INTEGRATED METHOD FOR URINARY PROSTATE SPECIFIC ANTIGEN N-GLYCOSYLATION PROFILING BY CAPILLARY ELECTROPHORESIS

Abstract

In the present invention a method is provided for capillary electrophoresis-based (CE-based) comprehensive N-glycan analysis of urinary PSA. The method is useful for analysis of PSA glycans from urine and for diagnosis of PSA-related conditions.

Claims

1. A method for PSA-analysis from urine sample, preferably from urine sample of a male mammal, wherein said method comprises the steps of providing urine sample, preparing said urine sample for analysis, separating PSA from said urine sample by affinity separation using PSA-binding molecules to obtain enriched PSA, releasing glycans from the enriched PSA to obtain released PSA-glycans, analyzing the released glycans by separation with capillary electrophoresis (CE) wherein detection of the separated released glycans is carried out.

2. The method according to claim 1, wherein detection of the separated released glycans is carried out by fluorescent detection, preferably detected by laser induced fluorescence detection, wherein preferably between the releasing step and the analyzing step the following step is carried out: labeling the released PSA glycans by a fluorescent dye, to obtain fluorophore labeled released PSA glycans, and the fluorophore labeled glycans are detected.

3. The method according to claim 1, wherein affinity separation is affinity partitioning.

4. The method according to claim 1, wherein preparing said urine sample comprises pre-concentration of the sample.

5. The method according to claim 1, wherein in the affinity-separation step the PSA-binding molecules are nanobodies, in particular single-domain antibodies.

6. The method according to claim 5, wherein the affinity partitioning column is a chromatography microcolumn, wherein the nanobodies have a tag and the chromatography column has a matrix binding said tags.

7. The method according to claim 1, wherein releasing glycans from the PSA to obtain released PSA-glycans is carried out enzymatically, preferably with endoglycosidases.

8. The method according to claim 1, wherein the method comprises analyzing the fluorophore labeled released PSA glycans by separating at least the fucosylated and non-fucosylated glycan structures of the fluorophore labeled released PSA glycans.

9. The method according to claim 1, wherein the method comprises analyzing the fluorophore labeled released PSA glycans by separating at least the ?2,3- and ?2,6-sialylated isomers of the fluorophore labeled released PSA glycans.

10. A diagnostic method for diagnosing PSA-related condition from urine sample of a subject, preferably of a male subject, said method comprising the steps of carrying out the method according to claim 1 for PSA-analysis, determining a glycosylation pattern according to the invention in the sample of a patient (patient glycosylation pattern), comparing the glycosylation pattern with a normal glycosylation pattern, and if there is a difference between the patient glycosylation pattern and the normal glycosylation pattern the patient is considered as having a PSA-related condition.

Description

BRIEF DESCRIPTION OF THE FIGURES

[0135] FIG. 1. Capillary electrophoresis separation of the APTS-labeled standard PSA N-glycome. Separation conditions: 50 cm total (40 cm effective) length, 50 ?m i.d. bare fused silica capillary with HR-NCHO gel buffer. Voltage: 30 kV (0.5 min ramp); Temperature: 25? C.; Injection: 5 psi/5 see water pre-injection followed by 5 kV/2 sec sample. Structures corresponding to peaks are listed in Table 1.

[0136] FIG. 2. Exoglycosidase based sequencing of standard PSA N-glycome. Standard PSA (A), after Sialidase A digestion (B), after Sialidase A, ?-Galactosidase digestion (C), after Sialidase A, ?-Galactosidase and ?-N-Acetyl Hexosaminidase digestion (D). Separation conditions were the same as in FIG. 1. Structures corresponding to peaks are listed in Table 1.

[0137] FIG. 3. Schematic of the single domain anti-PSA production. Genes of strong PSA binder N7 and C9 SdAb aPSA variants were incorporated into the pET23b expression vector and the proteins produced in SHuffle T7 Express E. coli cells. SHuffle ensured the proper formation of disulfide bonds, thus appropriate folding of aPSA, resulting in soluble and functional antibody for selective capture of PSA from body fluids.

[0138] FIG. 4. Urinary PSA analysis workflow. 1a) Immobilization of aPSA sdAbs to Ni-IMAC column; 1b) Concentration of urine via spinfilters; 2) PSA capture from urinary matrix; 3) Elution of antibody-PSA complex; 4) Desalting and concentration; 5) Denaturation of PSA; 6)N-glycan release by PNGase F digestion; 7) Fluorescent glycan labeling; 8) CE-LIF analysis

[0139] FIG. 5. Capillary electrophoresis separation of the sdAb captured male urinary PSA N-glycome (B) and female urine control (A). Separation conditions: 30 cm total (20 cm effective) length, 50 ?m i.d. bare fused silica capillary with HR-NCHO gel buffer. Voltage: 30 kV (0.5 min ramp); Temperature: 25? C.; Injection: 5 psi/5 sec water pre-injection followed by 2 kV/2 sec sample. Structures corresponding to peaks are listed in Table 1.

DETAILED DESCRIPTION OF THE INVENTION

[0140] The present inventors have developed a high-throughput capillary electrophoresis (CE) based glycan analysis workflow capable to process and analyze urinary prostate-specific antigen (PSA). The demonstrated technology utilizes a selective, high yield single domain antibody based PSA capture followed by preconcentration and capillary electrophoresis coupled by laser-induced fluorescence detection separation steps resulting in high resolution N-glycan profiles. The patient PSA glycan profile is compared with a commercially available PSA standard revealing that the suggested methodology provides reliable data even capable to differentiate ?2,3- and ?2,6-sialylated isomers. The latter plays an important role in classification between indolent, significant, and aggressive forms of prostate cancer.

[0141] The novel, integrated workflow was established for urinary PSA N-glycosylation analysis utilizing highly selective sdAb-based capture from a biological matrix, high throughput preconcentration, enzymatic release of N-glycans from PSA, fluorescent carbohydrate labeling, high resolution capillary electrophoresis separation and comprehensive glycan structure elucidation.

[0142] In this invention, a new method step was introduced to isolate PSA from urine samples by an affinity purification or partitioning column comprising immobilized PSA-specific binding molecules. The skilled person is aware that several types of such binding molecules may be used. Such binding molecules may include proteins of antibody derived protein scaffolds, like antibody fragments, single-domain antibodies, single chain antibody fragments (scFv), Fab fragments, nanobodies etc. (Muyldermans, 2021).

[0143] Alternatively, a binding molecule can be entirely artificially made, e.g. as a synthetic peptide. Thus, binding molecules include proteins of non-antibody protein scaffolds like fibronectin, lipocalins, anticalins, (?krlec et al., 2015; Stern et al., 2013; Gebauer & Skerra, 2019).

