Human Umbilical Cord Blood Serum–derived α-Secretase

Alzheimer’s disease (AD) is an age-related disorder that affects cognition. Our previous studies showed that the neuroprotective fragment of amyloid procurer protein (APP) metabolite, soluble APPα (sAPPα), interferes with β-site APP-cleaving enzyme 1 (BACE1, β-secretase) cleavage and reduces amyloid-β (Aβ) generation. In an attempt to identify approaches to restore sAPPα levels, we found that human cord blood serum (CBS) significantly promotes sAPPα production compared with adult blood serum (ABS) and aged blood serum (AgBS) in Chinese hamster ovary cells stably expressing wild-type human APP. Interestingly, CBS selectively mediated the α-secretase cleavage of human neuron-specific recombinant APP695 in a cell-free system independent of tumor necrosis factor-α converting enzyme (TACE; a disintegrin and metalloproteinase domain-containing protein 17 [ADAM17]) and ADAM. Subsequently, using 3-step chromatographic separation techniques (i.e., diethylaminoethanol, size-exclusion, and ion-exchange chromatography), we purified and ultimately identified a CBS-specific fraction with enhanced α-secretase catalytic activity (termed αCBSF) and found that αCBSF has more than 3,000-fold increased α-secretase catalytic activity compared with the original pooled CBS. Furthermore, intracerebroventricular injection of αCBSF markedly increased cerebral sAPPα levels together with significant decreases in cerebral Aβ production and abnormal tau (Thr231) phosphorylation compared with the AgBS fraction with enhanced α-secretase activity (AgBSF) treatment in triple transgenic Alzheimer’s disease (3xTg-AD) mice. Moreover, AgBSF administered intraperitoneally to transgenic mice with five familial Alzheimer’s disease mutations (5XFAD) via an osmotic mini pump for 6 weeks (wk) ameliorated β-amyloid plaques and reversed cognitive impairment measures. Together, our results propose the necessity for further study aimed at identification and characterization of α-secretase in CBS for novel and effective AD therapy.


Introduction
The neuropathological hallmarks of Alzheimer's disease (AD) that differentiate it from other types of dementia include extracellular b-amyloid plaques composed largely of amyloid-b (Ab) peptides 1 and intracellular neurofibrillary tangles (NFTs) composed of the hyperphosphorylated microtubule-associated protein tau 2 . Successive cleavage of amyloid procurer protein (APP) by band g-secretases produces Ab peptides of variable length (Ab x-40, 42 ), soluble APPb, membrane-bound b-C-terminal fragment (b-CTF, C99), and APP intracellular cytoplasmic/C-terminal domain (AICD) 3 . The Ab peptide fragments, which accumulate as plaques in the brain, induce neuroinflammation, synaptic dysfunction, and neuronal cell death that affects cognitive function 4 . In contrast, most of the APP is cleaved by aand g-secretases that not only preclude Ab generation but also produce a secreted soluble APPa (sAPPa), membrane-bound a-CTF (a-CTF, C83), P3 peptide, and AICD 5 . Overall, sAPPa has been shown to be involved in numerous physiological functions in the brain, which appear to be interrupted in AD. Several studies have shown that neurotrophic fragment sAPPa not only prevents Ab generation 6 and tau phosphorylation 7 but is also known to be a neuroprotective APP metabolite including but not limited to proliferation, neurite outgrowth, and long-term potentiation [8][9][10] . Thus, we hypothesized that therapeutic interventions or approaches that have potential to produce sAPPa markedly could improve AD pathology and cognitive function.
Several studies have shown that human umbilical cord blood cells (HUCBCs) have therapeutic potential in numerous age-related neuroinflammatory conditions including AD. In line with those studies, we showed that single as well as multiple low-dose infusion of HUCBC significantly reduced amyloidogenic APP processing, Ab and b-amyloid plaque accumulation, glial neuroinflammation, and cognitive impairments in preclinical AD mouse models 11,12 . Additionally, HUCBC treatment changed microglial phenotypes from pro-inflammatory to anti-inflammatory, increased microglial Ab phagocytosis, increased anti-inflammatory cytokines in the brain (i.e., interleukin-10, transforming growth factor b1, and nerve growth factor), and reduced CD40 receptor-CD40 ligand (CD40-CD40L) interaction that is important for Ab-induced pro-inflammatory microglial activation 13 . To identify the specific HUCBC responsible for this neuroprotective effect, we found that cord bloodderived monocyte reduces b-amyloid pathology and improves cognition with much more effectively than monocyte-deficient cord blood in AD mouse model 13 . In line with the findings of above studies, several recent reports underscored the role of young blood and/or plasma in aging and age-associated neurodegenerative conditions. Among those, Wyss-Coray and other labs have reported that exposing old mice to a young systemic environment by parabiosis increased synaptic plasticity, improved pathology, and behavioral recovery such as contextual fear conditioning and spatial learning in old mice. More interestingly, they also found that it is not the blood cells rather the soluble factors that are getting into the mice brain. They pooled plasma from young mice as well as from young human and injected into old mice, which successfully rejuvenated old mice brain structure and cognition tested by Barnes maze memory test [14][15][16][17][18] . In a follow-up study, they showed that human cord blood plasma (CBP) as well as plasma enriched in tissue inhibitor of metalloproteinases 2 improves synaptic plasticity and hippocampaldependent cognitive function in old mice 19 .
Based on our preliminary laboratory findings, we hypothesized that human cord blood serum (CBS) possesses novel APP-specific a-secretase-like enzyme, reflected by marked increase in sAPPa level. As CBS contains many different small molecules, growth factors, proteins, inhibitors, hormones, enzymes, and other unknown substances, we also hypothesized that infusion of characterized CBS fraction will ameliorate AD-like pathology and cognitive impairments in mouse models. Here, we show that CBS markedly enhanced the level of sAPPa in Chinese hamster ovary (CHO) cells stably expressing wild-type human APP (CHO/APPwt cells) as well as mediated a-secretase cleavage of human neuron-specific APP 695 (fAPP 695 ) in a cellfree system, which effects are not seen with normal adult or aged blood serum (ABS or AgBS). Additionally, we have been successfully able to characterize a CBS fraction with enhanced a-secretase-like catalytic activity (refer to aCBSF) using sequential diethylaminoethanol (DEAE)-affinity column, size-exclusion, and anion-exchange chromatographic fractionation processes. Moreover, we found that aCBSF infusion increased sAPPa levels, decreased Ab production/ b-amyloid plaque formation, prevent neuronal loss and abnormal tau (Thr 231 ) phosphorylation in the cortex, and improved cognitive function in Alzheimer's mouse models.
Our findings indicate that aCBSF holds immense therapeutic potential for treatment of AD.

