Islet Amyloid Polypeptide (IAPP): A Second Amyloid ...

29 downloads 39 Views 651KB Size Report
Bryan, Texas; 2Department of Physiology and Neuroscience, PO-Box 7, St George's University, St Georges, Grenada,. West Indies; 3Department of ...... 26(1): 272-81 (2012). [53]. Blalock EM, Geddes JW, Chen KC, Porter NM, Markesbery WR,.
Send Orders for Reprints to [email protected] Current Alzheimer Research, 2014, 11, 000-000

1

Islet Amyloid Polypeptide (IAPP): A Second Amyloid in Alzheimer’s Disease Janelle N. Fawver1, Yonatan Ghiwot1, Catherine Koola1, Wesley Carrera1, Jennifer Rodriguez-Rivera3, Caterina Hernandez3, Kelly T. Dineley3, Yu Kong4, Jianrong Li4, Jack Jhamandas4, George Perry6 and Ian V.J. Murray1,2,* 1

Department of Neuroscience and Experimental Therapeutics, College of Medicine, Texas A&M Health Science Center, Bryan, Texas; 2Department of Physiology and Neuroscience, PO-Box 7, St George’s University, St Georges, Grenada, West Indies; 3Department of Neurology, Mitchell Center for Neurodegenerative Diseases, University of Texas Medical Branch, Galveston, Texas; 4Department of Veterinary Integrative Biosciences, Texas A&M University; 5Division of Neurology, Department of Medicine, University of Alberta; 6Dept of Biology, University of Texas at San Antonio Abstract: Amyloid formation is the pathological hallmark of type 2 diabetes (T2D) and Alzheimer’s disease (AD). These diseases are marked by extracellular amyloid deposits of islet amyloid polypeptide (IAPP) in the pancreas and amyloid  (A) in the brain. Since IAPP may enter the brain and disparate amyloids can cross-seed each other to augment amyloid formation, we hypothesized that pancreatic derived IAPP may enter the brain to augment misfolding of A in AD. The corollaries for validity of this hypothesis are that IAPP [1] enters the brain, [2] augments A misfolding, [3] associates with A plaques, and most importantly [4] plasma levels correlate with AD diagnosis. We demonstrate the first 3 corollaries that: (1) IAPP is present in the brain in human cerebrospinal fluid (CSF), (2) synthetic IAPP promoted oligomerization of A in vitro, and (3) endogenous IAPP localized to A oligomers and plaques. For the 4th corollary, we did not observe correlation of peripheral IAPP levels with AD pathology in either an African American cohort or AD transgenic mice. In the African American cohort, with increased risk for both T2D and AD, peripheral IAPP levels were not significantly different in samples with no disease, T2D, AD, or both T2D and AD. In the Tg2576 AD mouse model, IAPP plasma levels were not significantly elevated at an age where the mice exhibit the glucose intolerance of pre-diabetes. Based on this negative data, it appears unlikely that peripheral IAPP cross-seeds or “infects” A pathology in AD brain. However, we provide novel and additional data which demonstrate that IAPP protein is present in astrocytes in murine brain and secreted from primary cultured astrocytes. This preliminary report suggests a potential and novel association between brain derived IAPP and AD, however whether astrocytic derived IAPP cross-seeds A in the brain requires further research.

Keywords: Alzheimer's disease, amyloid beta, amylin, blood, brain, immunohistochemistry, metabolic dysfunction, oligomers, plaques, type 2 diabetes. INTRODUCTION There is interest in “infectious” non-prion amyloid proteins, especially if such cross-seeding or “infection” underlies the mechanistic linkage between the extracellular amyloids of type 2 diabetes (T2D) and Alzheimer’s disease (AD). There are several studies linking T2D and AD, indeed the link being described as diabetes of the brain, or type 3 diabetes [1-4]. This idea is based on two known observations. First, that disparate amyloid proteins can cross-seed or catalyze each other’s misfolding to augment pathology in neurodegenerative diseases [5-13], and such an interaction was first demonstrated in vitro, and recently in vivo [5-11, 14-19]. Second, several publications have demonstrated that amyloids are transmissible similar to infectious prions [12, 20-27]. However, the majority of the above studies were concerned with brain derived amyloids. *Address correspondence to this author at the Current: Department of Physiology and Neuroscience, PO-Box 7, St George’s University, St Georges, Grenada, West Indies; Tel: 473 444-4175 Ext 3733; Fax: 473-444-4673; E-mail: [email protected] 1567-2050/14 $58.00+.00

One obvious test of this “infectious” hypothesis would be to investigate the interaction of the extracellular amyloids of T2D and AD. Islet amyloid polypeptide (IAPP) is generated in the pancreas, co-secreted along with insulin from storage vesicles into the blood stream, and forms extracellular amyloid deposits in the pancreatic islets in T2D. Amyloid  (A) is the amyloid protein that misfolds and accumulates as extracellular amyloid plaques in AD. We would expect that elevation of peripherally generated IAPP, as in the early stages of T2D, upon entering the brain could augment, crossseed, or “infect” the misfolding of the brain derived amyloid A. Several lines of evidence support a mechanistic link between these two amyloids: IAPP and A. IAPP levels are elevated with metabolic dysfunction [28-33], a known risk factor for AD [34]. Dr. Kapourniotu’s group at Technische Universitat Munchen has demonstrated that IAPP directly interacts with A [35], and stabilizes the oligomeric conformation of A [36] - a form of A known to be toxic [37-41]. Thus it is possible that IAPP cross-seeds A to augment AD pathology. © 2014 Bentham Science Publishers

2 Current Alzheimer Research, 2014, Vol. 11, No. 10

We formulated the hypothesis that islet amyloid polypeptide (IAPP) involved in T2D could cross-seed and augment A misfolding to exacerbate AD pathology. However, in order for IAPP to play a role in AD pathology, we first needed to demonstrate several corollaries of this hypothesis: 1. IAPP is capable of entering the brain, 2. IAPP is able to misfold A, 3. IAPP is associated with amyloid plaques, and 4. plasma levels of IAPP correlate with AD. For the first corollary, there is increasing evidence supporting the presence of IAPP in the brain. IAPP, like insulin, is known to be able to cross the blood brain barrier, albeit it has been observed that a maximum of 0.11% of IAPP measured over only a 15 min time period crossed the blood brain barrier [42, 43]. Furthermore, IAPP immunoreactivity and receptors have been demonstrated in the brain and in AD tissues [44-48]. During the preparation of this manuscript, IAPP was identified in AD brain tissues by another laboratory [49], and two other reports suggest that IAPP enters the brain to alter cognition [50, 51]. Equally important, and less well known, is that IAPP may also be generated in the brain, as IAPP mRNA has been identified in AD brain tissues (Westermark and unpublished observations made in references [44, 45, 52], and mined from dataset in [53] at gene expression omnibus GEO http://www.ncbi.nlm.nih.gov/ geo/tools/ profileGraph.cgi?ID=GDS810:207062_at). Furthermore IAPP has known functions in the brain. IAPP exerts an anorectic effect via the hypothalamus and also affects memory [32, 54, 55]. IAPP along with A appear to exert their toxic effects mediated through the amylin receptor [56, 57]. In this study, we determined whether IAPP enters the brain in non-transgenic murine models and also determined the levels of IAPP in human CSF from AD and non-AD cases. Second, IAPP augments A misfolding. While IAPP interacts with A [35], the literature of the effects of IAPP on A misfolding is sparse. A prior study indicated that IAPP did not augment A misfolding, however this study used high concentrations of pre-aggregated amyloids to evaluate fibril elongation [14]. As such cross-seeding of physiological concentrations of monomeric amyloids was not evaluated. A separate experimental study indicated that IAPP augmented A oligomerization [36], and an in silico study demonstrates hetero-assembly of IAPP and A into oligomers [58]. These studies also have drawbacks, in silico may not reflect in vivo conditions, and the former study did not use a form of IAPP competent to misfold to adopt an amyloid conformation. In contrast to the other studies, we used near physiological concentrations of misfolding competent A and IAPP, starting with the proteins in a monomeric conformation. We should mention that a recent publication suggest that IAPP may clear A from the brain [50]. Third, there is only one recent publication demonstrating the association of IAPP with amyloid plaques [49], published during the preparation of our manuscript. However, we are the first reporting IAPP association with amyloid plaques in 2 mouse models of AD, as well as in familial AD. Last and for the fourth corollary, there was no prior data correlating IAPP plasma levels with AD pathology, with two such studies only published during the preparation of this manuscript [50, 51].

Fawver et al.

