Abstract
All States screen for biotinidase deficiency and galactosemia, and X-linked adrenoleukodystrophy (X-ALD) has recently been added to the Recommended Uniform Screening Panel (RUSP). We sought to consolidate these tests by combining them into a single multiplex tandem mass spectrometry assay as well as to improve the current protocol for newborn screening of galactosemia. A 3 mm punch of a dried blood spot (DBS) was extracted with organic solvent for analysis of the C26:0-lysophosphatidylcholine biomarker for X-ALD. An additional punch was used to assay galactose-1-phosphate uridyltransferase (GALT) and biotinidase. All assays were combined for a single injection for analysis by liquid chromatography-tandem mass spectrometry (LC-MS/MS) (2.3 min per sample). The GALT LC-MS/MS assay does not give a false positive for galactosemia if glucose-6-phosphate dehydrogenase is deficient. The multiplex assay shows acceptable reproducibility and provides for rapid analysis of X-ALD, biotinidase deficiency, and galactosemia. The throughput and ease of sample preparation are acceptable for newborn screening laboratories. We also show that the LC-MS/MS assay is expandable to include several other diseases including Pompe and Hurler diseases (enzymatic activities and biomarkers). Because of consolidation of assays, less manpower is needed compared to running individual assays on separate platforms. The flexibility of the LC-MS/MS platform allows each newborn screening laboratory to analyze the set of diseases offered in their panel.
Keywords: lysosomal storage disease, inborn errors of metabolism, enzyme deficiency
Introduction
X-ALD have recently been added to the Recommended Uniform Screening Panel (RUSP) in the USA. Several State newborn screening programs are either live for this disease or are in the validation phase. Newborn screening for X-ALD is also occurring outside of the USA. Most newborn screening laboratories also screen for biotinidase deficiency and galactosemia. It is thus of interest to consolidate all of these tests into a single multiplex panel assay. This 3-plex assay can also be combined with assays for lysosomal enzymes and biomarkers for lysosomal storage diseases.
Most newborn screening laboratories assay GALT in DBS using the quantitative Beutler fluorimetric assay [1], which is outlined in Figure 1. The DBS extract is mixed with buffer containing galactose-1-phosphate, UDP-glucose, and NADP+. GALT generates UDP-galactose and glucose-1-phosphate. Phosphoglucomutase converts glucose-1-phosphate to glucose-6-phosphate, which is then converted to 6-phospho-D-glucono-1,5-lactone by glucose-6-phosphate dehydrogenase and then to ribulose-5-phosphate by 6-phosphogluconate dehydrogenase. These latter 3 enzymes are provided by the DBS, and the latter two enzymatic steps generate 2 equivalents of NADPH, which is detected by its fluorescence. It is important to note that this fluorimetric GALT assay may give rise to a false positive with patients who are deficient in glucose-6-phosphate dehydrogenase [2]. This deficiency is common in certain parts of Africa, Asia, the Mediterranean, and the Middle East. As part of our effort to develop a multiplex assay involving GALT, we sought to provide a new assay that is not affected by glucose-6-phosphate dehydrogenase deficiency.
Figure 1.
Shown is the biochemical pathway from galactose-1-phosphate plus UDP-glucose to ribulose-5-phosphate, where the first enzymatic step is catalyzed by GALT. Two equivalents of NADPH are formed, and this is detected via its fluorescence. Also shown is the LgtC-catalyzed transfer of the galactosyl group from GALT product UDP-galactose to GALT-substrate to form gal-GALT-product, which is detected by LC-MS/MS with the use of an internal standard (not shown).
Methods
Materials
All patient samples were obtained with IRB approval. The following reagents were synthesized as described in Supplemental Material: Biotinidase-substrate, biotinidase-internal standard, and GALT-substrate. MPS-I-substrate, MPS-I-internal standard, Pompe-substrate, and Pompe-internal standard were synthesized as described [3,4]. Sources of other reagents are given in Supplemental Material. The LgtC enzyme was purified as described [5], and its concentration was measured by protein assay using the BCA kit from ThermoFisher (Cat. 23252) using bovine serum albumin as a standard.
Preparation of reagent stock solutions and assay cocktails are given in Supplemental Material.
X-ALD, GALT, and biotinidase assays
The assay uses two 3 mm DBS punches, one for X-ALD (organic solvent extraction) and one for GALT and biotinidase (enzymatic assays in aqueous buffer). To each well of a 96-well, polypropylene, shallow-well plate was added a 3 mm DBS punch followed by 30 μL of biotinidase/GALT assay cocktail (Supplemental Material), and then 10 μL of LgtC working solution (Supplemental Material). The plate was sealed with film-type sealing film or with a silicone-sealing matt. To a second plate was added a second 3 mm DBS punch followed by 100 μL of 18.75 nM d4-C26-LPC in methanol, and the plate was sealed as above. Both plates were shaken on an orbital platform at 250 rpm at 37 °C for 3 hr.
