Tandem Mass Spectrometric Assay of N-Acetylglucosamine-6-sulfatase for Multiplex Analysis of Mucopolysaccharidosis-IIID in Dried Blood Spots

Abstract

Previously we developed a multiplex liquid chromatography-tandem mass spectrometry (LC-MS/MS) assay for all subtypes of mucopolysaccharidoses (MPS) except MPS-IIID. Here we show that the MPS-IIID enzyme N-acetylgalactosamine-6-sulfatase (GNS) is inhibited in dried blood spot (DBS) extracts, but activity can be recovered if the extract is diluted to reduce the concentrations of endogenous inhibitors. The new GNS assay displays acceptable characteristics including linearity in product formation with incubation time and amount of enzyme, low variability, and ability to distinguish MPS-IIID-affected from healthy patients using DBS. The assay can be added to the LC-MS/MS multiplex panel for all MPS subtypes requiring ~2 min per newborn for the LC-MS/MS run.

1. Introduction.

The mucopolysaccharidoses (MPSs) are a set of lysosomal storage diseases caused by deficiency of one of several enzymes required for the degradation of glycosaminoglycans in lysosomes. Newborn screening (NBS) of these diseases is of interest because several of them are now treatable either by enzyme replacement therapy, hematopoietic stem cell transplantation, or a combination of both. MPS-I and MPS-II are now on the Recommended Newborn Screening Panel in the USA, and several states and non-USA countries now screen for these diseases. We have developed liquid chromatography-tandem mass spectrometry (LC-MS/MS) for the multiplex analysis of all MPS disorders (except the very recently discovered subtype MPS-X [1]) in which all enzymatic activities are obtained by quantification of enzymatic products in a single LC-MS/MS run per newborn lasting 1.9 min [2]. Assay of all of the relevant enzymes except N-acetylglucosamine-6-sulfatase (GNS) was demonstrated using dried blood spots (DBS). For GNS, activity was detected only in fibroblasts; no activity was detected when a 3 mm punch of a DBS was used.

In this study we further analyzed the possibility of detection of GNS using DBS. These studies show that DBS contains one or more endogenous inhibitors of GNS, and activity was readily detected if the DBS extracted is diluted to reduce the concentration of the relevant inhibitor(s). With this method in place, the multiplex analysis of all MPS subtypes (except MPS-X) is achieved at 1.9 min per sample using LC-MS/MS.

2. Materials and Methods

2.1. Materials

We have previously reported the syntheses of the GNS enzymatic assay substrate and internal standard (GNS-S and GNS-IS) (Figure 1). All chemicals used were reagent grade, unless specified otherwise. All experiments with human blood samples were approved by the University of Washington IRB. DBS were stored at −20°C in sealed plastic bags containing desiccant. All MPS-IIID patients, whose DBS was used in this study, were previously diagnosed by observing low GNS enzymatic activity and by genotyping.

Figure 1.

Substrate and Internal standard for the GNS enzymatic assay.

2.2. Methods

Western blot analysis of GNS: Either 10 – 50 ng recombinant human GNS (R&D Systems, Cat No. 2484-SUC-010) or serial dilutions of 10 μL pooled frozen whole blood sample were loaded per well of the gel. Samples were mixed 1:1 with 2x Laemmli buffer (Bio-Rad, Cat No. 1610737) and 5% β-mercaptoethanol, boiled 95 °C for 5 min prior to loading (Bio-Rad, Cat No. 4561025). A protein ladder (Bio-Rad, Cat No. 1610374) was used for size comparison. Proteins were transferred to a nitrocellulose membrane (Bio-Rad, Cat No. 1620112) using semi-dry transfer (Trans-Blot SD Semi-Dry Transfer Cell, Bio-Rad) after SDS-PAGE. The membrane was washed in TBST (Tris-buffered saline, 0.1% Tween-20), and blocked in blocking buffer (LI-COR, Cat No. 927–80001) for 1 h at 4 °C. Western blot was performed using a 1:1000 dilution of GNS Recombinant Rabbit Monoclonal Antibody (Thermo Fisher Scientific, Cat No. MA5–41050) overnight at 4 °C, followed by detection with a 1:10000 dilution of IRDye 800CW Goat anti-Rabbit IgG Secondary Antibody (LI-COR, Cat No. 926–32211) for 1 h at room temperature. The membrane was scanned (Odyssey 9120 Imager, LI-COR) and processed with ImageJ software.