This report represents a comprehensive analysis of several aspects of IAPP: IAPP entry into the brain, misfolding of A, association with plaques and correlation of peripheral IAPP levels with AD diagnosis. We demonstrated that IAPP is present in the brain, augments A oligomerization in vitro, and colocalizes with amyloid plaques in AD and 2 mouse models of AD. However, our data does not support an “infectious” hypothesis whereby peripheral IAPP enters the brain to cross-seed A misfolding. In short, peripheral levels of IAPP are not altered with AD pathology in African Americans or in Tg2576 AD mice. We provide evidence for IAPP protein within and secreted by astrocytes, suggesting that elevations of brain IAPP may augment A misfolding. Further studies are required to identify the mechanistic role of brain IAPP in AD pathology. MATERIALS AND METHODS Materials Lyophilized A42 at >95% purity was purchased from rPeptide (Athens, GA). Human IAPP was purchased from Bachem (Torrance, CA), rodent IAPP was purchased from Abbiotec (San Diego, CA), and human 5-FAM-IAPP (1-37) amide was purchased from Anaspec (Fremont, CA). All peptides were stored desiccated at -80o C. Stock solutions (0.5 mg/mL or 1mg/mL) were prepared by dissolving A and IAPP in hexafluoroisopropanol (HFIP). All water used in experiments was purified to 18 m using an ion exchanger and reverse osmosis (Purlab Ultra, Elga, Lowell, MA). IRB All human tissues and tissue sections received Institutional Review Board (IRB) exemption by the Texas A&M Institutional review board (IRB 2008-0095). This research involved the use of publically available, existing and archived tissues and the information recorded in a manner that subjects cannot be identified. IACUC All animal procedures and protocols were in accordance with the respective institutions animal use protocols (AUP) and received institutional animal care and use committee (IACUC) approval. Tissue Distribution of IAPP Non-transgenic C57/B6 strain mice were injected with 5mg/kg (concentrations described previously (37)) of human IAPP, rodent IAPP, and fluorescently labeled FAM IAPP. Tissues (blood, brain, liver, kidney, heart, lungs, omental adipose, spleen, and pancreas) were collected at 2 (n=1), 6 (n=2), and 24 hours (n=2). The average ages and weights of the mice were ~ 3 months of age and 22.1 ± 3.35 gm, respectively. Tissues from the non-transgenic mice above were frozen on dry ice and stored at -80oC until extraction. The tissues were weighed and sonicated on ice (3X for 8-10 seconds each with a probe tip sonicator) in 4 times volume per tissue weight of RIPA extraction buffer (5X stock: 250 mM Tris base, 750 mM NaCl, 5% NP-40, 25 mM EDTA, 2.5 % Na

Amyloid Beta and Islet Amyloid Polypeptide

deoxycholate, 0.5% SDS), with HALT protease buffer (Pierce, 1:100 dilution) and PMSF (1:500 dilution). Note: the following tissues (of mice injected with IAPP) were diluted in order to fall within the range of the standard curve of the ELISA: pancreas 10,000; spleen 1,000-10,000; lung 1,000-10,000; liver 12.5; heart 12.5; adipose 12.5; kidney 12.5. Also, levels of IAPP were high in the control pancreas and lung, which we had to dilute 1,000 times. Transgenic Mouse Model of Alzheimer’s Disease (Tg2576 and CRND8) Mouse brain tissues from 13 month old Tg2576 (n=10) and control (n=6) mice were obtained from Dr. Kelly T. Dineley at University of Texas Medical Branch (UTMB). The animal colony was maintained as described previously [59]. Brains were also obtained from 12-month old Tg CRND8 mice (Jhamandas laboratory) for immunohistochemical studies. Postfixed whole brains were sucrose protected overnight at 4°C, and stored at -80oC. The brains were then post-fixed again in 4% paraformaldehyde/saline overnight prior to paraffin embedding. The brains were sectioned sagittally at 5 m by the Texas A&M Histology Core and used for the following experiments. Mouse Blood from Tg2576 AD Transgenic Mice Blood was drawn from 9 month old Tg2576 mouse model of Alzheimer’s disease via retro-orbital bleed, and allowed to clot for ~2 hours on ice. Serum was collected following centrifugation (2000xg 20 min) and stored at 80oC until use. Human Cerebrospinal Fluid (CSF) Human CSF (500 μL) was obtained in cryovials from the Kathleen Bryce brain bank at Duke University, and stored at -80oC until use. The brain bank is supported by the NIA AG05128 grant. Normal CSF were classified according to CERAD as normal 1A (n=4) or 1B (n=3) via CERAD criteria. FTD, and AD cases were 2 cases each of Braak stages III, IV, stage VI and 1 case of Braak stage IV. (Details of the cases can be found in Supplemental Table 2). Human Plasma We obtained plasma from an African American cohort as this population has increased susceptibility of insulin resistance, T2D, and AD [60-67], allowing us to more easily detect a correlation between T2D and AD. Human plasma from African American patients is part of the Minority Aging Research Study cohort and were obtained via collaboration with Drs Fawn Cothran and Lisa Barnes at Rush University’s Alzheimer’s disease Center. The samples were stored at 80oC until use. The Minority Aging Research Study is supported by the NIH R01 AG022018 grant to Dr Lisa Barnes. Histochemistry Slides were deparaffinized by twice immersion in xylenes for 5 minutes each, followed by rehydration through decreasing concentrations of ethanol (100%, 95%, 80%, and 70%) for 1 min each and ending in water. Histopathological staining was performed on hippocampal AD brain regions

Current Alzheimer Research, 2014, Vol. 11, No. 10 3

obtained from Case Western Reserve, as well as brains from the Tg2576 and TgCRND8 AD mouse models obtained from Drs. Kelley Dineley and Jack Jhamandas, respectively. Antibody immunofluorescence was performed as previously described [68]. Briefly, the tissue sections were rehydrated and immunostained with the mouse monoclonal antibodies: the anti-A antibodies 6E10 (mouse monoclonal, epitope 4-9, Covance) and NAB 228.3 (epitope 1-11, a gift from Dr Virginia M.Y. Lee [69]), the anti-A oligomeric antibodies NU4 (oligomeric epitope, a gift from Drs Pascal Lacor and William Klein [70]) and 11A1 (oligomeric epitope, antigen 1035 of A mutant E22P, Cat# 10379, Immuno-Biological Laboratories [71]) or the anti-glial fibrillary protein (GFAP, clone GA5, Cat #MAB360, Millipore). The rabbit polyclonal antibody used for IAPP was from Abbiotec (Cat# 250470, Abbiotec, San Diego, CA). All antibodies were used at a 1:200 dilution incubated overnight at 4oC in a humidified chamber. Antigen retrieval of the tissues was not performed for any of the antibody staining. Following a washing step, fluorescent secondary antibodies (Alexa Fluor Red 594 or Oregon Green 488) specific to the primary antibodies (mouse or rabbit, 1:200 dilution) were also incubated with the slides overnight at 4oC. After immunostaining, the slides were washed in 100 mM Tris buffered saline, water, mounted in fluorescence mounting medium without 4',6diamidino-2-phenylindole (Vectashield, Vector labs), and visualized via fluorescent microscopy (n= 16 Tg2576 mice). ELISA Measurement of IAPP IAPP in human CSF, human and mouse plasma, and murine tissue extracts were measured using a commercially available IAPP ELISA kit (human Amylin ELISA, EZHA52K, Millipore). The ELISA kit is specifically designed to measure IAPP in the physiological range, with quality control samples in the normal (QC1) and high range (QC2). This kit is suitable for the measurement of mouse IAPP, as noted by the manufacturers (Millipore). The human CSF was diluted 4/5 with wash buffer included in the IAPP kit. The human plasma was used undiluted, while the Tg2576 mouse serum was diluted 1/10. For tissue extracts from animals injected IAPP, the tissues were already diluted 1/5 from the RIPA extraction, and then diluted further so that the IAPP would fall within the standard curve as follows (final dilution): blood (1/5-1/12.5), cardiac (1/12.5), liver (1/12.5), adipose (1/12.5), brain (1/10-1/12.5), kidney (1/12.5- 1/100), lung and spleen (1/1000- 1/10,000), and pancreas (1/10,000). Tissues extracts from animals not injected with IAPP, already diluted 1/5 from the RIPA extraction, and were used without further dilution. (Except pancreas and lung). Colorimetric Measurement of Glucose Glucose was measured using a commercially available glucose oxidase colorimetric kit (Sigma, GAGO20-1KT). The assay was adapted for use in a 96 well plate and using 50 μL final volume of both samples and standards. ELISA Measurement of Insulin Insulin in human CSF was measured using a commercially available human insulin ELISA Kit (EZHIASF-14k, Millipore). Quality control samples were provided with