After incubation, the biotinidase/GALT plate was quenched with 100 μL of ethyl acetate/methanol (1/1), and the plate was centrifuged in a swinging bucket rotor at ~900×g for 5 min at room temperature. Thirty μL of supernatant was transferred to a 96-well, shallow-well, black, fluorimeter plate, and to each well in this plate was added 200 μL of water. The remaining liquid in the biotinidase/GALT plate was transferred to a polypropylene deep-well plate. To each well was added 400 μL of ethyl acetate and 200 μL of 0.5 M NaCl in water. The well contents were mixed by aspirating up and down with a pipetor ~10 times, then the plate was centrifuged as above. A portion of the upper ethyl acetate layer (200 μL) was transferred to a new 96-well, polypropylene, shallow-well plate. After 3 hr incubation, the plate containing the d4-C26-LPC was centrifuged as above, and 50 μL was transferred to the plate containing the 200 μL ethyl acetate extract. Solvent was removed with a jet of nitrogen or oil-free air for ~30 min at room temperature. To the residue in each well was added 100 μL of methanol/water (65/35) with 5 mM ammonium acetate. This plate was wrapped with aluminum foil (to prevent solvent evaporation) and placed in the autosampler of the LC-MS/MS instrument (autosampler chamber at 8 °C to prevent solvent loss).
X-ALD, GALT, biotinidase Pompe, and Hurler assays
This 5-plex assay was carried out using a total of three 3 mm DBS punches or two punches. First, we give the 3-punch method. The X-ALD and GALT/biotinidase plates were setup as above. A third plate was setup with a third 3 mm DBS punch per well. To each well was added 30 μL of MPS-I/Pompe assay cocktail (Supplemental Material), and the plate was sealed as above.
After 3 hr incubation, the biotinidase/GALT plate was quenched with 100 μL of ethyl acetate/methanol (1/1), and the plate was centrifuged in a swinging bucket rotor at ~900×g for 5 min at room temperature. Thirty μL of supernatant was transferred to a 96-well, shallow-well, black, fluorimeter plate, and to each well in this plate was added 200 μL of water. A second portion (75 μL) from the centrifuged plate was transferred to the plate containing the MPS-I/Pompe reactions; this transfer served to quench these reactions. All of the liquid in each well was transferred to a 96-well, polypropylene, deep-well plate. To each well was added 400 μL of ethyl acetate and 200 μL of 0.5 M NaCl in water. The well contents were mixed by aspirating up and down with a pipetor ~10 times, then the plate was centrifuged as above. A portion of the upper ethyl acetate layer (200 μL) was transferred to a new 96-well, polypropylene, shallow-well plate. After incubation, the plate containing the d4-C26-LPC was centrifuged as above, and 50 μL was transferred to the plate containing the 200 μL ethyl acetate extract. Solvent was removed with a jet of nitrogen or oil-free air for ~30 min at room temperature. To the residue in each well was added 100 μL of methanol/water (65/35) with 5 mM ammonium acetate. This plate was wrapped with aluminum foil (to prevent solvent evaporation) and placed in the autosampler of the LC-MS/MS instrument (autosampler chamber at 8 °C to prevent solvent loss).
The above 5-plex assay was also carried out using a total of two 3 mm DBS punches as follows. To a 96-well, polypropylene, shallow-well plate (plate A) containing a single 3 mm DBS punch per well was added 30 μL of 0.9% NaCl in water. The plate was sealed as above and shaken at 37 °C and 250 rpm for 30 min. The plate seal was removed, and 15 μL was transferred from each well to the well of a second plate (plate B). To plate A with the DBS punch was added 15 μL of 2X-MPS-I/Pompe assay cocktail (Supplemental Material). To plate B (without the DBS punch) was added 15 μL of 2X-biotinidase/GALT assay cocktail (Supplemental Material) and 10 μL of LgtC working solution (Supplemental Material). The third plate was prepared with a second 3 mm DBS punch with d4-C26-LPC as above. All 3 plates were sealed and incubated for 3 hr at 37 °C with shaking at 250 rpm. The workup of the 3 plates was exactly as described above for the 3-punch method.
Fluorimetry
The 96-well, black, fluorimetry plate was placed in a plate fluorimeter (PerkinElmer Victor 3V) with excitation at 355 nm and emission at 460 nm. Blank fluorimetric assays were carried out as described in Supplemental Materials.
LC-MS/MS
LC-MS/MS was carried out with a Waters Xevo-TQ MS/MS instrument coupled to a Waters Aquity binary solvent system for LC. The LC column and guard column were from Waters (XBridge C8 with guard, Cat. 186006041 and 186007781) and held at 40 °C in the column oven. The flow rate was 0.45 mL/min. Solvent A was water/methanol (1/1) with 5 mM ammonium acetate, and solvent B was methanol/acetonitrile (1/1) with 5 mM ammonium acetate. It was convenient to prepare a 5 M ammonium acetate stock in water and to use this to make the mobile phases. The solvent program was 25% solvent B from 0–0.8 min, then a linear gradient to 100% solvent B from 0.8–0.85 min, then hold at 100% solvent B from 0.85–1.75 min, then jump back to 25% solvent B at 1.75 min and hold at 25% solvent B until 1.85 min. The inject-to-inject was 2.3 min when using the “load ahead” function and a “loop offline time” of 0.5 min. The weak and strong needle wash solvents for the autosampler were water/acetonitrile (9/1) and methanol/isopropanol/water (47.5/47/5/5), respectively. Ten μL were injected per sample using the full-loop method with a loop overfill of 1.5X. MS/MS parameters are given in Supplemental Tables 5 and 6. A standard diversion valve was programed to divert the LC void volume and the post-C26-LPC LC eluant to waste rather than to the electrospray source so as to help maintain source cleanliness.
