CSF and Tissue Samples: Human brain and serum/plasma samples were obtained from Maine Medical

Center Research Institute BioBank (Scarborough, ME). Human CSF samples were obtained from Johns

Hopkins School of Medicine, Alzheimer’s Disease Research Center (Baltimore, MD).

Preparation of Protein Lysates and Digested Peptides: Tissue was pulverized under liquid nitrogen using a

Bessman press, and transferred into Urea Lysis Buffer (ULB, 8 M Urea, 20 mM HEPES pH 8.0, 1 mM β-

glycerophosphate, 1 mM sodium vanadate, 2.5 mM sodium pyrophosphate). The tissue slurry was

homogenized using a mini-beadbeater and sonicated 3 times for 20 s each at 15 W output power with a 1 min

cooling on ice between each burst. Lysates were centrifuged 15 min at 4 °C at 20,000× g to remove insoluble

debris. CSF was added to dry aliquots of RapiGest (Waters) and ammonium bicarbonate (pH8.0) to a final

concentration of 0.1% and 50 mM, respectively. Soluble protein was reduced with 4.5 mM DTT (30 min, 40

°C) and alkylated with 10 mM iodoacetamide (15 min, RT) in the dark. Samples were diluted 1:4 with 200

mM ammonium bicarbonate (pH 8.0) and digested overnight with trypsin-TPCK (1:75, w:w, Promega) in 1

mM HCl. Other protease digestions were performed using the manufacturer’s recommended protocol (LysC,

AspN, GluC, ArgC, Promega and NEB). Digested peptide lysates were desalted over 360 mg SEP PAK

Classic C18 columns (Waters, Richmond, VA, USA, #WAT051910). Peptides were eluted with 50%

acetonitrile in 0.1% TFA, dried under lyophilization conditions, and stored at −80 °C in 0.1 – 1.0 mg

aliquots. TAU and H3K9acetyl BAMS assays were performed using 100 g of protein from human brain.

Beta-amyloid BAMS assay was performed using 100 L of undigested CSF.

BAMS Assay - Bead Preparation: Each antibody was conjugated to NHS-activated magnetic agarose beads

in a 10 μL slurry (7 g of antibody/100 beads, 375 - 420 micron diameter) in PBS buffer, for 3-12 h (4 °C),

and quenched with Tris HCl (100 mM, pH 8.0) for 1 h at RT. Unbound antibody was removed with three 400

μL washes of cold PBS (2 min, 4 °C). Beads were stored in PBS and 0.02% sodium azide (4 °C).

BAMS Assay – Target Peptide Binding: Target peptide enrichment was performed using 10 – 1000 µg of

digested peptides (or soluble protein) with typically 3 replicate beads/target. Multiplex peptide enrichment

was performed using 10 – 1000 µg of purified peptides (or soluble protein) with typically 3 replicate

beads/target in a volume of 50 – 200 µL. Peptides and affinity capture beads were incubated in binding

buffer (1M KCl, 100 mM Tris HCl in deionized water, pH 8.0) for a period of between 3 – 12 hrs at 4 °C in

Thermomixer (Eppendorf). Beads were washed sequentially in PBS (700 uL), ammonium bicarbonate (700

uL, 10 mM, pH 8.0) and deionized water (700 uL) to remove any nonspecific bound peptides (2 min, 4 °C).

BAMS Assay – Target Peptide Elution: Washed BAMS beads are transferred to the hydrated wells to settle

into the micro-wells of the BAMS plate assembly with gentle agitation and centrifugation (5 min, 200 x g).

After centrifugation, the sample chamber gasket is removed, leaving the micro-well gasket fixed in place on

the slide. The bead array is exposed to an aerosol of elution buffer using a Matrix Sprayer, containing 0.5

mg/mL α-cyano-4-hydroxycinnamic acid (CHCA) in 50% acetonitrile and 0.4% trifluoroacetic acid (TFA)

for approximately 15 min. Once the matrix is dry, the silicone gasket is lifted off the slide and any remaining

dry agarose beads are removed by compressed air leaving an array of spots containing purified and

concentrated target peptides for subsequent MALDI MS measurement.