[0144] In a highly preferred embodiment immobilized antiPSA nanobodies are used. In a highly preferred embodiment such antibodies are applied on IMAC-Ni microcolumns. In the present exemplary embodiment the sdAbs were produced in-house and were engineered with a special linker tag to help their anchoring to the column matrix. The usage of sdAbs over mAbs may facilitate extension of the application in future clinical tests as their production even at larger quantities is possible with lower costs. The selective capturing procedure contributed both to remove other glycoprotein contaminants from the biological matrix and pre-concentrating the PSA content, resulting in higher detection signal.

[0145] Sample concentration is an important feature of the process. In a preferred embodiment the eluted samples are concentrated. In a preferred embodiment a filter with an appropriate cut-off is used, said cut-off is preferably between 5 to 20 kDa, highly preferably about 10 kDa. In a preferred embodiment spinfilters are applied and centrifugation is used to concentrate the sample. Centrifugation may be carried out typically with 5000 to 20000 g, preferably 10000 to 15000 g, for 2 to 20 minutes, as required, preferably for 5 to 15 minutes, more preferably for 8 to 12 minutes at 10000 to 15000 g. The samples are cooled as needed, e.g. to 1 to 10? C. In a preferred embodiment the samples are dried and resolved again. The volume of the samples is preferably 1 to 20 ?l, more preferably about 10 ?l.

[0146] Sample preparation preferably and usually includes denaturation of the protein. The use of magnetic beads supports the removal of excess fluorescent dye from the labeling reaction mixture. The dye shall be used in a relatively large excess in the reaction. In aqueous medium the glycans are solution together with the fluorescent dye whereas in the organic phase the labeled glycans attach to the surface of the magnetic beads. As the dye does not attach to the surface of the beads it can be removed by washing.

[0147] The denaturation solution typically comprises one or more types of detergents, e.g. nonionic and/or ionic detergent(s) or mixture thereof. The ionic detergent may be anionic or cationic detergents. A typical anionic detergent comprises an alkyl-sulphate, e.g. dodecyl-sulphate. The counter ion is a positive or a negative ion, respectively. In an example the mixture may be that of a nonionic detergent, like Nonidet and of an anionic detergent like SDS.

[0148] The denaturation solution may also comprise a reducing agent to keep amino acids, like cysteine, in a reduced state. Such reducing agents may comprise DTT and/or beta-mercaptoethanol, among others.

[0149] Analysis of the oligosaccharides of PSA require a glycan derivatization step: glycans may be derivatized to introduce a chromophore or fluorophore, facilitating detection after chromatographic or electrophoretic separation. Here fluorescent labeling of glycans was applied. Today fluorescent labeling methodology can be considered as a routine for a person skilled in the art and several labeling kits are available like, merely as examples, labeling kits from Glycan labeling products from Merck like GlycoProfile? Labeling Kit. In the examples the Fast Glycan Labeling and Analysis Kit was from SCIEX (Brea, CA, USA) including the tagging dye of 8-aminopyrene-1,3,6-trisulfonic acid (APTS).

[0150] In an example the labeling solution may comprise a mixture of the tagging dye, e.g. APTS and aqueous solvent mixture comprising an organic acid, e.g. AcOH, an organic solvent, e.g. THF, and a reducing agent, e.g. a borohydride derivative, like Sodium cyanoborohydride in an appropriate ratio.

[0151] Magnetic beads are used for removing excess dye. Magnetic bead suspension: the solvent was removed from 200 ?l of magnetic bead suspension from the Fast Glycan Labeling and Analysis Kit on a magnetic stand, the beads were resuspended in 200 ?l of water and the solvent was removed again on a magnetic stand, the beads were resuspended again in 20 ?l of water

[0152] The glycans were removed from PSA by using enzymatic removal technologies. The enzyme should be one which results in a deaminated protein or peptide and a free glycan, e.g. an enzyme which cleaves between the innermost GlcNAc and asparagine residues of a glycoprotein (e.g. PSA) and thus cleaves high mannose, hybrid, and complex oligosaccharides from N-linked glycoproteins and glycopeptides. An example is Peptide:N-glycosidase F, commonly referred to as PNGase F, an amidase of the peptide-N4-(N-acetyl-beta-glucosaminyl)asparagine amidase class.

[0153] Sequencing should also use a set of exoglycosidase enzymes, including e.g. sialidase (e.g. Sialidase A), galactosidase (e.g. ?-Galactosidase), hexosaminidase (?N-Acetyl Hexosaminidase), etc. and also an endoglycosidase (PNGase F). Upon sequencing, repeated cleavage steps are carried out in a well defined order to explore the sequence of sugar moieties in the glycan structure.

[0154] Capillary electrophoresis and methods for its application to N-glycosylation analysis of prostate-specific antigen are known in the art in general and reviewed e.g. by Reider et al. (Reider et al., 2020). The present inventors have found that while using a longer capillary the separation improves, but the method can be carried out with a relatively short capillary as well.

[0155] In a particular embodiment capillary gel electrophoresis is carried out with a capillary of at least 5 cm and at most 100 cm effective length, preferably at least 10 cm and at most 60 cm effective length, highly preferably at least 15 cm and at most 50 cm effective length. Preferably, a longer capillary of 30 to 50 cm effective length is used for standard PSA glycan analysis and sequencing and a shorter one with 10 to 30 cm effective length for captured urinary PSA glycan analysis. In a particular embodiment, a fused silica capillary is used, preferably a bare fused silica capillary.

[0156] Separation gel buffers are known in the art, in a particular solution HR-NCHO separation gel buffer (high resolution N-linked carbohydrate; SCIEX) is used.

[0157] The applied electric field is a reversed polarity field of 20 to 40 kV, preferably 25 to 35 kV, highly preferably 28 to 32 kV.

[0158] The separation temperature can be optimized, however, in a particular method it is 20-37? C., preferably 25 to 35?, highly preferably 28 to 32? C.

[0159] The CE-LIF analysis of the released glycans from urinary PSA resulted in 19 structures out of the 30 obtained from standard PSA. Each identified glycan structure originated from native PSA was sialylated, suggesting that the mild conditions of both the capture and analysis procedure preserved the labile sugar residues, e.g., sialic acids. Due to the high resolving power of CE, both the fucosylated and non-fucosylated structures as well as the ?2,3- and ?2,6-sialylated isomers were separated. Since the most common cancer related alterations are reportedly associated with these structures, it was key for any diagnostic application to reliably detect any possible changes in these residues. Furthermore, as not only glycan types but individual structures were identified more complex alterations could be discovered, which occurs in certain glycans only. Our results showed that the developed workflow was suitable for comprehensive N-glycan analysis of urinary PSA and could be a basis for future endeavors aiming to detect the alterations in PSA glycosylation caused by certain diseases like prostate cancer or benign prostatic hyperplasia.

