CIRCULATING SERUM CELL-FREE DNA BIOMARKERS AND METHODS

BACKGROUND OF THE INVENTION

1. Field of the Invention

[0001] The present invention relates to identification and utility of circulating cell-free DNA in serum as diagnostic biomarkers in Parkinson’s disease to diagnose the disease and assist the clinicians to determine the treatment options for a subject.

2. Brief Description of the Background Art

[0002] Parkinson’s disease (PD), the second most common neurodegenerative disease, is a movement disorder characterized by the demise of dopaminergic neurons. Due to unknown etiology and lack of clinical biomarker the current treatment is only for symptomatic relief. L-dopa treatment in addition to other drug combinations alleviates the motor symptoms but cannot reverse or halt the process of neuronal cell death. There are neither any objective tests nor any established biochemical biomarkers for the diagnosis of PD. Further, the heterogeneity, subtypes and the progression of the disease makes it even complex to develop specific therapeutic candidates. Thus it is imperative to diagnose disease at the early stage to increase the efficacy of therapeutic agents as well as to employ new therapies that can be beneficial to patients.

[0003] The cell-free DNA (cfDNA) was first detected in blood plasma by Mandel and Metais in 1948 (1). It took many years before the application of cfDNA as a tool for diagnostic purpose. Initial and arguably most successful application of cfDNA was in fetal DNA-based prenatial testing that ranged from sex-determination to detect various genetically linked developmental and other diseases (2). This also points to the fact that the cfDNA found in blood has chimeric origin of diseased as well as healthy cells. cfDNA is highly fragmented, double stranded DNA is mostly l50bp in length and found freely circulating in the blood. Most fragments of cfDNA correspond to length of nucleosome units, the primary building block of nuclear DNA. This suggests the cell death as major source of cfDNA in blood. This property of cfDNA is key to its application as a diagnostic biomarker especially in diseases associated with cell death or apoptosis. The cfDNA amounts in patient samples could differ and its function remains largely elusive after 70 years since initial discovery. Since there are factors like sample collection, blood cell lysis that can affect the cfDNA yield in plasma samples, serum can be an alternative source for biomarker discovery.

SUMMARY OF THE INVENTION

[0004] It is an object of the present invention to identify serum cfDNA sequences relevant to patients suffering from Parkinson’s disease.

It is another object of the present invention to provide methods for determining patients suffering from Parkinson’s disease.

These objects and others are achieved by the present invention, which provides circulating cfDNA biomarkers that may be used singly, in pairs or in combination to determine patients suffering from Parkinson’s disease.

DETAILED DESCRIPTION OF THE INVENTION

[0005] Unless defined otherwise, all technical and scientific terms used herein have the meaning commonly understood by a person skilled in the art to which this invention belongs. The following references provide one of skill with a general definition of many of the terms used in this invention: Singleton et ak, Dictionary of Microbiology and Molecular Biology (2nd ed. 1994); The Cambridge Dictionary of Science and Technology (Walker ed., 1988); The Glossary of Genetics , 5th Ed., R. Rieger et al. (eds.), Springer Verlag (1991); and Hale & Marham, The Harper Collins Dictionary of Biology (1991). As used herein, those following terms have the meanings ascribed to them unless specified otherwise.

METHODS

[0006] Serum samples handling and classification

All patients and controls participated in the Norwegian ParkWest project, which are ongoing prospective population-based longitudinal cohort studies investigating the incidence, neurobiology and prognosis of PD. The Norwegian ParkWest study is a prospective longitudinal muiticenter cohort study of patients with incident Parkinson's disease (PD) from Western and Southern Norway. Between November 1st 2004 and 31st of August 2006 it was endeavored to reemit all new cases of Parkinson Disease within the study area. Since the start of the study 212 of 265 (80 %) of these patients and their age-/ sex-matched control group have been followed. Further information about this project can be found at http:/Avww.parkvest.no.

[0007] All possible efforts were undertaken to establish an unselected and population- representative cohort of patients with PD. Patients were included if they had provided serum at study entry and fulfilled diagnostic criteria for PD of the National Institute of Neurological Disorders and Stroke

UK Brain Bank (http://www.nchi .nlm.nih.gov/proiects/gap/cgi- bin/GetPdf. cg ?id ^::: phd00Q042) at latest follow-up. Patients with secondary

parkinsonism at study entry were excluded from this study. Control subjects were recruited from multiple sources, including friends, spouses, and public organizations for elderly and were included in this study if they had provided serum. All patients and controls were Caucasian.

In this study of possible biomarkers for PD we utilized serum from 6 patients and 3 controls which were selected at random. Serum samples were collected at the same day as the clinical examinations and then stored frozen at -70 degrees Celsius until transported to the facilities in New York on dry ice.

