LC/MS Method for Analysis of Guanine Deaminase: Activity, Kinetics and Effects of Inhibitors Justin Godinho1, Ben Libert1, Chuping Luo1 and Barry Boyes1,2 1Advanced Materials Technology Inc., Wilmington, DE; 2Complex Carbohydrate Research Center, University of Georgia, Athens, GA Presented at ASMS 2018 Poster MP437 Introduction Efficient Separation of Highly Polar Purine Metabolites using Kinetic Analysis of Recombinant and Bovine Brain Guanine Deaminase Guanine deaminase, also known as “nedasin “ or “cypin”, catalyzes the purine catabolic commitment Standard Microplate Enzyme Assay and LC/MS Conditions step from guanine (G), through xanthine (X), to the elimination product, uric acid. In rabbit and human, Superficially Porous Particle HPLC Columns Shimadzu Nexera G u a n i n e D e a m i n a s e , B o v i n e B r a i n the enzyme appears to exist predominantly cytoplasmic as a homodimer, with catalytic domains for HALO 90Å AQ-C18, 2.7µm, 2.1x75mm G u a n i n e D e a m i n a s e , B o v i n e B r a i n 8 - A z a G u a n i n e A=0.1%Formic Acid +2 uV(x100,000) 6 0 8 0

the Zn -dependent hydrolytic deamination of guanine to xanthine plus ammonia. Genomic details for 1.75 B=Acetonitrile

)

) U HALO 90Å, AQ-C18, 2.7 µm U

the GDA gene are well mapped, and expression profiling in certain tissues and organisms has been A selection of C18 phases stable in 0.5ml/min, 35°C, 265nm, MTP, autosampler 25°C m

1.50 m (

( 6 0

initiated, although the complement of various transcript variants is incomplete. 9 100% Aqueous conditions were Gradient: 0%B 0-1.5min to 70%B @1.7-2.2min to 0%B @2.3-4.2min

y

y t

4 0 t i

1.25 i v

screened. The HALO 90Å, AQ-C18, Thermo Orbitrap Velos Pro ETD v i

3 5 i t

• Structure: c. 50 kDa subunits with sequences that vary at internal and terminal sites, due to exon t c

2.7 µm column was chosen for its MS Run Time (min): 2.20; divert first 0.8 min from MS source c 4 0 A

1.00 Best Fit Parameters A Best Fit Parameters

selection; at least 4 significant forms predicting proteins of various lengths are known. e 4 ITMS (-) 0.7 to 1.7min; ITMS (+) 1.7 to 2.0min; ITMS (-) 2.0 to 2.2 min e

1,2 ability to resolve the highly polar 2 0 Vmax: 47.1 ± 0.6 Vmax: 146 ± 18 m

• Interactions: tubulin, snapin, and the post-synaptic domain protein 95 (PSD-95). 0.75 HESI Source Type; Capillary 350°C; Heater 325°C; Sheath Gas 40; Aux Gas Flow 10 m y

y 2 0 z

6 7 8 purine metabolites and inhibitors Km: 0.307 ± 0.021 z Km: 155 ± 27 n

• PSD-95 binding is through the PDZ binding motifs present at the C-terminus of the dimeric structure. ITMS Full AGC 50K; ITMS SIM AGC 100K; ITMS MSn AGC 50K 2 n 2 E 0.50 R : 0.980; 34 d.f. E R : 0.985; 34 d.f. • Sequence variants occur mostly at the protein binding domains, although minor variants lack the of interest. Under the current use Source Voltage (-kV) 2.70; Ion Trap Full Max Ion Time (ms) 200; 0.25 conditions, the AQ-C18 column is Segment (SEG) Information/Scan Event Details: 0 0 deaminase catalytic site. 0 5 0 1 0 0 0 5 0 1 0 0 ITMS - c low injrf=70.0 ·(151.00000)->oS(106.0-110.0) 0.00 stable for many thousands of [ G ]  M [ A z G ]  M • In mammalian brain, high enzymatic levels are in telencephalic brain regions; very low levels in MS/MS: AT CID CE 30.0% Q 0.350 Time 10.000 IsoW 0.8, CV = 0.0V white matter and cerebellum; moderate levels in liver and certain other organs; low levels in 0.25 0.50 0.75 1.00 1.25 1.50 1.75 min samples for at least 4 weeks of use. uV (x100,000) Assay Conditions The rate analyses for partially purified Bovine brain GDA, and bacterially expressed human recombinant GDA, both exhibit plasma/serum, notably altered by liver dysfunction. 2.5 Competitor (A) 2.6 µm SPP Aqueous C18 LC Conditions 50 mM Bicine pH 7.8, 10 µg/mL BSA, varying Guanine or 8-azaGuanine. G u a n i n e D e a m i n a s e , r e c H u m a n Reaction at room temperature with agitation for >15 minutes, typical volume of 100 µL Michaelis-Menten kinetics with high confidence. Each • 1 5 0 0 Actual role of guanine deaminase in specialized organ metabolism and synaptic physiology is Columns: 2.1 x 100 mm 2.0 Stop reaction by addition of 0.1 volume of 10% Formic Acid. ) preparation exhibits Vmax consistent with a semi-purified uncertain, and relatively little is known about the enzyme characteristics outside of rabbit, with U Mobile Phase: 0.1% Formic Acid enzyme preparation, and a highly purified recombinant. The sub-