4 Current Alzheimer Research, 2014, Vol. 11, No. 10

ranges of 4.8-10.1 uU/mL (QC1) and 30.9-64.2 μU/mL (QC2). Amyloid Aggregation In vitro In order to determine if increasing concentrations of human and rodent IAPP cross-seeded A aggregation, aliquots of stock A42 and IAPP in HFIP were co-lyophilized and then resuspended in 100 mM phosphate buffered saline (calcium and magnesium free). The final A concentration was 2 μM which is below the minimal concentration for aggregation of ~6.3 μM [72]. Increasing concentrations of human and rodent IAPP, as indicated, were added to A to determine if they augmented A aggregation. Samples (50 μL in a 500 μL microfuge tube) were agitated in an Eppendorf thermomixer (37o C, 1400 rpm, ThermoFisher Scientific, Pittsburg, PA) for 24 hrs and overlaid with oil to reduce evaporation. To assure that A does not augment IAPP aggregation, the converse experiments were performed- where increasing concentrations of A was added to human and rodent IAPP. Western Blotting For western blotting analysis, A and IAPP were electrophoresed on 16.5% Tris-Tricine PAGE gels and transferred onto 0.2 m nitrocellulose membranes. All membranes were blocked with 5% fat-free milk (BLOTTO, VWR) in TTBS (Tween Tris Buffered Saline, 0.1% Tween 20, 150mM NaCl, 20mM Tris, pH 7.6) for 1 hour and incubated overnight at 4oC with primary antibodies in 5% fat-free milk in TTBS. Anti-A murine monoclonal antibody 6E10 (epitope 4-9) was purchased from Covance (Princeton, NJ) and used at 1:2000 dilution. The anti-IAPP antibody was purchased from Bachem (Cat# T-4149) and used at 1:1000 dilutions. Blots were washed in TTBS and then incubated for 30 minutes at RT with the secondary anti-mouse or -rabbit HRP detection antibodies (Invitrogen) diluted to 1:2000. All membranes were blotted with regular ECL reagents (Pierce, Thermo Scientific) and then exposed to X-ray film (Pierce, Thermo Scientific). Astrocyte Cultures Enriched astrocyte cultures were prepared from forebrains of 1- to 2-day-old Sprague-Dawley rat pups as described previously [73]. Briefly, dissociated cells were plated onto poly-D-lysine (PDL)-coated T75 flasks or 96 well culture plates directly and were fed every other day with D10S (DMEM with 10% FBS) for 7–10 days. Enriched astrocytes (>90%) were isolated from the flasks after shaking overnight at 200 rpm. Astrocytes were treated with vehicle, LPS (10 ng/ml), TNF  (30 ng/mL), or A (10 and 20 μM) for 48 hrs. The conditioned media (100 μL) were collected for ELISA analysis. Prior to immunostaining, the cells were fixed with 4% paraformaldehyde in PBS for 10 min at room temperature, thrice washed in 100 mM Tris pH 7.4, and then blocked in Tris with 2% goat serum and 0.1% Triton X-100. The cells within the 96 well culture plate were then coimmunostained using an antibody to the astrocyte marker glial fibrillary acidic protein (chicken anti-GFAP, AB5541, Millipore) and the anti-IAPP antibody (Rabbit polyclonal, Cat# 250470, Abbiotec, San Diego, CA) both at 1/2000 dilu-

Fawver et al.

tion and incubated overnight in a humidified chamber at 4°C. The cells were washed with Tris buffer, blocked with Tris/goat serum as above, and then incubated at 4°C overnight with the fluorescence secondary antibodies specific for the primary antibodies from Life technologies: goat antichicken Alexa Fluor 488 (A11039) and anti-rabbit Alex Fluor 594 (A11012) used at 1/2000 dilution. The cells were then thrice washed, and coverslipped following the application of Vectashield without DAPI. Florescent photomicrographs are shown for the enriched astrocyte cultures. RESULTS IAPP is Significantly Elevated in Human AD CSF To more accurately evaluate if IAPP associates with AD pathology, IAPP levels were measured in human CSF samples of normal, non-AD dementia, and AD patients. IAPP levels in the human CSF and mouse tissue extracts were below the sensitivity of detection by western blots (data not shown) and thus were analyzed by ELISA. The CSF IAPP levels were elevated in AD (Fig. 1). While the difference between AD and controls were significant as determined by one way ANOVA and Holm-Sidak analysis (p =0.013), this should be interpreted cautiously as the small sample sizes resulted in an underpowered study (alpha 0.695).

Fig. (1). IAPP levels are significantly elevated in CSF from AD patients. IAPP is present in CSF at picomolar levels. Human CSF (500 μL) was obtained in cryovials from the Kathleen Bryce brain bank at Duke University, and stored at -80oC until use. IAPP was measured using the human Amylin ELISA, EZHA-52K kit from Millipore. Normal CSF were classified according to CERAD as normal 1A (n=4) or 1B (n=3) via CERAD criteria. The normal, frontotemporal dementia (FTD), and Alzheimer’s disease (AD) cases are described in detail in Supplemental Table 1. CSF IAPP significantly increased in AD, as compared to normals, (*2-tailed ttest, p=0.024). We should note that this statistic should be interpreted with caution as the study was underpowered.

The elevated IAPP CSF levels could just reflect a leaky blood brain barrier in AD. As such, insulin, which is cosecreted with IAPP from the pancreas [74] and can also enter the brain [42], should exhibit similar changes if the blood brain barrier is compromised. However, the CSF- see note insulin levels were not significantly different among groups (Normal 1.88 ± 0.34, FTD dementia 2.03 ± 0.21, and AD 2.06 ± 0.45, data are mean ± std with n = 7, 7, and 9, respectively). Since peripheral glucose levels also increase IAPP

Amyloid Beta and Islet Amyloid Polypeptide

levels, we measured the CSF glucose levels. These values were also not significantly different (Normal 126.8 ± 27.6, FTD dementia 160.4 ± 73.8, and AD 146.6 ± 58.5, data are mean ± std with n = 7, 7, and 9, respectively) via a one way ANOVA (p = 0.661). These latter negative results should be interpreted cautiously as they were underpowered. However, taken together, these findings suggest that the elevation of IAPP in CSF from AD cases is specific and that IAPP is present in the brain. To our knowledge, this is the only published measurement of IAPP levels in human CSF.

Current Alzheimer Research, 2014, Vol. 11, No. 10 5

IAPP (human or rodent IAPP, Fig. 2B). This in vitro data suggests that such an IAPP mediated oligomerization of A could occur in the human brain.

Tissue Distribution of Exogenous IAPP While it is increasingly recognized that peripheral IAPP enters the brain, we mention preliminary data for tissue distribution of i.p. injected IAPP for two reasons: 1). Tissue distribution is of importance if IAPP it to be used as a therapeutic as suggested by recent findings where IAPP improved cognition in mouse models [50, 51] and 2). There was a 106 fold difference between injected dose and (M vs pM) IAPP levels in the tissue compartments, which suggested complex tissue pharmacokinetics, and possible trapping of injected IAPP in other tissue compartments. In our preliminary data, albeit with very small sample sizes, IAPP levels were highest in the pancreas, then spleen, followed by kidney > blood > lung > heart> adipose > blood  brain. Our preliminary data is concordant with several published studies that indicate IAPP binds to tissues such as the kidney (acts upon and is reabsorbed in the proximal tubules [75, 76]), lung [19, 77,78], intestinal villi [77], and high pancreatic IAPP (IAPP is stored in secretory granules estimated to be 1-4 mM [79]). Obviously, future studies of IAPP tissue distribution with appropriate sample sizes are needed. IAPP Augments A Oligomerization IAPP is known to interact with A in vitro [35, 36, 80], and may augment the misfolding of A in vivo as well. There is a paucity of literature on cross seeding of A by IAPP, likely because an early study demonstrated that preaggregated IAPP was unable to cross-seed A [14]. We revaluated this cross-seeding hypothesis importantly using lower concentrations of IAPP and A in their monomeric form. This was to mimic conditions expected to occur in vivo. Incubating A with increasing concentrations of human IAPP in vitro resulted in marked dose dependent misfolding of A to form low molecular weight (MW) oligomers, especially tetramers to pentamers as determined by western blot analyses (Fig. 2A). Importantly, IAPP augmented A aggregation occurred at concentrations where A alone did not misfold, and at lower than the critical concentration (~6.3 μM) required for A42 misfolding [72]. Our data indicates that this IAPP dose dependent cross-seeding of A was specific for only human IAPP. First, rodent IAPP did not have an effect on A misfolding. This is important as only human IAPP is reported to misfold to form -sheet amyloid structures, while amino acid substitutions by -sheet breaking prolines in rodent IAPP prevent it from misfolding [44]. Secondly, the converse experiment did not yield similar results, as increasing concentrations of A was unable to augment the misfolding of

Fig. (2). IAPP augments misfolding of A. In vitro incubation of A with increasing concentrations of human, but not rodent IAPP, with shaking in an Eppendorf Thermomixer augments A oligomerization as observed via western blotting with A antibody 6E10 (A). Lanes 1 and 2 represent 2 M of unaggregated (Unagg) and aggregated (Agg) A repectively. A (2 M) was incubated with increasing concentrations of amyloidogenic human IAPP (Lanes 3-5) and non-amyloidogenic rodent IAPP (A). Incubation of human and rodent IAPP with increasing concentrations of A does not increase oligomerization of IAPP via blotting with the antiIAPP antibody (Bachem) (B). Lanes 1,2 represent 11.5 M of unaggregated (Unagg) and aggregated (Agg) amyloidogenic human IAPP, and lanes 3,4 represent unaggregated (Unagg) and aggregated (Agg) non-amyloidogenic rodent IAPP. Human IAPP (Lanes 5-7) and rodent IAPP (lanes 8-10) at a fixed concentration of 2 M were incubated with increasing concentrations of A and nonamyloidogenic rodent IAPP (B). In both of these in vitro experiments, oligomerization was performed using relatively low concentrations IAPP and A and started with monomeric proteins. HMW = high molecular weight. MW = molecular weight, Unagg = unaggregated, and Agg = aggregated.