[0008] Example 1: Analyses of differential levels of human cfDNA by NGS

cfDNA Isolation from serum samples and QC

After thawing on ice, nine (three control, six PD samples) serum samples were spun down for 5 mins at 3000xg to remove debris. The supernatant was used to perform cfDNA isolation using E-Z Nucleic Acid (E.Z.N.A.®) circulating DNA Isolation Kit (Omega Bio-tek, GA). Before DNA Isolation, the samples were spiked with O.lpg/ul of spike-in control DNA (L34, Zea mays). The remaining part of the RNA isolation was performed following manufacturer’s protocol. The isolated cfDNA was quantified on a Qubit 4 Fluorometer (Thermo Scientific, MA) and quality of the cfDNA was assessed by Alu assay and High sensitivity Bioanalyzer DNA assay (Agilent, CA). The cfDNA was then used for library preparation.

[0009] Library prep. QC and sequencing

The isolated cfDNA (l5ng) from nine patient serum samples were subjected to sequencing library preparation using Accel-NGS® 2S Plus DNA Library Kit (Swift Biosciences, MI) following manufacturer’s protocol. The prepared libraries were quality checked and quantified using Qubit 4 Fluorometer (Thermo Scientific, MA), KAPA Library Quantification Kits (KAPA Biosystems, MA) and Agilent 4200 TapeStation System (Agilent, CA). The spike-in control was recovered at consistent levels for all the samples. The cfDNA libraries were sequenced using HiSeq 4000 Paired-End 100 or PE150 Cycle lanes (Illumina, CA) at NYU School of Medicine’s Genome Technology Center (https://med.nvu.edu/research/scientific-cores-shared- resources/genome-iechnoloax'-center). The fastq.gz files obtained from sequencing runs were imported into Partek Flow (Partek, MO) for analysis.

[0010] NGS data analysis

Paired end sequencing data was imported into Partek Flow for data analysis. Quality control analysis demonstrated a read depth range of 47,627,391-100,811,153, with a mean depth of 83,311,383. The data was also checked to ensure appropriate sequence quality with regards to average base score quality, the percent of missing bases, and GC content. Alignment of sequencing reads to the hgl9 version of the human genome assembly was performed using default parameters of the MEM algorithm in the BWA aligner version 0.7.15 (3, 4). Post alignment quality control demonstrated an average alignment rate of 98.74% (min: 98.36%, max: 99.22%), primarily composed of alignments mapping to a unique location (mean: 94.65%, min: 93.00%, max: 95.94%). The coverage of the genome ranged from 88.02%-92.08% (mean: 91.15%), with an average coverage depth of 8.60 (min: 5.08, max: 10.26). The aligned reads were then filtered to remove duplicate reads in the data set. DNA copy number analysis was performed independently for each sample utilizing Control-FREEC version 11.0, using default parameters and a window of 5kb (5). The resulting segment ratios of regions of copy number imbalance were imported into Partek Genomics Suite version 6.6, and data was filtered to exclude identified regions of gain and loss found in control sample. Recurrent regions of gain and loss were then identified to determine regions conserved across all cases and those unique to either moderate or severe groups. These regions of copy number imbalance were then annotated with regards to their proximity to gene content utilizing the RefSeq.

[0011] Differentially expressed human cfDNA sequences

The differentially expressed human cfDNA sequences in Parkinson’s disease patients’ serum samples from The Norwegian ParkWest study were determined employing NGS. Table 1 below illustrates the cfDNA sequences with statistically significant differential levels in PD patient serum samples obtained by methods explained in [00010] The identified chromosomes that were used for data analysis are known to those of ordinary skill herein from the human genome sequence (hgl9 - Genome Reference Consortium Human Build 37 (GRCh37)) found at

https://www ncbi.nhm nih.gov/assembly/GCF 000001405.13/ [0012]

