Analytica Chimica Acta 936 (2016) 139e148

Contents lists available at ScienceDirect

Analytica Chimica Acta

journal homepage: www.elsevier.com/locate/aca

UPLC-MS/MS detection of disaccharides derived from

glycosaminoglycans as biomarkers of mucopolysaccharidoses
Christiane Auray-Blais a, *, Pamela Lavoie a, Shunji Tomatsu b, Vassili Valayannopoulos c,
John J. Mitchell d, Julian Raiman e, Maxime Beaudoin a, Bruno Maranda a, Joe T.R. Clarke e
a
Division of Medical Genetics, Department of Pediatrics, Faculty of Medicine and Health Sciences, Universit

e de Sherbrooke, CR-CHUS, Hospital Fleurimont,
3 001, 12th Avenue North, Sherbrooke, QC, J1H 5N4, Canada
b
Nemours/Alfred I DuPont Hospital for Children, 1600 Rockland Road, Wilmington, DE, 19803, United States
c
Centre de R
ef

erence Maladies M
etaboliques, Ho^pital Universitaire Necker-Enfants Malades et Institut IMAGINE, 149 Rue de S evres, 75015, Paris, France
d
McGill University Health Center, 1001 Boulevard D
ecarie, Montreal, QC H4A 3J1, Canada
e
Division of Clinical and Metabolic Genetics, Department of Pediatrics, Hospital for Sick Children, 555 University Avenue, Toronto, ON, M5G 1X8, Canada

h i g h l i g h t g r a p h i c a l a b s t r a c t

A mass spectrometry method useful

and reliable for monitoring treated
MPS patients.

Analysis of dermatan sulfate, heparan
sulfate, keratan sulfate, chondroitin
sulfate.

Urine analysis approach of patients
affected with MPS I, II, III, IVA, IVB
and VI.

GAG MS/MS analysis for high-risk
screening, diagnosis, and monitoring
MPS patients.

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

Article history: Mucopolysaccharidoses (MPSs) are a group of disorders resulting from primary defects in lysosomal
Received 4 May 2016 enzymes involved in the degradation of glycosaminoglycans (GAGs). Depending on the specific enzyme
Received in revised form defect, the catabolism of one or more GAGs is blocked leading to accumulation in tissues and biological
21 June 2016
fluids. GAG measurements are important for high-risk screening, diagnosis, monitoring treatment effi-
Accepted 29 June 2016
Available online 2 July 2016
cacy, and patient follow up. The dimethylmethylene blue (DMB) spectrophotometric method commonly
used in most biochemical genetics laboratories relies on a non-specific total GAG analysis which has led
to false positive results, and even false negative results (mainly for MPS III and IV patients). The main
Keywords:
Mucopolysaccharidoses
objective of our project was to devise and validate a reliable tandem mass spectrometry multiplex
Tandem mass spectrometry analysis for the urine quantitation of four GAGs (dermatan sulfate (DS), heparan sulfate (HS), keratan
Glycosaminoglycans sulfate (KS), and chondroitin sulfate (CS)) for an eventual technological transfer to the clinic. The
Dermatan sulfate developed methodology is rapid (7 min) and our results showed good intraday and interday precision
Heparan sulfate (RSDs  8.7%) and accuracy (Biases range: 12.0%e18.4%). Linearity was good (r2 > 0.995) for DS, HS, CS,
Keratan sulfate and KS calibration curves. In comparison with the DMB spectrophotometric method, this multiplex
Chondroitin sulfate tandem mass spectrometry method allows GAG fractionation, thus a differentiation of MPS types, except
for MPS I and II which are characterized by the same GAG profile. The devised method is a useful and

* Corresponding author.
E-mail address: [email protected] (C. Auray-Blais).

http://dx.doi.org/10.1016/j.aca.2016.06.054
0003-2670/© 2016 Elsevier B.V. All rights reserved.
140 C. Auray-Blais et al. / Analytica Chimica Acta 936 (2016) 139e148

reliable tool for diagnosis of MPS patients, as well as their monitoring and follow up, as shown by
longitudinal studies.
© 2016 Elsevier B.V. All rights reserved.

1. Introduction years, the DMB test was shown to be unreliable for MPS detection
because of interfering substances, such as hemoglobin, which may
Mucopolysaccharidoses (MPSs) are a group of disorders result- lead to false-positive results. The possibility of false-negative re-
ing from primary defects in lysosomal enzymes involved in the sults was also reported owing to the formation of aggregates by
degradation of glycosaminoglycans (GAGs). Depending on the electrostatic interactions with collagen, glycoproteins, albumin or
specific enzyme defect, the catabolism of one or more GAGs is other serum proteins causing modifications in the physicochemical
blocked leading to the accumulation of corresponding substrate(s), properties of GAGs [13]. False negative results can also be related to
such as dermatan sulfate (DS), heparan sulfate (HS), keratan sulfate the natural history of the disease with GAG levels in some MPSs
(KS), chondroitin sulfate (CS) and hyaluronan [1e3]. Table 1 shows decreasing as growth terminates. With the advancement of tech-
enzyme deficiencies associated to the accumulation of various nologies, tandem mass spectrometry (MS/MS) methodologies have
GAGs in different MPSs. Marked phenotypic and genotypic het- been shown to be more reliable for the analysis of different GAGs in
erogeneity leads to a wide-spectrum of clinical manifestations, age a single multiplex analysis [14e23]. Our group, as well as others,
onset of symptoms, as well as severity and progression of disease. recently developed and validated MS/MS methods based on a
This contributes to the difficulty in detecting patients early, and methanolysis reaction aiming depolymerisation of GAGs before
possible false negative results. Common progressive multi-systemic analysis of DS, HS, and CS-related disaccharides in various biolog-
symptoms include eye involvement, cardiac, respiratory, and ical fluids, such as urine [14e18] and cerebrospinal fluid [19,23].
cognitive disorders, hepatosplenomegaly, umbilical and inguinal These methods are useful for detection, diagnosis and monitoring
hernias, dysostis multiplex, hearing loss, macroglossia and a char- of patients affected with MPS I, II, III (Sanfilippo syndrome subtypes
acteristic facies [1e3]. Currently, treatment is available for some IIIA, OMIM 252900, IIIB, OMIM 252920, IIIC, OMIM 252930, and
MPSs and varies from hematopoietic stem cell transplantation to IIID, OMIM 252940), IVA, VI, and VII (Sly syndrome, OMIM 253220).
intravenous enzyme replacement therapy (ERT) for MPS I (mainly However, to our knowledge, the feasibility to analyze KS-related
Hurler syndrome, OMIM 607014, and Hurler/Scheie syndrome, disaccharides as part of this multiplex method was never investi-
OMIM 607015, while Scheie syndrome, OMIM 607016 is less se- gated. As a consequence, the detection of patients with Morquio B
vere), MPS II (Hunter syndrome, OMIM 309900), MPS IVA (Morquio syndrome (MPS IVB, OMIM 253010) was not possible. Moreover, KS
A syndrome, OMIM 253000), and MPS VI (Maroteaux-Lamy syn- is the main GAG elevated in MPS IVA patients, and it has been
drome, OMIM 253200). Although these therapies do not result in shown that patients with other MPS types, especially MPS II pa-
cure [4], they often alter the progression of the disease. Other tients, may also have abnormal KS levels [22]. Therefore, the
treatment strategies are being developed, including, among others, analysis of KS is important to improve the detection, diagnosis and
substrate reduction therapy and gene therapy [5]. monitoring of these patients.
Urinary GAG analysis has been employed for decades to detect The main objective of the current research study was thus to
MPS patients early (high-risk screening), and to monitor the effi- devise and validate a quantitative methodology using a MS/MS
cacy of ERT [6]. Some methodologies were qualitative using a urine multiplex approach focusing on a wider spectrum of GAG profiles,
spot test [7,8], others were quantitative based on a colorimetric test specifically targeting DS, HS, CS, and KS disaccharides. A specific
using the dimethylmethylene blue (DMB) reaction [9e11], or a KS-related disaccharide was chosen, parameters of the method
nuclear magnetic resonance spectroscopy analysis [12]. Over the optimized, and analysis performed as part of a multiplex