Reagents and Antibodies
CBS was obtained from Lee Biosolutions (St. Louis, MO, USA, and human umbilical CBP was obtained from STEM-CELL Technologies Inc. (Vancouver, British Columbia, Canada). Human cord blood is aspirated from the umbilical cord vein into a cord blood collection bag containing citratephosphate-dextrose as an anticoagulant. Individual lot of CBP is prepared from a single cord blood sample. Three to five different lots of CBP samples were pooled in as "pooled CBP." CBP is separated from umbilical cord blood centrifugation at 3,500 rpm for 5 to 10 min. CBP is aliquoted and frozen at À20 C first and then transferred to a À80 C freezer after 24 h at 4 C. There is no placement into À80 C for a snap freeze. Frozen CBP is not heat inactivated. No analysis was performed to determine the number of platelets in each sample; therefore, the plasma cannot be specifically characterized as "low-platelet" or "platelet-poor" plasma. Frozen CBP samples were thawed in a 37 C water bath before being used in experiment. CBS was collected from umbilical cord blood; and it is the blood that remains in the placenta and in the attached umbilical cord after the cord has been detached from the newborn at the time of childbirth. CBS is separated from umbilical cord blood by allowing it to clot for 5 to 10 h in red top tubes with no anticoagulation followed by centrifugation at 3,500 rpm for 5 to 10 min at 4 C. CBS sample was passed through a filter membrane with a pore size of 0.22 mm. Individual CBS was prepared from a single sample. More than 10 serum samples of CBS were pooled in as "pooled CBS." Normal human adult blood serum (ABS, 25 to 30 years old) and AgBS (>75 years old) as well as their plasma (ABP and AgBP, respectively) were obtained from Florida Blood Services (Tampa, FL, USA).  Antibodies include specific anti-sAPPa monoclonal antibody (2B3; IBL, Minneapolis, MN, USA, Cat# 11088 RRID:  AB_494690), anti-APP C-terminal polyclonal antibody  (pAb751/770; EMD Millipore, La Jolla, CA,

Cell Culture
CHO cell line with stable expression of human wild-type APP (CHO/APPwt) was a generous gift from Drs. Stefanie Hahn and Sascha Weggen (University of Heinrich Heine, Düsseldorf, Germany). At the beginning of the experiment, CHO/APPwt cells were genotyped and confirmed the genetic makeup. The cells were cultured in Dulbecco's modified Eagle's medium (DMEM) with 10% fetal bovine serum (FBS), 1 mM sodium pyruvate, and 100 U/mL of penicillin/ streptomycin. For treatment, the cells were plated in a 24-well plate at 2 Â 10 5 cells/well for overnight incubation, washed and treated with CBS (0% to 10%), inact CBS (5%), ABS (0% to 10%), AgBS (0% to 10%), or aCBSF (0% to 1%) in DMEM. After treatment, supernatants were collected and the cells were washed with ice-cold phosphate-buffered saline (PBS) 3X and lysed with cell lysis buffer (20 mM Tris, pH 7.5, 150 mM NaCl, 1 mM EDTA, 1 mM EGTA, 1% v/v Triton X-100, 2.5 mM sodium pyrophosphate, 1 mM b-glycerophosphate, 1 mM Na 3 VO4, 1 mg/mL leupeptin, and 1 mM phenylmethane sulfonyl fluoride; Cell Signaling Technology, Danvers, MA, USA). Both cell supernatants and lysates were used for sAPPa analysis by ELISA. For immunoprecipitation, 100 ng of fAPP 695 was incubated with aCBSF at 37 C for 1 h, and then the sAPPa/aCBSF-derived immune complex was immunoprecipitated using 2B3 sAPPa-specific antibody or anti-DDK antibody. The supernatants were then collected and used for treatment of CHO/APPwt cells.

CBS Fractionation
Next, in order to purify and characterize the a-secretase in CBS or AgBS, the Econo-Pac Serum IgG Purification Kit, and 10DG columns (Bio-Rad, Philadelphia, PA, USA) were initially employed to remove highly abundant IgG and salts. The desalted serum was applied to DEAE Affi-Gel Blue columns, and residual IgG was eluted according to the instructions. Then, 20 additional protein fractions were collected by eluting with an ionic strength gradient of NaCl buffer ranging from 0.1 M to 2.0 M. The remaining proteins on the column were eluted by the regeneration buffer included in the kit and collected as the regeneration fraction. The 0.8 M NaCl-eluted protein fractions were combined together and sent to Moffitt Cancer Center Protein Purification Core (Tampa, Fl, USA) for further separation by sizeexclusion chromatography, employing analytic Superdex 200 columns and eluting with PBS, and ion-exchange chromatography, employing Q-Sepharose columns and eluting with 500 mM NaCl, 50 mM Tris, pH 7.6. The final enzyme containing fractions was exchanged to PBS by Ultracel-10 membranes (10 kDa, Merck Millipore, Billerica, MA, USA) for further experimentation and referred to as aCBSF or AgBSF.

Animal Models
Both 5XFAD (MMRC Stock No: 34840-JAX; RRID IMSR_JAX: 006554) and 3xTg-AD (MMRC Stock No: 34830-JAX) mice of male and female were purchased from Jackson Laboratory (Bar Harbor, Maine, USA). In this preclinical study to investigate whether CBS fractionation changes AD-like pathology and associated behavioral deficits, 5-month-old 5XFAD mice were used that harbor 5 mutations ( 20 and rapidly develop AD-like pathology including accumulation of high levels of extracellular b-amyloid plaques, neurodegeneration, and behavioral impairments. In order to investigate whether CBS fractionation administration changes both Ab and tau phosphorylation, 3xTg-AD mice which harbor presenilin-1 (PS1/M146V), APP (KM670/671NL), and tau (P301L) mutants were used. These mice progressively develop b-amyloid and NFT pathology, which potentially synergize to accelerate neurodegeneration at the age of 6 months (6 mo) 21 . At the beginning of the experiment, all mice were confirmed as mutant by polymerase chain reaction. One male and 4 female mice were housed in a single cage separately. All animal experiments were performed in accordance with the guidelines of the National Institutes of Health and were approved by University of South Florida (USF) Institutional Animal Care and Use Committee (IACUC reference number: IS00000438). Transgenic mice used for aging studies may exhibit signs such as ruffled hair coat, hair loss, excessive weight gain, and/or loss. When one of these signs observed, mice were monitored more closely and weighed twice weekly. Mice exhibited multiple clinical signs or showing >20% weight loss were excluded from the study. As per our previous practice, if a mouse appears overtly sick or in pain as indicated by ruffled, matted, or dull hair; hunched back or head pressing; failure to move about the cage; failure to respond to stimuli, rapid, shallow, labored breathing, twitching or trembling; or failure to experience seizure, a veterinarian was consulted in order to ensure timely intervention and treatment or removal from the study. All mice were maintained on a 12-h light/ 12-h dark cycle at ambient temperature and humidity and housed in the Morsani College of Medicine Animal Facility at the USF with ad libitum access to food and water.