Table 1

Average

Relative Fold

Copy change in

Seq. ID. Chromosome Seq. Start Seq. Stop Number levels

1 2 37960000 37965000 3.658 1.829

2 2 37965000 37970000 3.658 1.829

3 2 37970000 37975000 3.658 1.829

4 2 37975000 37980000 3.658 1.829

5 2 37980000 37985000 3.658 1.829

6 2 37985000 37990000 3.658 1.829

7 2 37990000 37995000 3.658 1.829

8 2 37995000 38000000 3.658 1.829

9 2 38000000 38005000 3.658 1.829

10 2 89070000 89075000 4.132 2.066

11 2 89075000 89080000 4.132 2.066

12 2 89080000 89085000 4.132 2.066

13 2 95470000 95475000 5.136 2.568

14 2 95475000 95480000 5.136 2.568

15 2 98125000 98130000 4.297 2.149

16 2 98130000 98135000 4.297 2.149

17 3 40245000 40250000 7.275 3.637

18 3 46160000 46165000 7.435 3.718

19 4 5315000 5320000 8.866 4.433

20 4 27695000 27700000 4.963 2.482

21 6 29685000 29690000 4.569 2.285

22 6 162295000 162300000 8.784 4.392

23 6 170705000 170710000 7.853 3.927

24 8 144750000 144755000 4.551 2.275

25 8 7095000 7100000 1.282 0.641

26 8 7100000 7105000 1.325 0.663

27 8 7105000 7110000 1.325 0.663

28 8 7110000 7115000 1.325 0.663

29 8 7115000 7120000 1.325 0.663

30 8 7120000 7125000 1.325 0.663

31 8 7125000 7130000 1.325 0.663

32 8 7130000 7135000 1.325 0.663

33 8 7135000 7140000 1.325 0.663

34 8 7140000 7145000 1.325 0.663

35 8 7145000 7150000 1.325 0.663

36 8 7150000 7155000 1.325 0.663 Average

Relative Fold

Copy change in

Seq. ID. Chromosome Seq. Start Seq. Stop Number levels

37 8 7155000 7160000 1.325 0.663

38 8 7160000 7165000 1.325 0.663

39 8 7165000 7170000 1.325 0.663

40 8 7170000 7175000 1.325 0.663

41 8 7175000 7180000 1.325 0.663

42 8 7180000 7185000 1.325 0.663

43 8 7185000 7190000 1.325 0.663

44 8 7190000 7195000 1.325 0.663

45 8 7195000 7200000 1.325 0.663

46 9 69680000 69685000 4.481 2.240

47 9 69685000 69690000 4.481 2.240

48 9 40815000 40820000 1.116 0.558

49 9 40820000 40825000 1.116 0.558

50 9 40825000 40830000 1.116 0.558

51 9 40830000 40835000 1.116 0.558

52 10 119940000 119945000 4.837 2.419

53 11 106580000 106585000 5.750 2.875

54 11 119610000 119615000 4.945 2.473

55 12 38150000 38155000 2.653 1.327

56 12 38155000 38160000 2.653 1.327

57 12 38160000 38165000 2.653 1.327

58 12 38165000 38170000 2.653 1.327

59 12 38170000 38175000 2.653 1.327

60 12 38175000 38180000 2.653 1.327 61 12 38180000 38185000 2.653 1.327 62 12 38185000 38190000 2.653 1.327

63 12 38190000 38195000 2.653 1.327

64 12 38195000 38200000 2.653 1.327

65 12 38200000 38205000 2.653 1.327

66 12 38205000 38210000 2.653 1.327

67 12 38210000 38215000 2.653 1.327

68 12 38215000 38220000 2.653 1.327

69 12 38220000 38225000 2.653 1.327

70 12 38225000 38230000 2.653 1.327

71 12 38230000 38235000 2.653 1.327

72 12 38235000 38240000 2.653 1.327

73 12 38240000 38245000 2.653 1.327

74 13 53685000 53690000 3.635 1.817

75 14 34815000 34820000 6.120 3.060 Average

Relative Fold

Copy change in

Seq. ID. Chromosome Seq. Start Seq. Stop Number levels

76 14 106780000 106785000 3.412 1.706

77 14 106785000 106790000 3.412 1.706

78 14 106790000 106795000 3.412 1.706

79 14 106795000 106800000 3.450 1.725

80 14 106800000 106805000 3.450 1.725 81 14 106805000 106810000 3.450 1.725 82 15 84855000 84860000 3.840 1.920

83 15 84860000 84865000 3.840 1.920

84 15 84865000 84870000 3.840 1.920

85 17 41535000 41540000 5.425 2.712

86 17 43590000 43595000 4.612 2.306

87 17 34670000 34675000 1.143 0.571

88 18 19790000 19795000 4.416 2.208

89 19 19885000 19890000 3.685 1.842

90 19 52135000 52140000 0.216 0.108

91 19 52140000 52145000 0.216 0.108

92 19 52145000 52150000 0.216 0.108

93 22 17235000 17240000 4.942 2.471

94 22 24275000 24280000 0.925 0.463

95 22 24280000 24285000 0.925 0.463

96 22 24285000 24290000 0.925 0.463

97 22 24290000 24295000 1.083 0.542

98 22 24295000 24300000 1.083 0.542

99 22 24300000 24305000 1.083 0.542 100 22 24305000 24310000 1.083 0.542 101 22 24310000 24315000 1.083 0.542 102 22 24315000 24320000 1.083 0.542

103 22 24320000 24325000 1.083 0.542

104 22 24325000 24330000 1.083 0.542

105 22 24330000 24335000 1.077 0.538

Note: Copy number is the average copy number in PD serum samples assuming the copy number for control samples as 2.