[0160] Therefore, the present method is useful for obtaining a glycosylation pattern from the urine sample. A glycosylation pattern may comprise data on individual glycan structures in a sample. The data comprise values typical of the glycan structures given or selected, e.g. on the ratio or level of individual glycans in the urine PSA sample. In a preferred embodiment the pattern comprises data on at least 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18 or 19 glycan structures from the 19 structures obtained by CE-LIF analysis of the present inventors.

[0161] Typical values which can be used to obtain a glycosylation pattern may include but not limited to the presence, any typical ratio or level, e.g. concentration, percentage, composition of one or more selected glycans. The pattern necessarily comprises structure information on the glycans, e.g. comprise also the sequence of the glycans.

[0162] Thus, the glycosylation pattern may comprise data on multiple glycan structures. Preferably it comprises data regarding at least fucosylated and non-fucosylated structures. Moreover it also comprises data regarding at least the ?2,3- and ?2,6-sialylated isomers. Moreover it also comprises data regarding both.

[0163] By analyzing samples of healthy subjects (or patients) a normal glycosylation pattern can be established.

[0164] Preferably the glycosylation pattern data comprise ranges for each glycan. In case of a normal glycosylation pattern each range indicates the values (e.g. ratio or level) typical of the corresponding glycan which indicates the normal range. In case of a healthy subject (or patients) the values or at least the values for the majority of the glycans, preferably at least 80%, or 90% or 95% of the glycans fall into the normal range.

[0165] Quite similarly glycosylation pattern typical for various PSA-related conditions can be obtained by analyzing, using the method of the invention, samples of patients having the specific PSA-related condition. Such can be e.g. benign prostatic hyperplasia, prostate cancer including a metastatic cancer, or a precancerous condition e.g. a condition that is likely to progress to cancer.

[0166] Thereby a set of reference glycosylation pattern can be established.

[0167] In a diagnostic method the glycosylation pattern obtained by analyzing a patient sample can be compared with a typical glycosylation pattern, e.g. a glycosylation pattern typical of a PSA-related condition or a normal glycosylation pattern.

[0168] Comparing glycosylation patterns can include comparing any data determined by the methods of the present invention, including but not limited to the presence, concentration, percentage, composition or sequence of one or more selected glycans of the target glycoprotein, to the reference. This comparison allows diagnosis, staging, prognosis or monitoring of the PSA-related condition.

[0169] The invention also allows providing a database comprising a plurality of records.

[0170] The database, or e.g. the records, preferably each record can include the data as described above.

[0171] More specifically the database may comprise: [0172] reference glycosylation patterns, like [0173] normal glycosylation pattern and optionally one or more glycosylation patterns typical of a PSA-related condition.

[0174] In an embodiment records of the database may comprise one or more of the following: [0175] data on the glycosylation pattern of PSA associated with a disorder isolated from a sample from a subject obtained by determining the glycosylation profile of said subject by the method of the present invention; and optionally [0176] data on the status of the subject, e.g., whether the subject has cancer, a pre-cancerous condition, a benign condition, or no condition, and any clinical outcome data, e.g., metastasis, recurrence, remission, recovery, or death; [0177] data on any treatment administered to the subject; [0178] data on the subject's response to treatment, e.g., the efficacy of the treatment; personal data on the subject, e.g., age, gender, education, etc. and/or environmental data, such as the presence of a substance in the environment, residence in a preselected geographic area, and performing a preselected occupation. In some embodiments, the database is created by entering data resulting from determining the glycoprofile of a target glycoprotein in a sample from a subject using a method described herein.

[0179] The diagnostic method of the invention preferably comprises [0180] determining a glycosylation pattern according to the invention in the sample of a patient (patient glycosylation pattern), [0181] comparing the glycosylation pattern with a normal glycosylation pattern, and [0182] if there is a difference between the patient glycosylation pattern and the normal glycosylation pattern the patient is considered as having a PSA-related condition.

[0183] In a further embodiment, the diagnostic method of the invention preferably comprises [0184] determining a glycosylation pattern according to the invention in the sample of a patient (patient glycosylation pattern), [0185] comparing the glycosylation pattern with a reference glycosylation pattern typical of a PSA-related condition, and if [0186] the patient glycosylation pattern is identical with a reference glycosylation pattern the patient is diagnosed as having the PSA-related condition.

[0187] In general, the method of the invention comprises detecting alterations in the PSA-glycosylation pattern.

[0188] In an embodiment the invention comprises detecting alterations in the sialylation of the glycan moieties.

[0189] In an embodiment the invention comprises detecting alterations in the fucosylation of the glycan moieties.

[0190] In preferred embodiments of the invention the method comprises detecting alterations in both the sialylation and fucosylation patterns.

[0191] In preferred embodiments of the invention the method comprises a fingerprint analysis of the PSA-glycosylation pattern.

[0192] In a preferred embodiment the invention also relates to methods of determining whether a subject has prostate cancer, comprising determining whether a subject has an altered PSA alpha-2,3-sialylation pattern as compared to the alpha-2,3-sialylation pattern of PSA from a healthy subject, wherein an altered PSA alpha-2,3-sialylation pattern is indicative of prostate cancer.

[0193] In a preferred embodiment the method of the invention comprises measuring whether the level of alpha-2,3-sialylated glycans or the ratio of the level of alpha-2,3-sialylated glycans and a reference value (level or ratio for alpha-2,3-sialylated glycans, respectively) is/are increased as compared to a normal level or ratio typical of a healthy subject. The reference value can be a value or range, e.g. an amount or level calculated from the total amount or level of glycans or the amount or level of a given number of predetermined glycans, i.e. a set of glycans or a specific glycan. Alternatively, the reference can be the PSA-level or the level of another protein, e.g. a household protein. In a particular method a ratio of the level of alpha-2,3-sialylated glycans and a reference value is formed. In an embodiment the level of alpha-2,3-sialylated glycans is measured and compared to a normal value. The normal value can be a value or range, e.g. an amount or level calculated from a set of measurements for samples of healthy subjects by the assay method of the invention.

[0194] In a preferred embodiment if the level or the ratio for alpha-2,3-sialylated glycans is above a threshold level the subject is diagnosed as having prostate cancer or showing tendency for prostate cancer if the ratio is higher than a threshold value.