Calculations
Specific activities of enzymes (μmol/hr/L of blood) were calculated by multiplying the ratio of ion counts for each enzymatic product to that of the corresponding internal standard by the μmoles of internal standard in the assay well,, then dividing by the incubation time (3 hr) and by the volume of blood in a 3 mm DBS punch (3.2 μL for the 3-punch method and 1.6 μL for the 2-punch method). The concentration of C26-LPC in blood in nanomolar was obtained by taking the ratio of ion counts for C26-LPC to that of d4-C26-LPC and multiplying by the nmoles of d4-C26-LPCin the assay (0.001875 nmole) and dividing by the volume of blood in a 3 mm DBS punch (3.2 × 10−6 liters). The specific activity of GALT for the fluorimetric assay (Units per g hemoglobin) was obtained by comparing the fluorimetry reading of the sample to a standard curve made with low, medium, and high quality control GALT DBS obtained from the CDC (Y-axis is fluorimeter reading, X-axis in Units per g hemoglobin for the standards as reported by the CDC (https://www.cdc.gov/labstandards/pdf/nsqap/GALT_Certification_Set_2_2017.pdf).
Results
Description of the enzymatic assays
Figure 1 shows the enzymatic reactions including GALT and those that generate fluorescent NADPH that is detected in the standard fluorimetric GALT assay. In addition, we show a novel reaction in which the bacterial enzyme α-1,4-galactosyltransferase from Neisseria meningitidis (LgtC) transfers a galactosyl group from UDP-galactose to the lactosyl group of GALT-substrate to form the trisaccharide gal-GALT-product (Figure 2 shows the structures). Note that GALT-substrate is not a direct substrate for GALT but is used to measure GALT activity via the coupled assay with LgtC. LgtC displays no enzymatic activity with UDP-glucose as shown below, and thus gal-GALT-product is formed only when UDP-galactose is generated via the action of GALT on galactose-1-phosphate plus UDP-glucose. Detection of gal-GALT-product by LC-MS/MS constitutes an assay for GALT enzymatic activity that does not depend on glucose-6-phosphate dehydrogenase and is thus not affected by deficiency of this latter enzyme (Figure 1). Detection is quantitative by the use of the internal standard UDP-[13C]galactose. This is converted into [13C]gal-GALT-product by LgtC in the incubation mixture. The heavy isotopic trisaccharide is detected along with the non-isotopic trisaccharide gal-GALT-product by LC-MS/MS. Note, any incomplete conversion of UDP-galactose to gal-GALT-product by LgtC is accounted for by using UDP-[13C]galactose as the internal standard since the latter has to undergo the same LgtC-catalyzed step.
Figure 2.
Structures of GALT-substrate and gal-GALT-product.
The new biotinidase assay is shown in Figure 3. Biotinidase acts on biotinidase-substrate to produce the substituted aniline biotinidase-product, which is quantified by LC-MS/MS with use of a structurally identical but isotopically substituted biotinidase-internal standard (Figure 3). All previously reported substrates used to detect biotinidase in DBS contain an aromatic-amine as the leaving group, presumably because substrates with aliphatic-amines are more sluggish substrates (poorer leaving group).
Figure 3.
Biotinidase catalyzed conversion of biotinidase-substrate (top) to biotinidase-product (middle). As shown is biotinidase-internal standard (bottom).
Testing for X-ALD is based on the method of Haynes et al. in which the elevated biomarker C26-LPC is quantified by LC-MS/MS with use of a structurally identical, but isotopically substituted, internal standard d4-C26-LPC [6]. Haynes and colleagues detected C26-LPC in negative ion mode. In our multiplex, we used positive ionization mode since the other analytes are also detected in positive ion mode. In positive ion mode, other species in DBS that are isobaric with C26-LPC are detected, but they are fully resolved from C26-LPC by LC (Supplemental Figure 1)[6]. It is also possible to switch from positive to negative ionization mode just prior to the elution of C26-LPC from the LC column, but this is not needed.
In some studies we also added assays for the lysosomal enzymes relevant to MPS-I and Pompe disease. Substrates and internal standards are as reported previously [4,7]. Again, use is made of structurally-identical, isotopically-substituted internal standards.