Table 1
Enzyme deficiencies leading to GAG elevations in mucopolysaccharidoses.
 C. Auray-Blais et al. / Analytica Chimica Acta 936 (2016) 139e148 141

methodology. Moreover, a KS internal standard was synthesized in- at 1 mg/mL. Stock and intermediate solutions were stored at 20  C
house. This improved method was validated and compared with and were stable for at least 9 months.
the DMB spectrophotometric method. The urine analysis of pa-
tients with various MPSs and controls was performed to establish
2.4.2. Standard curve and quality control working solutions
reference values according to age. In addition, the analysis of
Working solutions were a mixture of DS, HS, and KS prepared by
pathological control samples from others lysosomal storage disor-
diluting stock and intermediate solutions with water according to
ders (LSDs) was included.
Supplementary Table 1. A total of seven standard curve solutions (0,
8, 20, 40, 80, 200, and 400 mg/mL) and three quality control working
2. Materials and methods
solutions (low: 30, medium: 140, and high: 300 mg/mL) were pre-
pared. Working solutions were stable at e 20  C for at least 9
2.1. Ethics approval
months.
Since the disaccharide derived from CS (GlcA-GalNAc) is also
This project was approved by the Research Ethics Board (REB) at
found in DS, a calibration curve of CS was prepared separately
the Centre hospitalier universitaire de Sherbrooke (CHUS), as well
(0e400 mg/mL) and used in order to determine the concentration of
as by the institutional REBs of all the collaborators.
the GlcA-GalNAc disaccharide derived from DS in the calibration
curve and QC working solutions. These measured concentrations
2.2. Urine specimen collection
(0, 0.9, 2.4, 5.1, 10.4, 26.7, 54.2 mg/mL for the calibration curve and
3.8, 18.7, and 40.5 mg/mL for the low, medium and high QCs) were
Urine specimens were collected at random from patients having
then used to quantitate the GlcA-GalNAc disaccharide attributed to
different MPSs and in whom the diagnosis was confirmed by
CS, using the DS standard and thus avoiding interferences.
enzymology and/or identification of a pathogenic mutation. Urine
specimens from 58 MPS patients were analyzed: 12 patients had
Hurler, Hurler/Scheie, or Scheie syndromes (age range: 9 months to 2.4.3. Internal standard (IS) stock solutions
20 years, median age: 9.5 years); 10 patients had Hunter syndrome Deuterated internal standards were prepared in-house by deu-
(age range: 5e18 years, median age: 10.0 years); 4 patients had teriomethanolysis of DS, HS, and KS standards as previously
Sanfilippo syndrome (age range: 2e14 years, median age: 2.9 described [16]. Stock solutions were prepared in water at 600 mg/
years); 25 patients had Morquio A Syndrome (age range: 2e36 mL and stored at e 20  C. No degradation was observed after 9
years, median age: 12.0 years); and 7 patients had Maroteaux-Lamy months in these storage conditions.
syndrome (age range: 4e42 years, median age: 15 years). Nine
patients with MPS I were treated, either by bone marrow transplant
(BMT) or ERT, while 6 patients with MPS II, 16 patients with MPS 2.4.4. Resuspension solution
IVA, and five patients with MPS VI were treated by ERT treatment. A resuspension solution containing a mixture of DS, HS, and KS
In addition, three MPS patients (MPS I, MPS II, and MPS IVA) were internal standards at 9, 2, and 12 mg/mL, respectively, was prepared
part of a longitudinal study and had urine samples collected over when needed by diluting stock solutions in a 90:10 acetonitrile:-
time. Liquid urine specimens were also obtained from 79 healthy water solution.
reference controls (age range: 1 month to 58 years, median age: 10
years) in order to establish normal reference values for di-
2.5. Sample preparation
saccharides resulting from the methanolysis of DS, HS, CS, and KS.
Finally, urine samples from patients having different LSDs other
2.5.1. Processing urine samples from patients
than MPSs (Gaucher disease, Fabry disease, alpha-mannosidosis,
Urine samples were processed as described previously [16], with
mucolipidosis II, Pompe disease, GM1 gangliosidosis, aspartylglu-
slight modifications. Twenty-five microliters of homogenized urine
cosaminuria, and Schindler disease) were collected in order to
sample were deposited in a disposable borosilicate glass culture
verify the specificity of the devised methodology. All samples were
tube fitted with a screw-cap (16  100 mm). After evaporation of
stored at 20  C until analysis.
the urine sample under a stream of nitrogen, 500 mL of a com-
mercial methanolic HCl 3 N solution was added. Tubes were cap-
2.3. Reagents
ped, vortexed, and incubated at 65  C for 1 h. After incubation,
samples were immediately evaporated under a stream of nitrogen,
Optima® liquid chromatography-mass spectrometry (LC-MS)
then suspended in 200 mL of the resuspension solution. Samples
grade water (H2O) was from Fisher Scientific (Fair Lawn, NJ). High-
were transferred to ultra-performance liquid chromatography
performance liquid chromatography (HPLC) grade methanol
(UPLC) vials, and centrifuged prior to the injection of 2 mL in the
(MeOH) and LC-MS grade acetonitrile (ACN) were from EMD
UPLC-MS/MS system.
Chemicals Inc. (Darmstadt, Germany). Dermatan sulfate, chon-
droitin sulfate A, and heparan sulfate calibration standards were
from Sigma Aldrich (Saint-Louis, MO). Keratan sulfate calibration 2.5.2. Processing calibrators and quality controls
standard came from Amsbio (Abingdon, UK). Methanolic hydro- Twenty-five microliters of each of the seven standard curve and
chloric acid (HCl) 3 N, methanol-d4 99.8% purity, and ammonium three quality control working solutions were deposited in a
acetate 98% purity were also purchased from Sigma Aldrich. Acetyl disposable borosilicate glass culture tube fitted with a screw-cap
chloride was purchased from Fluka (Milwaukee, WI). Synthetic (16  100 mm), along with 25 mL of synthetic human urine
human urine was purchased from Bioreclamation (Hicksville, NY). (which is an artificially synthesized substance that simulates the
composition and chemical properties of urine). After evaporation
2.4. Preparation of standard solutions under a stream of nitrogen, processing was performed in a manner
similar to that employed for patients. It is noteworthy to mention
2.4.1. Stock and intermediate solutions that standards must be processed simultaneously with patient
Stock solutions of DS, HS, CS, and KS were prepared individually samples, and that the methanolysis time must be monitored
in water at 10 mg/mL, while intermediate solutions were prepared carefully.
142 C. Auray-Blais et al. / Analytica Chimica Acta 936 (2016) 139e148