6 Current Alzheimer Research, 2014, Vol. 11, No. 10

Endogenous IAPP Associates with A Oligomers in Murine Brain We have demonstrated that IAPP can cross the blood brain barrier to enter the brain. We next wished to determine if IAPP is associated with A pathology in Alzheimer’s disease transgenic mice. Since IAPP augments A oligomers formation in vitro, we determined if IAPP is spatially associated with oligomeric A in the Tg2576 AD transgenic brain tissues. The endogenous rodent IAPP (red, Fig. 3A and B, IAPP) immunostaining colocalized to A amyloid plaques (green, Fig. 3A and B, A [Nab 228.3 and NU4]) pathology in serial sections of brain tissue from the Tg2576 transgenic mouse model of AD. Colocalization of IAPP immunostaining with that of A, observed as yellow fluorescence, using the anti-A oligomer antibodies NAB 228.3, and even more so with NU-4 (Fig. 3A and B, Both). Such staining was not observed in controls: secondary antibody alone and nontransgenic mice (data not shown). However no colocalization was observed for IAPP and the anti-A oligomeric antibody 11A1 (Fig. 3C). This suggests the IAPP colocalizes with A in a specific oligomeric conformation that is histochemically detected by NU-4 and not 11A1. IAPP localization to A plaques was also observed in cortex of 12-month old TgCRND8 mice (Supplemental Fig. 1). Antibody Characterization To determine if these antibodies NAB228.3, NU-4, and 1A11 differentially recognized A conformations, we further characterized them via western blotting (Figs. 3D and E). As such, we utilized different A proteins which adopt primarily dimeric (A42 MetOx, OX), oligomeric (A42 E22, ), and high molecular weight (A42 wild type, Wt) conformations as described previously [81]. NAB 228.3 and NU-4 similarly immunodetected unaggregated A (Fig. 3D). Interestingly, the pentameric band was detected with NU-4, less so with NAB 228.3, and not identified at all by 1A11 in the unaggregated A blot (Fig. 3D). For the aggregated A blot, NU-4 does not detect high molecular weight aggregated A and oligomeric species (Fig. 3E). Further experiments are required to identify the specific A oligomeric species that interacts with IAPP, but it is likely a pentameric A species based on our western data. IAPP Colocalizes with Amyloid Plaques in Swedish FAD We immunostained several sporadic AD cases and did not observe any colocalization of IAPP with amyloid plaques with a representative image shown in Fig. (4). However, IAPP immunostained amyloid plaques in the 2 Swedish FAD cases with a representative image shown in Fig. (4). In addition to colocalization with A in amyloid plaques, there was also punctuate IAPP staining. While we did not further characterize this punctuate staining, it may be either astrocytic dystrophic neurite immunostaining. The IAPP colocalization with amyloid plaques in the Swedish FAD is consistent with the colocalization in the Tg2576 transgenic mouse, which contains the Swedish familial AD mutation. No Correlation Between Peripheral IAPP Levels with Type 2 Diabetes (T2D) and AD Diagnosis We demonstrated that IAPP enters the brain, augments A misfolding, and colocalizes with A in amyloid plaques.

Fawver et al.

To satisfy the last corollary of the “infective” hypothesis, we next determined whether peripheral IAPP levels correlated with AD using an African American cohort with and without diabetes and AD. We reasoned that elevation of IAPP during prediabetes or early diabetes may translate into more IAPP entering the brain to augment A oligomerization and AD pathology. We performed an ELISA for IAPP in serum samples from African American patients from the Alzheimer’s cohort at Rush University. There was no significant difference in IAPP levels among the groups (Ctl, T2D, AD, and T2D plus AD (6.3 ± 4.08, 5.87 ± 4.46, 6.16 ± 4.12, and 5.30 ± 3.80 pM; with data representing mean ± std and n= 18, 19, 18, and 7 respectively), with a trend towards lower IAPP in the diabetic groups. Additionally, there was no significant difference in the body mass index (data not shown). While indicators of diabetes, hemoglobin A1c (HbA1c) and glucose levels were elevated in the diabetic groups, there was no significant difference in IAPP levels between the groups. Indeed, there was a trend of lower IAPP in the T2D group. Glycosylated hemoglobin (HbA1c), a marker of diabetes, and glucose levels, were significantly different in the diabetic cases (data not shown). Thus peripheral IAPP levels do not correlate with AD diagnosis in the African-American cohort. No Correlation Between Peripheral IAPP and AD in a Mouse Model Tg2576 mice naturally develop metabolic dysfunction (89 months) and reduced glucose tolerance [59, 82] before or concomitant with the development of amyloid plaques (~9 months) [83]. These mice are hyperinsulinemic and have normal fasting blood glucose levels [59]. This would mimic prediabetes, and thus we expected that blood IAPP would be increased in the Tg2576 mice with metabolic dysfunction. We further evaluated the effect of the antidiabetic thiazolidinedione drug, rosiglitazone (RSG) which normalized hyperinsulinemia and glucose intolerance [59]. There was no significant difference in the IAPP levels (pM) between controls and AD transgenic mice, or mice treated with RSG (non-Tg no drug 27.89 ± 11.57, non-Tg RSG 26.83 ± 8.56, Tg2576 no drug 27.20 ± 11.27, Tg2576 RSG 26.83 ± 6.20, with data mean ± stdev and n=11, 12, 7 and 9, respectively. Non-Tg = non transgenic; Tg 2576 =transgenic Tg2576 mouse model of AD). Thus there was no correlation between peripheral plasma IAPP and AD genotype. Astrocytes Express IAPP in Brain Sections of the AD Transgenic Mice Since IAPP appeared to immunostain cells with astrocyte like morphology surrounding the central amyloid plaque in the Tg2576 (red, Fig. 5, IAPP), we further confirmed that IAPP immunostained cells were indeed reactive astrocytes as indicated by GFAP immunolabeling (green, Fig. 5, GFAP). Activated astrocytes surrounded amyloid plaques, as well as the periphery of the amyloid plaque (green, Fig. 5, GFAP) in brains of 12 month old Tg256 AD. IAPP immunostained similar structures (red, Fig. 5, IAPP), in the same section, and colocalization evident by yellow fluorescence (green, Fig. 5, Both). Importantly, IAPP protein was present in the cytoplasm of cultured primary astrocytes (red) and immunostaining colocalized with the GFAP (green) as yellow

Amyloid Beta and Islet Amyloid Polypeptide

Current Alzheimer Research, 2014, Vol. 11, No. 10 7

Fig. (3). IAPP immunostains amyloid plaques and colocalizes with A oligomers. Sagittal and fixed brain sections from the Tg2576 mouse model of AD were co-immunostained using two antibodies with detect oligomeric A conformations (NAB228.3 and NU4, green) and IAPP (red). Colocalization of endogenous rodent IAPP (red, IAPP) with A immunostaining (green, NAB228.3 and NU4) was visualized yellow (Both, A and B) in cortical plaques of sagittal brain tissue sections from the Tg2576 mouse model of AD. This is evidenced with colocalization of A (NU4) with IAPP occurring more on the periphery of the plaques, leaving non colocalized IAPP (red) in the center of the plaques. This data is consistent with our independent in vitro experiments. However, using a separate anti-oligomeric antibody 1A11, no colocalization of IAPP was observed for oligomeric A and IAPP (C). In this case A localized to the center of the plaque and IAPP to the periphery. Astrocyte-like immunostaining also was observed for IAPP (red), and such immunostaining did not colocalize with A immunostain (green) (IAPP and Both A and B). Tissues were fixed in paraformaldehyde and paraffin embedded. Scalebar =50 um. Antibody characterization. The antibodies NAB228.3, NU4 and 1A11 were characterized with unaggregated (D) and aggregated (E) A proteins. Different A proteins were used which adopt primarily dimeric (A42 MetOx, OX), oligomeric (A42 E22, ) and high molecular weight (A42 wildtype, Wt) conformations as described previously (81). Immunostaining with the antibodies are markedly different for unaggregated and aggregated A. Antibodies: 1A11 – anti-oligomeric A antibody, NU4- anti-oligomeric A antibody specific for A -derived diffusible ligands (ADDLs), NAB228.3. Synthetic A proteins: wildtype A42 (Wt) = A42 without any modifications and is able to adopt monomeric, oligomeric and HMW conformations. MetOx E22 (Ox) = oxidized Ab42, which primarily adopts a dimeric conformation as previously described (81) A42 E22 () is a mutant A protein identified in a Japanese AD group (105) where the deletion of E22 results in this A primarily forming oligomers (81;105).

fluorescence (Fig. 5, Both, cultured astrocytes). To our knowledge, this is the first demonstration of IAPP protein present in astrocytes. Secretion of IAPP from Activated Cultured Primary Astrocytes Our above data demonstrated that IAPP can be produced by astrocytes in the Tg2576 mouse model and enriched primary astrocytes. Levels of IAPP secreted into the culture media from astrocytes were measured at 24 hours after

stimulation. There was a significant increase in secreted IAPP from lipopolysaccharide (LPS, 10 ng/mL, p=0.034), and TNF-stimulated (30 ng/mL, p=0.049) astrocytes cultures as compared to controls (Control 0.99 ± 0.38, LPS 1.56 ± 0.17 TNF  1.49 ± 0.25; n= 4, 4 and 5 respectively with power of 86.3 and 73.3% for LPS and TNF ). However, stimulation with synthetic A42 did not increase IAPP production (data not shown). These results suggest that IAPP protein can be secreted under inflammatory conditions, and compliments our histochemical data whereby IAPP protein is present in astrocytes.

8 Current Alzheimer Research, 2014, Vol. 11, No. 10

Fawver et al.

immunostaining along with GFAP colocalization in Tg2576, and demonstration that cultured astrocytes produce and secrete IAPP.

Fig. (4). IAPP immunostains amyloid plaques in Swedish FAD and not sporadic AD. A imunostained with the antibody 6E10 (green, A) decorated plaques in both sporadic AD and familial AD (Sporadic AD, and Swedish FAD. green, A) brain tissue sections. IAPP immunostaining (red, IAPP) exhibited background immunoreactivity in Sporadic AD (Sporadic AD. Red, IAPP), and appeared to stain amyloid like structures in the Swedish familial, AD case (Swedish FAD. Red, IAPP). A immunostaining (green) only colocalized with IAPP (red) in cortical plaques in the Swedish FAD and not the Sporadic AD brain tissue sections (Both, Swedish FAD vs Sporadic AD). Tissues were fixed in paraformaldehyde and paraffin embedded. Scale bar =50 um.