GNS enzymatic assay: The GNS assay cocktail was prepared by mixing an appropriate amount of GNS substrate (GNS-S) and internal standard (GNS-IS) in 50 mM sodium acetate buffer (pH 5.0) with cerium acetate (5 mM) to give a concentration of 2.5 mM GNS-S and 2.5 μM GNS-IS. GNS-S and GNS-IS were added from methanol stock solutions, and methanol was completely removed in a Speed-Vac centrifugal concentrator prior to the addition of buffer. A single 3 mm DBS punch was placed in each well of a 96-deep well plate (Corning, Cat No. CLS3959–100EA). To each well 15 μL of assay cocktail and additional 285 uL of assay buffer were added to give a final concentration of 0.125 mM GNS-S and 0.125 μM GNS-IS. The plate was sealed with a polyester plate seal and was shaken for 40 h at 37 °C using an orbital shaker (2̴50 rpm) incubator, followed by post-assay sample processing.

Post-assay sample processing: The assay was quenched with the addition of 100 μL 1:1 methanol-ethyl acetate mixture, followed by the addition of 400 μL ethyl acetate and 200 μL 0.5 M aqueous sodium chloride. After mixing up and down ~10 times with the pipettor, the plate was centrifuged (Allegra X- 12R, Beckmann) for 5 min at 3000 g to accelerate the solvent layer separation. A 150 μL aliquot of the top layer was pipet transferred to a shallow 96-well plate (Greiner, Cat No. M8185–100EA), and the solvent was dried under a stream of nitrogen at room temperature. The resulting residue was reconstituted in 150 μL 7:3 acetonitrile-water mixture with 0.1% formic acid. After mixing up and down ~10 times with the pipettor, samples were subjected to LC-MS/MS.

2.3. LC-MS/MS

MS/MS method development was carried out with a Waters Xevo-TQ MS/MS instrument. An electrospray ionization (ESI) source was used in the positive polarity mode. Parent ions were scanned in the first quadrupole of the instrument, fragmented in the second quadrupole from collision-induced dissociation (CID), and the daughter ions were scanned in the third quadrupole. ESI source parameters and MRM transition parameters are given in Supporting Information (Table S1 and S2).

LC-MS/MS was carried out with a Waters Xevo-TQ MS/MS instrument coupled to a Waters Acquity binary solvent system for UPLC. The UPLC column and guard column were from Waters (ACQUITY CSH C18 1.7 μm, 2.1×50 mm, Cat. 186005296) with a pre-column (Cat. 18605303) and held at 22 °C in the column oven. Solvent A was water with 0.1% formic acid, and solvent B was acetonitrile with 0.1% formic acid. The gradient started with 30% B, increased linearly to 56% B at 1.4 min, then linearly to 99% B at 1.41 min, held at 99% B until 1.70 min and then switched back to 30% B for re-equilibration, with a flow rate of 0.4 mL/min. The weak and strong needle wash solvents for the autosampler were water/acetonitrile (9:1) with 0.1% formic acid and acetonitrile with 0.1% formic acid. All solvents were Optima grade from Fisher Scientific.

A typical LC-MS/MS chromatogram is shown in Fig. 2. GNS-IS and GNS-P eluted at 0.9 min. The GNS-S peak eluted at 1.6 min, which was fully separated from the product peak (not shown). To minimize contamination of the ESI source, the void volume eluent was diverted to waste via a switching valve after 1.3 min.

Figure 2.