Assay procedure
The assay procedure is summarized in Figure 4. One 3 mm DBS punch is needed for methanol extraction for X-ALD. This solvent is denaturing to enzymes, and thus cannot be used for enzymatic assays. A second 3 mm DBS punch is used in a single buffer to assay GALT and biotinidase. After a 3 hr incubation, the samples are worked up and analyzed with a single injection into the LC-MS/MS instrument. A typical work schedule would consist of setting up the multiplex assay in the afternoon of day 1 (after receipt of the DBS in the morning). Setting up the plates for 3 hr incubation requires about 30 min. After incubation, the pre-LC-MS/MS sample processing requires about 1 hr. The autosampler plate is thus loaded on the LC-MS/MS instrument at the end of day 1, and the data is available in the morning of day 2. As an option, an aliquot of the biotinidase/GALT assay can be read on a standard fluorimetry plate reader in the afternoon of day 1, requiring ~30 min, if a same-day GALT result is required. Note that each multiplex involves one LC-MS/MS per newborn since all incubations are combined into a single well of the autosampler plate.
Figure 4.
Panel A shows the assay using three punches from the DBS, and panel B is for the 2 punch method. See text for more information.
We also combined the above 3-plex with assays for two lysosomal storage diseases, Pompe and Hurler diseases. In this case a separate buffer at low pH is needed for the lysosomal enzymes. This can be carried out either with a third 3 mm DBS punch or by pre-extracting a 3 mm DBS into saline and using a portion of the extract for GALT/biotinidase and a portion for the lysosomal enzymes. All assays are combined for a single injection into the LC-MS/MS instrument. In this study we used a 3 hr incubation for all enzymes so that the LC-MS/MS instrument is loaded with samples at the end of day 1.
The liquid-liquid extraction with ethyl acetate is recommended since essentially all of the buffer components including the sodium taurocholate detergent remain in the aqueous phase and are thus not injected onto the LC column. Note that only 10–15% of the sample in the autosampler well is injected, and thus the same sample can be re-injected should a problem occur with the overnight LC-MS/MS run. Thus, an additional DBS punch and sample processing are not needed.
Supplemental Figure 1 shows the LC-MS/MS selective ion traces for the one variation of this method, a 5-plex in which 5 analytes (C26-LPC and products for GALT, biotinidase, Pompe and Hurler enzymes) are quantified along with their respective internal standards. The inject-to-inject time is 2.3 min. Note that two product peaks are seen in the case of Pompe disease. The earlier peak elutes at the retention time of the Pompe-substrate and is due to a small amount of substrate-to-product conversion in the heated electrospray ionization source. This peak is of no concern since it is well resolved from the later eluting product peak, which is due to the enzymatically-generated product. The MPS-I-substrate also breaks down in the electrospray source, but no second product peak is observed since the substrate elutes in the void volume. Breakdown of GALT-substrate is of no concern since it cannot generate gal-GALT-product. Biotinidase-substrate breaks down to biotinidase-product in the heated electrospray source, but the breakdown is barely detectable, and the substrate elutes well before the enzymatically-generated product (Supplemental Figure 1). All enzymatically-generated products elute at the same retention time as their respective internal standards, as expected since internal standards and products are structurally identical. This is ideal since any loss of MS/MS signal due to ion suppression of the product in the electrospray source is accounted for by an identical degree of suppression of the internal standard.
Assay validation and performance
We carried out a series of control experiments to evaluate the new GALT assay. The product gal-GALT-product was not observed if any one of galactose-1-phosphate, UDP-glucose or LgtC was omitted from the assay, but was observed if NADP+ was omitted (not shown). This is the expected results since formation of gal-GALT-product requires only the formation of UDP-galactose, and NADP+ is required only for enzymes downstream of GALT (Figure 1). This result shows that GALT would read normal if glucose-6-phosphate dehydrogenase were deficient. The control results also shows that the LgtC enzyme does not utilize UDP-glucose since formation of the trisaccharide with glucose attached to GALT-substrate would have been seen by LC-MS/MS since it is isobaric with gal-GALT-product. GALT activity was measured as a function of the amount of LgtC added, and results indicate that addition of 0.6 μg of LgtC per assay is sufficient such that no further increase in activity is seen if additional LgtC is used (not shown). Thus, conversion of GALT-substrate to gal-GALT-product is limited by the GALT step rather than the LgtC step as desired.
Next, we carried out reproducibility studies using the 5-plex assay (GALT, biotinidase, X-ALD, Pompe and Hurler). Tables 1–5 show the values of coefficient of variation (CV) for 3–5 repeats of the 5-plex LC-MS/MS assay using multiple punches from a single DBS. These studies were carried out with filter paper only (no blood control), a DBS from a healthy adult, and a DBS from a healthy adult spiked with 0.5 μM C26-LPC (to mimic an X-ALD patient). Table 1 gives the X-ALD component of the 5-plex, and Table 2 gives the biotinidase component. Table 3 gives additional GALT data by LC-MS/MS for the CDC quality control low, medium, and high GALT DBS. Table 4 gives the MPS-I result, and Table 5 gives the Pompe result. Finally, Table 6 gives the GALT data measured with the fluorimetric assay. Assay reproducibility for all 5 tests was acceptable. For biotinidase and GALT, the specific activity of the enzymes (μmole/hr/L) were slightly lower for the 2-punch method compared to the 3-punch method suggesting that not all of the enzymes were extracted into saline for the 2-punch method. On the other hand, the X-ALD, MPS-I, and Pompe disease results were very similar for the 2- and 3-punch methods. Figures 5 and 5 show good linearity for the GALT assays (LC-MS/MS and fluorescence) versus the amount of GALT in the CDC quality control standards (in units of Units/g hemoglobin) as provided by the certification sheets on the CDC website (see Methods). Since the blank fluorescence for all the blank samples are similar (Table 6), one may choose to ignore the blank and read the GALT activity in U/g hemoglobin from the fluorescence reading of the newborn DBS compared to those from the standard curve (Figure 6) without blank correction.
