Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Presenilin 1 transgene addition to amyloid precursor protein overexpressing transgenic rats increases amyloid beta 42 levels and results in loss of memory retention

Presenilin 1 transgene addition to amyloid precursor protein overexpressing transgenic rats... Background: We previously reported the production of transgenic rats (APP21 line) that over-express human amyloid precursor protein (APP) containing Swedish and Indiana mutations. In order to generate a better model for Alzheimer’s disease (AD), the APP21 rat line was used to generate double transgenic line that over-expressed Preseni- lin 1 (PS1) with L166P mutation in addition to APP transgene (APP + PS1 line). Results: Thirty-two double transgenic founders were generated and the ultimate transgenic founder was selected based on PS1 transgene copy number and level of amyloid-beta (Aβ) peptide. The APP + PS1 double transgenic rats had 38 times more PS1 in brains compared to APP rats. Behavioral assessment using Barnes maze showed that APP + PS1 rats exhibited a larger learning and memory deficit than APP21 rats. Double transgenic rats also produced more Aβ . Histological examination of the brains showed that the APP21 rat line displayed neurofibrillary tangles and in contrast, the APP + PS1 line showed chromatolysis in hippocampal neurons and neuronal loss in CA3 region of hippocampus. Conclusions: Due to the separate segregation of APP and PS1 transgenes in APP + PS1 double transgenic rats, this transgenic line may be a valuable model for studying the effects of various levels of APP and PS1 transgenes on vari- ous aspects of brain pathologies associated with the AD phenotype. Background Amyloid peptides are produced via the cleavage of APP Alzheimer’s disease (AD) is one of the most devastat- by ϒ-secretase, a protease complex of four proteins that ing and costly diseases that affect approximately 12 % of includes PS1 or PS2, which are both homologous pro- the population over the age of 65 [2]. Alzheimer’s dis- teins that contain the catalytic site of the enzyme [7]. ease is associated with the over-production and reduced ϒ-secretase cleaves the β-amyloid protein into two major clearance of amyloid-beta (Aβ) peptides [19]. Cleavage forms of Aβ polypeptides, Aβ and Aβ . The relative 40 42 of amyloid-β precursor protein (APP) by the preseni- amount of the longer form of Aβ, Aβ , is particularly lin (PS) 1 enzyme results in the release of Aβ peptides, critical for AD progression because it is more prone to which aggregate and form Aβ plaques in the brain. These aggregation than the shorter Aβ peptide [4, 10, 23]. Aβ plaques and neurofibrillary tangles are the two pri - In addition, the ratio of Aβ /Aβ correlates with the 42 40 mary pathological manifestations of AD in the brain. load of Aβ plaques in the brain [8, 14]. Furthermore, the reduction of PS1 mRNA using siRNA against PS1 reduced Aβ in cultured cells [11] which suggests that *Correspondence: agcay@missouri.edu 1 lower levels of PS1 lead to decreased production of Aβ . Department of Veterinary Pathobiology, College of Veterinary Medicine, University of Missouri, 1600 East Rollins Street, Room W191, Columbia, We previously generated rats that overexpress human MO 65211, USA APP containing the Swedish and Indiana mutations [1]. Full list of author information is available at the end of the article © 2016 The Author(s). This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/ publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Agca et al. BMC Neurosci (2016) 17:46 Page 2 of 10 The APP21 line expressed high levels of APP, but failed gonadotropins (PMSG) followed by IP injections of to produce Aβ plaques in the brain. However, intrahip- 15  IU human chorionic gonadotropin 52  h after PMSG. pocampal seeding of dilute Alzheimer brain extracts con- Twenty-four hours after mating, rats were euthanized taining aggregated Aβ led to the production of plaques using CO and zygotes were collected from the oviducts. in APP21 rats [20]. This finding indicates that plaque Morphologically normal zygotes having two pronuclei formation depends on environmental challenges, condi- and a sperm tail were used for lentiviral vector injections tions which laboratory rats are not exposed to routinely. into perivitelline space as described earlier [1]. Lentiviral In order to increase the pathogeny of overexpressed APP, vector-injected zygotes were then transferred into the we generated double transgenic rats from the APP21 line oviducts of 8–10 week-old Sprague–Dawley pseudopreg- that over-express both the PS1 and the APP transgenes. nant recipient rats. Our hypothesis was that the addition of a PS1 transgene would increase more toxic Aβ polypeptide and thus, Genomic DNA isolation and polymerase chain reaction would produce a more suitable model for examining AD. Genomic DNA from tail-snips was isolated using the Wiz- ard Genomic DNA Purification Kit (Promega, Madison, Methods WI). Screening of transgenic rats was conducted by PCR Animals, construct design, and preparation of vesicular using primers annealing the Ubi-C promoter (Forward; stomatitis virus glycoprotein pseudotyped lentivirus TGTCCGCTAAATTCTGGCCGTT) and PS1 transgene The homozygous APP21-transgenic rat line [1 ], con- (Reverse; AGCATTCAGAATTGAGTGCAGGGC). The taining human APP with