Typical LC-MS/MS chromatograms of GNS-IS and GNS-P channels for the MPS-IIID assay with a DBS from a random healthy newborn and an MPS-IIID patient.

2.4. GNS Activity Calculation

GNS activity in DBS (μM/h) was calculated by multiplying the amount of GNS-IS by the ion ratio of GNS-P to GNS-IS (blank subtracted), then dividing by the incubation time (h) and volume of blood per 3 mm DBS punch (3.2 μL).

3. Results and Discussion

3.1. Quantification of GNS in human blood

In the previous study, we were able to detect GNS activity in fibroblasts, however initial attempts to detect GNS enzyme activity in DBS were unsuccessful. When a 3 mm DBS punch was incubated in 30 μL of assay cocktail, no enzymatic product could be detected by LC-MS/MS. This outcome indicates either a sufficiently low amount of GNS enzyme such that activity is below the limit of GNS-P detection in DBS or potential GNS inhibitors. Herein, we present the quantitative analysis of GNS levels in frozen human whole blood using western blot analysis, aiming to address the first possibility.

Freeman et al. [3] reported the purification and characterization of human liver GNS. GNS exists in two major forms: Form A with a single subunit of 78 kDa, and form B with two subunits of 48 and 32 kDa. Robertson et al. [4] determined the identical N-terminal sequence between the 78 kDa Form A and the 32 kDa Form B, suggesting that Form A polypeptide undergoes internal peptidase cleavage to yield a 32 kDa N-terminal and a 48 kDa C-terminal species.

Our western blot analysis confirmed the presence of both Form A and Form B (Fig. 3). Form A was observed in both recombinant human GNS and blood extracts, while Form B was exclusively present in blood extracts. Lysosomal enzyme precursors undergo a series of post-translational modifications during transit through pre-lysosomal organelles, however the high-level expression of lysosomal enzymes in cell culture favors the extracellular secretion pathway. As a result, the recombinant GNS exists in a precursor form that hasn’t undergone proteolytic cleavage during transit to the lysosome [5].

Figure 3.

Western blot analysis.

Additional minor diffuse bands around 60 kDa and 120 kDa were also observed, likely representing either incompletely deglycosylated or glycosylated forms of GNS protein [6]. The diffuseness of the band was likely due to heterogeneity in the N-linked oligosaccharides of the glycoprotein and has been observed in other recombinant lysosomal enzymes [7].

We were able to quantify the precursor form of GNS in frozen whole blood. The protein bands were quantified, and a calibration curve for recombinant GNS was established (Supplementary Fig. 1), with a good correlation between the band intensity and the amount of protein. Since the relative amounts of recombinant GNS are known, we determined that there are 54 ng GNS of the precursor form per μL of blood sample, and 600 ng GNS of the mature form.

Next, a standard curve for rhGNS activity versus the amount of protein was established (Fig. 4). The activity of rhGNS was readily detectable, and there was a clear linear correlation between the amount of enzyme used in the assay and the product formation. Taking into account both the precursor and mature forms of the enzyme, we estimated the amount of GNS in a 3 mm DBS punch to be 2 μg, which should be sufficient for generating enough product through the enzymatic reaction for LC-MS/MS quantification, and we would expect a LC-MSMS signal 600 times higher than the no-blood blank.

Figure 4.

Correlation between rhGNS activity and the amount of enzyme used in the assay.

3.2. Optimization and validation of GNS enzymatic assay

Known GNS enzyme inhibitors include inorganic sulfates and phosphates [8]. We introduced cerium (III) acetate into the assay buffer to precipitate trace amounts of sulfates and phosphates, however this approach failed to rescue enzyme activity in DBS, suggesting that the inhibitors are not merely inorganic ions.