MALDI TOF Linear Instrument Settings: Matrix Assisted Laser Desorption Ionization Time of Flight

(MALDI-TOF) MS data was acquired on Bruker Daltonics (Billerica MA) Autoflex Speed or rapifleX

MALDI-TOF/TOF mass spectrometer using FlexControl software. Unless otherwise indicated, the autoflex

speed acquisition conditions in the positive linear mode were, 750-7000 m/z mass range (2 kHz, 10000

spectra/spot using the random walk method). The voltage settings were 19.50 kV (ion source 1), 18.35 kV

(ion source 2) and 6.0 kV (lens). The pulsed ion extraction was 130 ns. The detector gain voltage was 4.0X

or 2910 V. The acquisition settings on the rapifleX for automated runs in positive linear mode were 700-7000

m/z mass range, 10 kHz laser repetition rate, 4000 laser shots.

MALDI TOF Analysis: Mass spectra were processed and analyzed using FlexAnalysis software. Peaks

were detected that had a signal-to-noise ratio of at least 3. In some cases, baseline subtraction procedure was

applied to individual mass spectra. Unless otherwise indicated, peaks in the mass spectra are labeled using

either average or monoisotopic m/z values using the peak picking algorithms available in the software.

mMass was utilized for data review.

INTRODUCTION METHODS Proteomic studies critically rely on multi-dimensional separation methods such as 2D-LC to simplify

the complexity of protease digested biological specimens before subsequent MS & MS/MS to

efficiently monitor multiple proteins of interest. The time and expertise required to implement LCMS

methods can often be a barrier to using targeted proteomic applications in translational research. Here

we present further development of a microarray analytical platform called Bead Assisted Mass

Spectrometry (BAMS), which integrates multiplex immuno-affinity capture with MALDI MS to create

customized targeted proteomic assays for translational biology research. We demonstrate efficient

monitoring of several core Alzheimer’s disease pathological biomarkers including tau and amyloid beta

peptides with a sensitivity greater than that of LCMS as well as histone epigenetic modifications.

METHODS

RESULTS

CONCLUSIONS • BAMS assays have been configured to enable multiplexed MALDI MS monitoring of both

unmodified and phosphorylated sites within TAU in human brain & CSF samples.

• A single on-bead, multiplex BAMS assay has been configured monitor up to 18 Beta-Amyloid

fragments in undigested CSF.

• BAMS assays have been configured for epigenetic applications to monitor Histone H3 K9-

acetylation and associated combinatorial PTMs.

• Data processing of BAMS assays is facilitated using software that performs peak detection,

clustering, and export of spectral features for library matching and statistical analysis.

• BAMS assays can be utilized to perform a wide range of targeted proteomic applications and

is amenable for high-throughput screening.

1 G.M.Hamza, V.B.Bergo, S.Mamaev, D.M.Wojchoswki, P.Toran, C.R.Worsfold, M.P.Castaldi & J.C.Silva, International Journal of

Molecular Sciences, (2020) 21(6):1-36 (https://www.mdpi.com/1422-0067/21/6/2016). 2 US patent 9,618,520 by inventor V.Bergo, titled Devices and methods for producing and analyzing microarrays 3 US patent 10,101,336 by inventor V.Bergo, titled Eluting analytes from bead arrays 4 US patent application 16/125164 by inventor V.Bergo, titled Multiplexed bead arrays for proteomics

ACKNOWLEDGEMENTS • North Shore InnoVentures (https://nsiv.org/) and its corporate sponsor sand the

Massachusetts Life Science Center (http://www.masslifesciences.com/) for grant support.

Figure 1. BAMS Workflow for Targeted Proteomics. Standard bottom-up methods are used to

generate peptides for targeted monitoring of proteins using BAMS (lysis, reduction, alkylation,

digestion). Digested peptides are incubated overnight with BAMS affinity capture beads in Eppendorf

tube or 96-well plate. For middle-down applications, such as CSF, beads are incubated directly for

peptide binding (1). Magnetic agarose beads are transferred, sequentially into wash buffers (PBS,

ammonium bicarbonate, DDW) before placed onto the BAMS slide (2). Washed beads are transferred

onto the BAMS slide to settle beads into pico-wells (3), captured peptides from each BAMS bead are

eluted into the pico-well using a matrix sprayer (4), eluted, dry peptides are analyzed after

disassembling gaskets and placing into slide adapter for MALDI MS measurement (5).

