as described forthe LC–MS/MS method [7,14–17,19–22], suggesting that chromato-graphic separation of disaccharides is not essential for the detectionand identification of ΔDiHS-0S and ΔDiHS-NS, 2) that HS assays byHT-MS/MS can distinguish untreated patients with MPS II or III fromhealthy controls, showing the potential use of this technology as a diag-nostic platform, 3) that patient groups treated by ERT or HSCT had re-duced HS levels, suggesting that HS measured by HT-MS/MS can beused as a biomarker for monitoring therapeutic effect, 4) that the runtime of 6–8 s per sample provides 10–100 fold higher throughputwhen compared to traditional LC–MS/MS, indicating that HT-MS/MSmay be useful formass screening applications forMPS, and 5) that levelsof HS in newborn DBS were elevated in all MPS I, II and III patients, sug-gesting that HT-MS/MS could be used in a NBS to identify these patients.

Time

Series of dilutions in duplicate

Inte

nsity

(cou

nts

per s

ec.)

SampleSample

Fig. 1.MRMs of HT-MS/MS. Multiple injections of ΔDiHS-0S (Q1; 378.2, Q3; 174.3) with a series of dilutions (15.625, 31.25, 62.5, 125, 250, 500, and 1000 ng/ml) in duplicate show sevengradient peaks per set of dilutions (surrounding samples are also shown).

a) b)

y = 1.2263x + 5.4158R² = 0.9409

0.0

50.0

100.0

150.0

200.0

250.0

300.0

350.0

400.0

450.0

0.0 50.0 100.0 150.0 200.0 250.0 300.0 350.0

Conc

. (ng

/ml)

in L

C-M

S/M

S

Conc. (ng/ml) in HT-MS/MS

ΔDiHS-0Sy = 1.025x - 10.669

R² = 0.8898

0.0

20.0

40.0

60.0

80.0

100.0

120.0

140.0

160.0

180.0

0.0 50.0 100.0 150.0 200.0

Conc

. (ng

/ml)

in L

C-M

S/M

S

Conc. (ng/ml) in HT-MS/MS

ΔDiHS-NS

Fig. 2. Correlation in HS concentration between LC–MS/MS and HT-MS/MS. a) ΔDiHS-0S; b) ΔDiHS-NS. Coefficient of determination is r2 = 0.9409 and r2 = 0.8898 for ΔDiHS-0S andb) ΔDiHS-NS, respectively; showing a strong correlation in ΔDiHS-0S and ΔDiHS-NS between LC–MS/MS and HT-MS/MS in human blood samples (n = 228).

95T. Shimada et al. / Molecular Genetics and Metabolism 113 (2014) 92–99

Author's personal copy

In MPS, the undegraded GAGs are stored in lysosomes, secretedinto the circulation and excreted into the urine. In this study, bloodHS was digested to disaccharides by heparitinase using the samemethod as described for the LC–MS/MS method [7,14,15]. Twosubclasses of HS (ΔDiHS-0S and ΔDiHS-NS) were detected by HT-MS/MS simultaneously. The levels of ΔDiHS-0S and ΔDiHS-NS in

blood samples from both control and patient subjects, determined byHT-MS/MS, correlated strongly with those determined by LC–MS/MS,both in this and previous studies [15,17]. The calculated valuesof ΔDiHS-0S and ΔDiHS-NS using both methods were similar, indi-cating that LC–MS/MS and HT-MS/MS are comparable in sensitivityand specificity.

b)a)

d)c)

f)e)

1

10

100

1000

0 20 40 60 80 100

ΔD

iHS-

0S (n

g/m

l)

Age (year)0 20 40 60 80 100

Age (year)

0 20 40 60 80 100Age (year)

0 20 40 60 80 100Age (year)

0 20 40 60 80 100Age (year)

0 20 40 60 80 100Age (year)

Control Untreated MPS II severeERT treated MPS II severe HSCT treated MPS II severeUntreated MPS II a�enuated ERT treated MPS II a�enuated

1

10

100

1000

ΔD

iHS-

NS

(ng/

ml)

Control Untreated MPS II severeERT treated MPS II severe HSCT treated MPS II severeUntreated MPS II a�enuated ERT treated MPS II a�enuated

1

10

100

1000

ΔD

iHS-

0S (n

g/m

l)