Table 1.
LC-MS/MS assay. X-ALD analysis
| DBS Sample | Method | C26-LPC (ion counts)1 | d4-C26-LPC (ion counts)1 | C26-LPC (nM)2 | C26-LPC (nM) Average (% CV) |
|---|---|---|---|---|---|
| filter paper (no blood) | 3-punch | 4.0 | 2069 | essentially 0 | |
| normal adult | 3-punch | 124 | 2,241 | 32.5, 40.1, 31.0, 30.4, 21.4 | 31.1 (19) |
| normal adult + 0.5 μM C26-LPC | 3-punch | 2,714 | 2,838 | 560, 470, 556, 527, 528 | 528 (6) |
| filter paper (no blood) | 2-punch | 4,8 | 1,450 | essentially 0 | |
| normal adult | 2-punch | 181 | 3,486 | 30.4, 24.0, 18.6, 23.0, 21.5 | 23.5 (16) |
| normal adult + 0.5 μM C26-LPC | 2-punch | 1,950 | 3,256 | 351, 343, 417, 374, 430 | 383 (9) |
Ion counts for a typical run are shown.
Results from multiple punches from the same DBS are shown.
Table 5.
5-Plex LC-MS/MS assay. Pompe disease analysis.
| DBS Sample | Method | Product (ion counts)1 | Internal Standard (ion counts)1 | Enzymatic Activity (μM/hr/L)2 | Average (% CV) |
|---|---|---|---|---|---|
| filter paper (no blood) | 3-punch | 3,756 | 401,890 | 0.014, 0.007, 0.005, 0.007, 0.016 | 0.010 (44) |
| normal adult | 3-punch | 123,659 | 282,614 | 0.64, 0.62, 0.64, 0.65, 0.73 | 0.66 (6) |
| normal adult + 0.5 μM C26-LPC | 3-punch | 106,605 | 248,268 | 0.63, 0.63, 0.64, 0.65, 0.73 | 0.66 (6) |
| 30 Random Newborns | 3-punch | 0.42 – 1.72 | |||
| filter paper (no blood) | 2-punch | 781 | 244,997 | 0.009, 0.009, 0.014, 0.007, 0.028 | 0.013 (57) |
| normal adult | 2-punch | 44,457 | 187,432 | 0.67, 0.57, 0.62, 0.65 | 0.64 (7) |
| normal adult + 0.5 μM C26-LPC | 2-punch | 31,580 | 168,100 | 0.55, 0.65, 0.64, 0.65, 0.84 | 0.67 (14) |
| Galactosemia #1 | 2-punch | 0.42 – 1.72 | |||
| Galactosemia #2 | 2-punch | 0.11, 0.11, 0.11 | 0.11 (0) | ||
| Galactosemia #3 | 2-punch | 0.36 | |||
| Galactosemia #4 | 2-punch | 0.41, 0.39, 0.44 | 0.41 (4) | ||
| Biotinidase deficient #1 | 2-punch | 1.16 |
Ion counts for a typical run are shown.
Results from multiple punches from the same DBS are shown.
Table 2.
LC-MS/MS assay. Biotinidase analysis
| DBS Sample | Method | Product (ion counts)1 | Internal Standard (ion counts)1 | Enzymatic Activity (μM/hr/L)2 | Average (% CV) |
|---|---|---|---|---|---|
| filter paper (no blood) | 3-punch | 355 | 19,084 | 0.011, 0.013, 0.015, 0.011, 0.012 | 0.012 (11) |
| normal adult | 3-punch | 7,859 | 10,290 | 0.45, 0.50, 0.38, 0.40, 0.45 | 0.44 (9) |
| normal adult + 0.5 μM C26-LPC | 3-punch | 14,911 | 20,895 | 0.42, 0.61, 0.56, 0.54, 0.60 | 0.54 (13) |
| 30 Random Newborns | 3-punch | 0.028 – 0.149 | |||
| Galactosemia #1 | 3-punch | 0.15, 0.15, 0.14 | 0.14 (4) | ||
| Galactosemia #2 | 3-punch | 0.32 | |||
| Galactosemia #3 | 3-punch | 0.36, 0.27, 0.31 | 0.31 (12) | ||
| Biotinidase deficient #1 | 3-punch | 0.0029 | |||
| filter paper (no blood) | 2-punch | 250 | 13,542 | 0.022, 0.021, 0.021, 0.019, 0.020 | 0.020 (4) |
| normal adult | 2-punch | 3,296 | 13,671 | 0.28, 0.29, 0.25, 0.28, 0.27 | 0.27 (6) |
| normal adult + 0.5 μM C26-LPC | 2-punch | 3,336 | 13,455 | 0.29, 0.38, 0.23, 0.24, 0.38 | 0.30 (21) |
| 30 Random Newborns | 2-punch | 0.010 – 0.040 | |||
| Galactosemia #1 | 2-punch | 0.063, 0.041, 0.062 | 0.055 (18) | ||
| Galactosemia #2 | 2-punch | 0.12 | |||
| Galactosemia #3 | 2-punch | 0.097, 0.059, 0.11 | 0.087 (23) | ||
| Biotinidase deficient #1 | 2-punch | 0.0076 |
Ion counts for a typical run are shown.