RESULTS

Vladislav B. Bergo1, Ghaith M. Hamza2,3, Sergei Dikler4, Abhay Moghekar5, Manor Askenazi1,6, Don M. Wojchowski3, Sergey Mamaev1, Camilla Worsfold1, Jeffrey C. Silva1

1Adeptrix Corporation, Beverly MA 01915; 2AstraZeneca, Boston MA 02451; 3University of New Hampshire, Durham NH 03824; 4Bruker Scientific, Billerica MA 01821; 5Johns Hopkins University School of Medicine, Baltimore MD 21287; 6Biomedical Hosting, Arlington MA 02140

Multiplexed Analytical Platform using Affinity Capture and MALDI MS Enables Novel Assay Development for Screening Biomarkers in Neurological Diseases

Figure 2. BAMS Assay Validation. BAMS affinity beads are individually validated by localized

peptide elution and direct, MALDI MS measurement of the target peptide. Targeted MS/MS acquisition

can be performed to verify the sequence of the captured peptide(s). A MALDI MS spectral library is

generated for each protease digestion condition with each Affi-BAMS bead (A). The BAMS assay can

accommodate thousands of target peptides (unmodified & protein PTM) in a single experiment on a

slide for identification and quantification of the target proteins in the configured assay panel (B). Jeffrey C. Silva | Adeptrix Corporation | 100 Cummings Center, Suite 438Q | Beverly, MA 01915 | www.adeptrix.com

jsilva@adeptrix.com

RESULTS

Figure 4. TAU BAMS Assays with LysC Digested Brain Sample. A) N-terminus of Tau (aa1-24) with

loss of methionine and acetylation at the n-terminal alanine, B) N-terminus of Tau (aa45-67), including a

phosphorylated site within the amino acid region 45-67, C) Internal TAU (aa88-130), D) Phosphorylated

TAU (pT181) containing amino acids 175-190. E) Unmodified and phosphorylated TAU within amino

acids 191-224, a region containing multiple sites of phosphorylation (pS198, pS199, pS202 & pT205). F)

Phosphorylated TAU (pT231) with nonphosphorlyated and phosphorylated LysC peptide readout.

Figure 3. Regions of TAU Protein Configured for BAMS Assay. TAU protein sequence (isoform Tau-F)

with yellow highlighted areas corresponding to LysC peptide sequences captured by BAMS assay (A)

shown in Figure 4. List of 8 antibodies configured for BAMS assays for the TAU protein. Alternative

proteases can be used to monitor the designated regions of TAU as specified by the antibody. Validated

captured peptide sequences are shown for LysC, Trypsin and ArgC protease digestion conditions (B).

Alternative digestions conditions can be used to monitor areas of the protein outside the epitope region if

the resulting proteolytic peptide preserves the epitope sequence for efficient antibody binding.

Figure 5. TAU BAMS Assays with Normal (BLACK) and Disease (RED) Brain. A) Mono- and Di-

phosphorylated TAU (singly pT231 and doubly pT231 & pS235 phosphorylated), Normalized to pT231,

B) Phosphorylated TAU containing amino acids 191-224 containing multiple sites of phosphorylation

(pS198, pS199, pS202 & pT205), normalized to singly phosphorylated pS199.

Figure 6. Beta-Amyloid BAMS Assay with Undigested Normal (RED) and Disease (BLACK)

CSF. Overlaid MALDI MS spectrum from an on-bead multiplexing BAMS assay targeting the specific

region of beta amyloid spanning from aa672-aa713 from triplicate bead assays of four patients (2

normal & 2 disease). The multiplex assay retrieves 19 specific beta-amyloid peptides between aa672-

aa713. Normalizing MS spectra to the Aβ1-40 peptide (BLUE BOX), clear separation of NORMAL

and DISEASE patient samples is seen for specific beta-amyloid peptides (RED BOX).

Figure 7. Histone H3 K9acetyl BAMS Assay with Normal (BLACK) and Disease (RED) Brain.

MS spectra from the H3K9acetyl BAMS assay of LysC digested normal and disease brain (TOP,

PANELS A-E). Captured targets include H3K9-acetylated peptides within aa1-23 of the H3 tail.