[0195] In a particular embodiment the present invention relates to a method for diagnosis of a subject for prostate cancer comprising determining the ratio of the amount of glycan in which the terminal sialic acid residue of the glycan is +(2,3)-linked to a second galactose residue from the terminal of the glycan, to a reference value and the subject is diagnosed as having prostate cancer or showing tendency for prostate cancer if the ratio is higher than a threshold value.

[0196] In a preferred embodiment the invention also relates to methods of determining whether a subject has prostate cancer, comprising determining whether a subject has an altered PSA alpha-2,6-sialylation pattern as compared to the alpha-2,6-sialylation pattern of PSA from a healthy subject, wherein an altered PSA alpha-2,6-sialylation pattern is indicative of prostate cancer.

[0197] In a preferred embodiment the method of the invention comprises measuring whether the level of alpha-2,6-sialylated glycans or the ratio of the level of alpha-2,6-sialylated glycans to alpha-2,3-sialylated glycans and a reference value (level of alpha-2,6-sialylated glycans or ratio of alpha-2,6-sialylated glycans to alpha-2,3-sialylated glycans, respectively) is/are increased as compared to a normal level or ratio typical of a healthy subject. The reference value can be a value or range, e.g. an amount or level calculated from the total amount or level of glycans or the amount or level of a given number of predetermined glycans, i.e. a set of glycans or a specific glycan. In a particular method a ratio of the level of alpha-2,6-sialylated glycans to alpha-2,3-sialylated glycans and a reference value is formed. In an embodiment the level of alpha-2,6-sialylated glycans is measured and compared to a normal value. The normal value can be a value or range, e.g. an amount or level calculated from a set of measurements for samples of healthy subjects by the assay method of the invention.

[0198] In a preferred embodiment if the level or the ratio of alpha-2,6-sialylated glycans to alpha-2,3-sialylated glycans is above a threshold level, the subject is diagnosed as having prostate cancer or showing tendency for prostate cancer.

[0199] In a preferred embodiment the invention also relates to methods of determining whether a subject has prostate cancer, comprising determining whether a subject has an altered PSA fucosylation pattern as compared to the fucosylation pattern of PSA from a healthy subject wherein an altered PSA fucosylation pattern is indicative of prostate cancer.

[0200] In a preferred embodiment the method of the invention comprises measuring whether the level of fucosylation or the ratio of the level of fucosylated glycans and a reference value (level or ratio for fucosylated glycans, respectively) is/are increased as compared to a normal level or ratio typical of a healthy subject. The reference value can be a value or range, e.g. an amount or level calculated from the total amount or level of glycans or the amount or level of a given number of predetermined glycans, i.e. a set of glycans or a specific glycan. Alternatively, the reference can be the PSA-level or the level of another protein, e.g. a household protein. In a particular method a ratio of the level of fucosylated glycans and a reference value is formed. In an embodiment the level of fucosylated glycans is measured and compared to a normal value. The normal value can be a value or range, e.g. an amount or level calculated from a set of measurements for samples of healthy subjects by the assay method of the invention.

[0201] In a preferred embodiment if the level or the ratio for fucosylated glycans is above a threshold level the subject is diagnosed as having prostate cancer or showing tendency for prostate cancer.

[0202] Below the invention is illustrated further by way of Examples which are, however, non-limiting and provide description of particular embodiments for illustrative purposes.

EXAMPLES

Materials and Methods

Chemicals and Reagents

[0203] Water (HPLC grade), acetic acid (glacial), acetonitrile (MeCN), sodium cyanoborohydride (1 M in THF), sodium chloride, imidazole, 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES) and DTT (dithiothreitol) were obtained from Sigma Aldrich (St. Louis, MO, USA). SDS (sodium dodecyl sulfate) and Nonidet P-40 were from VWR (West Chester, PA, USA). The Fast Glycan Labeling and Analysis Kit was from SCIEX (Brea, CA, USA) including the tagging dye of 8-aminopyrene-1,3,6-trisulfonic acid (APTS), magnetic beads for excess dye removal, HR-NCHO separation gel-buffer system, the bracketing standards of maltose (DP2) and maltopentadecaose (DP15). The exoglycosidase enzymes of Sialidase A (Arthrobacter ureafaciens), ?-Galactosidase (Jack bean) and ?N-Acetyl Hexosaminidase (Jack bean) were from ProZyme (Hayward, CA, USA). The endoglycosidase PNGase F was from Asparia Glycomics, (San Sebastian, Spain). 20 ml volume 10 kDa cut-off spinfilters were from Pall (New York, NY, USA), 500 ?l ml volume 10 kDa cut-off spinfilters were from VWR. PhyTip Ni-IMAC microcolumns (40 ?l) were provided by PhyNexus (San Jose, CA). Buffer A: 100 mM HEPES, 500 mM NaCl, 50 mM imidazole, pH=8.0. Buffer B: 100 mM HEPES, 500 mM NaCl, 500 mM imidazole, pH=8.0. Denaturation solution: mixture of Nonidet P-40:DTT:SDS=6:1:1. Digestion solution: 75 ?unit/?l of PNGase F in 16.7 mM ammonium acetate. Labeling solution: 5.7 mM of APTS in the mixture of H.sub.2O:AcOH:THF:NaBH.sub.3CN (1 M in THF)=5:5:8:2. Magnetic bead solution: the solvent was removed from 200 ?l of magnetic bead suspension from Fast Glycan Labeling and Analysis Kit on a magnetic stand, the beads were resuspended in 200 ?l of water and the solvent was removed again on a magnetic stand, the beads were resuspended again in 20 ?l of water.

Gene Construction

[0204] Two antiPSA (aPSA) coding polypeptide sequences, N7 and C9 were taken from (Saerens et al., 2004), back translated and codon optimized for E. coli and the genes with flanking 5-NdeI and 3-XhoI cleavage sites were synthetized by Genscript (Piscataway, New Jersey, United States). After amplification both genes were cloned into pET23b (Novagen, Merck, Darmstadt, Germany) between its NdeI and XhoI sites, thus fusing a C-terminal 6-histidine coding sequence to the gene of aPSA. Proper incorporation of the insert into the plasmid was verified by sequencing (Macrogen Europe B.V., Amsterdam, the Netherlands).