2.6. Instrumentation and parameters Table 3

UPLC Acquity I-Class parameters.

The analysis of DS, HS, KS, and CS disaccharides and respective Parameters Description
internal standards was performed simultaneously on a Xevo TQ-S Column BEH amide
(Waters Corp., Milford, MA) MS/MS combined to an Acquity I- ID  Length 2.1  50 mm
Class (Waters) UPLC system. Electrospray (ESI) was in positive ion Particle size 1.7 mm
mode, and signals were acquired during a multiple reaction Temperature 30  C
Weak wash 90:10 ACN:H2O
monitoring (MRM) experiment. Mass spectrometry acquisition
Strong wash 50:50 MeOH:H2O
parameters are presented in Table 2. Injection volume 2 mL
A BEH Amide UPLC column (2.1  50 mm, 1.7 mm particle size) Injection mode Partial Loop with Needle Overfill (PLNO)
with an on-line pre-filter (0.2 mm) was used for chromatographic Autosampler 10  C
Mobile phase (MP) A 90:10 ACN:H2O þ 10 mM CH3COONH4
separation. The method run-time and the total analysis time be-
Mobile phase (MP) B 90:10 H2O:ACN þ 10 mM CH3COONH4
tween injections were 7 and 8 min, respectively. In order to Flow rate 0.3 mL/min
maximize the column durability, the flow state was set to waste for Gradient 0.00e1.50 min 0% MP B
the first 90 s of the analysis, then redirected into the UPLC system. 1.50e2.25 min 0e10% MP B (linear)
Liquid chromatography parameters are presented in Table 3. 2.25e4.50 min 10% MP B
4.50e5.50 min 40% MP B
5.50e7.00 min 0% MP B
2.7. Quantification of disaccharides resulting from the methanolysis
of DS, HS, CS, and KS
Quantification of selected disaccharides was achieved using a
CS is a linear polysaccharide comprised of repeating units of N- seven-point calibration curve. The best results were obtained using
acetylgalactosamine (GalNAc) linked by a glycosidic bond to glu- a quadratic calibration curve for DS, KS, and CS, and a linear cali-
curonic acid (GlcA). Epimerization of any of the GlcA moieties to bration curve for HS. Weighting 1/x was applied and calibration
iduronic acid (IdoA) constitutes the formation of dermatan sulfate origin was excluded. Data were acquired over the entire analysis
[24]. Different sulfation patterns are possible either on the uronic and quantification was performed using the response factor. Data
acid (O-2 and/or O-3 positions free or sulfated) or hexoamine res- processing was achieved using TargetLynx® Application Manager,
idues (O-6 and/or O-4 positions free or sulfated), but these are lost an option with MassLynx™ (version 4.1 SCN810) Software (Waters).
during methanolysis. Consequently, structural information is lost in Variations due to urine concentrations were minimized by
the process; however, a lower limit of quantification is achieved normalizing biomarkers as ratios to creatinine. The analysis of
owing to the fact that all these forms are analyzed together as a urinary creatinine was performed according to a previously pub-
single peak. This observation is valid for all GAGs under study. The lished methodology [27].
specific IdoA-GalNAc disaccharide was used to quantitate DS, while
GlcA-GalNAc was used to quantitate CS. This latter disaccharide is
also produced as a result of the methanolysis of DS, but to a lesser 2.8. Method validation
extent.
Heparan sulfate is a polymer containing either GlcA or IdoA Intraday (4 replicates in a day) and interday (5 different days)
alternating with glucosamine (GlcN). Again, different levels of accuracy (% Bias) and precision (% RSD) were assessed using quality
sulfation are possible on the GlcN moiety (O-6 and O-3 positions controls at three levels of concentration (low, medium, and high).
free or sulfated, and sulfation or acylation on the amino group is Intraday and interday precisions were also assessed using a MPS
possible) and on the uronic acid (O-2 position can be sulfated or patient urine specimen. The linearity of the seven-point calibration
not) [25]. The IdoA-GlcN disaccharide resulting from the meth- curve was evaluated for DS, HS, CS, and KS disaccharides. Limits of
anolysis is used for quantification. detection (LODs) and limits of quantification (LOQs) were defined
Keratan sulfate is a polymer of sulfated N-acetyl lactosamine as three and ten times the standard deviation of the analyte
[26]. Three major sulfation patterns are found; 6-O sulfation on response at low concentration (n ¼ 8), respectively, divided by the
either galactopyranose (Gal), N-acetylglucosamine (GlcNAc), or analyte calibration curve slope. Considering the endogenous pres-
both moieties. The Gal-GlcNAc disaccharide is used for ence of the analytes in reference control urine specimens, matrix
quantification. effects were evaluated as follows. Twenty-five microliters from six
Fig. 1 shows the methanolysis and deuteriomethanolysis (syn- healthy reference urine specimens (creatinine concentrations
thesis of internal standards) of four glycosaminoglycans under ranging from 2.2 to 12.2 mmol/L) were deposited in borosilicate
study, leading to targeted disaccharides of interest. It is noteworthy tubes in duplicates for a total of 12 samples. One urine sample from
to mention that specific disaccharides were selected and optimized each pair was spiked with 25 mL of the medium-concentration
for the UPLC-MS/MS analysis. Oligosaccharides of different lengths quality control working solution at 140 mg/mL. When dried, sam-
might also have been produced during the methanolysis process, ples were processed as usual. After analysis, concentrations
but were not taken into account. measured in the unspiked urine specimens were subtracted from

Table 2
ESI-MS/MS acquisition parameters for the quantitative analysis of disaccharides obtained after the methanolysis of DS, HS, CS, and KS. IS: internal standard.