Results from multiple punches from the same DBS are shown.
Table 3.
LC-MS/MS assay. GALT analysis
| DBS Sample | Method | Product (ion counts)1 | Internal Standard (ion counts)1 | Enzymatic Activity (μM/hr/L)2 | Average (% CV) |
|---|---|---|---|---|---|
| filter paper (no blood) | 3-punch | 6,155 | 102,845 | 0.70, 0.41, 0.80 | 0.6 (31) |
| normal adult | 3-punch | 1,030,366 | 34,864 | 346, 279, 285, 286 | 292 (10) |
| normal adult + 0.5 μM C26-LPC | 3-punch | 917,160 | 29,142 | 323, 369, 362, 353, 345 | 350 (4) |
| CDC GALT low | 3-punch | 1,745 | 53,037 | 0.4, 0.4, 1.8 | 0.8 (77) |
| CDC GALT medium | 3-punch | 390,434 | 48,674 | 90, 94, 95 | 93 (2) |
| CDC GALT high | 3-punch | 1,167,212 | 37,649 | 309, 300, 313 | 307 (2) |
| 30 Random Newborns | 3-punch | 92 – 300 | |||
| Galactosemia #1 | 3-punch | 3.0, 2.4, 2.7 | 2.7 (10) | ||
| Galactosemia #2 | 3-punch | 5.6 | |||
| Galactosemia #3 | 3-punch | 16.1, 9.3, 14.3 | 12(22) | ||
| Biotinidase deficient #1 | 3-punch | 112 | |||
| filter paper (no blood) | 2-punch | 3,616 | 69,122 | 0.4, 1.3, 1.2, 1.2, 1.1 | 1.0 (34) |
| normal adult | 2-punch | 404,961 | 39,386 | 241, 229, 201, 200, 187 | 212 (9) |
| normal adult + 0.5 μM C26-LPC | 2-punch | 383,507 | 39,122 | 230, 296, 206, 214, 275 | 244 (14) |
| CDC GALT low | 2-punch | 7,442 | 72,162 | 2.4, 2.6, 4 | 3 (23) |
| CDC GALT medium | 2-punch | 184,200 | 65,219 | 66, 63, 65 | 65 (2) |
| CDC GALT high | 2-punch | 523,826 | 64,508 | 190, 200, 238 | 210 (10) |
| 30 Random Newborns | 2-punch | 62 – 257 | |||
| Galactosemia #1 | 2-punch | 4.5, 3.1, 2.8 | 3.5 (21) | ||
| Galactosemia #2 | 2-punch | 4.1 | |||
| Galactosemia #3 | 2-punch | 4.6, 4.2, 4.6 | 4.4 (4) | ||
| Biotinidase deficient #1 | 2-punch | 65 |
Ion counts for a typical run are shown.
Results from multiple punches from the same DBS are shown.
Table 4.
5-Plex LC-MS/MS assay. MPS-I analysis.
| DBS Sample | Method | Product (ion counts)1 | Internal Standard (ion counts)1 | Enzymatic Activity (μM/hr/L)2 | Average (% CV) |
|---|---|---|---|---|---|
| filter paper (no blood) | 3-punch | 5,218 | 242,571 | 0.019, 0.013, 0.009, 0.013, 0.022 | 0.016 (31) |
| normal adult | 3-punch | 67,140 | 134,412 | 0.44, 0.44, 0.50, 0.45, 0.41 | 0.45 (6) |
| normal adult + 0.5 μM C26-LPC | 3-punch | 66,511 | 113,734 | 0.51, 0.49, 0.47, 0.50, 0.57 | 0.51 (6) |
| filter paper (no blood) | 2-punch | 1,755 | 185,908 | 0.017, 0.015, 0.016, 0.014, 0.033 | 0.019 (38) |
| normal adult | 2-punch | 23,827 | 81,872 | 0.51, 0.45, 0.40, 0.46, 0.47 | 0.46 (8) |
| normal adult + 0.5 μM C26-LPC | 2-punch | 18,759 | 66,009 | 0.50, 0.56, 0.57, 0.58, 0.66 | 0.57 (9) |
| 30 Random Newborns | 2-punch | 0.19 – 1.0 | |||
| Galactosemia #1 | 2-punch | 0.21, 0.20, 0.22 | 0.21 (5) | ||
| Galactosemia #2 | 2-punch | 0.31 | |||
| Galactosemia #3 | 2-punch | 0.52, 0.46, 0.52, 0.50 | 0.50 (6) | ||
| Biotinidase deficient #1 | 2-punch | 0.57 |
Ion counts for a typical run are shown.