Control Untreated MPS IIIAUntreated MPS IIIB HSCT-treated MPS IIIB

1

10

100

1000Δ

DiH

S-N

S (n

g/m

l)Control Untreated MPS IIIAUntreated MPS IIIB HSCT treated MPS IIIB

1

10

100

1000

ΔD

iHS-

0S (n

g/m

l)

Control Untreated MPS IVA

ERT treated MPS IVA Untreated MPS IVB

1

10

100

1000

ΔD

iHS-

NS

(ng/

ml)

Control Untreated MPS IVA

ERT treated MPS IVA Untreated MPS IVB

Fig. 3. HS levels in the blood of patients with MPS and control subjects. a) ΔDiHS-0S in patients with MPS II, b) ΔDiHS-NS in patients with MPS II, c) ΔDiHS-0S in patients with MPS III,d) ΔDiHS-NS in patients with MPS III, e) ΔDiHS-0S in patients with MPS IV, and f) ΔDiHS-NS in patients with MPS IV. Results of all specimens from patients and control subjects wereplotted on a semilogarithmic scale with respect to age (years).

96 T. Shimada et al. / Molecular Genetics and Metabolism 113 (2014) 92–99

Author's personal copy

Overall, the levels of ΔDiHS-0S and ΔDiHS-NS obtained in the bloodof normal controls and MPS patients by HT-MS/MS were equivalent tothose obtained by LC–MS/MS, indicating that the use of the HT-MS/MSplatform did not affect the specificity and sensitivity of the assay. TheHT-MS/MS system does not only acquire data much faster than tradi-tional LC-MS/MS systems, but also provide data of a similar qualitywithout using a time-consuming separation process.

MPS II results from a deficiency of iduronate-2-sulfatase, leading toaccumulation of HS and DS. In this studyΔDiHS-0S andΔDiHS-NS levelsincreased in untreated patients with a severe form of MPS II comparedwith those in the control groups, as described previously using theLC–MS/MS method [7,15]. There was a significant increase in the levelof ΔDiHS-0S in patients with a severe form of this MPS compared tothose with an attenuated form, and the level of ΔDiHS-0S in the ERT-or HSCT-treated group was significantly lower compared with that inthe untreated group. Furthermore, the level of ΔDiHS-0S was signifi-cantly lower in the blood from patients treated by HSCT than in theblood from patients treated by ERT. Levels of ΔDiHS-NS were alsolower in the blood from the HSCT treated patients but the differencewas not statistically significant. Consistent with our results, Tanakaet al. reported that HSCT given during early stages of MPS II provides ef-fectiveness toward brain and heart disorders with a decrease of urinaryGAG level and that patients treated by ERT at a similar stage showed lessreduction of urinary GAGs [36]. For ERT, the timing of the blood drawnafter a round of therapy is likely to affect blood levels of GAGs, so moredetailed studies are needed to determinewhether HSCT or ERT is betterat reducing GAGs. In future studies it will be of great interest to explorewhether differences inΔDiHS-0S and/orΔDiHS-NS levels are associatedwith clinical severity and/or improvement of CNS involvement.

MPS IIIA and MPS IIIB result from deficiencies of enzymes heparanN-sulfatase and α-N-acetylglucosaminidase, respectively, leading tothe accumulation of HS. The levels of ΔDiHS-0S and ΔDiHS-NS in un-treated patients with MPS IIIA or MPS IIIB were significantly raised

compared with those in normal controls, as described previouslyusing LC–MS/MS [7,15]. The frequency of MPS IIIB is higher in a regionaldistrict in Japan, based upon the founder effect [37]. It is noteworthythat one Japanese MPS IIIB patient who received successful HSCTwhen under two years of age had ΔDiHS-0S and ΔDiHS-NS levels inthe normal range. After over 13 years of follow-up, the level ofneurocognition in this patient is far better than that in age-matcheduntreated patients, although neurological involvement still remainsobvious. The finding that the primary storage material, HS, is withinthe normal range is consistent with the less severe clinical course inthis patient, showing that both ΔDiHS-0S and ΔDiHS-NS should begood biomarkers for monitoring the clinical course and therapeutic effi-cacy in MPS III.

MPS IVA and MPS IVB are caused by the deficiency of N-acetylgalactosamine-6-sulfate sulfatase (GALNS) and β-galactosidase(β-Gal), respectively, and consequently KS but not HS should accumu-late in the blood. In this study we found that HS levels in patients withMPS IV were the same as those in the control group, consistent withthe predicted result and previous reports [15,17].