Compound Transition (m/z) Dwell time (s) Cone voltage (V) Collision energy (V)

DS/CS 426 > 236 0.1 36 12

DS IS/CS IS 432 > 239 0.1 36 12
HS 406 > 245 0.1 22 24
HS IS 412 > 251 0.1 22 24
KS 420 > 258 0.1 30 22
KS IS 423 > 261 0.1 30 22
 C. Auray-Blais et al. / Analytica Chimica Acta 936 (2016) 139e148 143

Fig. 1. Methanolysis (conditions A) and deuteriomethanolysis (conditions B) processes for four glycosaminoglycans under study. Conditions A are HCl 3 N in CH3OH (65  C, 60 min)
and conditions B are DCl in CD3OD (65  C, 60 min), where D ¼ 2H. X ¼ H (conditions A), or D (conditions B, synthesis of internal standards). * GlcA might also be present in the
polymer, but to a lesser extent.

concentrations measured on the corresponding spiked urine sam- Fig. 2.

ples. Influence of matrix effects on precision (% RSD) and accuracy The precision and accuracy results for intraday (n ¼ 4) and
(% BIAS) was calculated for the resulting values. Stability over time interday (n ¼ 5) assays using low, medium and high quality con-
of DS, HS, CS, and KS in a MPS patient urine sample kept at room trols, and using a urine sample of a patient affected with MPS II are
temperature (22  C), stored in a refrigerator (4  C), in a freezer (e shown in Supplementary Table 2.
20  C) and after three freeze/thaw cycles was evaluated. Intraday precision assays were good with RSDs 7.6% and the
methodology showed good accuracy with biases ranging
2.9. Other GAG analyses performed for comparison purposes from 12.0% to 18.4%. Regarding interday precision and accuracy
assays, RSDs were 8.7% and biases ranged from 12.8% to 14.2%.
Total glycosaminoglycan measurements were performed by the The linearity was good for DS, HS, CS, and KS disaccharides with a
DMB-based spectrophotometry technique according to a previ- coefficient of determination (r2) > 0.995. LODs for DS, HS, KS, and
ously published methodology [11]. The analysis of KS by enzymatic CS disaccharides were 0.8, 0.2, 0.8, and 0.1 mg/mL, respectively,
digestion with keratanase II was also performed according to a whereas LOQs were 2.5, 0.7, 2.7, and 0.5 mg/mL. The stability study
recently published methodology by our group [17]. Results are revealed that the compounds were stable in urine for 20 h at room
given as the sum of two major disaccharides obtained after enzy- temperature (22  C), 20 h in a refrigerator (4  C), at least 9 months
matic digestion of KS, Galb1-4GlcNAc(6S), also called LacNAc (S1) in a freezer (e 20  C), and after three freeze/thaw cycles. Matrix
and Gal(6S)b1-4GlcNAc(6S), or LacNAc (S2), normalized to effects were evaluated and precision for DS, HS, KS, and CS di-
creatinine. saccharides (n ¼ 6) was acceptable with RSDs of 2.6%, 16.3%, 10.4%,
and 12.5%, respectively. Accuracy ranges (n ¼ 6) were the following:
DS: 5.7e14.2%, HS: 28.4e9.3%, KS: 6.2e28.3%, CS: 33.0e3.0%.
3. Results

3.1. Method validation 3.2. Normal reference values

A representative UPLC chromatogram of selected disaccharides Following the analysis of healthy control urine samples, normal
from DS, HS, KS, CS, and respective internal standards is shown in reference values were established. Five age ranges were
144 C. Auray-Blais et al. / Analytica Chimica Acta 936 (2016) 139e148

Fig. 2. Chromatographic separation of a standard at 400 mg/mL over a 7-min run. DS: dermatan sulfate; CS: chondroitin sulfate; HS: heparan sulfate; KS: keratan sulfate; IS: internal
standard; cps: counts per second.

determined: < 12 months, 1e3 years, 4e9 years, 10e17 years and (Table 5, Assay 2). The analysis of two KS disaccharides resulting
>18 years of age. Outliers were defined as values > 3 interquartile from the degradation of KS by the enzyme keratanase II was also
ranges from the 75th percentile of the distribution and were performed by UPLC-MS/MS (Table 5, Assay 3). All results are shown
excluded from normal reference calculations. Two standard de- in Table 5.
viations were added to the mean to determine normal reference The present UPLC-MS/MS multiplex methodology based on GAG
values. Reference values are shown in Table 4. methanolysis (Table 5, Assay 1) has the advantage that each GAG is
analyzed separately. This GAG profile is especially useful for diag-
3.3. GAG analyses in untreated MPS patients nostic purposes, as well as for the follow up of patients and the
evaluation of treatment efficacy. According to the results obtained
Urine samples from untreated MPS patients were analyzed us- in the current study, the analysis of KS by this multiplex assay
ing the devised multiplex methodology based on GAG meth- seems to be less sensitive in older MPS IVA patients (see Table 5,
anolysis. Results are shown in Fig. 3, and Table 5 (Assay 1). patients with IDs 18, 19, 20).
As expected, the analysis of urine samples from MPS I and II The DMB spectrophotometry methodology (Table 5, Assay 2)
patients reveals an elevation of DS and HS. The measured concen- does not need sophisticated instrumentation, such as mass spec-
tration of CS is also elevated, since a part of the measured GlcA- trometers, and is therefore more accessible to clinical laboratories
GalNAc CS disaccharide results from the methanolysis of DS. The for diagnostic purposes. However, this methodology may lead to
urinary profile of patients with MPS III shows an increase in HS, false-positives, or borderline cases, thus ambiguous and unreliable
while DS and CS are normal or at the upper limit of normal. MPS results mainly due to interfering substances, such as hemoglobin.
IVA patients present an elevation of urinary KS and CS only. Finally, False-negative cases may arise owing to the formation of aggre-
MPS VI patients have an elevation of urinary DS and CS. As previ- gates by electrostatic interactions with collagen, glycoproteins,
ously reported [18,21,22,28e30], KS is not only found in MPS IV proteins (albumin or other serum proteins) which may modify
patients, but it might also be elevated, or present at borderline physicochemical properties of GAGs [13]. In fact, false-negative
levels, in urine of patients having all MPS types, especially MPS II. results have often been reported, especially for some MPS III and
The present study confirms this observation. IV cases which are confirmed by the observations from this study.
For comparison purposes, these samples were also analyzed Our results show that the DMB spectrophotometry method has led
using the DMB spectrophotometry assay for total GAG quantitation to normal results for one MPS II, one MPS III, 4 MPS IV, and one MPS
VI patients (see Table 5, patients with IDs 5, 10, 16, 18, 19, 20, and 22)
and borderline results for two MPS III and one MPS IV patient (see
Table 4 Table 5, patients with IDs 8, 11, and 15). Moreover, differentiation of
Normal reference values for dermatan sulfate (DS), heparan sulfate (HS), keratan each GAG is not possible with this methodology.
sulfate (KS) and chondroitin sulfate (CS) selected disaccharides according to
The UPLC-MS/MS analysis of KS after enzymatic digestion
different age ranges. Standard deviations are shown in brackets.
(Table 5, Assay 3) seems to be very sensitive for the diagnosis of
Age range (years) n Reference values (mg/mmol creatinine) MPS IVA, even in older patients. In fact, all MPS IVA patients clearly
DS HS KS CS had abnormal results. However, considering that KS is not a very
<1 6 25 (4.8) 5 (1.2) 225 (52.0) 36 (8.5)
specific biomarker, positive results can be obtained for patients
1e3 13 14 (3.8) 3 (0.6) 84 (15.9) 19 (3.2) with other MPSs. We suggest that this test should thus be done as a
4e9 20 11 (1.9) 2 (0.4) 52 (9.7) 12 (2.4) second-tier test following the analysis by the multiplex GAG assay,
10e17 23 11 (2.7) 2 (0.4) 29 (6.4) 7 (1.9) especially in the presence of a clinical suspicion of Morquio
>18 17 9 (1.2) 1 (0.2) 27 (6.3) 2 (0.8)
syndrome.
 C. Auray-Blais et al. / Analytica Chimica Acta 936 (2016) 139e148 145