DISCUSSION We hypothesized that elevations of a peripherally generated amyloid, islet amyloid polypeptide (IAPP), would lead to greater amounts of IAPP entering the brain and catalyzing the cross-seeding of the amyloid A involved in Alzheimer’s disease (AD). This would serve as a potential mechanism underlying the linkage between type 2 diabetes (T2D) and AD. We demonstrate that IAPP is present in the brain, augments A misfolding, and colocalizes to pathological amyloid plaques in 2 mouse models of AD as well as in familial AD. Additionally, systemic administration of IAPP is primarily sequestered in the spleen and only a small percentage injected IAPP enters the blood and brain. We next determined that peripheral levels of IAPP in blood/plasma do not correlate with AD diagnosis. Thus while our present data does not support a mechanism whereby peripheral IAPP augments A misfolding or the “infective” hypothesis, it does suggest a novel involvement of brain derived IAPP in AD pathology.

Fig. (5). IAPP is present in astrocytes. In brain sections from 12 mo old Tg256 AD, activated astrocytes were immunostained using an anti-GFAP antibody (Tg2576, green, GFAP). GFAP immunostaining was present in astrocytes at the plaque periphery and also within the plaque itself. The plaque is at the center of the image (not stained for A). IAPP (red) immunostained similar structures, as well as the plaque periphery (red, IAPP), and in the same section colocalized with the green GFAP immunostaining as yellow fluorescence (Both). Scalebar =50 m. We independently verified IAPP protein in astrocytes using cultured and purified primary astrocytes. In cultured astrocytes, IAPP (red) immunoreactivity was localized to the cytoplasm of GFAP+ (green) cells (Both, cultured astrocytes). Scale bar =100 um.

Peripheral IAPP is known to enter the brain (42, 43). Furthermore, two recent studies using i.p. injected IAPP or IAPP analogues resulted in improved cognition in an AD mouse models, which suggest entry of peripheral IAPP into the brain (50, 51). Interestingly, we noted that while we injected micromolar concentrations of IAPP intraperitoneally, the resulting blood and brain tissue levels of IAPP were extremely low, only in the picomolar range (a 10-6 fold difference). We pursued this further to demonstrate that significant amounts of injected exogenous IAPP accumulate in other tissues (e.g. spleen), concordant with previous publications (19, 76, 77, 84-86). This latter finding of IAPP clearance is of importance and needs to be further characterized if IAPP or IAPP analogues are to be used as AD therapy.

IAPP is Present in the Brain

There is precedence for pancreatic proteins entering the brain, as well as associating with AD pathology. It is known that IAPP and other pancreatic proteins, such as pancreatic thread protein and insulin, do cross the blood brain barrier to enter the brain (42, 43) and are associated with AD pathology [87-89]. A study published during the preparation of this manuscript also demonstrated that IAPP amyloid is present in AD brain tissues and occasionally colocalized with A [49]. Thus IAPP enters the brain and colocalizes with amyloid pathology.

We used three independent methods to demonstrate that IAPP is present in the brain- CSF IAPP measurements, IAPP

While IAPP is generally thought to be produced peripherally in the pancreas and not in the brain, both IAPP mRNA

Amyloid Beta and Islet Amyloid Polypeptide

Current Alzheimer Research, 2014, Vol. 11, No. 10 9

and protein are present in the brain [52] (unpublished observations by Dr Westermark [44]), and data mined from dataset ([53] Gene expression omnibus GEO http://www.ncbi.nlm.nih.gov/geo/tools/profileGraph.cgi?ID= GDS810:207062_at). Indeed, there was a trend towards an elevation of IAPP mRNA expression with increasing severity of AD when the RNA was specifically extracted from astrocytes isolated by laser capture dissection (http://www.ncbi.nlm.nih.gov/geo/tools/profileGraph.cgi?ID =GDS4135:207062_at [90]). A recent publication indicated that IAPP is not generated in the brain [49], however brain IAPP mRNA may have subtle sequence differences and thus not detected by the IAPP (amylin) probe Hs00169095 in that study. Alternately, production of IAPP by a small subset of cells within the brain may not be detected in total brain extracts. Importantly, IAPP immunoreactivity is found in many brain regions including the hippocampus and cortex [45], precisely the regions where A deposition occurs in AD pathology. There is precedence for the argument as to whether a pancreatic protein is also produced in the brain. For example, it was previously thought that insulin present in the brain was only derived from the periphery, however it is now known that the brain produces insulin [91]. While the source of IAPP is important, we clearly demonstrate that IAPP can enter the brain.

With regards to IAPP and A, interactions occur at nanomolar affinities [35, 36]. Also cross-seeding appears to require the use of unaggregated, monomeric amyloid, as fibrillar A and IAPP are unable to cross-seed [14]. More importantly, Dr Soto has indicated that crossing of transgenic IAPP mice with the Tg2576 model resulted in the augmentation of both T2D and AD pathologies (personal communication). While the coimmunostaining of brain tissue does not demonstrate cause and effect, taken with the above indicates a role in pathogenesis but further work is required in this new research area.

While IAPP is present in CSF, it is currently unknown if the source of the IAPP is either from astrocytes in the brain or due to greater amounts of peripheral IAPP entering the brain. Regardless, there are clear implications of IAPP in the brain. For example, it is possible that the elevation of IAPP in the CSF could underlie the weight loss associated with AD [92], as IAPP induces weight loss and reduced food intake [32]. Additionally, since centrally administered IAPP induces memory loss [55], it would be interesting to evaluate the role of CSF IAPP in memory loss in humans and AD. Last and importantly, it remains to be determined if IAPP in the brain directly augments A misfolding in vivo.

However, the association between peripheral levels of IAPP in blood/plasma and AD is not straightforward. Indeed, while IAPP levels are elevated in pre-diabetes/insulin resistance and obesity [28-32], IAPP levels are actually reduced with T2D and aging [31, 97, 98]. Thus IAPP levels may better correlate with pre-AD stages such as mild cognitive impairment (MCI), where IAPP levels are likely elevated to augment pathology, and not with AD years later. We tested this scenario using Tg2576 mice at a stage where prediabetes or insulin resistance is known to occur. Peripheral IAPP levels were not significantly different between AD and non-AD genotype or in the presence of an anti-diabetic drug.

IAPP Association with Plaques and Oligomers We demonstrate that IAPP associates with amyloid plaques in 2 mouse models of AD, as well as amyloid plaques in familial AD with the Swedish mutation. We did not observe such colocalization in sporadic AD cases. Since IAPP augments A oligomer formation both in vitro and in silico, we determined that IAPP associates with A oligomers in the Tg2576 mouse models of AD. Our in vitro data suggests that IAPP may stabilize A oligomers within the CSF. This is supported by an independent publication as modified IAPP augments A oligomer formation [36]. The formation of A oligomers is of importance as oligomeric A is thought to be the toxic species in AD (37-41). For example, while levels of monomeric A are generally reported to decrease in the CSF with AD severity, the levels of toxic A oligomers actually increases in AD CSF [93]. Testing the idea that increased levels of IAPP stabilize A in the CSF will require the development of more sensitive A oligomer assays, as few studies detect A oligomers in the CSF [93-96]. Could such cross-seeding occur in vivo? As mentioned in the introduction, there is a precedence for cross-seeding between other disparate amyloids in vitro and in vivo [5-11].

Role in Pathology Since elevated levels of IAPP in the blood could lead to IAPP entering the brain, and IAPP is elevated with obesity and the early insulin resistance stages of T2D (28-32), we determined if blood/plasma levels of IAPP correlated with AD diagnosis. If so, this would represent what some researchers in the field have termed an “infective amyloid” mechanism and has been demonstrated for peripherally injected A [12, 13, 22, 25-27). However, our data demonstrate that peripheral IAPP levels were not significantly elevated in blood from an African American cohort or plasma from the Tg2576 mouse model of AD. Thus our data does not support an “infective” or cross-seeding hypothesis.