Stereotaxic Intracerebroventricular (i.c.v.) Injection
In order to determine whether aCBS fraction could modify Ab and tau pathology, cohorts of 17 (n ¼ 17, 9\/8_) triple transgenic 3xTg-AD mice were arbitrary anesthetized with isoflurane (2% to 3% induction, 1% maintenance). After reflexes were checked to ensure that mice were unconscious, they were positioned on a stereotaxic frame (Stoelting's Lab Standard™, Wood Dale, IL, USA) with ear bars positioned and jaws fixed to a biting plate. The axis coordinates were taken from a mouse brain atlas, and the needle of a Hamilton microsyringe was implanted into the left lateral ventricle delimited from the stereotaxic coordinates (coordinates relative to bregma: À0.6 mm anterior/posterior, þ1.2 mm medial/lateral, and À3.0 mm dorsal/ventral) using the stereotaxic device. aCBSF (0.5 mg/mouse, n ¼ 6, 3\/3_), AgBSF (0.5 mg/mouse, n ¼ 6, 3\/3_), and PBS (1 mL/mouse, n ¼ 5, 3\/ 2_) were administered at 1 mL/min. After administration, the syringe was removed slowly to prevent bleeding and further brain damage. The lesions were closed with 1 to 2 staples and observed until anesthesia had cleared. Seventy-two hour after the i.c.v. injections, animals were killed with isoflurane, then transcardially perfused with ice-cold PBS, and brains were harvested for biochemical, histochemical, and immunohistochemical analyses.

Intraperitoneal (i.p.) Administration With an Osmotic Mini Pump
Mice were labeled using tail tattooing by veterinarian who was blinded about the entire experiment. In order to determine whether CBS fractionation changes AD-like pathology and associated behavioral deficits, cohort of even-number labeled 5XFAD mice was randomly assigned to 2 experimental groups of 6 mice each, receiving aCBSF (n ¼ 6, 3\/3_) or AgBSF (n ¼ 6, 3\/3_) treatment by an Alzet ® osmotic mini pump (Alzet 2004, DURECT Corporation, Cupertino, CA, USA). A third group of wild-type (WT) control mice received aCBSF (n ¼ 6) through the same administration route. Mice were briefly anesthetized with isoflurane as described previously, an area of the abdomen was shaved, a 1-cm abdominal incision was made and an Alzet ® osmotic mini pump was filled with 100 mL of CBSF, or AgBSF was implanted i.p.. The pump delivered these fractionated sera at a constant rate of 0.15 mL/h for 6 wk, yielding a treatment dose of 1 mg/kg/day or 30 mg/ mouse/day. At the end of 4-to 5-weeks (wk) treatment (6 mo of age), cognitive evaluations were conducted in these mice with our established behavioral battery. After 6 wk treatment, mice were killed with isoflurane, then transcardially perfused with ice-cold PBS, and brains were removed to assess b-amyloid plaque pathology.

Behavioral Assessments
Novel object recognition test. Novel object recognition is based on the spontaneous tendency of a mouse to explore a new object compared with an old object. At first, during the habituation phase (day 1), each mouse was acclimatized with the testing apparatus box for 10 min. Next, during the training day (day 2), each mouse was familiarized with 2 similar objects (4 cm Â 4 cm Â 4 cm) for 10 min. Finally, during the testing day (day 3), one of the objects was replaced with a new object and tested for 10 min. The amount of time spent exploring the new and old objects during the test phase was quantified by video tracking (ANY-Maze; Stoelting's Lab Standard™, Wood Dale, IL, USA) and provides an index of recognition memory. Discrimination index was calculated as the frequency of exploration of new versus original objects.
Y-maze test. Y-maze test was performed as described previously 22 . This task was used to assess basic mnemonic processing by spontaneous percentage alternation and exploratory activity of mice placed into a black Y-maze. The arms of this maze were 21-cm long and 4-cm wide with 40cm high walls. Each mouse was placed in one of the arms and allowed one 5-min trial of free exploration of the 3 alleys in the maze. The numbers of total arm choices and sequence of arm choices were recorded.

Histochemical and Immunohistochemical Analyses
Mice were euthanized with isoflurane and then transcardially perfused with ice-cold PBS. Brains were rapidly isolated, and one hemisphere was frozen immediately in liquid nitrogen and stored at À80 C. For molecular analysis, brain hemispheres were sonicated in radioimmunoprecipitation assay buffer (Cell Signaling Technology, Danvers, MA, USA), centrifuged at 14,000 rpm for 1 h at 4 C, and supernatants were isolated for WB analyses. The other hemisphere was placed in 4% paraformaldehyde for cryostat sectioning. The 25-mm free-floating coronal sections were collected and stored in PBS with 100 mM sodium azide at 4 C. Immunohistochemical staining was performed using various primary antibodies in conjunction with the VECTASTAIN Elite ABC kit (Vector Laboratories, Burlingame, CA, USA) coupled with diaminobenzidine substrate. A set of sections without adding primary antibody were used as negative staining control. Sections were also stained with Congo red dye and Thioflavin-S fluorescence dye for detecting fibrillary Ab species as described previously 23,24 . Images of five 25-mm sections (150 mm apart) through hippocampus and neocortex were captured, and a threshold optical density was obtained that discriminated staining from background. Data are reported as percentage of immunolabeled area captured (positive pixels divided by total pixels captured). Quantitative image analysis was performed by a single examiner (T.M.) blinded to sample identities.

WB Analysis and ELISA
WB analyses and quantification were performed as previously described 25 . Briefly, the proteins from the cell-free suspensions, cell lysates, and homogenized tissue were electrophoretically separated using 10% bicine/Tris gel (8 M urea) for proteins less than 5 kDa or 10% Tris/sodium dodecyl sulfate (SDS) gels for larger proteins. Electrophoresed proteins were transferred to nitrocellulose membranes (Bio-Rad), washed, and blocked for 1 h at room temperature in Tris-buffered saline containing 5% (w/v) nonfat dry milk (TBS/NFDM). After blocking, membranes were hybridized overnight with various primary antibodies, washed, and incubated for 1 h with the appropriate horseradish peroxidase-conjugated secondary antibody in TBS/NFDM. Blots were developed using the luminol reagent (Thermo Fisher Scientific, Waltham, MA, USA). The sAPPa ELISA (IBL) was performed according to manufacturer's instruction.

Statistical Analysis
Comparison between 2 groups were performed by Student's t-test analysis. For more than 2 groups, one-way analysis of variance followed by Least Significance Difference (LSD) post hoc analysis was used to compare each other for statistical significance. The a was set at p < 0.05 for all analyses.
The significance level of P value was set at < 0.05 for all analyses. All the mice experiment were repeated 3 times in parallel to attain the above significant difference. Data are expressed as mean + standard error of the mean. The statistical package for the social sciences released by IBM SPSS Version 23.0 (IBM, Armonk, NY, USA) was used for all data analyses.