Results from multiple punches from the same DBS are shown.
Table 6.
Fluorescence assay of GALT.
| DBS Sample | Method | Fluorimeter Reading | Average (% CV) | GALT Activity (U/g Hb) |
|---|---|---|---|---|
| blank (normal adult) | 3-punch | 283,310,276,281,302 | 290 (4) | |
| blank (CDC GALT low) | 3-punch | 292, 293 | ||
| blank (CDC GALT medium) | 3-punch | 370, 307 | ||
| blank (CDC GALT high) | 3-punch | 304, 269 | ||
| blank (Galactosemia #1) | 3-punch | 381, 351 | ||
| filter paper | 3-punch | 198,187,187,201,220 | 200 (7) | |
| normal adult | 3-punch | 10,430, 8,734, 5,920, 8,573,8,247 | 8381 (17) | 21, 18, 12, 17, 17 |
| normal adult + 0.5 μM C26-LPC | 3-punch | 7,194, 11,080, 9,943, 10,456, 9550 | 9644 (14) | 15, 22, 20, 21, 19 |
| CDC GALT low | 3-punch | 374, 358, 447 | 393 (10) | 1.4 |
| CDC GALT medium | 3-punch | 805, 1,277, 1,429 | 1170 (23) | 3.8 |
| CDC GALT high | 3-punch | 4,704, 3,355, 4,900 | 4319 (16) | 9.1 |
| 30 Random Newborns | 3-punch | 2,965 – 7,593 | 5.3 – 12.0 | |
| Galactosemia #1 | 3-punch | 605, 545, 590 | 579 (5) | 2.0, 1.9, 2.0 |
| Galactosemia #2 | 3-punch | 573 | 2.0 | |
| Galactosemia #3 | 3-punch | 719, 711, 736 | 722 (1) | 2.2, 2.2, 2.2 |
| Biotinidase deficient #1 | 3-punch | 2,306 | 4.4 | |
| blank (normal adult) | 2-punch | 284,352,243,309,334 | 304 (13) | |
| blank (CDC GALT low) | 2-punch | 278, 322 | 300 | |
| blank (CDC GALT medium) | 2-punch | 362, 309 | 335 | |
| blank (CDC GALT high) | 2-punch | 317, 413 | 365 | |
| blank (Galactosemia #1) | 2-punch | 385, 566 | 475 | |
| filter paper | 2-punch | 255,261,211,239,206 | 234 (10) | |
| normal adult | 2-punch | 3,252, 2,897, 2,446, 2,596, 1,863 | 2,611 (18) | 19, 17, 14, 15, 11 |
| normal adult + 0.5 μM C26-LPC | 2-punch | 2,773, 3,666, 3,240, 2,689, 3,393 | 3152 (12) | 16, 22, 19, 16, 20 |
| CDC GALT low | 2-punch | 354, 362, 357 | 358 (1) | 1.4 |
| CDC GALT medium | 2-punch | 786, 701, 797 | 761 (6) | 3.8 |
| CDC GALT high | 2-punch | 1,430, 1,580, 1,806 | 1605 (2) | 9.1 |
| 30 Random Newborns | 2-punch | 1,338 – 3,293 | 6.0 – 15.0 | |
| Galactosemia #1 | 2-punch | 464, 438, 443 | 448 (2) | 1.9, 1.8, 1.8 |
| Galactosemia #2 | 2-punch | 487 | 2.1 | |
| Galactosemia #3 | 2-punch | 516, 547, 629 | 564 (8) | 2.2, 2.3, 2.7 |
| Biotinidase deficient #1 | 2-punch | 1,209 | 5.5 |
Figure 5.
LC-MS/MS GALT assay using CDC GALT Low, Medium, and High DBS standards and the 3-punch method (filled circles) or the 2-punch method (open circles). The Y-axis is the activity measured by LC-MS/MS, and the X-axis are the activities for the quality control standards determined by the CDC (https://www.cdc.gov/labstandards/pdf/nsqap/GALT_Certification_Set_2_2017.pdf).
Figure 6.
Same as Figure 5 but using the Fluorimetric GALT assay.
Also given in the data tables is the 5-plex data for DBS from 30 random newborns as well as three newborns previously diagnosed with classical galactosemia (GALT deficient) and one newborn previously diagnosed with biotinidase deficiency. The patients showed a clear deficit in the level of the relevant enzyme. We did not submit DBS from patients confirmed to have MPS-I or Pompe disease to the 5-plex study since we have extensively shown that such DBS give vanishingly low enzymatic activities using the same substrates [4,7], and these DBS are in short supply. For X-ALD, we used DBS from non-affected donors as well as DBS made from blood spiked with levels of C26-LPC found in typical X-ALD patients [6]. The C26-LPC method used in our 5-plex is essentially identical to the previous method [6], which has been validated with affected-patient DBS [6].