Absence of mono-, di- and tri-methylated H3K9acetyl peptides (2041, 2395 & 2625) are observed in

diseased brain (D & E). MS spectra from H3K9acetyl BAMS assays with equal quantities (100 g)

ArgC digested normal and disease brain (BOTTOM, PANELS F-I). Captured targets include H3 K9-

acetylated peptides within aa3-21 of the H3 tail, including combinatorial H3 tail peptides with

acetylation, methylation and deimination of arginine (from conversion of arginine to citrulline), all

containing lysine acetylation at K9 as directed by the site specific affinity capture of the BAMS assay.

a) trypsin

b) chymotrypsin

c) others…

B A

1. Enrich

2. Wash

3. Assemble Bead Array 4. Elute Target Peptides 5. Scan BAMS Plate

As little as 10 g of total soluble

protein

Overlay of 28-plex

BAMS assay

Individual MALDI MS from

replicate BAMS assays

K.VAVVRtPPK.S MH+ (Calc, mono) = 1046.58

K.VAVVRtPPKsPSSAK.S MH+ (Calc, mono) = 1683.82

pT231

pT231 & pS235

K.SGDRSGYssPGsPGtPGSRSRTPSLPTPPTREPK.K MH+ (Calc, mono) = 3533.68

1P

1P

K

80 80 80 2P 2P 3P

2P 3P

195

5.1

9

432

9.9

1

182

7.0

6

206

8.3

3

156

1.9

2

119

6.6

5

413

1.6

0

169

8.9

8

231

4.5

6

326

2.4

5

132

5.7

0

423

0.7

6

378

7.0

5

367

3.8

9

451

3.7

7

216

7.0

5

246

1.5

9

313

3.9

1

0

2

4

6

6x10

Inte

ns. [a

.u.]

1000 1500 2000 2500 3000 3500 4000 4500m/z

MH+ = 1196.50, M.DAEFRHDSGY.E, aa672-681, Aβ1-10

MH+ = 1325.54, M.DAEFRHDSGYE.V, aa672-682, Aβ1-11

MH+ = 1561.67, M.DAEFRHDSGYEVH.H, aa672-684, Aβ1-13

MH+ = 1698.73, M.DAEFRHDSGYEVHH.Q, aa672-685, Aβ1-14

MH+ = 1826.78, M.DAEFRHDSGYEVHHQ.K, aa672-686, Aβ1-15

MH+ = 1954.88, M.DAEFRHDSGYEVHHQK.L, aa672-687, Aβ1-16

MH+ = 2067.96, M.DAEFRHDSGYEVHHQKL.V, aa672-688, Aβ1-17

MH+ = 2167.03, M.DAEFRHDSGYEVHHQKLV.F, aa672-689 Aβ1-18

MH+ = 2314.10, M.DAEFRHDSGYEVHHQKLVF.F, aa672-690, Aβ1-19

MH+ = 2461.17, M.DAEFRHDSGYEVHHQKLVFF.A, aa672-691, Aβ1-20

MH+ = 3133.44, M.DAEFRHDSGYEVHHQKLVFFAEDVGSN.K, aa672-698, Aβ1-27

MH+ = 3261.53, M.DAEFRHDSGYEVHHQKLVFFAEDVGSNK.G, aa672-699, Aβ1-28

MH+ = 3672.78, M.DAEFRHDSGYEVHHQKLVFFAEDVGSNKGAIIG.L, aa672-704, Aβ1-33

MH+ = 3785.87, M.DAEFRHDSGYEVHHQKLVFFAEDVGSNKGAIIGL.M, aa672-705, Aβ1-34

MH+ = 4073.00, M.DAEFRHDSGYEVHHQKLVFFAEDVGSNKGAIIGLMVG., aa672-708, Aβ1-37

MH+ = 4130.02, M.DAEFRHDSGYEVHHQKLVFFAEDVGSNKGAIIGLMVGG.V, aa672-709, Aβ1-38

MH+ = 4229.09, M.DAEFRHDSGYEVHHQKLVFFAEDVGSNKGAIIGLMVGGV.V, aa672-710, Aβ1-39

MH+ = 4328.16, M.DAEFRHDSGYEVHHQKLVFFAEDVGSNKGAIIGLMVGGVV.I, aa672-711, Aβ1-40

MH+ = 4512.28, M.DAEFRHDSGYEVHHQKLVFFAEDVGSNKGAIIGLMVGGVVIA.T, aa672-713, Aβ1-42

Aβ1-10 Aβ1-11 Aβ1-13

Aβ1-16 Aβ1-28 Aβ1-34

Aβ1-34 Aβ1-37 Aβ1-38

Aβ1-39 Aβ1-40 Aβ1-42

267

7.1

75

133

9.9

04

Group-1, 2677.2, N-terminus (aa1-24) 806503/835203

0

1

2

3

4

5

5x10

Inte

ns. [a

.u.]