In the United States, nearly 4.2 million neonates undergo NBS annu-ally [38]. There are no specific incidence studies of MPS in the USA, butwe estimate that approximately 200 newborns are affected with MPSannually in the USA [39–45]. In this study, we show that the levels ofΔDiHS-0S and ΔDiHS-NS from DBS of newbornMPS I, II and III patientsare higher than those of newborn control. De Ruijter et al. also reportedthat the LC–MS/MS method could show elevation of ΔDiHS-0S in thepatients with these forms of MPS using newborn DBS samples [20].Our results lead to the possibility of applying HT-MS/MS to NBS forMPS, although further studies with more samples are required. Thefact that HT-MS/MS can process the same samples with throughputsup to 10-100 folds faster than conventional LC–MS/MSmakes this plat-form a more appropriate system, due to time-saving and potential costreductions for NBS.

HS can be elevated in the blood due to a release from the endotheliumduring ischemia/reperfusion injury and severe inflammation [46,47]. Itcan also be elevated in the blood after virus-infection and septic shock[48,49]. Thus, HT-MS/MS could be a novel high throughput system to

Table 1Levels of ΔDiHS-0S and ΔDiHS-NS of patients with MPS and control subjects over 3 yearsof age.

ΔDiHS-0S (ng/ml) ΔDiHS-NS (ng/ml)

Control 27.5 ± 8.7 22.7 ± 4.8Untreated MPS II severe 144.5 ± 77.9a 89.4 ± 100.1a

ERT treated MPS II severe 116.1 ± 101.0a 61.0 ± 41.4HSCT treated MPS II severe 40.3 ± 11.1b,c 34.1 ± 5.9Untreated MPS II attenuated 49.2 ± 21.9b 32.2 ± 11.5ERT treated MPS II attenuated 30.5 22

Number of each group is as follows; control (n = 59), untreatedMPS II (severe; n = 24),ERT-treatedMPS II (severe; n = 15), HSCT-treatedMPS II (severe; n = 6), untreatedMPSII (attenuated; n = 9), and ERT-treated MPS II (attenuated; n = 1).

a Significantly different from the control at p b 0.05.b Significantly different from the untreated MPS II severe at p b 0.05.c Significantly different from the ERT-treated MPS II severe at p b 0.05.

0

50

100

150

200

250

300

Before A�er

ΔD

iHS-

0S (n

g/m

l)

0

20

40

60

80

100

120

Before A�er

ΔD

iHS-

NS

(ng/

ml)

a) b) **

Fig. 4. HS levels in serum of seven patients with MPS II who underwent ERT for more than two years. a) ΔDiHS-0S; b) ΔDiHS-NS. Each line describes an individual patient. *: showssignificant difference between pre-treatment and post-treatment.

Table 2Levels of ΔDiHS-0S and ΔDiHS-NS in patients with MPS III and control subjects over3 years of age.

ΔDiHS-0S (ng/ml) ΔDiHS-NS (ng/ml)

Control 27.5 ± 8.7 22.7 ± 4.8Untreated MPS IIIA 126.6 ± 30.0a 75.0 ± 11.5Untreated MPS IIIB 161.3 ± 35.6a 95.0 ± 22.3a

HSCT treated MPS IIIB 43.6 40.5

Number of each group is as follows, control (n = 59), untreated MPS IIIA (n = 6),untreated MPS IIIB (n = 11), and HSCT treated MPS IIIB (n = 1).

a Significantly different from the control at p b 0.05.

97T. Shimada et al. / Molecular Genetics and Metabolism 113 (2014) 92–99

Author's personal copy

identify biomarkers not only for MPS but also for these more commondisorders. This manuscript has focused on assay of HS, but CS, DS andKS assays could also be developed for HT-MS/MS to screen for otherforms of MPS and other diseases.

In conclusion, we have demonstrated a unique HT-MS/MS methodfor the measurement of HS that can distinguishMPS patients from con-trols with a similar reproducibility, sensitivity and specificity asestablished by LC–MS/MSmethods, and that themethod can be appliedto analysis of newborn DBS. The HT-MS/MS should provide both a rapidthroughput and economic advantages for monitoring, diagnosis, andNBS for MPS and has the potential of being applied to other disorders.

Conflict of interest

All the authors have contributed to this “Original Article” and haveno conflict of interest with any other party.