the Swedish double missense PCR reactions were carried out using 50  ng genomic mutation (K595M/N596L) and Indiana single missense DNA, 100  ng of each primer and 1.5 U Biolase taq (Bio- mutation (V642F), were used for the generation of APP- line, Randolph, MA), and amplified PCR products were PS1 double transgenic rats. Inbred Fischer 344 rats were size separated through a 1 % agarose gel stained with eth- used as the background strain for the APP21 rat line. The idium bromide for visualization. human PS1 cDNA coding region corresponding to 285- 1688 bases of PS1 variant 1 (GenBank Accession num- ELISA measurement of Aβ and Aβ 40 42 ber NM_000021.3)  with L166P mutation was cloned Serum samples were analyzed for Aβ and Aβ using 40 42 into the Lentivial vector (pLV) containing ubiquitin-C commercial ELISA kits (Genetics Company, Schlieren, (Ubi-C) and enhanced green fluorescence protein (eGFP; Switzerland). Serum samples were diluted in assay buffer pLVU-eGFP) cassette [17] in place of the eGFP coding and processed according to the manufacturer’s recom- sequence. The new vector was designated as pLVU-PS1. mended protocols. Briefly, samples and standards were The Ubi-C promoter, which is upstream of PS1, controlled incubated in capture wells overnight at 4  °C with either PS1 transcription. The pLVU-PS1 was a self-inactivating a biotinylated Aβ or an Aβ -specific antibody. After 40 42 vector, composed of the woodchuck hepatitis virus post- several rinses, the enzyme-conjugated detection rea- transcriptional regulatory element (WRE) to increase gent was added to the wells for 30  min. After additional transcription levels and minimize position effects. Addi - rinses, wells were incubated with the chromogen solution tionally, a human immunodeficiency virus-1-flap element for 30 min at room temperature and were shielded from was inserted between the 5′ long terminal repeat and the light. After the addition of the stop solution, the wells internal promoter, which increased the titer. The prepara - were read for absorption at 450  nm and the Aβ concen- tion of vesicular stomatitis virus glycoprotein pseudotyped tration in the samples was calculated based on standard Lentivirus was as described by Lois et  al. [17]. All animal curves. studies were performed in accordance with the University of Missouri’s Animal Care and Use Committee guidelines Southern blot analysis and the Institute for Laboratory Animal Research Guide Southern blot analysis was conducted to determine the for the Care and Use of Laboratory Animals. The rats were copy number of the transgene in founder rats. Genomic housed in conventional cages at 20–25  °C in a controlled DNA was digested with EcoRI, which cut the junction of lighting environment and provided free access to water the PS1 and the WRE transgene. The digestion products and standard pellet rodent chow. were size-separated through a 0.8 % agarose gel and trans- ferred to a Genescreen plus membrane (Perkin Elmer, Zygote collection, microinjection of lentiviral vector, Wellesley, MA) overnight. The 524  bp PS1 probe tem - and embryo transfer plate was prepared by amplification of pLVU-PS1 using Homozygous APP21-transgenic female rats (28–30  day- forward (AGGTCCACTTCGTATGCTGGTTGA) and old) were super ovulated by administering intraperi- reverse (TGATGGAGATTGGAAGAGTGGCA) prim- toneal (IP) injections of 15  IU pregnant mares’ serum ers. The P labeled probe was generated using the probe Agca et al. BMC Neurosci (2016) 17:46 Page 3 of 10 template, Ready-To-Go DNA Labeling Beads (Amersham lumins were hung above the platform served to brightly Biosciences, Piscataway, NJ) and [α- P]-dCTP (Perkin light the maze in order to create a potentially aversive Elmer, Wellesley, MA). The membranes were prehybrid - environment to help motivate the rats to escape from the ized in 10  % dextran sulfate, 6  ×  saline sodium sitrate brightly lit, open surface in favor of the dark environment (SSC), 1  % sodium dodecyl sulfate (SDS) for 2  h and of the escape box. One light hung 68.5 on side of the hybridized using the P labeled probe overnight before maze while the other hung 137 cm from the other side. exposure to BioMax MS-1 Autoradiography Film (Kodak, Each rat was assigned an escape hole number under Rochester, NY). which the escape box was placed on each test trial; assigned hole numbers were alternated across rats to Northern blot analysis eliminate odor cues for consecutively tested rats. The Total RNA was isolated using Trisure (Bioline, CA) from escape box location remained constant for any individual founder APP21 + PS1 and APP21 transgenic rats. Brain, rat across test trials. Behavioral testing consisted of two heart, kidney, liver, and lung tissues were used in North- pretraining trials on the first day, and 6 evaluation trials ern blot analysis. Total RNA was size-separated through (2 trials/day) over a period of 3  days. Each day, the ani- 1  % agarose gel before transferring to the Genescreen mals were transferred in their home cage from their col- plus membrane overnight. The membrane was prehy - ony room to the testing room 30 min prior to the start of bridized in 10  % dextran sulfate, 5 ×  SSPE, 50  % forma- testing. On the pretraining trial, the rat was preexposed mide, 5  ×  Denhardt’s, 1  % SDS at 42  °C for 6  h prior to to the goal box (with the goal box placed