1000 2000 3000 4000 5000 6000m/z

2390.8

64

2470.7

40

0.0

0.5

1.0

1.5

2.0

2.5

4x10

Inte

ns. [a

.u.]

1000 2000 3000 4000 5000 6000m/z

4401.9

23

2201.5

49

3957.0

10

0.00

0.25

0.50

0.75

1.00

1.25

5x10

Inte

ns. [a

.u.]

1000 2000 3000 4000 5000 6000m/z

115

3.0

79

345

3.9

77

179

1.1

72

358

2.3

39

366

2.4

26

0

1

2

3

4

5

5x10

Inte

ns. [a

.u.]

1000 2000 3000 4000 5000 6000m/z

345

3.9

77

358

2.3

39

353

3.9

37

366

2.4

26

K

79.96 80.09

0

1

2

3

4

5x10

Inte

ns. [a

.u.]

3400 3450 3500 3550 3600 3650 3700m/z

166

7.6

41

0.0

0.5

1.0

1.5

2.0

2.5

4x10

Inte

ns. [a

.u.]

1000 2000 3000 4000 5000 6000m/z

168

3.7

58

104

7.1

29

0

2

4

6

8

4x10

Inte

ns. [a

.u.]

1000 2000 3000 4000 5000 6000m/z

168

3.7

58

176

3.5

42

160

3.7

77

79.98 79.78

0.0

0.2

0.4

0.6

0.8

1.0

5x10

Inte

ns. [a

.u.]

1575 1600 1625 1650 1675 1700 1725 1750 1775 1800m/z

2390.8

64

2470.7

40

79.88

0

1

2

3

4x10

Inte

ns. [a

.u.]

2360 2380 2400 2420 2440 2460 2480m/z

A B C

D E F

MH+ (calc) = 2766.26 -.MAEPRQEFEVMEDHAGTYGLGDRK.D

MH+ (calc) = 2391.03 K.ESPLQTPTEDGSEEPGSETSDAK.S

MH+ (calc) = 4401.09 K.QAAAQPHTEIPEGTTAEEAGIGDTPSLEDEAAGHVTQARMVSK.S

MH+ (calc) = 1667.80 K.TPPAPKtPPSSGEPPK.S

MH+ (calc) = 3581.81

K. SGDRSGYSSPGSPGTPGSRSRTPSLPTPPTREPKK.V

MH+ (calc) = 1683.82

K.VAVVRtPPKsPSSAK.S

Figure 8. Optimized Assay Data Processing. Spectral features are extracted from each MALDI MS

as centroids and clustered into mass bins. A reference library for the specific BAMS assay is then

generated by assigning each peptide identification to the corresponding measured mass and quantitation

is performed using the apex centroid within each clustered centroid group (A-B). Unassigned clustered

centroids can be flagged for further characterization, C).

Figure 9. Hierarchical Clustering of Beta-Amyloid BAMS Assay Features. Hierarchical clustering

using all standardized features (regardless of identification) from the multiplexed BAMS assays yields an

overall grouping by disease status (Normal vs. Disease) with internal grouping by biological replicate (A).

Focusing on features with known assignments to Beta-amyloid fragments further improves the separation of

Normal vs. Disease groups (B).

1570.6

04

2041.1

73

215

4.3

55

1546.5

91

1076.2

62

2395.5

21

1925.0

61

2624.1

63

0

1

2

3

4

5x10

Inte

ns. [a

.u.]

1200 1400 1600 1800 2000 2200 2400 2600m/z

107

6.2

60

0.00

0.25

0.50

0.75

1.00

1.25

4x10

Inte

ns. [a

.u.]

1060 1065 1070 1075 1080 1085 1090 1095m/z

157

0.6

09

0.0

0.5

1.0

1.5

2.0

2.5

5x10

Inte

ns. [a

.u.]

1555 1560 1565 1570 1575 1580 1585 1590m/z

204

1.1

73

215

4.3

55

205

5.1

64

192

5.0

61

206

9.7

77

13.99 14.61

0.0

0.2

0.4

0.6

0.8

1.0

5x10

Inte

ns. [a

.u.]

1925 1950 1975 2000 2025 2050 2075 2100 2125 2150m/z

239

5.5

33

240

9.6

04

242

3.9

45

262

5.2

88

249

7.6

69

263

9.2

81

252

5.3

76

251

1.2

40

265

3.1

56

14.07 14.34 14.1413.8813.99

13.57

0.0

0.5

1.0

1.5

2.0

4x10

Inte

ns. [a

.u.]