Fig. 3. Results of GAG analysis using the devised UPLC-MS/MS multiplex assay based on a methanolysis process. Urinary concentrations of: A) DS; B) HS; C) KS; and D) CS selected
disaccharides for patients affected by different MPS types are shown. Results are displayed as ratios to normal reference values. Ctrls: healthy controls. Dotted line: normal reference
value.

Table 5
Analysis of glycosaminoglycans (GAGs) in urine samples from untreated MPS patients. Units are as followed: Assay 1: mg/mmol creatinine, Assays 2 and 3: mg/g creatinine.
Values in bold are higher than normal reference values. DS: dermatan sulfate; HS: heparan sulfate; KS: keratan sulfate; CS: chondroitin sulfat; Ref: normal reference values; N/
A: not available.

ID MPS type Age (y) Gender Assay 1 Assay 2 Assay 3

DS, HS, KS, CS by methanolysis (MS/MS) Total GAGs by DMB KS by

spectrophotometry enzymatic
digestion (MS/
MS)

DS Ref HS Ref KS Ref CS Ref Total GAGs Ref KS Ref

1 I 0.8 M 272 25 478 5 194 225 69 36 N/A 218 7.2 11.5

2 I 10.0 F 67 11 121 2 42 29 18 7 251 75 3.5 3.1
3 I 20.0 F 25 9 73 1 22 27 9 2 85 50 1.1 0.6

4 II 9.0 M 40 11 62 2 64 52 13 12 201 85 3.8 3.1

5 II 10.0 M 7 11 3 2 83 29 9 7 57 75 8.7 3.1
6 II 12.8 M 53 11 183 2 47 29 9 7 N/A 75 2.8 2.7
7 II 18.0 M 39 9 163 1 41 27 9 2 251 64 3.2 0.6

8 III 2.0 M 11 14 551 3 59 84 18 19 167 156 3.5 11.5

9 III 2.9 F 14 14 1678 3 111 84 27 19 484 156 6.8 11.5
10 III 3.0 M 6 14 241 3 39 84 11 19 99 156 3.8 4.7
11 III 14.0 M 4 11 270 2 15 29 2 7 69 64 0.7 2.7

12 IVA 2.0 F 12 14 2 3 132 84 59 19 278 156 86.1 11.5

13 IVA 3.0 M 7 14 0 3 139 84 54 19 530 156 79.7 4.7
14 IVA 5.0 M 8 11 3 2 164 52 45 12 184 142 77.7 4.7
15 IVA 5.0 F 6 11 1 2 105 52 34 12 146 142 53.2 4.7
16 IVA 5.0 F 7 11 1 2 69 52 27 12 84 142 31.3 4.7
17 IVA 15.0 M 8 11 2 2 48 29 21 7 95 64 27.3 2.7
18 IVA 16.0 F 6 11 1 2 24 29 3 7 4 64 3.4 0.6
19 IVA 31.0 F 9 9 1 1 29 27 6 2 33 50 3.8 0.6
20 IVA 36.0 M 4 9 1 1 26 27 5 2 21 50 6.0 0.6

21 VI 10.0 F 85 11 3 2 71 29 56 7 420 75 5.5 3.1

22 VI 42.0 M 16 9 1 1 25 27 12 2 40 50 0.8 0.6
146 C. Auray-Blais et al. / Analytica Chimica Acta 936 (2016) 139e148

3.4. Treated MPS patients of 6 months. When the first urine specimen was collected, this
patient had been treated for 2.2 years. Baseline GAG levels are
Urine samples from treated MPS patients were analyzed using unknown.
the devised multiplex methodology based on GAG methanolysis.
Results are shown in Table 6. These results show that GAG levels 3.6. Specificity regarding different pathologies
were normalized or slightly above normal reference values in some
patients treated with enzyme replacement therapy (see Table 6, The specificity of the proposed methodology was examined for
patients with IDs 5, 6, 7, 17, 18, 22, 28, 33, 35, and 36). However, patients affected with various lysosomal storage disorders. Results
most patients had GAG levels that were still above reference values. are shown in Table 7. In general, DS, HS, and CS urinary levels were
It is noteworthy to mention that baseline GAG levels are not known slightly above or within normal limits. However, KS levels were
for these patients. abnormal in the majority of patients. As expected, the patient with
GM1 gangliosidosis had abnormal levels of KS, related to deficiency
3.5. Longitudinal follow up of b-galactosidase enzyme activity catalyzing the hydrolysis of b-
galactosides. As reported before in the literature, the patient with
Two patients had urine samples collected before and after the mucolipidosis II had an elevation of KS, but also of CS [18,28]. All
beginning of treatment with ERT, to assess fluctuations of urinary three patients with alpha-mannosidosis had an important eleva-
GAGs. Results are shown in Fig. 4. Patient A was an 8 month-old tion of KS and this was the only GAG abnormality detected. Finally,
MPS I patient who was followed up over a period of 5 years, patients with Gaucher disease, Fabry disease, Pompe disease,
while Patient B was a 12 year-old MPS II patient followed up for a aspartylglucosaminuria, and Schindler disease had KS urinary
period of 2 years. These two patients showed a diminution of DS, levels slightly above normal, or within normal limits. In the lights of
and HS urinary levels after ERT treatment which was sustained over these results, an elevation of urinary KS should be interpreted with
time. A reduction of KS level was also observed in patient A, but this caution, taking into account the entire clinical profile of the patient.
diminution was not relevant when compared to age-matched
reference values. 4. Conclusions
A 14 year-old patient with Morquio A syndrome also had urine
samples collected after beginning of treatment with ERT. Results GAG biomarker measurements are important for high-risk
are shown in Fig. 5. This patient was followed up closely for a period screening, diagnosis, and monitoring the efficacy of ERT. The