Tsutomu Shimada, Joan Kelly, William A LaMarr, Naomi van Vlies,Eriko Yasuda, Pravin Patel, Robert W. Mason, William Mackenzie,Francyne Kubaski, Roberto Giugliani, Yasutsugu Chinen, Seiji Yamaguchi,Yasuyuki Suzuki, Kenji E. Orii, Toshiyuki Fukao, Tadao Orii, and ShunjiTomatsu declare that they have no conflict of interests.

Informed consent

Informed consent was obtained from the patients and/or theirguardians through the physicians who recruited the patients withMPS at each local institute, and approved by the local InstitutionalReview Board (IRB).

Contributions to the project

Tsutomu Shimada contributed to the concept of the project, theplanning, execution of experiments (LC–MS/MS), data analysis, andreporting of the work described in the article.

Joan Kelly contributed to the concept of the project, the planning, ex-ecution of experiments (HT-MS/MS), data analysis, and reporting of thework described in the article.

WilliamA LaMarr contributed to the concept of the project, the plan-ning, execution of experiments (HT-MS/MS), data analysis, andreporting of the work described in the article.

Naomi vanVlies contributed to collecting samples, data analysis, andreporting of the work described in the article.

Eriko Yasuda contributed to collecting samples, data analysis, andreporting of the work described in the article.

Robert W. Mason contributed to the planning, execution of LC–MS/MS, data analysis, and reporting of thework described in the article.

William G. Mackenzie contributed to collecting samples, data analy-sis, and reporting of the work described in the article.

Francyne Kubaski contributed to collecting samples, data analysis,and reporting of the work described in the article.

Roberto Giugliani contributed to collecting samples and reporting ofthe work described in the article.

Yasutsugu Chinen contributed to collecting samples and reportingof the work described in the article.

Seiji Yamaguchi contributed to collecting samples and reporting ofthe work described in the article.

Yasuyuki Suzuki contributed to collecting samples and reportingof the work described in the article.

Tadao Orii contributed to collecting samples and reporting of thework described in the article.

Toshiyuki Fukao contributed to collecting samples and reporting ofthe work described in the article.

Kenji E. Orii contributed to collecting samples and reporting of thework described in the article.

Shunji Tomatsu is a Principal Investigator and is responsible for theentire project. He has contributed to the concept of the project, plan-ning, analysis of data, and reporting of the work described in the article.

Acknowledgment

This work was supported by grants from the Austrian MPS Society,International Morquio Organization (Carol Ann Foundation) andDelaware Bioscience Center for Advanced Technology. This work wasalso supported by the Japanese MPS Family Society. R.W.M. and S.T.were supported in part by an Institutional Development Award (IDeA)from the National Institute of General Medical Sciences of NIH undergrant number P20GM103464. The content of the article has not been in-fluenced by the sponsors. F.K. was supported by Conselho Nacional deDesenvolvimento Científico e Tecnológico from Brazil (CNPq). Editorialassistance to the manuscript was provided by Michelle Stofa atNemours/Alfred I. duPont Hospital for Children. We thank MichaelPrinsen and Mary Campbell (Center for World Health & Medicine,Saint Louis University) for helpful earlier discussions about theRapidFire technology.

References

[1] S. Tomatsu, A.M.Montaño, V.C. Dung, et al., Enhancement of drug delivery: enzyme-replacement therapy for murine Morquio A syndrome, Mol. Ther. 18 (2010)1094–1102.

[2] R.W. Farndale, D.J. Buttle, A.J. Barrett, Improved quantitation and discrimination ofsulphated glycosaminoglycans by use of dimethylmethylene blue, Biochim. Biophys.Acta 883 (1986) 173–177.

[3] J.G. de Jong, R.A. Wevers, R. Liebrand-van Sambeek, Measuring urinary glycosamino-glycans in the presence of protein: an improved screening procedure formucopolysaccharidoses based on dimethylmethylene blue, Clin. Chem. 38 (1992)803–807.

[4] C.B.Whitley, R.C. Spielmann, G. Herro, S.S. Teragawa, Urinary glycosaminoglycan ex-cretion quantified by an automated semimicro method in specimens convenientlytransported from around the globe, Mol. Genet. Metab. 75 (2002) 56–64.

[5] P. Whiteman, The quantitative determination of glycosaminoglycans in urine withAlcian Blue 8GX, Biochem. J. 131 (1973) 351–357.