All purine reagent stocks are maintained in 15 mM NaOH/0.15 mM MgCl2 until dilution. m ( significant detail available detail on gene expression patterns, but not at the protein level. Available 4 9 Enzyme incubations are conducted in standard 96 well reaction plates, which are directly loaded in y µM Km of the bovine enzyme is lower than expected, potentially 1.5 Flow Rate: 0.5 mL/min t 1 0 0 0

1,2 3 5 i v

protein expression surveys have suggested extensive post-translational modifications, that have not the Nexera autosampler, for LC/MS analysis of reaction products. i reflecting the particular buffer system and analysis method Temperature: 35 °C t c employed. We note 2-4 fold lower Vmax values observed in yet been detailed with certainty. 1.0 A Best Fit Parameters Detection: UV 254 nm 6 7 8 50 50 e phosphate buffers. The comparably high Km observed for • 5 0 0 Vmax: 1019 ± 18 Few useful inhibitors have been described for the enzyme. m

Injection: 1 µL y Km: 17.0 ± 1.1 recombinant GDA is in the range of values reported by others, 0.5 z Instrument: Shimadzu Nexera 40 n 2 and may reflect sequence effects in the fusion polypeptide, near 40 E R : 0.991; 34 d.f.

0.0 0 the subunit interaction surfaces, or could suggest structural 30 0 5 0 1 0 0 1 5 0 2 0 0 features, or an activiation state of brain enzyme that is not yet 0.00 0.25 0.50 0.75 1.00 1.25 1.50 1.75 min 30 [ G ]  M known. AzG exhibits a comparably higher Vmax, but with a much uV (x100,000) Peak Identities 20 20 lower Km than for deamination of G to X. Competitor (B) 2.7 µm SPP Aqueous C18 1: 5-Amino-imidazole-4-carboxamide 1.50 9 2: Azepinomycin 10 10

Purpose 1.25 Analysis of Enzyme Inhibitors pmol pmol X of Produced 3 5 3: Guanine - Substrate pmol X of Produced • Previous enzyme assays are problematic, using either the subtle spectral shift from G to X in 1.00 4: 2,6-Diaminopurine 0 0 Compounds that are potential inhibitors of guanine deaminase have been investigated using the LC/MS assay absorbance spectroscopy, or coupled enzyme assay with capture of X by Xanthine Oxidase to 0.75 5: Uric Acid 0 500 1000 described, with bovine brain enzyme. As shown for the example of the known competitor inhibitor, 5-amino- 1,2 4 6 7 8 0 100 200 300 generate H2O2, with a fluorescent dye readout of oxidation products. 6: Xanthine - Product -6 imidazole-4-carboximide (AICA), inhibition is concentration dependent, and exhibits typical enzyme kinetic 0.50 Enzyme Dilution (x10 ) Time (min) • Spectrophotometric high throughput assay uses ammonium liberation, but has uncertain specificity 7: 8-Azaxanthine response (apparent Km increase, with no change in Vmax). The effective Ki values for three structurally-related and requires guanine as substrate near the limit of aqueous solution solubility. Although useful for 0.25 8: 8-AzaGuanine Enzyme activity is linear over 4 orders of dilution, in this example, an enzyme concentration of 1 in purinergic compounds are shown. Azepinomycin is confirmed to be the most effective inhibitor known to date. 12,000 (50 mU/rx), X production is linear with incubation time to 300 minutes, consuming less than screening purposes, lacks specificity and dynamic range. Km or Ki ± SEM (µM) 0.00 9: Allopurinol 10% of substrate at 100 uM G. Observed for 10 uL injection; LOQ for X =20 fmol; LOD for X = 5 fmol; C o m p e t i t i v e I n h i b i t o r A n a l y s i s