Table 6
Analysis of glycosaminoglycans (GAGs) in urine samples from treated MPS patients. Values in bold are higher than normal reference values. DS: dermatan sulfate; HS: heparan
sulfate; KS: keratan sulfate; CS: chondroitin sulfate; Ref: normal reference values; N/A: not available.

ID MPS type Age (y) Gender Treatment Time since ERT initiation (y) GAGs (mg/mmol creatinine)

DS Ref HS Ref KS Ref CS Ref

1 I 4 F ERT N/A 14 11 11 2 48 52 13 12
2 I 7 F ERT N/A 24 11 19 2 40 52 12 12
3 I 8 M ERT N/A 91 11 199 2 88 52 40 12
4 I 9 F BMT N/A 18 11 10 2 82 52 15 12
5 I 9 F ERT N/A 7 11 8 2 18 52 5 12
6 I 12 M ERT N/A 10 11 5 2 25 29 8 7
7 I 14 M ERT 0.8 9 11 3 2 29 29 7 7
8 I 15 M ERT 9.0 15 11 16 2 20 29 6 7
9 I 16 M ERT 13.0 9 11 26 2 40 29 6 7

10 II 5 M ERT 2.1 10 11 19 2 30 52 9 12
11 II 6 M ERT 0.3 17 11 33 2 39 52 9 12
12 II 7 M ERT 6.6 13 11 17 2 36 52 9 12
13 II 10 M ERT 6.5 12 11 14 2 41 29 9 7
14 II 11 M ERT 9.0 5 11 16 2 48 29 6 7
15 II 13 M ERT 11.0 11 11 27 2 36 29 9 7

16 IVA 6 M ERT 0.2 12 11 2 2 93 52 44 12

17 IVA 8 M ERT 2.7 9 11 2 2 47 52 19 12
18 IVA 8 F ERT 3.2 7 11 1 2 43 52 16 12
19 IVA 10 F ERT 2.6 17 11 2 2 47 29 15 7
20 IVA 11 M ERT 2.2 9 11 2 2 69 29 19 7
21 IVA 11 M ERT 3.2 10 11 1 2 65 29 24 7
22 IVA 12 F ERT 2.1 9 11 1 2 34 29 5 7
23 IVA 12 F ERT 1.7 11 11 2 2 55 29 11 7
24 IVA 13 F ERT 1.6 12 11 1 2 38 29 8 7
25 IVA 13 M ERT 2.7 11 11 2 2 44 29 20 7
26 IVA 15 M ERT 2.7 11 11 1 2 54 29 18 7
27 IVA 16 M ERT 4.0 5 11 1 2 71 29 22 7
28 IVA 17 M ERT 1.7 8 11 1 2 29 29 7 7
29 IVA 18 M ERT 4.0 5 9 1 1 59 27 17 2
30 IVA 19 F ERT 1.6 12 9 1 1 50 27 5 2
31 IVA 21 F ERT 2.1 12 9 1 1 36 27 5 2

32 VI 4 M ERT N/A 18 11 1 2 43 52 18 12
33 VI 6 M ERT N/A 13 11 1 2 29 52 10 12
34 VI 15 M ERT N/A 31 11 2 2 55 29 14 7
35 VI 15 F ERT N/A 11 11 1 2 22 29 5 7
36 VI 18 F ERT N/A 12 9 1 1 21 27 3 2
 C. Auray-Blais et al. / Analytica Chimica Acta 936 (2016) 139e148 147

Table 7
UPLC-MS/MS analysis of DS, HS, KS, and CS disaccharides in urine of patients with
other LSDs. Values in bold are higher than normal reference values.

Disorder Age GAGs (mg/mmol creatinine)

DS Ref HS Ref KS Ref CS Ref

Gaucher disease 69 y 3 9 2 1 38 27 2 2
Fabry disease 75 y 6 9 0 1 33 27 2 2
Fabry disease 24 y 4 9 1 1 17 27 1 2
Alpha-mannosidosis 5y 3 11 1 2 186 52 12 12
Alpha-mannosidosis 17 y 2 11 1 2 103 29 1 7
Alpha-mannosidosis 7y 2 11 1 2 170 52 3 12
Mucolipidosis II 6d 13 25 4 5 287 225 44 36
Pompe disease 34 y 8 9 1 1 34 27 3 2
GM1 gangliosidosis 3m 11 25 7 5 785 225 22 36
Aspartylglucosaminuria 8y 3 11 2 2 56 52 8 12
Schindler disease 3y 0 14 1 3 44 84 0 19