Protein Expression and Purification

[0205] For protein expression Shuffle T7 Express E. coli (New England Biolabs, Ipswich, Massachusetts, US) cells were transformed with the N7-aPSA-pET23b and C9-aPSA-pET23b plasmids, according to the supplier's protocol. 5 mL sterile Luria Broth (LB) media containing 100 ?g/mL Ampicillin (Amp) was inoculated from a freshly prepared LB/Amp plate and grown at 30? C. with vigorous shaking until OD600 was between 0.4-0.6 and kept at 4? C. overnight. The cells were then separated from the supernatant by centrifugation, resuspended in 5 mL freshly prepared LB and 2 mL of it was used to inoculate 1 L LB/Amp. Cell culture was grown at 37? C. for two hours and then at 30?, 140 rpm in baffled flask until OD600 fell into 0.6-1 and then the temperature was decreased down to 22? C. After 15 min of cooling protein expression was induced by the addition of 0.5 mM isopropyl ?-D-1-thiogalactopyranoside (IPTG, final concentration) and further incubated overnight. Cells were harvested by centrifugation at 6,000 g for 30 min, washed with buffer A and centrifuged again at 10,000 g for 30 min. Cells were resuspended again on ice in 20 ml buffer A containing two EDTA-free Mini Complete protease inhibitor tablets (Roche, Basel, Switzerland) and disrupted by sonication (10?30 s, 50% amplitude). The suspension was centrifuged at 30,000 g and the supernatant filtered through a 0.45 ?m syringe filter. Sample was loaded on a pre-equilibrated 5 ml HiTrap Chelating (GE Healthcare, Chicago, Illinois, US) nickel saturated column and purified using isocratic elution and the pure protein was eluted at 50% B buffer. In order to get rid of the high salt and imidazole content of the protein solution, it was dialyzed against a buffer solution. Purity of aPSA was confirmed on 15% SDS PAGE gels and protein concentration calculated using the following parameters given by ProtParam (Gasteiger et al., 2005): N7-aPSA-6His, 14.5 kDa, 27,180 c.sup.?1M.sup.?1 and C9-aPSA-6His, 14.3 kDa, 21,555 c.sup.?1M.sup.?1, respectively.

Biological Specimens

[0206] Urine samples were collected with the appropriate Ethical Permissions (approval number: 23580-1/2015/EKU (0180/15)) and Informed Patient Consents in the Semmelweis Hospital (Miskolc, Hungary). Samples were taken from male and female (blind control) healthy volunteers (population of seven Caucasian, age average: 28.3, age median 27) and kept at 4? C. until processing.

PSA Quantitation in Urine

[0207] All standard ELISA tests were carried using a UniCel D?I 800 Access Immunoassay System, kindly provided by the central laboratory of Csolnoky Ferenc Hospital (Veszprem, Hungary). Urine samples were analyzed directly, without any sample preparation.

PSA Capture from Urine

[0208] 300 ml of male urine was cooled to room temperature and it was concentrated to 500 ?l by 20 ml volume 10 kDa cut-off spinfilters (13,500 g for 30 min at 6? C. for each consecutive 20 ml concentration). The concentrated urine was diluted and centrifuged (13,500 g for 10 min at 6? C.) twice with 3-3 ml of buffer A then it was transferred to an Eppendorf vial. The filter was washed two times with 250 ?l buffer A each and both were added to the transferred urine. The final, 1 ml mixture was vortexed and divided into two 500-500 ?l, one for PSA capture and one for control. Two Ni-IMAC microcolumns were washed by 200 ?l buffer A for 5 min, by connecting the tips to an automated pipette and the buffer was continuously aspirated and dispensed through the tips with the flow rate of 2 ml/min. In the next step, the tips were washed by 50 ?g/ml C9 sdAb in 100 ?l buffer A for 10 min. For control sample 100 ?l buffer A without sdAb was used instead. Both of the tips were washed again by 200 ?l buffer A for 5 min. Then, the tips were washed by 500-500 ?l concentrated urine samples for 10 min. After that, the tips were washed again 3 times with 200 ?l of buffer A, 5 min each, then washed by 100 ?l buffer B for 5 min. The eluted buffers (containing the targets) were transferred onto 500 ?l volume 10 kDa cut-off spinfilters and the solvent was centrifuged (13,500 g for 10 min at 6? C.). The filter was washed twice by 400 ?l of water, both times the water was centrifuged (13,500 g for 10 min at 6? C.). The samples were taken up into 80 ?l water (washing the filter twice with 40-40 ?l of water), transferred into an Eppendorf vial, then they were dried in a SpeedVac under reduced pressure at 70? C. for 15 min. The dried sample was resolved again in 10 ?l of water.

Sample Preparation

[0209] The sample preparation process was identical for all captured PSA, control and standard PSA (150 ?g/ml) samples. 2 ?l of denaturation solution was added to 10 ?l sample, and it was heated up from 30? C. to 80? C. in 8 min 20 sec and was incubated at 80? C. for 1 min 40 sec. Then 20 ?l of digestion solution was added and the sample was heated up from 40? C. to 60? C. in 20 min. After that, labeling solution was added and sample was incubated at 37? C. overnight in an open cap vial to let the solvent evaporate. The dried sample was then resolved in 20 ?l of magnetic bead solution, then 185 ?l of MeCN was added and solvent was removed from the magnetic beads on a magnetic stand. The beads were suspended in 20 ?l of water, 185 ?l of MeCN was added and the solvent was removed from the magnetic beads on a magnetic stand again. This step was repeated two additional times (total of 4 washes with the first one adding the beads). Finally, beads were diluted with 60 ?l of water and, on a magnetic stand, 50 ?l of sample was transferred into a new vial to CE analysis.

Glycan Structure Identification

[0210] Structural elucidation of separated, asparagine linked PSA glycans was utilized by direct mining of GU database entries (GUcal.hu), exoglycosidase digestion based carbohydrate sequencing and some earlier published literature data on the same subject matter (Kammeijer et al., 2018; M?sz?ros et al., 2020). Exoglycosidase sequencing was utilized in an automated fashion as reported in Szigeti & Guttman, 2017. Shortly, the released N-glycans were consequently digested by monomer and anomericity specific exoglycosidase enzymes such as Sialidase A, ?-Galactosidase and ?-N-Acetyl Hexosaminidase. Native, i.e., undigested and all digested pools were then separated by CE-LIF and the structural information was derived from the GU value shifts of the individual peaks as the result of the consecutive exoglycosidase treatments (Guttman & Ulfelder, 1997).

Capillary Gel Electrophoresis

[0211] A PA800 Plus Pharmaceutical Analysis System (SCIEX) with laser induced fluorescence detection (?ex=488 nm/?em=520 nm) was used for all capillary gel electrophoresis separations employing the HR-NCHO separation gel buffer in a 40 cm effective length (50 cm total length, 50 ?m ID) bare fused silica capillary for standard PSA and sequencing, and 20 cm effective length (30 cm total length, 50 ?m ID) bare fused silica capillary for captured urinary PSA. The applied electric field strength was 30 kV in reversed polarity mode (cathode at the injection side, anode at the detection side). The separation temperature was set at 30? C. A three-step electrokinetic sample injection was applied: 1) 3.0 psi for 5.0 sec water pre-injection, 2) 1.0 kV for 1.0 sec sample injection and 3) 1.0 kV for 1.0 sec bracketing standard (BST, DP2 and DP15). The 32 Karat (version 10.1) software package (SCIEX) was used for data acquisition and interpretation.