Our IAPP plasma data contrast that of two recent publications which observed a positive correlation between IAPP and AD [50, 51]. Indeed, in these publications, IAPP or IAPP analogs cleared A plaques [50] and improved cognitive measures in transgenic mice [50, 51]. The mechanism whereby elevated levels of peripheral IAPP removes A from the brain to cause elevations of blood A levels [50] may be analogous to the “peripheral sink” mechanism described for amyloid antibody treatments [99]. Or speculatively, it may lead to the development of anti-IAPP antibodies that can cross-react with A. Perhaps the A/IAPP cospecies may be more rapidly cleared from the brain? Also, completion for the insulin degrading enzyme (IDE), which can cleave both IAPP and A [100], could also account for elevation of A following IAPP i.p. injection. There are many avenues for research in this new field. Regardless, the association of peripheral IAPP and AD does not support the “infectious amyloid” hypothesis. Indeed, a recent study by the Trojanowski group also indicated that non-prion amyloids are not infectious [101]. Astrocytic IAPP Protein IAPP protein is present in cultured murine astrocytes, as well as in activated astrocytes in proximity to plaques in the

10 Current Alzheimer Research, 2014, Vol. 11, No. 10

Tg2576 mice. It may be that inflammation around the plaques activates IAPP expression. Indeed, there was a trend towards an elevation of IAPP mRNA expression with increasing severity of AD when the RNA was specifically extracted from astrocytes isolated by laser capture dissection (http://www.ncbi.nlm.nih.gov/geo/tools/profileGraph.cgi?ID =GDS4135:207062_at, [90]). Our novel data indicates that IAPP was secreted into the media upon stimulation of astrocytes by LPS and TNF, suggesting that astrocytes are likely the source of IAPP in the CSF. The presence of IAPP in astrocytes is intriguing as astrocyte activation is an early feature of AD, as ascertained by imaging studies [102]. CONCLUSION Our current data do not support a correlation between peripheral levels of IAPP in blood/plasma and AD pathology, ruling out a mechanistic or “infectious” link between peripheral changes of IAPP in Type 2 diabetes and AD. However to our knowledge, this is the first demonstration of a potential novel association between brain derived IAPP and AD. This suggests that IAPP plays a direct role in A oligomerization and AD pathology, and this potential connection requires further research. There are translational impacts of this finding since glucose dysfunction/insulin resistance occurs in the brain of AD patients (termed type 3 diabetes) [3, 103], and brain insulin resistance represents a therapeutic target in AD [104]. Since IAPP is known to play a role in glucose homeostasis [44], and thus akin to insulin, IAPP resistance may be a novel player in glucose dysfunction in type 3 diabetes of AD, and conversely pre-existing antidiabetic non-amyloidogenic analogues of IAPP may represent potential novel therapies for AD [50, 51]. ABBREVIATIONS AD

= Alzheimer’s disease

A

= Amyloid , A42- Amyloid  42

DAPI

= 4',6-diamidino-2-phenylindole

DMEM

= Dulbecco's Modified Eagle Medium

FBS

= Fetal bovine serum

GFAP

= glial fibrillary acidic protein

IAPP

= islet amyloid polypeptide

i.p.

= intraperitoneal

LPS

= lipopolysaccharide

T2D

= Type 2 diabetes

TNF 

= Tumor necrosis factor 

Fawver et al.

were obtained via collaboration with Drs. Fawn Cothran and Lisa Barnes at Rush University’s Alzheimer’s disease Center (Minority Aging Research Study is supported by the NIH R01 AG022018 grant to Dr Lisa Barnes). The anti-A oligomeric antibodies NAB 228.3 and NU-4 were gifts from Dr. Virginia M.Y. Lee, and Drs. Pascal Lacor and William Klein respectively. Human CSF and accompanying data was obtained from the Kathleen Bryce brain bank at Duke University. The Kathleen Bryce brain bank is supported by the NIA AG05128 grant. FUNDING This study was supported by funds from the Alzheimer’s Art Quilt Initiative (AAQI), the Texas Brain and Spine Institute Award, and startup funds from the Texas A&M Health Science Center. AUTHOR CONTRIBUTIONS Ms Janelle N. Fawver performed the research/study, collected and analyzed data, as well as wrote and proof read manuscript. Yonatan Ghiwot, Catherine Koola, and Wesley Carrera collected the data and performed experiments. Drs Jennifer Rodriguez-Rivera, Caterina Hernandez, and Kelly T. Dineley contributed important reagents, the blood from transgenic Tg2576 and non-transgenic littermates. Yu Kong, and Dr. Jianrong Li provided astrocyte cultures and culture media of treated cells. Dr. Jack Jhamandas provided us with the IAPP antibody, and performed the independent immunostaining of brain tissue from the independent CRND8 AD mouse model. Dr. George Perry provided us with fixed human brain tissues, and contributed to writing the manuscript. Dr. Ian V.J. Murray designed all of the experiments, performed some of the experiments, collected and analyzed data, and wrote and proof read the manuscript. I would also like to dedicate this manuscript to good friends who passed away during the preparation of this manuscript; especially my mentor, friend and colleague, Dr. Luis V. Colom of U.T. Brownsville. SUPPLEMENTARY MATERIALS Supplementary material is available on the publishers web site along with the published article. REFERENCES [1] [2]

CONFLICT OF INTEREST

[3]

The authors confirm that this article content has no conflict of interest.

[4]

ACKNOWLEDGEMENTS

[5]

We would like to thank Sriram Shankar for editorial assistance in the preparation of this manuscript. Human plasma and the accompanying data from African American patients is part of the Minority Aging Research Study cohort and

[6]

Adeghate E, Donáth T, Adem A. Alzheimer disease and diabetes mellitus:do they have anything in common? Curr Alzheimers Res 10(6): 609-17 (2013). Malek-Ahmadi M, Beach T, Obradov A, Sue L, Belden C, Davis K, et al. Increased Alzheimer's disease neuroptahology is associated with type 2 diabetes and Apo E4 carrier status. Curr Alzheimers Res 10(6): 654-9 (2013). de la Monte SM, Wands JR. Alzheimer's disease is type 3 diabetesevidence reviewed. J Diabetes Sci Technol 2(6): 1101-13 (2008). de la Monte SM. Brain insulin resistance and deficiency as therapeutic targets in Alzheimer's disease. Curr Alzheimers Res 2012;9(1):35-66. Guo JP, Arai T, Miklossy J, McGeer PL. A and tau form soluble complexes that may promote self aggregation of both into the insoluble forms observed in Alzheimer's disease. Proc Natl Acad Sci USA 103(6): 1953-8 (2006). Giasson BI, Forman MS, Higuchi M, Golbe LI, Graves CL, Kotzbauer PT, et al. Initiation and synergistic fibrillization of tau and alpha-synuclein. Science 300(5619): 636-40 (2003).

Amyloid Beta and Islet Amyloid Polypeptide [7]

[8]

[9]

[10] [11]

[12] [13] [14] [15]

[16] [17]

[18]

[19]

[20]

[21] [22]

[23]

[24]

[25] [26]

[27] [28] [29]

[30]

Han H, Weinreb PH, Lansbury J. The core Alzheimer's peptide NAC forms amyloid fibrils which seed and are seeded by amyloid: is NAC a common trigger or target in neurodegenerative disease? Chem Biol 2(3): 163-9 (1995). Larsson A, Malmström S, Westermark P. Signs of cross-seeding: aortic medin amyloid as a trigger for protein AA deposition. Amyloid 18(4): 229-34 (2011). Morales R, Estrada LD, Diaz-Espinoza R, Morales-Scheihing D, Jara MC, Castilla J, et al. Molecular cross talk between misfolded proteins in animal models of Alzheimer's and prion diseases. J Neurosci 30(13): 4528-35 (2010). Ono K, Takahashi R, Ikeda T, Yamada M. Cross-seeding effects of amyloid -protein and -synuclein. J Neurochem 122(5): 883-90 (2012). Ma B, Nussinov R. Selective molecular recognition in amyloid growth and transmission and cross-species barriers. J Mol Biol 421(23): 172-84 (2012). Prusiner SB. A unifying role for prions in neurodegenerative diseases. Science 336(6088): 1511-3 (2012). Lahiri DK. Prions: A piece of the puzzle? Science 337(6099): 1172 (2012). O'Nuallain B, Williams AD, Westermark P, Wetzel R. Seeding specificity in amyloid growth induced by heterologous fibrils. J Biol Chem 279(17): 17490-9 (2004). Lasagna-Reeves CA, Castillo-Carranza DL, Guerrero-Muñoz MJ, Jackson GR, Kayed R. Preparation and characterization of neurotoxic tau oligomers. Biochem 49(47): 10039-41 (2010). Yan J, Fu X, Ge F, Zhang B, Yao J, Zhang H, et al. Cross-seeding and cross-competition in mouse apolipoprotein A-II amyloid fibrils and protein A amyloid fibrils. Am J Pathol 171(1): 172-80 (2007). Lundmark K, Westermark GT, Olsén A, Westermark P. Protein fibrils in nature can enhance amyloid protein A amyloidosis in mice: Cross-seeding as a disease mechanism. Proc Natl Acad Sci USA 102(17): 6098-102 (2005). Lee VMY, Giasson BI, Trojanowski JQ. More than just two peas in a pod: common amyloidogenic properties of tau and -synuclein in neurodegenerative diseases. Trends Neurosci 27(3): 129-34 (2004). Bhogal R, Smith DM, Purkiss P, Bloom SR. Molecular identification of binding sites for calcitonin gene-related peptide (CGRP) and islet amyloid polypeptide (IAPP) in mammalian lung: species variation and binding of truncated CGRP and IAPP. Endocrinology 133(5): 2351-61 (1993). Walker LC, Bian F, Callahan MJ, Lipinski WJ, Durham RA, LeVine H. Modeling Alzheimer's disease and other proteopathies in vivo: Is seeding the key? Amino Acids 23(1): 87-93 (2002). Walker L, LeVine H, Jucker M. Koch's postulates and infectious proteins. Acta Neuropathol 112(1): 1-4 (2006). Meyer-Luehmann M, Coomaraswamy J, Bolmont T, Kaeser S, Schaefer C, Kilger E, et al. Exogenous induction of cerebral amyloidogenesis is governed by agent and host. Science 313(5794): 1781-4 (2006). Lundmark K, Westermark GT, Nyström S, Murphy CL, Solomon A, Westermark P. Transmissibility of systemic amyloidosis by a prion-like mechanism. Proc Natl Acad Sci USA 99(10): 6979-84 (2002). Desplats P, Lee HJ, Bae EJ, Patrick C, Rockenstein E, Crews L, et al. Inclusion formation and neuronal cell death through neuron-toneuron transmission of -synuclein. Proc Natl Acad Sci USA 106(31): 13010-5 (2009). Eisele YS, Obermüller U, Heilbronner G, Baumann F, Kaeser SA, Wolburg H, et al. Peripherally applied A-containing inoculates induce cerebral ß-amyloidosis. Science 330(6006): 980-2 (2010). Morales R, Duran-Aniotz C, Castilla J, Estrada LD, Soto C. De novo induction of amyloid- deposition in vivo. Mol Psychiatry 17(12): 1347-53 (2011). Kim J, Holtzman DM. Prion-like behaviour of amyloid . Science 330(6006): 918-9 (2010). Boyle CN, Rossier MM, Lutz TA. Influence of high-fat feeding, diet-induced obesity, and hyperamylinemia on the sensitivity to acute amylin. Physiol Behav 104(1): 20-8 (2011). Johnson KH, Brien TD, Westermark P. Newly identified pancreatic protein islet amyloid polypeptide: what is its relationship to diabetes? Diabetes 40(3): 310-4 (1991). Takada K, Kanatsuka A, Tokuyama Y, Yagui K, Nishimura M, Saito Y, et al. Islet amyloid polypeptide/amylin contents in