However, no urine specimens from MPS IVB, and MPS VII patients
were available for confirmation in this study. This novel fully vali-
dated method offers a complete GAG analysis over a 7-min run-
time. This efficient GAG fractionation method enables the distinc-
tion of patients with different types of MPSs, except for MPS I and II
patients, which are characterized by the same excretion profile.
Using this devised multiplex methodology, all patients, except one
MPS IVA patient, presented abnormal GAG profile results specif-
ically related to their disease. This latter Morquio IVA patient who
had normal GAG results with the MS/MS method was a 16 year old
female who presented abnormal results using the enzymatic ker-
atanase method recently published by our group, which is a
method specific for KS [17]. The analysis of KS by the proposed
multiplex assay seemed to be less sensitive in older MPS IVA pa-
tients, and we recommend using the enzymatic keratanase method
Fig. 4. Variation of the urinary GAG excretion levels in: a) an 8 month-old MPS I pa-
tient over a period of 5 years; and B) a 12 year-old MPS II patient over a period of 2 or enzyme activity assay for these patients.
years before and after beginning of enzyme replacement therapy. The comparison of the mass spectrometry method with the
DMB spectrophotometric method showed that, among patients
under study, the latter test would have missed the detection of 7
objective of this research project was, among others, to develop a MPS patients out of 23 (30%) (one MPS II, one MPS III, 4 MPS IVA,
high resolution quantitative tandem mass spectrometry method, and one MPS VI patients) and has shown borderline results for 3
using internal standards, to analyze specifically urinary DS, HS, KS others (13%) (two MPS III and one MPS IVA patients), confirming
and CS-related disaccharides in a single, simultaneous analysis. To that this spectrophotometric total GAG test is not reliable for
our knowledge, this is the first methodology allowing the analysis screening MPS patients. Moreover, it does not offer quantitation of
of KS simultaneously with DS, HS, and CS related disaccharides individual GAG since their fractionation is not possible by spec-
using a methanolysis-based depolymerisation reaction. This trophotometry colorimetric reaction.
improved method now allows the efficient targeting of MPS I, MPS The MS/MS analysis of urine samples from ERT treated MPS
II, MPS III, MPS IVA, MPS IVB, MPS VI, and MPS VII patients. patients revealed that some patients had normalization or marginal

Fig. 5. Variation of the urinary GAG excretion levels in a 14 year-old MPS IVA patient treated with ERT over a period of 6 months.
148 C. Auray-Blais et al. / Analytica Chimica Acta 936 (2016) 139e148