[0013] Example 2

Measurement of levels of a combination of many cfDNA sequences in serum from patients can assist or improve the accuracy in distinctly differentiating between a potential PD patient and a healthy individual. A serum sample is obtained from blood withdrawn from patients suspected of PD. The serum is used for total cfDNA isolation and enrichment. This RNA would then be tested using NGS or qPCR to measure the levels of any two or more of the 105 cfDNA sequences mentioned in Example 1.

Detectable levels of any two or more of the 105 cfDNA sequences confirms the patient has PD. If desired, other sample fluids may be utilized, including plasma, venous or arterial blood, or CSF samples withdrawn by lumbar puncture. Such plasma, blood or CSF samples are processed as above. It will be understood that measurement of more than two cfDNA sequences in combination or a set of combinations used in a test matrix may desirably increase the accuracy of PD diagnosis. Similarly, practitioners of ordinary skill herein will further appreciate that shorter DNA sequences within any of the identified cfDNA sequences may desirably be utilized instead of the entire cfDNA sequence. These shorter DNA sequences are preferably unique to the SEQ ID NO. in which they are found, and may have lengths on the order of about lOOObp, 900bp, 800bp, 700bp, 600bp, 500bp, 400bp, 300bp, 200bp, lOObp, 50bp or 25bp.

[0014] Example 3

A microarray tray is provided containing labeled nucleotide sequences that are antisense to selected cfDNA SEQ ID NOS. among cfDNA SEQ ID NOS. 1-105. The labeled antisense nucleotide sequences may be antisense to shorter DNA sequences found with the selected cfDNA SEQ ID NOS., which shorter DNA sequences are preferably unique to the cfDNA sequences encompassing them. Alternatively, the microarray tray may contain labeled antibodies that specifically bind selected cfDNA SEQ ID NOS. among cfDNA SEQ ID NOS. 1-105. The labeled antibodies may specifically bind shorter DNA sequences found with the selected cfDNA SEQ ID NOS., which shorter DNA sequences are preferably unique to the cfDNA sequences encompassing them. The term "antibody" includes both polyclonal and monoclonal antibodies. The phrase “specifically (or selectively) binds” refers to a binding reaction between two molecules that is at least two times the background and more typically more than 10 to 100 times background molecular associations under physiological conditions. Specific binding is determinative of the presence of the DNA or cfDNA, in a heterogeneous population cfDNAs. Thus, under designated immunoassay conditions, the specified antibodies bind to a particular DNA sequence, thereby identifying its presence as well as the presence of the cfDNA encompassing it. Specific binding antibodies may be characterized by having specific binding activity (K _a ) of at least about 10 ⁵ M ¹ , 10 ⁶ M ¹ or greater, preferably 10 ⁷ M ¹ or greater, more preferably 10 ⁸ M ¹ or greater, and most preferably 10 ⁹ M ¹ or greater. The binding affinity of an antibody can be readily determined by one of ordinary skill in the art, for example, by Scatchard analysis (Scatchard, Ann. NY Acad. Sci. 51 : 660-72, 1949). Methods of making antibodies are well known to those skilled in the art (Winter and Harris, Immunol. Today 14:243-246 (1993); Ward et ak, Nature 341 :544-546 (1989); Harlow and Lane, supra, 1988; Hilyard et al., Protein Engineering: A practical approach (IRL Press 1992); Borrabeck, Antibody Engineering, 2d ed. (Oxford LTniversity Press 1995); each of which is incorporated herein by reference).

[0015] Example 4

Many neurodegenerative diseases are closely related to each other when it comes to symptoms as well as pathological markers. The circulating diagnostic markers for one neurodegenerative disease can be useful for diagnosis of other disease. A method to diagnose other neurodegenerative diseases like Dementia with Lewy body (DLB), Amyotrophic lateral sclerosis (ALS), Alzheimer’s disease (AD), Multiple system atrophy (MSA), CorticoBasal Degeneration (CBD), Progressive Supranuclear Palsy (PSP) can also be developed using similar cfDNA sequences measurements of candidates mentioned above. Disease specific kits can be developed similar to one mentioned in [0014] with various combinations of DNA or cfDNA sequences listed in [0012] and [0013]

[0016] Example 5

The absence or presence of one or more combinations of DNA or cfDNA sequences in PD patient samples as compared to control samples can be used to develop disease specific kit as mentioned in [0014] [0017] Example 6

The function of the cfDNA sequences which may cross blood brain barrier depending on the size of molecule is poorly understood but a disease specific sequence can be targeted for understanding PD etiology and to target them for therapy.

[0018] Example 7

Small nucleic acid molecules derived from cfDNA sequences mentioned in [0012] and [0013] will be designed to therapeutically intervene by specifically targeting genes in PD brains to achieve complete or partial remedy.