CBS Dose-Dependently Promotes a-cleavage in CHO/APPwt Cells
Our previous studies indicate that both single and multiple low-dose infusions of HUCBC as well as HUCBC-derived monocytes can significantly reduce b-amyloid plaques and cognitive impairments in AD mouse models. Having shown that HUCBC can reduce AD pathology, we next set out to determine whether human umbilical-derived CBS could also reduce b-amyloid pathology through alteration of APP processing. CHO/APPwt cells were treated with different concentrations (0% to 10%, 6 different doses) of CBS, ABS, or AgBS for 4 h ( Fig. 1A and B, left panel). The conditioned media were collected and subjected to sAPPa ELISA, and also sAPPa WB analysis using 2B3 sAPPa-specific antibody. CBS dose-dependently promoted sAPPa levels with greater than that elicited by ABS and AgBS. Similarly, CHO/APPwt cells were treated with 5% CBS, ABS, or AgBS for 6 different time points (0 to 4 h, Fig. 1A and B, right panel). CBS time-dependently promoted sAPPa levels with greater than that elicited by ABS and AgBS. To see whether the factor present in the serum mediating a-secretase activity is proteinaceous in nature, we treated CHO/ APPwt cells with heat inactivated serum (inact CBS) for 4 h. As expected, heat inactivation limited the sAPPa producing capacity of CBS, as shown by ELISA (Fig. 1C, upper panel) and WB (Fig. 1C, lower panel). Therefore, CBS possesses a-secretase, reflecting sAPPa level in a dose-and time-dependent fashion and the factor mediating this activity is heat-labile and most likely a protein. These results indicate that FBS also contains a heat sensitive a-secretase.
CBS Mediates a-Cleavage of Neuron-specific APP 695 Independent of ADAM Activity Next, we tested whether the a-secretase in CBS is mediated by TACE (ADAM17) or ADAM. Human recombinant fAPP 695 (100 ng)-tagged with C-terminal MYC/DDK was incubated with CBS, inactivated CBS (inact CBS), or AgBS at 37 C for 5 h in the presence or absence of different inhibitors (PI cocktail [1X], TAPI-0 [1 mM], or ADAM inhibitor [GM6001, 1 mM]; Fig. 2A). The reaction mixtures were subjected to sAPPa WB analysis using 2B3 sAPPa-specific antibody ( Fig. 2A, upper panel) and total APP analysis using 6E10 anti-Ab 1-17 antibody ( Fig. 2A, lower panel). PI cocktail significantly limited CBS-derived a-secretase catalytic activity, as reflected by sAPP 695 level, but this activity was not limited by TACE or ADAM inhibitors ( Fig. 2A, upper  panel). In addition, fAPP 695 (100 ng) was incubated with 5% CBS, FBS, or inactivated CBS for 1, 5, or 24 h. CBS a-secretase increased the level of sAPP 695 in a timedependent manner, measured by 2B3 antibody (Fig. 2B, upper panel). As shown, the level of sAPP 770 represents endogenous sAPPa.

Removal of High-and Low-abundance Proteins Increases Activity of CBS a-secretase
To purify and ultimately identify the target protein/complex mediating CBS a-secretase catalytic cleavage, 3-step chromatographic separation techniques were employed. Initially, removal of highly abundant immunoglobulins and desalting were performed using Bio-Rad Econo-Pac Serum IgG Purification Kit and 10DG columns. Desalted CBS was then applied to DEAE Affi-Gel Blue columns to completely remove IgGs and collect 20 additional protein fractions and eluted with increasing strengths of NaCl buffer. CHO/ APPwt cells were treated with each fraction for 2 h to determine a-secretase by ELISA. In addition, unfractionated whole and desalted CBS positive control as well as PBSnegative control was used to treat cells. These sequential approaches significantly increased a-secretase, with the fractions showing the highest a-secretase catalytic activity, as reflected by sAPPa level eluting around 0.6 to 0.9 M NaCl concentrations (Fig. 3A). As shown in Fig. 3C, 0.7 to 0.9 M NaCl fractions from 10 CBS lots increased sAPPa levels at least with maximum 5-fold higher than whole CBS. In addition, the fractionated and whole CBS was run in SDS- Conditioned media were subjected to sAPPa ELISA (Fig. 1C, top panel) and WB analyses (Fig. 1C, bottom panel) with 2B3 antibody. Data are presented as mean (+SD) of sAPPa produced (ng/mg or ng/mL) from 5 independent experiments in triplication. Human umbilical cord blood plasma (CBP) produced similar results (data not shown). APP a-secretase activity of pooled CBS or CBP or individualized CBS or CBP was similar (data not shown).
polyacrylamide gel electrophoresis (PAGE), demonstrating numerous proteins remaining in each CBS fraction (Fig. 3C, right upper panel). Therefore, we selected the 0.8 M NaCleluted fraction for further purification. As shown in Fig. 3B, the level of total protein concentration is represented in mg/ mL in CBS fraction.

Further Purification of CBS a-Secretase Using Size-exclusion and Anion-exchange Chromatography
The 0.8 M NaCl-eluted fraction of CBS was subjected to size-exclusion chromatography using Superdex 200 prep grade columns (XK 16/40, GE Healthcare, PA, USA) packed with cross-linked agarose and dextran. The mobile phase was 20 mg/mL acetone in distilled water, and the detection was performed at UV280 nm. Approximately 100 mg of protein from the 0.8 M NaCl fraction was applied to the column, and 48 fractions were eluted with PBS. The catalytic activity of a-secretase was greatly enhanced, as tested on CHO/APPwt cells by measuring sAPPa production. Fractions #11 and 12 produced sAPPa > 5-fold higher compared with the original 0.8 M NaCl-eluted fraction as well as all other fractions, as determined by WB (upper panel) and ELISA (Fig. 4A, lower panel). To confirm the enhancement of a-secretase, we determined the sAPPa in fractions #8 to 18 along with the original 0.8 M NaCl-eluted fraction from 3 different CBS lots. Fractions #10 to 13 showed a-secretase activity 15-fold more than the original 0.8 M NaCl-eluted fraction, as measured by ELISA (Fig. 4C). As determined by SDS-PAGE, the molecular mass of the #10 to 13 fractions was 177 to 275 kDa (Fig. 4C).
To examine the charge of a-secretase protein/complex in CBS, the size-exclusion fractions containing the highest a-secretase catalytic activities, as reflected by sAPPa levels (#10 to 13), were further subjected to anion-exchange chromatography using Q-Sepharose columns. Proteins from the size-exclusion fractions were applied to the column and 82 fractions were eluted with 0.5 M NaCl. Fractions #53 to 56 showed > 8-fold higher a-secretase catalytic activity, as reflected by sAPPa level than the original size-exclusion fraction, as determined in CHO/APPwt cells by WB and ELISA (fractions #2 to 4 in Fig. 5A, upper and lower panels, respectively). To further compare the enzymatic activity of CBS samples before and after anion-exchange chromatography, we collected fractions #1 to 5 and #23 from 3 different samples and determined the sAPPa level in each indirect  (7), CBS with PI (8), CBS with TAPI-0 (9), CBS with GM6001 (10), and AgBS (11). Lanes 12 to 14 represent (lanes 12 to 14; PI (12), TAPI-0 (13), and GM6001 (14) inhibitor and substrate control, respectively, without any serum sample. The reaction mixtures were subjected to soluble amyloid precursor protein a (sAPPa) Western blot (WB) analysis using 2B3 antibody (top panel) and total APP using 6E10 (an anti-Ab 1-17 antibody; lower panel). sAPP 770 refers to the endogenous a-secretase cleavage product of CBS or AgBS, whereas sAPP 695 refers to the a-secretase cleavage product of fAPP 695 . (B) 100 ng of fAPP 695 was incubated with 5% CBS for 1, 5, or 24 h, or 5% FBS or inact. FBS for 24 h, and then subjected to sAPPa and total APP WB analysis using 2B3 (top panel) and 6E10 (lower panel), respectively. sAPP 770 refers to the endogenous a-secretase cleavage product of CBS or AgBS, whereas sAPP 695 refers to the a-secretase cleavage product of fAPP 695 . measurement of CBS a-secretase activity. Combined fractions #2 to 4 produced sAPPa > 50-fold higher than the original eluted (#23) fraction, as measured by ELISA (Fig.  5C). Combined fractions #2 to 4, referred to hereafter as aCBSF, were therefore used for further analysis.