Discussion
The new multiplex reported in this study may be useful for newborn screening for galactosemia, biotinidase deficiency, and X-ALD. Further studies are needed to validate the assay in a newborn screening laboratory. One advantage of the new multiplex is that it allows assays to be consolidated with or without addition of lysosomal storage diseases thus reducing overall costs. The lack of interference from glucose-6-phosphate dehydrogenase deficiency is an important factor in regions where there are a relatively high incidence of this deficiency.
In our previous MS/MS assays of lysosomal enzymes we used an overnight incubation period, whereas in the current study we incubated for 3 hrs [3,7]. The overnight period is needed only if the Krabbe disease enzyme is included since this enzyme has a relatively low specific activity. If Krabbe disease is omitted, a short incubation period of 3 hr is more than sufficient.
The new multiplex assay is expected to be easily incorporated into previously developed MS/MS assays. For example, Supplemental Figure 2 shows an LC-MS/MS trace where X-ALD, biotinidase and GALT are multiplexed with 6 lysosomal enzymes (relevant to Gaucher, Fabry, Krabbe, and Niemann-Pick-A/B). This 9-plex requires ultra-high pressure LC (UPLC), and can be done on most modern LC systems (for example on the Waters Aquity system). More detailed studies are underway of this 9-plex to expand this assay to other lysosomal enzymes (for example, [8]) as well as biomarkers that cannot be analyzed in the standard MS/MS protocols used in newborn screening laboratories to analyze amino acids and acyl-carnitines.
Supplementary Material
Acknowledgments
We are grateful to Dr. Martin Sadilek for assistance with MS/MS experiments and to Prof. S. Withers (Univ of British Columbia) for providing the LgtC plasmid. This work was supported by a grant from the National Institutes of Health (DK67859).
Abbreviations
- C26-LPC
1-hexacosanoyl-2-hydroxy-sn-glycero-3-phosphocholine
- DBS
dried blood spot on a newborn screening card
- GALT
galactose-1-phosphate uridyltransferase
- LC-MS/MS
liquid chromatography-tandem mass spectrometry
- MPS-I
mucopolysaccharidosis-I
- UDP-galactose
uridine diphosphate-galactose
- UDP-glucose
uridine diphosphate-glucose
- X-ALD
X-linked adrenoleukodystrophy
Footnotes
Author contributions
M. H. Gelb conceived of the project and director the laboratory where the studies were carried out.
X. Hong did most of the assay studies in this paper.
A. B. Kumar synthesized all the new compounds
C.R. Scott helped direct the studies along with Gelb
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
References
- 1.Fujimoto A, Okano Y, Miyagi T, Isshiki G, Oura T. Quantitative Beutler test for newborn mass screening of galactosemia using a fluorometric microplate reader. Clin Chem. 2000;46:806–810. [PubMed] [Google Scholar]
- 2.Stuhrman G, Perez Juanazo SJ, Crivelly K, Smith J, Andersson H, Morava E. False-Positive Newborn Screen Using the Beutler Spot Assay for Galactosemia in Glucose-6-Phosphate Dehydrogenase Deficiency. JIMD Rep. 2017;36:1–5. doi: 10.1007/8904_2016_34. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Li Y, Brockmann K, Turecek F, Scott CR, Gelb MH. Tandem mass spectrometry for the direct assay of enzymes in dried blood spots: application to newborn screening for Krabbe disease. Clin Chem. 2004;50:638–640. doi: 10.1373/clinchem.2003.028381. [DOI] [PubMed] [Google Scholar]
- 4.Chennamaneni NK, Kumar AB, Barcenas M, Spacil Z, Scott CR, Turecek F, et al. Improved reagents for newborn screening of mucopolysaccharidosis types I, II, and VI by tandem mass spectrometry. Anal Chem. 2014;86:4508–4514. doi: 10.1021/ac5004135. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Wakarchuk WW, Cunningham A, Watson DC, Young NM. Role of paired basic residues in the expression of active recombinant galactosyltransferases from the bacterial pathogen Neisseria meningitidis. Protein Eng. 1998;11:295–302. doi: 10.1093/protein/11.4.295. [DOI] [PubMed] [Google Scholar]
- 6.Haynes CA, De Jesus VR. Improved analysis of C26:0-lysophosphatidylcholine in dried-blood spots via negative ion mode HPLC-ESI-MS/MS for X-linked adrenoleukodystrophy newborn screening. Clin Chim Acta. 2012;413:1217–1221. doi: 10.1016/j.cca.2012.03.026. [DOI] [PubMed] [Google Scholar]
- 7.Li Y. Direct Multiplex Assay of Lysosomal Enzymes in Dried Blood Spots for Newborn Screening. Clin Chem. 2004;50:1785–1796. doi: 10.1373/clinchem.2004.035907. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Liu Y, Yi F, Kumar AB, Kumar Chennamaneni N, Hong X, Scott CR, et al. Multiplex Tandem Mass Spectrometry Enzymatic Activity Assay for Newborn Screening of the Mucopolysaccharidoses and Type 2 Neuronal Ceroid Lipofuscinosis. Clin Chem. 2017;63:1118–1126. doi: 10.1373/clinchem.2016.269167. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.