elevations of GAG levels. However, most patients had GAG levels diagnosis of mucopolysaccharidoses d an evaluation, Ann. Clin. Biochem. 44
(2007) 360e363.
that were still increased. One limitation of the study was that we,
[10] J.R. Alonso-Fernandez, J. Fidalgo, C. Colon, Neonatal screening for mucopoly-
unfortunately, did not have baseline GAG levels. saccharidoses by determination of glycosaminoglycans in the eluate of urine-
This multiplex mass spectrometry method allowed the longi- impregnated paper: preliminary results of an improved DMB-based proce-
tudinal evaluation of GAGs in a MPS I and a MPS II patient prior- and dure, J. Clin. Lab. Anal. 24 (2010) 149e153.
[11] J.G.N. de Jong, R.A. Wevers, C. Laarakkers, B.J.H.M. Poorthuis, Dimethyl-
post ERT treatment. Our results show a reduced DS and HS level methylene blue-based spectrophotometry of glycosaminoglycans in un-
after ERT over time. A 6-month longitudinal study of a treated MPS treated urine: a rapid screening procedure for mucopolysaccharidoses, Clin.
IVA patient was conducted after ERT initiation (no data available Chem. 35 (7) (1989) 1472e1477.
[12] M. Hochuli, K. Wϋthrich, B. Beat Steinmann, Two-dimensional NMR spec-
prior to ERT) and showed that urinary GAG levels were stable over troscopy of urinary glycosaminoglycans from patients with different muco-
this period. The specificity of the proposed method which was polysaccharidoses, NMR Biomed. 16 (2003) 224e236.
evaluated in pathological samples of patients with other LSDs was [13] M. Piraud, I. Maire, M. Mathieu, Pitfalls of screening for mucopolysacchar-
idoses by the Dimethylmethylene Blue test, Clin. Chem. 39 (1) (1993) 163.
good for DS and HS in various LSDs, but showed a definite lack of [14] C. Auray-Blais, P. Bhe
rer, R. Gagnon, S.P. Young, H.H. Zhan, Y. An, J.T.R. Clarke,
specificity for KS in mucolipidosis II and alpha-mannosidosis. D.S. Millington, Efficient analysis of urinary glycosaminoglycans by LC-MS/MS
Overall, the multiplex mass spectrometry method reported here in mucopolysaccharidoses type I, II and VI, Mol. Genet. Metab. 102 (1) (2011)
49e56.
is a useful and reliable tool for the detection and diagnosis of MPS [15] C. Auray-Blais, P. Lavoie, H. Zhang, R. Gagnon, J.T.R. Clarke, B. Maranda,
patients, as well as monitoring and follow up of treated patients. S.P. Young, Y. An, D.S. Millington, An improved method for glycosaminoglycan
analysis by LC-MS/MS of urine samples collected on filter paper, Clin. Chim.
Acta. 413 (7e8) (2012) 771e778, 11.
Disclaimer
[16] H. Zhang, S.P. Young, D.S. Millington, Quantification of glycosaminoglycans in
urine by isotope-dilution liquid chromatography-electrospray ionization
No funding organization played a role in the design of study, tandem mass spectrometry, Curr. Protoc. Hum. Genet. 17 (2013) 12.
interpretation of data, and preparation of this manuscript. [17] C. Auray-Blais, P. Lavoie, B. Maranda, M. Boutin, Evaluation of urinary keratan
sulfate disaccharides in MPS IVA patients using UPLC-MS/MS, Bioanalysis 8 (3)
(2016) 179e191.
Acknowledgements [18] C. Auray-Blais, D.S. Millington, F. Wijburg, T. Wood, R. Giugliani, P. Harmatz,
K. Ellsworth, P. Lavoie, N. van Vlie, H. Zhang, N. Miller, Urine keratan sulfate
(uKS) elevation in lysosomal disorders (LSD): comparison of uKS levels in
This research was funded by grants-in-aid of research from Morquio/MPS IV versus non-Morquio LSD, Mol. Genet. Metab. 114 (2) (2015)
Shire. The funding organization played no role in the choice of S16.
enrolled patients, interpretation of data, or preparation of this [19] H. Zhang, S.P. Young, C. Auray-Blais, P.J. Orchard, J. Tolar, D.S. Millington,
Analysis of glycosaminoglycans in cerebrospinal fluid from patients with
manuscript. We are thankful to Waters Corporation for their tech- mucopolysaccharidoses by isotope-dilution ultra-performance liquid
nical and scientific support. We would like to thank Dr Paula Waters chromatography-tandem mass spectrometry, Clin. Chem. 57 (7) (2011)
and Mr Denis Cyr in the Biochemical Genetics Division at the 1005e1012.
[20] S. Tomatsu, F. Kubaski, K. Sawamoto, R.W. Mason, E. Yasuda, T. Shimada,
Sherbrooke Medical Centre for the DMB spectrophotometric total A.M. Montan ~ o, S. Yamaguchi, Y. Suzuki, T. Orii, Newborn screening and
GAG reference values. We would like to thank Dr Anais Brassier diagnosis of mucopolysaccharidoses: application of tandem mass spectrom-
from the Ho ^ pital Universitaire Necker-Enfants Malades et Institut etry, Nihon Masu Sukuriningu Gakkai Shi 24 (2014) 19e37.
[21] S. Tomatsu, T. Shimada, R.W. Mason, A.M. Montan ~ o, J. Kelly, W.A. LaMarr,
IMAGINE and Mr Mohammed Hussain from the Hospital for Sick
F. Kubaski, R. Giugliani, A. Guha, E. Yasuda, W. Mackenzie, S. Yamaguchi,
Children for specimen and data collection. We would also like to Y. Suzuki, T. Orii, Establishment of glycosaminoglycan assays for mucopoly-
thank all MPS patients for their cooperation. Without them this saccharidoses, Metabolites. 4 (3) (2014) 655e679, 11.
study would not have been possible. [22] T. Shimada, S. Tomatsu, R.W. Mason, E. Yasuda, W.G. Mackenzie, J. Hossain,
Y. Shibata, A.M. Montano, F. Kubaski, R. Giugliani, S. Yamaguchi, Y. Suzuki,
K.E. Orii, T. Fukao, T. Orii, Di-sulfated keratan sulfate as a novel biomarker for
Appendix A. Supplementary data mucopolysaccharidosis II, IVA, and IVB, J. Inherit. Metab. Dis. Rep. 21 (2015)
1e13.
[23] H. Naimy, K.D. Powell, J.R. Moriarity, J. Wu, T.G. McCauley, P.A. Haslett,
Supplementary data related to this article can be found at http:// A.J. Barbier, Y. Qiu, A novel LCeMS/MS assay for heparan sulfate screening in
dx.doi.org/10.1016/j.aca.2016.06.054. the cerebrospinal fluid of mucopolysaccharidosis IIIA patients, Bioanalysis 8
(4) (2016) 285e295.
[24] J.E. Silbert, G. Sugumaran, Biosynthesis of chondroitin/dermatan sulfate,
References IUBMB Life 54 (4) (2002) 177e186.
[25] S.B. Dulaney, X. Huang, Strategies in synthesis of heparin/heparan sulfate
[1] J. Muenzer, The mucopolysaccharidoses: a heterogeneous group of disorders oligosaccharides: 2000-present, Adv. Carbohydr. Chem. Biochem. 67 (2012)
with variable pediatric presentations, J. Pediatr. 144 (2004) S27eS34. 95e136.
[2] J. Muenzer, Overview of the mucopolysaccharidoses, Rheumatology 50 [26] V.H. Pomin, Keratan sulfate: an up-to-date review, Int. J. Biol. Macromol. 72
(Suppl. 5) (2011) v4e12. (2015) 282e289.
[3] R. Giugliani, A. Federhen, M.V. Rojas, T. Vieira, O. Artigal
as, L.L. Pinto, et al., [27] C. Auray-Blais, M. Boutin, R. Gagnon, F.O. Dupont, P. Lavoie, J.T.R. Clarke,
Mucopolysaccharidosis I, II, and VI: brief review and guidelines for treatment, Urinary globotriaosylsphingosine-related biomarkers for Fabry disease tar-
Genet. Mol. Biol. 33 (4) (2010) 589e604. geted by metabolomics, Anal. Chem. 84 (6) (2012) 2745e2753, 20.
[4] R. Giugliani, A. Federhen, F. Vairo, C. Vanzella, G. Pasqualim, L.M. da Silva, [28] S. Tomatsu, K. Okamura, H. Maeda, T. Taketani, S.V. Castrillon, M.A. Gutierrez,
L. Giugliani, A.P. de Boer, C.F. de Souza, U. Matte, G. Baldo, Emerging drugs for T. Nishioka, A.A. Fachel, K.O. Orii, J.H. Grubb, A. Cooper, M. Thornley, E. Wraith,
the treatment of mucopolysaccharidoses, Opin. Emerg. Drugs 21 (1) (2016) L.A. Barrera, L.S. Laybauer, R. Giugliani, I.V. Schwartz, G.S. Frenking, M. Beck,
9e26. S.G. Kircher, E. Paschke, S. Yamaguchi, K. Ullrich, M. Haskins, K. Isogai,
[5] Z. Banecka-Majkutewicz, J. Jako
bkiewicz-Banecka, M. Gabig-Cimin
ska, Y. Suzuki, T. Orii, N. Kondo, M. Creer, T. Okuyama, A. Tanaka, A. Noguchi,
A. We˛ grzyn, G. We˛ grzyn, Putative biological mechanisms of efficiency of Keratan sulfate levels in mucopolysaccharidoses and mucolipidoses, J. Inherit.
substrate reduction therapies for mucopolysaccharidoses, Immunol. Ther. Exp. Metab. Dis. 28 (2005) 187e202.
60 (6) (2012) 461e468. [29] S. Tomatsu, M.A. Gutierrez, T. Ishimaru, O.M. Pen ~ a, A.M. Montan ~ o, H. Maeda,
[6] L.A. Clarke, B. Winchester, R. Giugliani, A. Tylki-Szyman
ska, H. Amartino, S. Velez-Castrillon, T. Nishioka, A.A. Fachel, A. Cooper, M. Thornley, E. Wraith,
Biomarkers for the mucopolysaccharidoses: discovery and clinical utility, Mol. L.A. Barrera, L.S. Laybauer, R. Giugliani, I.V. Schwartz, G.S. Frenking, M. Beck,
Genet. Metab. 106 (4) (2012) 395e402. S.G. Kircher, E. Paschke, S. Yamaguchi, K. Ullrich, K. Isogai, Y. Suzuki, T. Orii,
[7] K.C. Huang, K. Sukegawa, T. Orii, Screening test for urinary glycosaminogly- A. Noguchi, Heparan sulfate levels in mucopolysaccharidoses and mucoli-
cans and differentiation of various mucopolysaccharidoses, Clin. Chim. Acta pidoses, J. Inherit. Metab. Dis. 28 (2005) 743e757.
151 (2) (1985) 147e156. [30] D.J. Rowan, S. Tomatsu, J.H. Grubb, A.M. Montan ~ o, W.S. Sly, Assessment of
[8] H.K. Berry, Screening for mucopolysaccharide disorders with the Berry spot bone dysplasia by micro-CT and glycosaminoglycan levels in mouse models
test, Clin. Biochem. 20 (5) (1987) 365e371. for mucopolysaccharidosis type I, IIIA, IVA, and VII, J. Inherit. Metab. Dis. 36 (2)
[9] G. Gray, P. Claridge, L. Jenkinson, A. Green, Quantitation of urinary glycos- (2013) 235e246.
aminoglycans using dimethylene blue as a screening technique for the