To measure tissue enzyme levels, follow purification processes, and assay enzyme kinetics and effects 0.00 0.25 0.50 0.75 1.00 1.25 1.50 1.75 min Linear Range for X is 20 fmol to >20 pmol injected. A I C A Guanine 0.31 ± 0.021 y

t 6 0 i

of inhibitors, highly specific LC/UV and LC/MS methods would be useful. Our assay measures direct v 0

i t

Selective Detection of Enzyme Reaction Products using SRM in the Ion Trap c 1  M A generation of product xanthine from the deamination of guanine, using rapid separation of these e 5  M s 4 0 8-AzaGuanine 155 ± 27 highly polar metabolites by a new reversed-phase material, HALO AQ-C18. To obtain high sensitivity a

ITMS [50-200 m/z] Product Ion 108 m/z n

Precursor Ion 151 m/z ITMS [106.00-110.00 m/z] i 2 0 M 28.0 120  260000 % 26000 m and selective detection, LC-MS/MS was employed in the SRM mode. The assay performance was XIC (-) 10pmol Xanthine inj. 25.0 151.01 XIC (-) %SRM MS^2 108.01 a 100 Panels A-D represent steps taken for selective detection e 150-152 m/z D 2 0 assessed across a range of substrate concentrations, and in the presence of known and potential 107.8- 151.00 200000 20.0 20000 e 80 of the enzyme reaction product, xanthine and 8- n AICA 4.1 ± 0.24 108.2 m/z @cid30 i competitive inhibitors. We have further compared the features of the LC-MS assay using two presently n 150000 15000 60 a 15.0 azaxanthine. Panel A shows a chromatogram of a 10 pmol u 0 known substrates for guanine deaminase, Guanine and 8-AzaGuanine. We compared kinetic properties G 100000 10.0 10000 40 injection of X or AzX using the extracted ions from 150 to 0 2 0 4 0 6 0 of a commercially available recombinant enzyme to one substantially purified from bovine brain. 20 [ G ]  M Azepinomycin 0.11 ± 0.043 50000 5.0 5000 152 m/z. Panel B shows the spectra with xanthine and 0 azaxanthine precursor ions [M−H]− at 151 m/z (X) or 152 0 0.0 m/z 0 m/z Substrates for Guanine Deaminase 0.875 1.000 1.125 1.250 120.0 140.0 160.0 180.0 0.875 1.000 1.125 1.250 55.0 75.0 100.0 125.0 150.0 (AzX) chosen for fragmentation. Panel C shows the Conclusions ITMS [50-200 m/z] Precursor Ion 152 m/z ITMS [106.00-110.00 m/z] Product Ion 109 m/z extracted ion chromatograms centered at dominant 200000 28.0 40000 120 XIC (-) 10pmol 8-Azaxanthine inj % XIC (-) % 109.08 fragment ions, and Panel D shows the product ions 108 A highly selective and sensitive LC/MS method is demonstrated for the analysis of the 25.0 100 151- 108.8- SRM MS^2 150000 30000 m/z monitored for X and 109 m/z for AzX. enzyme GDA. The HALO 90Å, AQ-C18, 2.7 µm column was chosen for its ability to resolve the 153 m/z 20.0 109.2 m/z 80 152.00 151.99 @cid30 very polar purine metabolites and potential inhibitors of interest. The column’s robust 15.0 60 100000 20000 stability in 100% aqueous conditions allowed for high throughput assays and the assessment 10.0 40 Xanthine Theoretical Mass [m/z] 10000 124.01 of hundreds of different measurements. Studying enzyme kinetics with highly selective 50000 20 152.033 Da monoisotopic, [151.025 m/z (-) ] 5.0 LC/MS methods avoid many of the limitations of previous methods including lack of 0 0 8-azaxanthine Theoretical Mass [m/z] 0 0.0 m/z m/z sensitivity in complex samples. Selection of the enzyme reaction product through SRM in the 153.029 Da monoisotopic, [152.021 m/z (-)] 0.875 1.000 1.125 1.250 120.0 140.0 160.0 180.0 0.875 1.000 1.125 1.250 55.0 75.0 100.0 125.0 150.0 ion trap allows for specific detection of the product without interference due to chemically A B C D similar potential inhibitors.