aCBSF Promotes Nonamyloidogenic APP Processing
Human recombinant fAPP 695 was incubated with 5 different concentrations (0% to 1%) of aCBSF at 37 C for 2 h. The reaction mixtures were subjected to sAPPa WB analysis using 2B3 antibody as well as fAPP and a-CTF analyses using pAPP751/770 antibody (an anti-APP C-terminal antibody; Fig. 6A). This analysis showed that aCBSF increases sAPPa as well as a-CTF fragments and decreases (fulllength) holo APP with increasing doses.
In order to confirm that aCBSF mediates novel a-secretase independent of TACE or ADAM, fAPP 695 was incubated with aCBSF in the absence or presence of ADAM (GM6001, 1 mM) or TAPI-0 (1 mM) for 1 h. aCBSF treatment alone increased the levels of APP-processing fragments such as sAPPa and a-CTF and decreased the level of holo APP as determined by WB, whose effects did not alter significantly by GM6001 and TAPI-0 combined treatment with aCBSF (Fig. 6B). In addition, TACE and ADAM10 enzymatic activities of aCBSF were measured by TACE and ADAM10 cleavage activity kits. TACE (25 mg/mL) in the presence or absence of TAPI-0 (1 mM) and ADAM10 (50 mg/mL) in the presence or absence of ADAM inhibitor (GM6001, 1 mM) were included as positive controls. Results suggest that aCBSF has very little TACE or ADAM10 activity (Fig. 6C and D).

Immunoprecipitation of fAPP 695 /CBS Specifically Limits APP a-secretase Cleavage
To determine whether immunoprecipitation could specifically limit a-secretase of aCBSF, 100 ng of fAPP 695 was incubated with 0.125 mg of aCBSF at 37 C for 1 h, and the sAPPa/ aCBSF immune complex was immunoprecipitated using anti-DDK antibody, 2B3, or nonspecific IgG. CHO/APPwt cells were treated in the FBS free condition for 2 h with the supernatants collected from the immune complex or PBS as reference control, and then conditioned media were analyzed by sAPPa WB using anti-N-terminal APP antibody (22C11; Fig.  7A) and sAPPa ELISA (Fig. 7B) to determine a-secretase in CBS. Immunoprecipitation of the sAPPa/aCBSF with anti-DDK antibody significantly reduced a-secretase activity of aCBSF, indicating that immunoprecipitation limits the CBS-mediated APP a-secretase cleavage. In contrast, immunoprecipitation of sAPPa/aCBSF with 2B3 did not reduce CBS a-secretase, indicating that aCBSF does not form an immune complex with sAPPa. In addition, there was no notable or significant difference in sAPPa production elicited by 0.5% supernatant from aCBSF IP with control IgG and 2.5% (equivalent to 0.5% Super) aCBSF alone (data not shown).
aCBSF Reduces b-cleavage, Promotes a-cleavage of APP, and Stabilizes Tau Phosphorylation in 3xTg-AD Mice To test whether aCBSF suppresses b-site APP-cleaving enzyme 1 (BACE1)-mediated APP processing in vivo, Fig. 3. Fractionation of amyloid precursor protein (APP)-specific a-secretase activity in cord blood serum (CBS). To purify and eventually identify the a-secretase activity in CBS, the Econo-Pac Serum IgG Purification Kit (Bio-Rad, Philadelphia, PA, USA) was initially employed to remove highly abundant IgG. CBS was then desalted using Econo-Pac 10DG columns. The desalted serum was applied to DEAE Affi-Gel Blue columns to remove residual IgG and collect 20 additional protein fractions, by eluting with an increasing ionic strength gradient of NaCl buffer ranging from 0.1 M to 2.0 M. The remaining proteins on the column were eluted by the regeneration buffer included in the kit and collected as the regeneration fraction (Reg). (A) Chinese hamster ovary cells stably expressing wild-type human APP (CHO/APPwt) cells were cultured in 24-well plates and treated with 10 mL of each protein fraction for 2 h. Conditioned media were then collected and analyzed by soluble amyloid precursor protein a (sAPPa) Western blot (upper panel) and ELISA (lower panel). 10 mL CBS, desalted CBS, and phosphatebuffered saline (PBS; Ctrl) were included under the same cell culture conditions as positive and negative controls, respectively. Cell lysates were also prepared from each fraction-treated cell culture as an additional reference to evaluate sAPPa production levels (data not shown). (B) Protein concentration of each fraction. (C) CHO/ APPwt cells were treated with the 0.6 to 1.0 M NaCl-eluted fractions from 10 different CBS lots, as well as whole CBS and PBS (Ctrl), for 2 h and the conditioned media were collected for sAPPa ELISA. The results were presented as mean (+SD) sAPPa produced (ng/mg protein). In addition, each protein fraction was subjected to sodium dodecyl sulfate polyacrylamide gel electrophoresis to assess total protein fractionation (C, right panel).

aCBSF Ameliorates b-amyloid Pathology in 5XFAD Mice
To determine the effect of aCBSF on b-amyloid pathology, transgenic 5XFAD mice at the age of 5-month-old were continuously treated with aCBSF or AgBSF via i.p. osmotic mini pump for 6 wk. Immunohistochemical staining using 4G8 antibody showed that aCBSF treatment substantially decreases cortical and hippocampal b-amyloid plaques (Fig.  9A, upper panel) and reduces fibrillary Ab species visualized by Thioflavin-S histochemical staining (Fig. 9A, lower panel) compared with AgBSF treatment. Moreover, the aCBSF-treated cohort also revealed less b-amyloid plaque pathology than the AgBSF-treated cohort, as determined by Congo red histochemical staining (Fig. 9A, middle panel). Quantitative analysis disclosed that aCBSF therapy significantly ameliorated b-amyloid pathology, as determined by 4G8 antibody staining in both neocortex and hippocampus regions compared with AgBSF treatment (Fig. 9B).