Current Alzheimer Research, 2014, Vol. 11, No. 10 11

[31]

[32]

[33] [34]

[35]

[36]

[37]

[38] [39] [40]

[41] [42] [43]

[44] [45]

[46] [47]

[48] [49]

[50]

[51]

pancreas change with increasing age in genetically obese and diabetic mice. Diabetes Res Clin Pract 33(3): 153-8 (1996). Hanabusa T, Kubo K, Oki C, Nakano Y, Okai K, Sanke T, et al. Islet amyloid polypeptide (IAPP) secretion from islet cells and its plasma concentration in patients with non-insulin-dependent diabetes mellitus. Diabetes Res Clin Pract 15(1): 89-96 (1992). Lutz TA. The role of amylin in the control of energy homeostasis. Am J Physiol Regul Integr Comp Phsyiol 298(6): R1475-R1484 (2010). Hull RL, Westermark GT, Westermark P, Kahn SE. Islet amyloid: A critical entity in the pathogenesis of type 2 diabetes. J Clin Endocrinol Metab 89(8): 3629-43 (2004). Murray IVJ, Proza JF, Sohrabji F, Lawler JM. Vascular and metabolic dysfunction in Alzheimer's disease: a review. Exp Biol Med (Maywood) 236(7): 772-82 (2011). Andreetto E, Yan LM, Tatarek-Nossol M, Velkova A, Frank R, Kapurniotu A. Identification of hot regions of the A-APP interaction interface as high-affinity binding sites in both cross- and self-association. Agnew Chem Int Ed Engl 49(17): 3081-5 (2010). Yan LM, Velkova A, Tatarek-Nossol M, Andreetto E, Kapurniotu A. IAPP mimic blocks a cytotoxic self-assembly: crosssuppression of amyloid toxicity of a and IAPP suggests a molecular link between Alzheimer's disease and type I diabetes. Agnew Chem Int Ed Engl 46(8): 1246-52 (2007). Benilova I, Karran E, De Strooper B. The toxic A oligomer and Alzheimer's disease: an emperor in need of clothes. Nat Neurosci 15(3): 349-57 (2012). Ferreira ST, Klein WL. The A oligomer hypothesis for synapse failure and memory loss in Alzheimer's disease. Neurobiol Learn Mem 96(4): 529-43 (2011). Larson ME, Lesné SE. Soluble A oligomer production and toxicity. J Neurochem 2012;120:125-39. Selkoe DJ. Soluble oligomers of the amyloid -protein impair synaptic plasticity and behavior. Behav Brain Res 192(1): 106-13 (2008). Walsh DM, Selkoe DJ. A beta oligomers - a decade of discovery. J Neurochem 101(5): 1172-84 (2007). Banks WA, Kastin AJ. Differential permeability of the bloodbrain barrier to two pancreatic peptides: insulin and amylin. Peptides 19(5): 883-9 (1998). Banks WA, Kastin AJ, Maness LM, Huang W, Jaspan JB. Permeability of the blood-brain barrier to amylin. Life Sci 57(22): 1993-2001(1995). Westermark P, Andersson A, Westermark GT. Islet amyloid polypeptide, islet amyloid, and diabetes mellitus. Physiol Rev 91(3): 795-826 (2011). Skofitseh G, Gubisch W, Wimalawansa SJ, Jacobowitz DM. Comparative immunohistochemical distribution of amylin-like and calcitonin gene related peptide like immunoreactivity in the rat central nervous system. Can J Physiol Pharmacol 3(7): 945-56 (1995). Sexton PM, Paxinos G, Kenney MA, Wookey PJ, Beaumont K. In vitro autoradiographic localization of amylin binding sites in rat brain. Neuroscience 62(2): 553-67 (1994). van Rossum D, Ménard DP, Fournier A, St-Pierre S, Quirion R. Autoradiographic distribution and receptor binding profile of [125I]Bolton Hunter-rat amylin binding sites in the rat brain. Journal of Pharmacology and Experimental Therapeutics 270(2): 779-87 (1994). Beaumont K, Kenney MA, Young AA, Rink TJ. High affinity amylin binding sites in rat brain. Mol Pharmacol 44(3): 493-7 (1993). Jackson K, Barisone GA, Diaz E, Jin Lw, DeCarli C, Despa F. Amylin deposition in the brain: A second amyloid in Alzheimer's disease? Ann Neurol 74(4): 517-26 (2013). Zhu H, Wang X, Wallack M, Li H, Carreras I, Dedeoglu A, et al. Intraperitoneal injection of the pancreatic peptide amylin potently reduces behavioral impairment and brain amyloid pathology in murine models of Alzheimer's disease. Molecular Psychiatry (2014);doi: 10.1038/mp.2014.17. Adler BL, Yarchoan M, Hwang HM, Louneva N, Blair JA, Palm R, et al. Neuroprotective effects of the amylin analogue pramlintide on Alzheimer's disease pathogenesis and cognition. Neurobiol Aging 35(4): 793-801 (2014).

12 Current Alzheimer Research, 2014, Vol. 11, No. 10 [52]

[53]

[54] [55]

[56]

[57] [58]

[59] [60] [61]

[62] [63]

[64] [65] [66]

[67] [68]

[69]

[70] [71]

[72]

[73]

[74] [75]

Szabó ÉR, Cservenák M, Dobolyi A. Amylin is a novel neuropeptide with potential maternal functions in the rat. FASEB J 26(1): 272-81 (2012). Blalock EM, Geddes JW, Chen KC, Porter NM, Markesbery WR, Landfield PW. Incipient Alzheimer's disease: Microarray correlation analyses reveal major transcriptional and tumor suppressor responses. Proc Natl Acad Sci USA 101(7): 2173-8 (2004). Lutz TA. Pancreatic amylin as a centrally acting satiating hormone. Curr Drug Targ 6(2): 1389-4501 (2005). Flood JF, Morley JE. Differential effects of amylin on memory processing using peripheral and central routes of administration. Peptides 13(3): 577-80 (1992). Fu W, Ruangkittisakul A, MacTavish D, Shi JY, Ballanyi K, Jhamandas JH. Amyloid  (A) peptide directly activates amylin-3 receptor subtype by triggering multiple intracellular signaling pathways. J Biol Chem 287(22): 18820-30 (1992). Jhamandas JH, Li Z, Westaway D, Yang J, Jassar S, MacTavish D. Actions of amyloid protein on human neurons are expressed through the amylin receptor. Am J Pathol 178(1): 140-9 (2011). Berhanu WM, Yasar F, Hansmann UHE. In silico cross seeding of A and amylin fibril-like oligomers. ACS Chem Neurosci 4(11): 1488-500 (2013). Rodriguez-Rivera J, Denner L, Dineley KT. Rosiglitazone reversal of Tg2576 cognitive deficits is independent of peripheral glucoregulatory status. Behav Brain Res 216(1): 255-61 (2011). Alzheimer's Association. Alzheimer's disease facts and figures. Alzheimers Dement 8(2): 131-68 (2012). Seeleang K. Genetic disparities in the development of type 2 diabetes among African Americans. J Am Acad Nurse Pract 23(9): 473-8 (2011). Kountz D. Special considerations of care and risk management for African American patients With type 2 diabetes mellitus. J Natl Med Assoc 104(5/6): 265-73 (2012). Cooper RS, Rotimi CN, Kaufman JS, Owoaje EE, Fraser H, Forrester T, et al. Prevalence of NIDDM among populations of the African diaspora. Diabetes Care 20(3): 343-8 (1997). Haffner SM. Epidemiology of Type 2 Diabetes: Risk Factors. Diabetes Care 21(3): C3-C6 (1998). Elbein SC. Evaluation of polymorphisms known to contribute to risk for diabetes in African and African-American populations. Curr Opin Clin Nutr Metab Care 10(4): 415-9 (2007). Gurland BJ, Wilder DE, Lantigua R, Stern Y, Chen J, Killeffer EHP, et al. Rates of dementia in three ethnoracial groups. Int J Geriat Psychiatry 14(6): 481-93 (1999). Miles TP, Froehlich TE, Bogardus ST, Inouye SK. Dementia and race: Are there differences between African Americans and Caucasians? J Am Geriatr Soc 49(4): 477-84 (2001). Giasson BI, Duda JE, Murray IVJ, Chen QP, Souza JM, Hurtig HI, et al. Oxidative damage linked to neurodegeneration by selective alpha-synuclein nitration in synucleinopathy lesions. Science 290(5493): 985-9 (2000). Lee EB, Leng LZ, Zhang B, Kwong L, Trojanowski JQ, Abel T, et al. Targeting amyloid-beta peptide (A beta) oligomers by passive immunization with a conformation-selective monoclonal antibody improves learning and memory in A beta precursor protein (APP) transgenic mice. J Biol Chem 281(7):4292-9 (2006). Lambert MP, Velasco PT, Chang L, Viola KL, Fernandez S, Lacor PN, et al. Monoclonal antibodies that target pathological assemblies of Abeta. J Neurochem 100(1): 23-35 (2007). Murakami K, Horikoshi-Sakuraba Y, Murata N, Noda Y, Masuda Y, Kinoshita N, et al. Monoclonal antibody against the turn of the 42-residue amyloid -protein at positions 22 and 23. ACS Chem Neurosci 1(11): 747-56 (2010). Tokuraku K, Marquardt M, Ikezu T. Real-time imaging and quantification of amyloid- peptide aggregates by novel quantumdot nanoprobes. PLoS ONE 4(12): e8492 (2009). Li J, Ramenaden ER, Peng J, Koito H, Volpe JJ, Rosenberg PA. Tumor necrosis factor  mediates lipopolysaccharide-induced microglial toxicity to developing oligodendrocytes when astrocytes are present. J Neurosci 28(20): 5321-30 (2008). Alessio DA, Ensinck JW, Fujimoto WY, Kahn SE, Porte DJ, Jr., Schwartz MW, et al. Evidence of cosecretion of islet amyloid polypeptide and insulin by beta-cells. Diabetes 39(5): 634 (1990). Harris PJ, Cooper ME, Hiranyachattada S, Berka JL, Kelly DJ, Nobes M, et al. Amylin stimulates proximal tubular sodium