Results

Identification of Standard PSA Glycosylate

[0212] The global N-glycosylation profile of standard PSA was analyzed in order to establish a reference point for the workflow development process. 40 cm effective length capillary was utilized to achieve high resolution glycan profile as depicted in FIG. 1, wherein the 30 most relevant peaks are annotated. The name, structure, migration time and GU values of all identified glycans are listed in Table 1. The exoglycosidase based glycan sequencing process is shown in FIG. 2, utilizing Sialidase A, ?-Galactosidase and ?-N-Acetyl Hexosaminidase, depicted by the corresponding traces. Sequence information was derived from the GU value shifts of the individual peaks as the result of the consecutive exoglycosidase treatments in (Szigeti & Guttman, 2017). As one can observe, all of the identified structures on the standard PSA sample were sialylated, emphasizing the surprising advantage of CE-LIF as a gentle separation technique being able to preserve sensitive glycan isomers. Orthogonal techniques, like MS often leads to de-sialylation due to in-source degradation (Kammeijer et al., 2017). Furthermore, the high resolution of CE-LIF was capable to readily differentiate the ?2,3- and ?2,6-sialylated isomers on mono- or multi-sialylated structures, which could be key for cancer diagnosis. Other PCa related alterations, like the degree of core fucosylation or branching could also be traceable, especially after exoglycosidase sequencing. Both, the number of identified glycans as well as the high resolution separation with reliably trackable ratios of the given structures proved the high potential of this workflow as a possible PCa diagnostic tool.

Expression and Purification of the aPSA Proteins

[0213] Several aPSA single domain antibody sequences were selected earlier by phage display using a variant library prepared from the variable domain of the heavy-chain antibody of dromedary immunized with PSA (Saerens et al., 2004). In order to be able to collect PSA from low concentration body fluids, we chose the two strongest PSA binder variants, namely N7 (K.sub.d=0.16 nM) and C9 (K.sub.d=4.7 nM) (FIG. 3). The sdAbs were expressed with fused histidine tags to facilitate easy immobilization for affinity based PSA capture from urine.

[0214] C9 and N7 aPSA variants contain 2 and 4 cysteines, respectively, which make proper protein folding less effective in the cytoplasm of standard E. coli strains (i.e. BL21 DE3 and its derivatives). Additionally, two cysteines from the N7 variant are located in the variable regions and supposedly form a stabilizing disulfide bond on the surface of the molecule. Periplasmic expression, which had been applied also for different aPSA variants (Saerens et al., 2004), while supporting disulfide bond formation, usually results in lower protein yields. For this reason, we decided to use Shuffle T7 Express cells for protein expression, which was an engineered E. coli B strain capable to promote disulfide bond formation in the cytoplasm. These cells can express the disufide bond isomerase DsbC, which promoted the correction of mis-oxidized proteins into their correct form (Bessette et al., 1999; Levy et al., 2001). As a result of applying our optimized production protocol, typical yield for the aPSA variants was 8-12 mg/L culture.

PSA Capturing Procedure

[0215] Preliminary experiment suggested that the limit of detection (LoD) was approximately 500 ng of standard PSA in 10 ?l sample for the CE analysis of released N-glycans. Taking the LoD value into consideration for selecting the optimal biological sample source, blood had to be excluded due to its very low PSA concentration. Standard PSA-ELISA tests of our urine samples resulted in an average concentration of 60 ng/ml with a maximum of 120 ng/ml and a minimum of 30 ng/ml, which was slightly lower comparing to literature data (Bolduc et al., 2007). Results varied on a high scale between the samples from different test subjects as well as from the same test subject but different donation dates. The PSA-ELISA method was developed for testing blood PSA level, thus a female urine sample was also processed as negative control, resulting in 0 ng/ml PSA. Female urine was also spiked with standard PSA for positive control, resulting in the expected concentration. Control results suggested that the urine matrix did not affect the accuracy of PSA-ELISA tests. Considering the lowest concentration, a minimum requirement was 150 ml of urine to reliably obtain sufficient quantity of PSA for capture and analysis.

[0216] To immobilize the aPSA sdAbs, Ni-IMAC microcolumns were chosen (FIG. 4/1a) as they showed strong affinity for the histidine tags of the nanobodies. In order to effectively introduce all of the biological matrix to the column, the volume of urine had to be reduced to the 1 ml scale (i.e., to fit in a 1000 ?L pipette tip based affinity columns). Simple evaporation of the water content of the samples was problematic as the higher temperature could lead to possible loss of sensitive sialic sugar structures and precipitation. Therefore, 10 kDa cut-off value filters were utilized (FIG. 4/1b), which could retain PSA effectively, while letting through the solvent and small contaminants (e.g., sugars). Changing the urinary matrix to buffer was beneficial in multiple levels. It contributed to avoid any possible undesirable effect on the sdAb capture due to the variable urinary pH as well as introduced imidazole as an inhibiting agent to prevent non-specific bindings of remaining (>10 kDa) urine components.

[0217] As capturing agent, both variants C9 and N7 of aPSA capturing capabilities were tested. Despite the different equilibrium binding constants, they resulted in equally high yields in terms of signal intensity in CE analysis, when their binding capacities were tested in preliminary experiments by standard PSA capture from buffer solution (data not presented). Eventually, our final choice was the C9 variant, as it provided slightly higher yields during both production and capture. Various methods were also tested during the elution step (FIG. 4/3) (denaturation, EDTA, etc.) and 500 mM imidazole proved to be the best for removing the nanobody-antigen complex from the column. Because E. coli is unable to glycosylate proteins, thus, sdAbs were not glycosylated, i.e., there was no need for their removal from the sample. However, changing the buffer for water was required after the elution (FIG. 4/4) as the high salt concentration interfered with the removal of excess labeling dye later in the sample preparation workflow.