[6] R.W. Burlingame, G.H. Thomas, R.L. Stevens, K. Schmid, H.W.Moser, Direct quantitationof glycosaminoglycans in 2 mL of urine from patients with mucopolysaccharidoses,Clin. Chem. 27 (1981) 124–128.

[7] T. Oguma, S. Tomatsu, A.M.Montaño, O. Okazaki, Analytical method for the determi-nation of disaccharides derived from keratan, heparan, and dermatan sulfates inhuman serum and plasma by high-performance liquid chromatography/turboionspray ionization tandem mass spectrometry, Anal. Biochem. 368 (2007) 79–86.

[8] R. Cappelletti, M. Del Rosso, V.P. Chiarugi, A new electrophoretic method for thecomplete separation of all known animal glycosaminoglycans in amonodimensionalrun, Anal. Biochem. 99 (1979) 311–315.

[9] J.J. Hopwood, J.R. Harrison, High-resolution electrophoresis of urinary glycosamino-glycans: an improved screening test for the mucopolysaccharidoses, Anal. Biochem.119 (1982) 120–127.

[10] S. Tomatsu, K. Okamura, H. Maeda, et al., Keratan sulphate levels inmucopolysaccharidoses and mucolipidoses, J. Inherit. Metab. Dis. 28 (2005) 187–202.

Table 3Levels of ΔDiHS-0S and ΔDiHS-NS in patients with MPS IV and control subjects over 3 -

years of age.

ΔDiHS-0S (ng/ml) ΔDiHS-NS (ng/ml)

Control 27.5 ± 8.7 22.7 ± 4.8Untreated MPS IVA severe 20.9 ± 10.3 20.6 ± 8.3ERT treated MPS IVA severe 15.2 ± 4.0 18.2 ± 2.1Untreated MPS IVB attenuated 13.1 ± 3.0 19.2 ± 5.5

Number of each group is as follows, control (n = 59), untreated MPS IVA (n = 11), ERT-treated MPS IVA (n = 6), and untreated MPS IVB (n = 5).

Table 4Levels of ΔDiHS-0S and ΔDiHS-NS in newborn DBS from control and patients with MPS.

(ng/spot) ΔDiHS-0S ΔDiHS-NS

Control 0.10 ± 0.02 0.11 ± 0.05MPS I 0.49 ± 0.13⁎ 0.29 ± 0.11⁎

MPS II 0.33 0.14MPS III 0.50 ± 0.10⁎ 0.30 ± 0.13⁎

Number of each group is as follows, control (n = 22), MPS I (n = 6), MPS II (n = 1), andMPS III (n = 5).⁎ Significantly different from the control at p b 0.05.

98 T. Shimada et al. / Molecular Genetics and Metabolism 113 (2014) 92–99

Author's personal copy

[11] S. Tomatsu, M.A. Gutierrez, T. Ishimaru, et al., Heparan sulfate levels inmucopolysaccharidoses and mucolipidoses, J. Inherit. Metab. Dis. 28 (2005) 743–757.

[12] A. Kinoshita, K. Sugahara, Microanalysis of glycosaminoglycan-derived oligosaccha-rides labeled with a fluorophore 2-aminobenzamide by high-performanceliquid chromatography: application to disaccharide composition analysis andexosequencing of oligosaccharides, Anal. Biochem. 269 (1999) 367–378.

[13] H. Yamada, S. Miyauchi, M. Morita, et al., Content and sulfation pattern of keratansulfate in hip osteoarthritis using high performance liquid chromatography, J.Rheumatol. 27 (2000) 1721–1724.

[14] T. Oguma, S. Tomatsu, O. Okazaki, Analytical method for determination of disaccha-rides derived from keratan sulfates in human serum and plasma by high-performance liquid chromatography/turbo-ionspray ionization tandem massspectrometry, Biomed. Chromatogr. 21 (2007) 356–362.

[15] S. Tomatsu, A.M. Montaño, T. Oguma, et al., Validation of disaccharide compositionsderived from dermatan sulfate and heparan sulfate in mucopolysaccharidoses andmucolipidoses II and III by tandem mass spectrometry, Mol. Genet. Metab. 99(2010) 124–131.

[16] S. Tomatsu, A.M. Montaño, T. Oguma, et al., Dermatan sulfate and heparan sulfate asa biomarker for mucopolysaccharidosis I, J. Inherit. Metab. Dis. 33 (2010) 141–150.