Fawver et al.

[76]

[77]

[78]

[79]

[80]

[81]

[82] [83]

[84] [85]

[86]

[87] [88]

[89]

[90]

[91]

[92] [93]

[94]

[95]

[96]

transport and cell proliferation in the rat kidney. Am J Physiol Renal Physiol 272(1): F13-F21 (1997). Wookey PJ, Cao Z, van Geenen RCI, Voskuil M, Darby IA, Komers R, et al. Increased density of renal amylin binding sites in experimental hypertension. Hypertension 30(3): 455-60 (1997). Stridsberg M, Tjalve H, Wilander E. Whole-Body Autoradiography of 123I-Labelled Islet Amyloid Polypeptide (IAPP) Accumulation in the lung parenchyma and in the villi of the intestinal mucosa in rats. Acta Oncol 32(2): 155-9 (1993). Bhogal R, Smith DM, Owji AA, Bloom SR. Binding sites for islet amyloid polypeptide in mammalian lung: species variation and effects on adenylyl cyclase. Can J Physiol Pharmacol 73(7): 1030-6 (1995). Jaikaran ET, Nilsson MR, Clark A. Pancreatic -cell granule peptides form heteromolecular complexes which inhibit islet amyloid polypeptide fibril formation. Biochem J 377: 709-16 (2004). Rezaei-Ghaleh N, Andreetto E, Yan LM, Kapurniotu A, Zweckstetter M. Interaction between amyloid beta peptide and an aggregation blocker peptide mimicking islet amyloid polypeptide. PLoS ONE 6(5): e20289 (2011). Fawver JN, Duong KT, Wise-Scira O, Schall HE, Coskuner O, Zhu X, et al. Probing and trapping a sensitive conformation: Amyloid beta fibrils, oligomers and dimers. J Alzheimers Dis 32(1): 197-215 (2012). Ho L, Qin W, Pompl PN, Xiang Z, Wang J, Zhao Z, et al. Dietinduced insulin resistance promotes amyloidosis in a transgenic mouse model of Alzheimer's disease. FASEB J 18: 902-4 (2004). Richardson JA, Burns DK. Mouse models of Alzheimer's disease: a quest for plaques and tangles. Institute of Laboratory for Animal Reseources 43(2): 89-99 (2002). Bhogal R, Smith DM, Bloom SR. Investigation and characterization of binding sites for islet amyloid polypeptide in rat membranes. Endocrinology 1992;130(2):906-13. Despa S, Margulies KB, Chen L, Knowlton AA, Havel PJ, Taegtmeyer H, et al. Hyperamylinemia contributes to cardiac dysfunction in obesity and diabetes. A study in humans and rats. Circ Res 110(4): 598-608 (2012). Gong W, Liu ZH, Zeng CH, Peng A, Chen HP, Zhou H, et al. Amylin deposition in the kidney of patients with diabetic nephropathy. Kidney Int 72(2): 213-8 (2007). Ozturk M, de la Monte SM, Gross J, Wands JR. Elevated levels of an exocrine pancreatic secretory protein in Alzheimer disease brain. Proc Natl Acad Sci USA 86(2): 419-23 (1989). O'Bryant SE, Xiao G, Barber R, Huebinger R, Wilhelmsen K, Edwards M, et al. A blood-based screening tool for Alzheimer's disease that spans serum and plasma: Findings from TARC and ADNI. PLoS ONE 6(12): e28092 (2011). De Felice FG. Alzheimer's disease and insulin resistance: translating basic science into clinical applications. J Clin Invest 123(2): 531-9 (2013). Simpson JE, Ince PG, Shaw PJ, Heath PR, Raman R, Garwood CJ, et al. Microarray analysis of the astrocyte transcriptome in the aging brain: relationship to Alzheimer's pathology and APOE genotype. Neurobiol Aging 32(10): 1795-807 (2011). Steen E, Terry BM, Rivera EJ, Cannon JL, Neely TR, Tavares R, et al. Impaired insulin and insulin-like growth factor expression and signaling mechanisms in Alzheimer's disease - is this type 3 diabetes? J Alzheimers Dis 7(1): 63-80 (2005). Sergi G, De Rui M, Coin A, Inelmen EM, Manzato E. Weight loss and Alzheimer's disease: temporal and aetiologic connections. Proc Nutr Soc 72(01): 160-5 (2013). Handoko M. Correlation of specific amyloid- oligomers with tau in cerebrospinal fluid from cognitively normal older adults. JAMA Neurology 70(5): 594-9 (2013). Haes AJ, Chang L, Klein WL, Van Duyne RP. Detection of a biomarker for Alzheimer's disease from synthetic and clinical samples using a nanoscale optical biosensor. J Am Chem Soc 127(7): 2264-71 (2005). Georganopoulou DG, Chang L, Nam JM, Thaxton CS, Mufson EJ, Klein WL, et al. Nanoparticle-based detection in cerebral spinal fluid of a soluble pathogenic biomarker for Alzheimer's disease. Proc Natl Acad Sci USA 102(7): 2273-6 (2005). Fukumoto H, Tokuda T, Kasai T, Ishigami N, Hidaka H, Kondo M, et al. High-molecular-weight -amyloid oligomers are elevated in

Amyloid Beta and Islet Amyloid Polypeptide

[97]

[98]

[99] [100]

[101]

Current Alzheimer Research, 2014, Vol. 11, No. 10 13

cerebrospinal fluid of Alzheimer patients. FASEB J 24(8): 2716-26 (2010). Dechenes CJ, Verchere CB, Andrikopoulos S, Kahn SE. Human aging is associated with parallel reductions in insulin and amylin release. Am J Physiol Endocrinol Metab 275(5): E785-E791 (1998). Kahn SE, Verchere CB, Andrikopoulos S, Asberry PJ, Leonetti DL, Wahl PW, et al. Reduced amylin release is a characteristic of impaired glucose tolerance and type 2 diabetes in Japanese Americans. Diabetes 47(4): 640-5 (1998). Zhang Y, Lee DHS. Sink hypothesis and therapeutic strategies for attenuating A levels. Neurocientist 17(2): 163-73 (2011). Kurochkin IV. Amyloidogenic determinant as a substrate recognition motif of insulin-degrading enzyme. FEBS Lett 427(2): 153-6 (1998) Irwin D.J., Abrams J.Y., Schonberger L.B., Leschek EW, Mills JL, Lee VMY, et al. Evaluation of potential infectivity of alzheimer

Received: ????????????????

[102]

[103]

[104] [105]

and parkinson disease proteins in recipients of cadaver-derived human growth hormone. JAMA Neurology 70(4): 462-8 (2013). Carter SF, Schöll M, Almkvist O, Wall A, Engler H, Långström B, et al. Evidence for astrocytosis in prodromal Alzheimer disease provided by 11C-deuterium-L-deprenyl: A multitracer PET paradigm combining 11C-Pittsburgh Compound B and 18F-FDG. J Nucl Med 53(1): 37-46 (2012). de la Monte SM, Tong M, Lester-Coll N, Plater J, Wands JR. Therapeutic rescue of neurodegeneration in experimental type 3 diabetes: Relevance to Alzheimer's disease. J Alzheimers Dis 10(1): 89-109 (2006 ). de la Monte SM. Therapeutic targets of brain insulin resistance in sporadic Alzheimer's disease. Front Biosci 4: 1582-605(2012). Tomiyama T, Nagata T, Shimada H, Teraoka R, Fukushima A, Kanemitsu H, et al. A new amyloid beta variant favoring oligomerization in Alzheimer's-type dementia. Ann Neurol 63(3): 377-87 (2008).

Revised: ????????????????

Accepted: ????????????????