[0218] The initial test of the capturing method was facilitated with standard PSA spiked into buffer solution. The glycan profile of captured PSA was corresponding to the standard PSA profile. To evaluate the matrix effect of urine, standard PSA was also spiked into and then captured from female urine, resulting similar yields in analytical signal compared to the spiked buffer, as well as, similar glycan profile as of the standard PSA, suggesting that sdAb were not affected negatively by the matrix. The results of concentrated and captured male urine were also corresponding with both spiked variants. To verify that all the resulted glycans were originated from urinary or spiked PSA, unspiked female urine was used as a control, in which case no glycan related peaks were detected. Each glycan profile is shown in FIG. 5. From concentrated and captured male urine 19 corresponding structure was identified and their peak area distribution was also similar to the standard. Please note that 20 cm effective length capillary was utilized for the separation of the captured urinary PSA glycans as we aimed the development of a fast, high-throughput method, which can be readily adapted into the clinic. No desialylated glycans were found, proving that both the analytical tool and the selective capture procedure provided sufficiently mild conditions to preserve sensitive sugar structures, e.g., sialic acids. As with the standard PSA, the ?2,3- and ?2,6-sialylated isomers were separated on both mono- and multi-sialylated structures, as well as, the core fucosylated or bisecting structures.

REFERENCES

[0219] Adamczyk, B., Tharmalingam, T., & Rudd, P. M. (2012). Glycans as cancer biomarkers. In Biochimica et Biophysica ActaGeneral Subjects (Vol. 1820, Issue 9, pp. 1347-1353). Elsevier. https://doi.org/10.1016/j.bbagen.2011.12.001 [0220] Arbabi-Ghahroudi, M. (2017). Camelid single-domain antibodies: Historical perspective and future outlook. Frontiers in Immunology, 8(Nov.), 1589. https://doi.org/10.3389/fimmu.2017.01589 [0221] Armbruster, D. A. (1993). Prostate-specific antigen: Biochemistry, analytical methods, and clinical application. In Clinical Chemistry. https://doi.org/10.1093/clinchem/39.2.181 [0222] Bessette, P. H., ?slund, F., Beckwith, J., & Georgiou, G. (1999). Efficient folding of proteins with multiple disulfide bonds in the Escherichia coli cytoplasm. Proceedings of the National Academy of Sciences of the United States of America. https://doi.org/10.1073/pnas.96.24.13703 [0223] Bolduc, S., Lacombe, L., Naud, A., Gr?goire, M., Fradet, Y., & Tremblay, R. R. (2007). Urinary PSA: A potential useful marker when serum PSA is between 2.5 ng/ml and 10 ng/ml. Journal of the Canadian Urological Association, 1(4), 377-381. https://doi.org/10.5489/cuaj.444 [0224] Carlsson, S. V., & Vickers, A. J. (2020). Screening for Prostate Cancer. In Medical Clinics of North America. https://doi.org/10.1016/j.mcna.2020.08.007 [0225] dos Santos Silva, P. M., Albuquerque, P. B. S., de Oliveira, W. F., Coelho, L. C. B. B., & dos Santos Correia, M. T. (2019). Glycosylation products in prostate diseases. In Clinica Chimica Acta. https://doi.org/10.1016/j.cca.2019.08.003 [0226] Ferens-Sieczkowska, M., Kowalska, B., & Kratz, E. M. (2013). Seminal plasma glycoproteins in male infertility and prostate diseases: Is there a chance for glyco-biomarkers? In Biomarkers. https://doi.org/10.3109/1354750X.2012.719035 [0227] Filella, X., Fern?ndez-Galan, E., Bonifacio, R. F., & Foj, L. (2018). Emerging biomarkers in the diagnosis of prostate cancer. In Pharmacogenomics and Personalized Medicine (Vol. 11, pp. 83-94). Dove Medical Press Ltd. https://doi.org/10.2147/PGPM.S136026 [0228] Gasteiger, E., Hoogland, C., Gattiker, A., Duvaud, S., Wilkins, M. R., Appel, R. D., & Bairoch, A. (2005). Protein Identification and Analysis Tools on the ExPASy Server. In The Proteomics Protocols Handbook. https://doi.org/10.1385/1-59259-890-0:571 [0229] Gebauer, M., & Skerra, A. (2019). Engineering of binding functions into proteins. Current Opinion in Biotechnology, 60, 230-241. https://doi.org/10.1016/j.copbio.2019.05.007 [0230] Gilgunn, S., Conroy, P. J., Saldova, R., Rudd, P. M., & O'Kennedy, R. J. (2013). Aberrant PSA glycosylationA sweet predictor of prostate cancer. In Nature Reviews Urology. https://doi.org/10.1038/nrurol.2012.258 [0231] Gilgunn, S., Murphy, K., St?ckmann, H., Conroy, P. J., Brendan Murphy, T., William Watson, R., O'kennedy, R. J., Rudd, P. M., & Saldova, R. (2020). Glycosylation in indolent, significant and aggressive prostate cancer by automated high-throughput n-glycan profiling. International Journal of Molecular Sciences. https://doi.org/10.3390/ijms21239233 [0232] Guttman, A., & Ulfelder, K. W. (1997). Exoglycosidase matrix-mediated sequencing of a complex glycan pool by capillary electrophoresis. Journal of Chromatography A, 781(1-2), 547-554. https://doi.org/10.1016/S0021-9673(97)00724-3 [0233] Kammeijer, G. S. M., Jansen, B. C., Kohler, I., Heemskerk, A. A. M., Mayboroda, O. A., Hensbergen, P. J., Schappler, J., & Wuhrer, M. (2017). Sialic acid linkage differentiation of glycopeptides using capillary electrophoresisElectrospray ionizationMass spectrometry. Scientific Reports. https://doi.org/10.1038/s41598-017-03838-y [0234] Kammeijer, G. S. M., Nouta, J., De La Rosette, J. J. M. C. H., De Reijke, T. M., & Wuhrer, M. (2018). An In-Depth Glycosylation Assay for Urinary Prostate-Specific Antigen. Analytical Chemistry. https://doi.org/10.1021/acs.analchem.7b04281 [0235] Kyselova, Z., Mechref, Y., A1 Bataineh, M. M., Dobrolecki, L. E., Hickey, R. J., Vinson, J., Sweeney, C. J., & Novotny, M. V. (2007). Alterations in the serum glycome due to metastatic prostate cancer. Journal of Proteome Research. https://doi.org/10.1021/pr060664t [0236] Levy, R., Weiss, R., Chen, G., Iverson, B. L., & Georgiou, G. (2001). Production