Albumin has been reported to be another known inhibitor [9], and notably, inhibition of the GNS enzymatic reaction was observed when the addition of bovine serum albumin (BSA) to rhGNS exceeded 0.1% (w/v), whereas human serum albumin (HSA) concentration in plasma typically ranges between 33~52 g/L [10]. The mechanism of albumin inhibition of GNS is unclear; however, albumin is known to bind to GlcNAc [11]. Furthermore, other serum proteins including conglutinin [12] and mannan-binding protein (MBP) [13], have been reported to show specific binding to GlcNAc with high affinity (K_d = 10⁻⁸ M). Thus, it is hypothesized that one or more components of DBS inhibit this enzymatic reaction, possibly due to inhibitors binding to to the GlcNAc-derived substrate. Upon addition of 10 μL of plasma to 20 ng of rhGNS, the enzymatic activity dropped to 25% of the control without plasma, and 30 μL of plasma completely inhibited the enzyme activity (Supplementary Table. 3).

To alleviate the inhibition and restore the enzyme activity, we sought to dilute the reaction mixture to lower the inhibitor’s concentration. When one punch of DBS was used, and the substrate concentration was held constant at 0.5 mM, the enzymatic reaction rate increased as the reaction mixture was diluted and started to plateau when the total volume reached 300 μL (Fig. 5). In these experiments the total amount of GNS was not varied, only the assay volume was increased. Notably, the dilution strategy increased the activity > 10-fold, compared to the original assay.

Figure 5.

GNS enzymatic activity in the presence of constant concentration of substrate and amount of enzyme as a function of the amount of buffer (dilution) added to the assay.

With the optimal dilution in hand, we investigated the variation of GNS enzymatic activity with substrate concentration. GNS activity in a random newborn DBS punch followed hyperbolic kinetics with increasing substrate concentration (Fig. 6). A K_M of 0.072 mM was determined from fitting the data to the Lineweaver-Burk plot (Supplementary Fig. 2).

Figure 6.

GNS enzymatic activity versus the concentration of substrate.

Next, we tested the GNS assay using various fractions of diluted random newborn DBS extract, and a linear correlation was observed for the GNS activity versus the percentage of one punch of DBS extract (Fig. 7). To assess the consistency of this assay, 6 different DBS punches from same random newborn were assayed, and the coefficient of variation (CV) is 6.0%, indicating high reproducibility. Additionally, product formation displayed linearity from 18 to 40 h (Supplementary Fig. 3).

Figure 7.

GNS activity as a function of the DBS extract fraction in the reaction mixture.

3.2. Application of GNS Enzymatic Assay

Having optimized and validated the GNS DBS assay, the assay was carried out with a 3 mm DBS punch from 54 random healthy newborns and 2 MPS-IIID patients. Two MPS-IIID patients had GNS activity of 0.007 and 0.013 μM/h, and the random newborns had activity in the range 0.022–0.168 μM/h, with a mean activity of 0.067 μM/h (Fig. 8). This outcome suggests that our enzymatic assay is capable of distinguishing MPS-IIID patients from healthy newborns.

Figure 8.

GNS activity in DBS from random healthy newborns and MPS-IIID patients.

As reported previously, GNS-P and GNS-IS can be detected in a single LC-MS/MS run that includes the appropriate analytes for all types of MPS disorders except the recently described MPS-X syndrome, which has not yet been studied in our laboratory. Thus, it will be possible to carry out a single multiplex, LC-MS/MS NBS assay for all of these MPS disorders using DBS. A multiplex panel assay can be setup for all MPS types except MPS-IIID, and a small aliquot of the mixture could be removed and diluted into a well with GNS-S and GNS-IS for a separate incubation. After incubation, the contents from both wells could be combined and submitted to a single LC-MS/MS run per newborn.

Abbreviations:

DBS

dried blood spot

GNS

N-acetylgalactosamine-6-sulfatase

GNS-IS

GNS internal standard

GNS-P

GNS product

GNS-S

GNS substrate

LC-MS/MS

liquid chromatography-tandem mass spectrometry

MPS-IIID

mucopolysaccharidosis-IIID

NBS

newborn screening