Neuroprotective Effects of aCBSF
5XFAD mice undergo neuronal loss in the neocortex and hippocampus that is associated with behavioral deficits. We examined whether continuous delivery of aCBSF by i.p. osmotic mini pump can elicit a neuroprotective effect in 5XFAD mice. Treatment with aCBSF partially prevented The results were presented as mean (+SD) sAPPa produced (ng/mg protein). In addition, each size fraction was subjected to sodium dodecyl sulfate polyacrylamide gel electrophoresis analysis to assess total protein fractionation (C, right panel).
neuronal loss in the neocortex region compared with AgBSF treatment, as demonstrated by NeuN antibody immunohistochemical staining, thus indicating that aCBSF may confer neuroprotective ability for AD brain (Fig. 10A and B).
aCBSF Improves Learning, Memory, and Cognitive Function in 5XFAD Mice 5XFAD mice received continuous treatment with aCBSF or AgBSF via i.p. osmotic mini pump for 6 wk and evaluated for cognitive function by novel object recognition and Ymaze tests during 4-5 wk of treatment. Novel object recognition test showed that aCBSF-treated 5XFAD mice spent more time with the novel versus old objects, whereas AgBSF-treated 5XFAD mice spent the same period of time with both novel and old objects (Fig. 11A). Thus, discrimination index (%) was enhanced by aCBSF compared with AgBSF treatment (Fig. 11B). Notably, improvement was complete because there was no significant difference (P > 0.05) from WT control mice (NTg). In addition, aCBSF treatment significantly increased the number of entries (Fig.  11C) and spontaneous alterations in 5XFAD mice compared with the AgBSF-treated cohort (Fig. 11D), as determined by Y-maze test, thus confirming that aCBSF treatment improved learning and working memory in this AD mouse model.