of correctly folded fab antibody fragment in the cytoplasm of Escherichia coli trxB gor mutants via the coexpression of molecular chaperones. Protein Expression and Purification. https://doi.org/10.1006/prep.2001.1520 [0237] Lynne, C. M., Aballa, T. C., Wang, T. J., Rittenhouse, H. G., Ferrell, S. M., & Brackett, N. L. (1999). Serum and semen prostate specific antigen concentrations are different in young spinal cord injured men compared to normal controls. Journal of Urology, 162(1), 89-91. https://doi.org/10.1097/00005392-199907000-00022 [0238] M?sz?ros, B., J?rv?s, G., Farkas, A., Szigeti, M., Kov?cs, Z., Kun, R., Szab?, M., Cs?nky, E., & Guttman, A. (2020). Comparative analysis of the human serum N-glycome in lung cancer, COPD and their comorbidity using capillary electrophoresis. Journal of Chromatography B: Analytical Technologies in the Biomedical and Life Sciences. https://doi.org/10.1016/j.jchromb.2019.121913 [0239] Munkley, J., Mills, I. G., & Elliott, D. J. (2016). The role of glycans in the development and progression of prostate cancer. In Nature Reviews Urology. https://doi.org/10.1038/nrurol.2016.65 [0240] Muyldermans, S. (2021). A guide to: generation and design of nanobodies. FEBS Journal, 288(7), 2084-2102. https://doi.org/10.1111/febs.15515 [0241] Peracaula Mir?, R., Aleixandre Cerarols, R. N., Barrab?s Vera, S., Comet Batlle, J., De Llorens Duran, R., Fahey, R., Ferrer Batall?, M. L., Escorihuela, E., Ramirez Malag?n, M., & Rudd, P. M. (2017). IN VITRO METHOD FOR DIAGNOSING PROSTATE CANCER (Patent No. WO2017077162A1). https://worldwide.espacenet.com/patent/search/family/058662301/publication/WO2017077162A1?q=pn % 3DWO2017077162A1 [0242] Pihikov?, D., Kas?k, P., & Tkac, J. (2015). Glycoprofiling of cancer biomarkers: Label-free electrochemical lectin-based biosensors. In Open Chemistry (Vol. 13, Issue 1, pp. 636-655). De Gruyter Open Ltd. https://doi.org/10.1515/chem-2015-0082 [0243] Pinsky, P. F., Kramer, B. S., Crawford, E. D., Grubb, R. L., Urban, D. A., Andriole, G. L., Chia, D., Levin, D. L., & Gohagan, J. K. (2006). Prostate volume and prostate-specific antigen levels in men enrolled in a large screening trial. Urology. https://doi.org/10.1016/j.urology.2006.02.026 [0244] Reider, B., Jarvas, G., Krenkova, J., & Guttman, A. (2020). Separation based characterization methods for the N-glycosylation analysis of prostate-specific antigen. Journal of Pharmaceutical and Biomedical Analysis, 113797. https://doi.org/10.1016/j.jpba.2020.113797 [0245] Saerens, D., Kinne, J., Bosmans, E., Wernery, U., Muyldermans, S., & Conrath, K. (2004). Single domain antibodies derived from dromedary lymph node and peripheral blood lymphocytes sensing conformational variants of prostate-specific antigen. Journal of Biological Chemistry. https://doi.org/10.1074/jbc.M409292200 [0246] Saldova, R., Fan, Y., Fitzpatrick, J. M., Watson, R. W. G., & Rudd, P. M. (2011). Core fucosylation and ?2-3 sialylation in serum N-glycome is significantly increased in prostate cancer comparing to benign prostate hyperplasia. Glycobiology. https://doi.org/10.1093/glycob/cwql47 [0247] Shriver, Z., Venkataraman, G., Sasisekharan, R., & Sundaram, M. (2004). GLYCAN MARKERS FOR DIAGNOSING AND MONITORING DISEASE (Patent No. WO2004066808A2). [0248] ?krlec, K., ?trukelj, B., & Berlec, A. (2015). Non-immunoglobulin scaffolds: A focus on their targets. Trends in Biotechnology, 33(7), 408-418. https://doi.org/10.1016/j.tibtech.2015.03.012 [0249] Stern, L. A., Case, B. A., & Hackel, B. J. (2013). Alternative non-antibody protein scaffolds for molecular imaging of cancer. Current Opinion in Chemical Engineering, 2(4), 425-432. https://doi.org/10.1016/j.coche.2013.08.009 [0250] Szigeti, M., & Guttman, A. (2017). Automated N-Glycosylation Sequencing of Biopharmaceuticals by Capillary Electrophoresis. Scientific Reports, 7(1). https://doi.org/10.1038/s41598-017-11493-6 [0251] Tkac, J., Bertok, T., Hires, M., Jane, E., Lorencova, L., & Kasak, P. (2019). Glycomics of prostate cancer: updates. In Expert Review of Proteomics. https://doi.org/10.1080/14789450.2019.1549993 [0252] Tkac, J., Gajdosova, V., Hroncekova, S., Bertok, T., Hires, M., Jane, E., Lorencova, L., & Kasak, P. (2019a). Prostate-specific antigen glycoprofiling as diagnostic and prognostic biomarker of prostate cancer. In Interface Focus. https://doi.org/10.1098/rsfs.2018.0077 [0253] Tkac, J., Gajdosova, V., Hroncekova, S., Bertok, T., Hires, M., Jane, E., Lorencova, L., & Kasak, P. (2019b). Prostate-specific antigen glycoprofiling as diagnostic and prognostic biomarker of prostate cancer. In Interface Focus (Vol. 9, Issue 2). Royal Society Publishing. https://doi.org/10.1098/rsfs.2018.0077 [0254] Vashist, S. K., & Luong, J. H. T. (2018). Antibody immobilization and surface functionalization chemistries for immunodiagnostics. In Handbook of lmmunoassay Technologies: Approaches, Performances, and Applications (pp. 19-46). Elsevier. https://doi.org/10.1016/B978-0-12-811762-0.00002-5 [0255] Yoneyama, T., Ohyama, C., Tobisawa, Y., Ishikawa, T., Date, M., & Nakamura, K. (2019). METHOD FOR DETERMINING PROSTATE CARCINOMA (Patent No. WO2019221279A1). [0256] Yoneyama, T., Tobisawa, Y., Ishikawa, T., Kurosawa, T., & Nakamura, K. (2017). Prostate carcinoma determination method (Patent No. WO2017130578A1). https://worldwide.espacenet.com/patent/search/family/059397997/publication/WO2017130578A1?q=pn % 3DWO2017130578A1 [0257] Yoon, K. Y., Son, Y. S., Hong, H. S., & Choi, H. (2011). KIT FOR DIAGNOSING PROSTATE CANCER AND DIAGNOSIS METHOD (Patent No. WO2011074802A2). https://worldwide.espacenet.com/patent/search/family/044167817/publication/WO2011074802A2?q=pn % 3DWO2011074802A2 [0258] Zhang, H., Meany, D. L., Chan, D. W.-Y., Zhang, Z., Li, Y., & Sokoll, L. J. (2010). DETECTION OF PROSTATE CANCER USING PSA GLYCOSYLATION PATTERNS (Patent No. WO2010011357A2).