[17] S. Tomatsu, A.M. Montaño, T. Oguma, et al., Validation of keratan sulfate level inmucopolysaccharidosis type IVA by liquid chromatography–tandem mass spec-trometry, J. Inherit. Metab. Dis. 33 (Suppl. 3) (2010) S35–S42 (Epub ahead of print).

[18] S. Tomatsu, T. Fujii, M. Fukushi, et al., Newborn screening and diagnosis ofmucopolysaccharidoses, Mol. Genet. Metab. 110 (2013) 42–53.

[19] S. Tomatsu, T. Shimada, R.W. Mason, et al., Assay for glycosaminoglycans by tandemmass spectrometry and its applications, J. Anal. Bioanal. Tech. S2 (2014) 006, http://dx.doi.org/10.4172/2155-9872.S2-006.

[20] J. de Ruijter, M.H. de Ru, T. Wagemans, et al., Heparan sulfate and dermatan sulfatederived disaccharides are sensitive markers for newborn screening formucopolysaccharidoses types I, II and III, Mol. Genet. Metab. 107 (2012) 705–710.

[21] J. de Ruijter, L. Ijlst, W. Kulik, et al., Heparan sulfate derived disaccharides in plasmaand total urinary excretion of glycosaminoglycans correlate with disease severity inSanfilippo disease, J. Inherit. Metab. Dis. 36 (2013) 271–279 (Epub ahead of print).

[22] M.H. de Ru, L. van der Tol, N. van Vlies, et al., Plasma and urinary levels of dermatansulfate and heparan sulfate derived disaccharides after long-term enzyme replace-ment therapy (ERT) in MPS I: correlation with the timing of ERT and with total uri-nary excretion of glycosaminoglycans, J. Inherit. Metab. Dis. 36 (2013) 247–255(Epub ahead of print).

[23] T. Shimada, S. Tomatsu, E. Yasuda, et al., Chondroitin 6-sulfate as a novel biomarkerfor mucopolysaccharidosis IVA and VII, J. Inherit. Metab. Dis. (2014) (in press).

[24] A.M. Montaño, H. Oikawa, S. Tomatsu, et al., Acidic amino acid tag enhances re-sponse to enzyme replacement in mucopolysaccharidosis type VII mice, Mol.Genet. Metab. 94 (2008) 178–189.

[25] S. Tomatsu, A.M. Montaño, A. Ohashi, et al., Enzyme replacement therapy in a mu-rine model of Morquio A syndrome, Hum. Mol. Genet. 17 (2008) 815–824.

[26] A.K. Quercia, W.A. LaMarr, J. Myung, C.C. Ozbal, J.A. Landro, K.J. Lumb, High-throughput screening by mass spectrometry: comparison with the scintillationproximity assay with a focused-file screen of AKT1/PKB alpha, J. Biomol. Screen.12 (2007) 473–480.

[27] X. Wu, J. Wang, L. Tan, et al., In vitro ADME profiling using high-throughputRapidFire mass spectrometry: cytochrome p450 inhibition and metabolic stabilityassays, J. Biomol. Screen. 17 (2012) 761–772.

[28] M.V. Leveridge, A.I. Bardera, W. LaMarr, et al., Lead discovery for microsomal prosta-glandin E synthase using a combination of high-throughput fluorescent-basedassays and RapidFire mass spectrometry, J. Biomol. Screen. 17 (2012) 641–650.

[29] S.E. Hutchinson, M.V. Leveridge, M.L. Heathcote, et al., Enabling lead discovery forhistone lysine demethylases by high-throughput RapidFire mass spectrometry, J.Biomol. Screen. 17 (2012) 39–48.

[30] M. Plant, T. Dineen, A. Cheng, A.M. Long, H. Chen, K.A. Morgenstern, Screening forlysine-specific demethylase-1 inhibitors using a label-free high-throughput massspectrometry assay, Anal. Biochem. 419 (2011) 217–227.

[31] K.A. Atkinson, E.E. Beretta, J.A. Brown, et al., N-benzylimidazole carboxamides as po-tent, orally active stearoyl CoA desaturase-1 inhibitors, Bioorg. Med. Chem. Lett. 21(2011) 1621–1625.

[32] M.K. Highkin, M.P. Yates, O.V. Nemirovskiy, et al., High-throughput screening assayfor sphingosine kinase inhibitors in whole blood using RapidFire® mass spectrom-etry, J. Biomol. Screen. 16 (2011) 272–277.