Discussion
Recent progress in HUCBC therapy for different neurological diseases 26,27 opened new opportunities for AD research 28 . We have previously found that multiple low-dose peripheral infusion of HUCBC reduced cerebral b-amyloid plaques, cerebral amyloid angiopathy, and astrocytosis, whereas these treatments improved cognitive impairments in the PSAPP AD mouse model 11 and enhanced neurogenesis in the aged rat brain 29 . In a subsequent study, Fig. 5. Further fractionation by anion-exchange chromatography. The size-exclusion fractions containing the highest amyloid precursor protein (APP) a-secretase activities (#10 to 13) were further subjected to anion-exchange chromatography using Q-Sepharose columns. Approximately 10 mg of protein from the size-exclusion fraction(s) was applied to the column, and 82 fractions were eluted with buffer containing 50 mM Tris, 500 mM NaCl, pH 7.6. (A) Chinese hamster ovary cells stably expressing wild-type human APP (CHO/APPwt) cells were treated with 40 mL of each protein fraction for 2 h and conditioned media and cell lysates were analyzed by soluble amyloid precursor protein a (sAPPa) Western blot (upper panel) and ELISA (lower panel). The original size exclusion-eluted fraction (#23) and phosphatebuffered saline (#24) were included as positive and negative controls respectively. In addition, sodium dodecyl sulfate polyacrylamide gel electrophoresis of #2 to 4 anion-exchange fractions showed the presence of multiple proteins (middle panel). (B) Protein concentration of each fraction. (C) CHO/APPwt cells were treated with the #1 to 4 anion-exchange fractions prepared from 3 independent experiments, and then conditioned media were subjected to sAPPa ELISA. The results are presented as mean (+SD) of sAPPa produced (ng/mg protein). The original size exclusion-eluted fraction (#23) was included as a positive control. Combined fractions #2 to 4, referred to as aCBSF, were used for further analysis.
we have reported that HUCBC-derived monocytes reduced cerebral b-amyloid pathology and cognitive impairments 13 . In addition, we have revealed that HUCBC-derived monocyte more effectively removed Ab by phagocytosis than the aged monocyte, whereas sAPPa enhanced Ab phagocytosis by the aged monocyte by forming a complex with Ab via the help of monocyte scavenger receptor 13 . In support of these findings, we demonstrated that overexpression of sAPPa significantly reduces both cerebral b-amyloid 6 and tau pathology in crossing Tg-sAPPa with PSAPP mice 7 . Meanwhile, using a sophisticated parabiosis mouse model, Wyss-Coray and colleagues showed that blood serum from old mice reduces neurogenesis and impairs cognitive functions when administered into young mice 30 . Subsequently, several groups reversed age-related cognitive impairments in aged mice by infusing plasma from young into old mice 14-16 as well as AD pathology 18 . More specifically, Wyss-Coray group have published several articles over the last few years 14,19 , showing the potential of young and/or umbilical cord plasma in ameliorating aged-associated cognitive impairments. In those experiments, they have either joined the young and aged mice through parabiosis or injected young and/or umbilical cord blood-derived plasma into aged mice via tail vein injection. Interestingly, only 3 to 4 injections within 3-to 4-wk period of time improved cognitive impairments in those experiments. These results encouraged us to determine whether human CBS could effectively reduce AD pathology in vitro (i.e., cell culture and Fig. 6. Cord blood serum-specific fraction with enhanced a-secretase catalytic activity (aCBSF) directly mediates a-cleavage of neuronspecific amyloid precursor protein (APP 695 ), but this activity is not mediated by a disintegrin and metalloproteinase domain-containing protein (ADAM) or tumor necrosis factor-a converting enzyme (TACE). (A) Human recombinant full-length APP 695 (fAPP 695 , 100 ng) was incubated with 0, 0.125, 0.25, 0.5, or 1 mg of aCBSF at 37 C for 2 h. The reaction mixtures were subjected to soluble amyloid precursor protein a (sAPPa) Western blot (WB) analysis using 2B3 (top panel) and holo APP and a-C-terminal fragment (a-CTF) analysis using pAPP751/770 antibody (lower panel). (B) fAPP 695 (100 ng) was incubated with 0.125 mg of aCBSF in the absence or presence of ADAM (GM6001, 1 mM) or TACE inhibitor (TAPI-0, 1 mM) for 1 h and then subjected to sAPPa, holo APP, and a-CTF WB analysis using 2B3 (top panel) and pAPP751/770 (lower panel). a-CTF of APP was further confirmed by an additional WB using an antibody specifically against Ab 17-cell-free systems) as well as in vivo (i.e., 3xTg-AD and 5XFAD mouse models) by enhancing sAPPa production.
Our preliminary findings indicate that CBS possesses a-secretase-like enzyme in cell culture and cell-free systems. In CHO/APPwt cells, CBS produces greater amount of sAPPa compared with ABS and AgBS in a concentrationand time-dependent manner ( Fig. 1A and B). Since a-secretase is proteinaceous and heat-labile, we hypothesized that a-secretase-like enzyme displayed by CBS is also inactivated by heat treatment. As expected, heat inactivation significantly limited the sAPPa-producing capacity of CBS (Fig. 1C). These results suggest that a-secretase-like enzyme of CBS is most likely a single complex protein that interacts with and cleaves APP. Subsequent study indicated that CBS mediates a-secretase cleavage of neuron-specific APP 695 in a cell-free system, further suggesting that this activity is mediated by an endogenous enzyme ( Fig. 2A and B).
To purify, characterize, and ultimately identify this a-secretase-like content in CBS, we employed 3-step affinity column, size-exclusion, and anion-exchange chromatography techniques in a sequential manner (Fig. 3-5). These sequential purification steps enhanced the catalytic activity more than 3,000-fold compared with original CBS. The fractions containing highest a-secretase catalytic activity, as reflected by sAPPa level, were combined and termed as "aCBSF" for the further study. SDS-PAGE analysis of the fractions from size-exclusion and anion-exchange chromatography yielding the highest a-secretase indicated size of our unknown enzyme could be around 177 to 275 kDa (Figs. 4 and 5). It is not easy for a 177-to 275-kDa protein to cross the blood-brain barrier through i.p. mini pump administration without any inhibition. We do not believe the protein is larger than 275 kDa based on the markers in for our gels. However, the SDS-PAGE also showed some low-molecularweight compounds that cannot be ruled out as well, which warrant further investigation. Interestingly, TAPI (ADAM17) and GM6001 (ADAM) inhibitors did not alter a-secretase in CBS, indicating that the enzyme is not TACE or ADAM, whereas the activity was dramatically reduced by PI/cocktail, confirming that the activity is mediated by a protease (Figs. 2 and 6). Moreover, immunoprecipitation of aCBSF with 2B3 antibody (anti-C-terminal of sAPPa) showed significant reduction in sAPPa levels, indicating that a-secretase-like enzyme aCBSF physically interacts with sAPPa (Fig. 7).
Previously, we and others have shown that sAPPa reduces b-amyloid pathology via inhibition of BACE1 6 . In a recent article, we have shown that sAPPa decreases tau phosphorylation via BACE1 inhibition and GSK-3bmediated inhibitory phosphorylation 7 . This study prompted us to investigate the functional efficacy of fractionated CBS (aCBSF) in 5XFAD and 3xTg-AD mouse models. We have shown that aCBSF significantly reduced Ab and tau phosphorylation (p-tau-Thr 231 ) in 3xTg-AD mice, whereas aCBSF enhanced a-secretase cleavage products (i.e., sAPPa and a-CTF), indicating that a-secretase-like content in CBS promotes APP nonamyloidogenic processing in vivo ( Fig.  8B and C). In 5XFAD mice with aggressive b-amyloid deposition and plaque formation, aCBSF reduced b-amyloid plaque pathology in both neocortex and hippocampus regions, and reduced neural loss in the neocortex region, compared with AgBSF-treated mouse brains ( Fig. 9 and  10). By carrying out sequential fractionation, we markedly enhanced CBS-derived a-secretase (termed aCBSF) and infused into 5XFAD mice via osmotic mini pump over the period of 6 wk. Behavioral analyses in 5XFAD mice indicate that aCBS-treated mice showed improved episodic memory, as determined by novel object recognition test (Fig. 11A and  B), as well as spatial working memory, as determined by Y-maze test (Fig. 11C and D), compared with AgBSF Fig. 7. Immunoprecipitation of full-length amyloid precursor protein (APP 695 )/cord blood serum-specific fraction with enhanced a-secretase catalytic activity (aCBSF) specifically limits APP asecretase activity of aCBSF. To determine whether immunoprecipitation could limit the ability of aCBSF to promote APP a-cleavage, we incubated 100 ng of fAPP 695 with 0.125 mg of aCBSF at 4 C for overnight and then immunoprecipitated (IP) the soluble amyloid precursor protein a (sAPPa)/aCBSF immune complex using an anti-DDK antibody (DDKAb), an sAPPa-specific antibody (2B3), or nonspecific IgG. Chinese hamster ovary cells stably expressing wild-type human APP cells were treated with the supernatants (Super.) from each immune complex, or phosphate-buffered saline as control, in the fetal bovine serum-free condition. Two hours after treatment, conditioned media were collected and analyzed by sAPPa Western blot (WB) using anti-N-terminal APP antibody (22C11, A) and sAPPa ELISA (B). For Panel (B), the results were presented as mean (+SD) of sAPPa production (ng/mL) in the conditioned media from 3 independent experiments with triplicates for each condition. There was no notable or significant difference in sAPPa production elicited by 0.5% supernatant from aCBSF immunoprecipitated with control IgG (Ctrl) and 2.5% (equivalent to 0.5% Super.) of aCBSF alone, as determined by sAPPa WB and ELISA analysis (data not shown).