2400 2450 2500 2550 2600 2650m/z

MH+ = 1076.2, K.QTARK(ac)STGGK.A, aa5-14

MH+ = 1546.8, -.ARTK(me)QTARK(ac)STGGK.A, aa1-14

MH+ = 1570.8, K.QTARK(ac)STGGK(ac)APRK.Q, aa5-18

MH+ = 1925.2, K.QTARK(ac)STGGK(ac)APRK(ac)QLA.T, aa5-21, TAIL CLIPPING

MH+ = 2041.4, -.ARTK(me)QTARK(ac)STGGK(ac)APRK.Q, aa1-18

MH+ = 2154.5, K.QTARK(ac)STGGK(ac)APRK(ac)QLATK.A, aa5-23

MH+ = 2395.8, -.ARTK(me)QTARK(ac)STGGK(ac)APRK(ac)QLA.T, aa1-21, TAIL CLIPPING

MH+ = 2625.1, , -.ARTK(me)QTARK(ac)STGGK(ac)APRK(ac)QLATK.A, aa1-23

B C

D E

Mono-, Di- & Tri-Methyl

(See Panel D)

Mono-, Di- & Tri-Methyl

(See Panel E)

A

2341.2

20

2397.6

05

2071.2

72

2354.9

62

2298.6

62

2113.4

28

2383.4

45

1913.3

44

2127.4

39

2441.2

27

2411.0

02

1686.0

48

2169.4

74

2368.7

46

1653.7

46

2312.2

77

1881.2

17

2214.4

55

2085.2

76

1927.1

89

1856.7

40

1956.3

48

2484.3

55

1700.1

99

2184.2

30

1630.1

64

2527.7

91

2141.4

96

1611.8

28

2269.9

89

1999.4

37

2570.7

00

2242.7

70

0.0

0.2

0.4

0.6

0.8

1.0

1.2

5x10

Inte

ns. [a

.u.]

1600 1700 1800 1900 2000 2100 2200 2300 2400 2500 2600m/z

2341.2

20

2397.6

05

2071.2

72

2354.9

62

2298.6

62

2113.4

28

2383.4

45

2127.4

39

2441.2

27

2411.0

02

2169.4

74

2368.7

46

2425.5

39

2455.8

57

2155.4

62

2312.2

77

2214.4

55

2085.2

76

2326.0

06

2227.8

85

2484.3

55

2498.7

41

2469.4

17

2184.2

30

2141.4

96

2269.9

89

2255.7

92

2512.1

22

2242.7

70

14.09 14.15 14.27

18.21

17.86

14.12

18.01

17.88

13.84 14.81 14.34

227.30

14.00 14.01 14.06 13.97 14.01 13.74 13.78

227.79

42.16 42.56

0.0

0.5

1.0

1.5

2.0

2.5

5x10

Inte

ns. [a

.u.]

2100 2150 2200 2250 2300 2350 2400 2450 2500m/z

1913.3

44

1899.2

26

1686.0

48

1671.9

57

1653.7

46

1881.2

17

1927.1

89

1856.7

40

1942.0

03

1956.3

48

1839.1

93

1700.1

99

1999.4

37

1714.4

71

1970.2

83

1984.4

01

14.00 14.01 14.06 13.97 14.01 13.74 13.78

227.79

42.16 42.56

17.86

18.21

14.09 14.15 14.27

17.88

18.01

14.12 13.84 14.81 14.34

227.30

0.00

0.25

0.50

0.75

1.00

1.25

1.50

5x10

Inte

ns. [a

.u.]

1650 1700 1750 1800 1850 1900 1950 2000m/z

MH+ = 1685.9, R.TK(me)QTARK(ac)STGGK(ac)APR.K

MH+ = 1913.2, -.ARTK(me)QTARK(ac)STGGK(ac)APR.K

MH+ = 2071.4, R.TKQTARK(ac)STGGKAPR(citr)KQLA.T

MH+ = 2298.7, -.ARTKQTARK(ac)STGGKAPR(citr)KQLA.T

+ AR (228)

+ AR (228)

❶ & ❷

❸ & ❹

❶ ❷ ❸ ❹ F G

H

I

Normalized

A

B

A B

C