[33] E.F. Langsdorf, A.Malikzay,W.A. Lamarr, et al., Screening for antibacterial inhibitors oftheUDP-3-O-(R-3-hydroxymyristoyl)-N-acetylglucosamine deacetylase (LpxC) usinga high-throughput mass spectrometry assay, J. Biomol. Screen. 15 (2010) 52–61.

[34] T.G. Holt, B.K. Choi, N.S. Geoghagen, et al., Label-free high-throughput screening viamass spectrometry: a single cystathionine quantitativemethod for multiple applica-tions, Assay Drug Dev. Technol. 7 (2009) 495–506.

[35] M. Jonas, W.A. LaMarr, C. Ozbal, Mass spectrometry in high-throughput screening: acase study on acetyl-coenzyme a carboxylase using RapidFire–mass spectrometry(RF–MS), Comb. Chem. High Throughput Screen. 12 (2009) 752–759.

[36] A. Tanaka, T. Okuyama, Y. Suzuki, et al., Long-term efficacy of hematopoietic stemcell transplantation on brain involvement in patients with mucopolysaccharidosistype II: a nationwide survey in Japan, Mol. Genet. Metab. 107 (2012) 513–520.

[37] Y. Chinen, T. Tohma, Y. Izumikawa, H. Uehara, T. Ohta, Sanfilippo type B syndrome: fivepatients with an R565P homozygous mutation in the alpha-N-acetylglucosaminidasegene from the Okinawa islands in Japan, J. Hum. Genet. 50 (2005) 357–359.

[38] L.L. McCabe, E.R. McCabe, Expanded newborn screening: implications for genomicmedicine, Annu. Rev. Med. 59 (2008) 163–175.

[39] D.A. Applegarth, J.R. Toone, R.B. Lowry, Incidence of inborn errors of metabolism inBritish Columbia, 1969–1996, Pediatrics 105 (2000) e10.

[40] F. Baehner, C. Schmiedeskamp, F. Krummenauer, E. Miebach, M. Bajbouj, C. Whybra,A. Kohlschütter, C. Kampmann, M. Beck, Cumulative incidence rates of themucopolysaccharidoses in Germany, J. Inherit. Metab. Dis. 28 (2005) 1011–1017.

[41] P.J. Meikle, J.J. Hopwood, A.E. Clague, W.F. Carey, Prevalence of lysosomal storagedisorders, JAMA 281 (1999) 249–254.

[42] J. Nelson, Incidence of the mucopolysaccharidoses in Northern Ireland, Hum. Genet.101 (1997) 355–358.

[43] J. Nelson, J. Crowhurst, B. Carey, L. Greed, Incidence of themucopolysaccharidoses inWestern Australia, Am. J. Med. Genet. A 123A (2003) 310–313.

[44] B.J. Poorthuis, R.A. Wevers, W.J. Kleijer, J.E. Groener, J.G. de Jong, S. van Weely, K.E.Niezen-Koning, O.P. van Diggelen, The frequency of lysosomal storage diseases inThe Netherlands, Hum. Genet. 105 (1999) 151–156.

[45] H.Y. Lin, S.P. Lin, C.K. Chuang, D.M. Niu, M.R. Chen, F.J. Tsai, M.C. Chao, P.C. Chiu, S.J.Lin, L.P. Tsai, W.L. Hwu, J.L. Lin, Incidence of the mucopolysaccharidoses in Taiwan,1984–2004, Am. J. Med. Genet. A 149A (2009) 960–964.

[46] A.W. Mulivor, H.H. Lipowsky, Inflammation- and ischemia-induced shedding ofvenular glycocalyx, Am. J. Physiol. Heart Circ. Physiol. 286 (2004) H1672–H1680.

[47] M. Rehm, D. Bruegger, F. Christ, et al., Shedding of the endothelial glycocalyx inpatients undergoing major vascular surgery with global and regional ischemia,Circulation 116 (2007) 1896–1906.

[48] A. Nelson, I. Berkestedt, M. Bodelsson, Circulating glycosaminoglycan species in sep-tic shock, Acta Anaesthesiol. Scand. 58 (2014) 36–43.

[49] F.M. Guven, H. Aydin, A. Kaya, et al., The role of heparan sulphate in pathogenesisof Crimean–Congo hemorrhagic fever disease, J. Vector BorneDis. 50 (2013) 133–136.

99T. Shimada et al. / Molecular Genetics and Metabolism 113 (2014) 92–99