treatment. Our work is in line with the work of Villeda et al. 14 and Castellano et al. 19 , where improvement of performance in cognitive impairment was found in aged mice treated with young plasma. Notwithstanding, we are not quite sure how CBS fraction (aCBSF) ameliorates b-amyloid pathology and cognitive functioning in 5XFAD and tau pathology in 3xTg-AD mouse model. The effect we observe may or may not be from CBS a-secretase-like enzyme. One of the plausible explanations for this effect may be a direct action from CBS a-secretase-like enzyme or could be an indirect effect through peripheral sink hypothesis which demand further investigation. Although we do not know the exact molecular mechanism, however, we believe that human cord blood-derived serum and/or plasma protein functions as a master regulator of several genes involved in the proliferation of cells, and blood vessels that might reduce neuroinflammation, Ab, and improve synaptic plasticity by affecting multiple pathways. Overall, our results show beneficial effects of aCBSF in ameliorating b-amyloid pathology and cognitive functioning in 5XFAD and reducing tau phosphorylation in 3xTg-AD mouse models. Fig. 8. Cord blood serum-specific fraction with enhanced a-secretase catalytic activity (aCBSF) promotes amyloid precursor protein (APP) a-secretase processing in vivo. 3xTg-AD female mice at 4 mo of age were treated with aCBSF, aged blood serum fraction with enhanced asecretase activity (0.5 mg/mouse; n ¼ 6), or phosphate-buffered saline control (1 mL/mouse; n ¼ 5 female) by i.c.v. injection and euthanized 72 h later. Mouse brain homogenates were then prepared from the right half of the brain (noninjection side). (A) Western blot (WB) analysis using Ab 1-17 antibody (6E10) shows total APP and Ab species. (B) WB analysis using a soluble amyloid precursor protein a (sAPPa)-specific antibody (2B3) or anti-N-terminal APP antibody (22C11) shows sAPPa or total APP, respectively. (C) WB analysis using pAb751/770 shows full-length APP (holo APP) and 2 bands corresponding to b-carboxy terminal fragment and a-C-terminal fragment. (D) Mouse brain cortices from each treatment group were stained with anti-phospho-tau (p-tau [Thr 231 ]) antibody. In addition, percentages (p-tau [Thr 231 ] positive area/total area; mean + SD) of anti-p-tau antibody positive cells were quantified by ImageJ (1.47v, NIH, USA) analysis (**P < 0.005; data not shown). WB data presented here are representative of results obtained from 5 to 6 female mice per group. Fig. 9. Cord blood serum-specific fraction with enhanced a-secretase catalytic activity (aCBSF) reduces b-amyloid plaques in 5XFAD mice. Five-month-old 5XFAD female mice were treated intraperitoneally with aCBSF (n ¼ 5 to 7) and aged blood serum fraction with enhanced asecretase activity (AgBSF; n ¼ 5 to 6) via osmotic mini pump at 30 mg/mouse/day for 6 wk. (A) Mouse brain sections from each group were stained with 4G8, Congo red, and Thioflavin-S. (B) Percentages of 4G8 positive areas were quantified by ImageJ (1.47v, NIH, USA) analysis for neocortex and hippocampus, showing that aCBSF treatment significantly reduced plaque area compared with AgBSF treatment (t-test for independent samples; *P < 0.05, **P < 0.01). Fig. 10. Neuroprotective effects of cord blood serum-specific fraction with enhanced a-secretase catalytic activity (aCBSF). 5XFAD mice at 5 mo of age were treated intraperitoneally with aCBSF (n ¼ 7) or aged blood serum fraction with enhanced a-secretase activity (AgBSF; n ¼ 6) via osmotic mini pump for 6 wk. (A) Mouse brain sections from aCBSFand AgBSF-treated groups were stained with anti-NeuN antibody. (B) Quantification of NeuN positive cells in the CA1, CA3, and neocortex revealed that aCBSF treatment significantly increased NeuN-positive cells compared with AgBSF treatment in neocortex (t-test for independent samples; *P < 0.05).
It is well known that members of the membrane-bound zinc-dependent metalloproteinase ADAM family are asecretase enzymes that cleave APP for the nonamyloidogenic pathway. In particular, 3 different members of this family, ADAM9, ADAM10, and ADAM17, possess APP a-secretase activity 31 . The ADAM family constitutes a large family of multidomain membrane proteins that have cysteine-rich, disintegrin, and zinc metalloprotease domains in their ectodomain 32 . The main function of ADAM family is to shed the ectodomain of different membrane proteins and has growth factors-like function via intracellular signaling cascade. However, it should be noted that numerous other substrates also have been linked to this ADAM family. These functions of ADAM family either protect against AD or promote AD pathogenesis via activation of different cytokines. One of the enzymes, ADAM17, is also known as TACE, responsible for secreting the main proinflammatory cytokine, TNFa 33 . Thus, TACE (ADAM17) is a therapeutic target for multiple diseases. Additionally, both ADAM10 and ADAM17 cleave various other membrane proteins and promote tumor in the cell 34 . ADAM10, in particular, cleaves many different kinds of transmembrane proteins in the vascular system, including the plateletactivating collagen receptor glycoprotein VI 35,36 , and endothelial proteins, including transmembrane chemokines (i.e., CX3CL1 and CXCL16) 37 . These 2 chemokines are known for angiogenesis, inflammation, and immune cell recruitment 38,39 . Likewise, ADAM9 cleaves and releases a number of molecules with important roles in tumorigenesis and angiogenesis. Taken together, whereas the known Fig. 11. Cord blood serum-specific fraction with enhanced a-secretase catalytic activity (aCBSF) improves cognitive function in 5XFAD mice. Both male and female 5XFAD mice at 5 mo of age and age-matched nontransgenic wild-type (WT) controls (NTg) were treated with aCBSF and aged blood serum fraction with enhanced a-secretase activity (AgBSF; 30 mg/mouse/day) intraperitoneally by osmotic mini pump for 6 wk as described in the Methods and Materials section. Each treatment groups as well as nontransgenic WT controls (NTg) were subjected to Novel Object Recognition and Y-maze behavioral testing. (A) Times spent exploring old and novel objects during the test phase of novel object recognition was recorded for each treatment group. (B) Discrimination index, calculated as the frequency of exploring new object versus original objects, was significantly reduced in 5XFAD mice treated with AgBSF, but not in those treated with aCBSF, compared with NTg controls. (C) Total number of arm entries for Y-maze was recorded for each treatment group. (D) Percentage alternations was significantly reduced in 5XFAD mice treated with AgBSF, but not in those treated with aCBSF, compared with NTg controls. Significance level determined by analysis of variance for a total of n ¼ 5, NTg mice; n ¼ 6, aCBSF-treated 5XFAD mice; and n ¼ 6 AgBSF-treated 5XFAD mice (***P < 0.001).
a-secretase enzymes, mainly ADAM10 and ADAM17 (TACE) and in some degree ADAM9, are involved in APP a-secretase cleavage, they are not APP-specific and cleave various substrates associated with inflammation, tumor formation, and progression. Thus, whereas AD is the only pathology in which an increased a-secretase activity has been proposed to be favorable, the nonspecific nature of the known a-secretases has made this strategy for AD treatment thus far unsuitable 40 .
In sum, our study has presumably discovered an umbilical cord blood-derived a-secretase that is independent of TACE or ADAM, thus making it a suitable candidate for the further study as a therapeutic target for AD treatment. This asecretase-like enzyme activity either directly or indirectly activates a-secretase or produces sAPPa in cell culture and AD animal models. In addition, we believe this a-secretase appears to be mediated by novel enzymes residing within the sera which decline with age. We expect that our study using fractionation, chromatographic separation, and massspectrometry (MS) techniques would identify the target enzyme as well as other interacting partners from CBS. However, identification of a target protein or enzyme with a particular function from a complex mixture of serum is a challenging task due to multiple factors including the high complexity and wide dynamic range of proteins as well as the presence of contaminating proteins of high abundance. Despite this, here we show that our purification techniques significantly enhanced the a-secretase of CBS. Further, MSbased sophisticated purification techniques will completely purify, identify, and characterize the factor mediating this asecretase in CBS.

Ethical Approval
The protocols in this study were approved by the relevant ethics committee (see Materials and Methods).

Statement of Human and Animal Rights
This article does not contain any studies with human subjects. All animal experiments were performed in accordance with the guidelines of the National Institutes of Health and were approved by University of South Florida (USF) Institutional Animal Care and Use Committee (IACUC reference number: IS00000438).

Statement of Informed Consent
There are no human subjects in this article and informed consent is not applicable.

Declaration of Conflicting Interests
The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: Paul R. Sanberg (PRS) is the coeditor in chief of Cell Transplantation. Neither PRS nor any of his colleagues were involved in the peer-review process or decision for this manuscript. PRS is also a cofounder, and JT is a consultant for Saneron CCEL Therapeutics, Inc. JT, DS, HH, and PRS are inventors on a patent application submitted by University of South Florida. PRS was not involved in any data acquisition and analysis. All other authors report no biomedical financial interests or potential conflicts of interest.

Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the NIH/NIA (R21AG055116, R01AG050253, R01AG049477, JT) and the Silver Endowment (JT).