in the same hybridization with the PS1 probe overnight. After expo- hole that would be used during subsequent training) by sure to BioMax MS-1 Autoradiography Film, membranes placing the rat in the goal box for 90  s. The rats’ escape were stripped to remove labeled cDNA probe by boil- from the goal box was prevented by covering the open- ing for 1 h in 1 %SDS and 0.1× SSC. The intensity of the ing to the maze surface with a grey, opaque start box (a bands was determined using Kodak 1D v 3.6.3 software 23  cm  ×  23  cm box). A training trial began by placing (New Haven, CT). the rat under the start box positioned in the center of the platform. After 30 s, the box was lifted and the rat had a Behavioral measures maximum of 5 min (300 s) to locate and enter the escape Behavioral assessment was conducted using APP  +  PS1 box. Latency (time it took for the rat to find and enter the (n =  12), APP (n =  11), and non-transgenic Fischer 344 escape box) and total errors (nose-pokes into non-escape (n  =  12) rats when their average age was approximately holes) were recorded. If the rat did not enter the escape 10  months. Prior to behavior studies, each rat was han- box within 5 min, it was gently guided there by the exper- dled at least 3 times to reduce stress due to handling dur- imenter’s hand. After 30  s, the rat was removed from ing the behavioral testing. the escape box and returned to its home cage. Rats were allowed to rest in their home cage in the testing room for Barnes maze 30  min before starting their second daily trial. The plat - The maze consisted of a grey circular platform 122 cm in form and escape box were cleaned after every trial with a diameter. The platform was surrounded by a wall that was 20 % ethanol solution. After the third day, testing abated 30.5 cm in height. The maze was elevated 83.8 cm above for 3  days, after which retention was evaluated for two the floor by a stand. Twenty holes measuring 10.2 cm in additional trials (1 day), conducted exactly like those just diameter were evenly spaced around the perimeter. A described. The following day, the rats were given reversal rectangular grey escape box (28 cm in length ×  12.7  cm training in which the rats were given training exactly like wide × 7.6 cm high at the area closest to the maze taper- those of original acquisition, except that the escape box ing to 16.5  cm high) could be placed beneath any hole. was located in a new location (in the opposite quadrant). The escape box included an entry ramp that provided Reversal training continued for a total of 3 days. easy entry access for the rat. Black curtains were hung around the maze and above the maze walls to surround Histopathology the apparatus and ensure that rats could only use the Nine of the female rats [3 of each: APP + PS1, APP, and visual cues provided in the maze, rather than the distal wild type (WT)] were euthanized at 18–19 months of age cues within the testing room. Proximal cues were more and their brains were collected. Tissues were fixed in 10 % likely to remain constant, within subjects and across sub- formalin for 48 h prior to paraffin embedding. Hematox - jects, during the course of training. Four visual cues con- ylin and eosin staining was used to determine histological sisting of various shapes (triangle, square, circle, cross) changes. The stained slides were analyzed using a Nikon were placed at evenly spaced intervals on the inside of the Eclipse E600 microscope (Melville, NY) and an Olympus maze walls. Two 86-W, 120-V floodlights producing 1690 DP72 Camera (Center Valley, PA). Agca et al. BMC Neurosci (2016) 17:46 Page 4 of 10 Statistical analysis Subsequent generations of APP + PS1 double transgenic Statistical analysis for PS1 mRNA expression, serum Aβ rat line level comparisons, and behavioral analyses were per- The APP21 line has been previously bred as a homozy - formed using general linear models in SAS (SAS 9.3, SAS gous transgenic line. However, homozygous APP  +  PS1 Institute Inc., Cary, NC, USA). For the Barnes maze, the rats failed to produce pups. Thus, the APP  +  PS1 dou- average of each animal’s daily score was used in the anal- ble transgenic line was retained as homozygous for APP ysis of overall performance during acquisition and rever- transgene and hemizygous for PS1 transgene. sal training. Comparisons of daily performance were Serum Aβ and Aβ levels were determined in 40 42 analyzed separately and differences were considered sig - F1 generation pups at 30  days of age and in F2 gen- nificant at p < 0.05 for all analyses. eration pups at 50  days of age (Fig.  3; Additional file  2: Table 2). The Aβ levels increased with age in both APP Results and APP  +  PS1 lines and both F1 and F2 generation APP21‑PS1 double transgenic founders APP +  PS1 rats had significantly greater (p  <  0.05) Aβ Forty-five rats were born from the lentiviral vector which levels compared to APP rats. Serum Aβ /Aβ levels 42 40 contained PS1 transgene-injected embryos and 32 of were approximately two times greater in the APP +  PS1 them were PCR positive for the PS1 transgene. Southern rats compared to the APP rats for both F1 and F2 gen- blot results showed that transgene copy number ranged erations (p  <  0.001), but serum Aβ levels were similar between 1 and 5 for each PS1 transgenic founder rat between PS1  +  APP and APP rats and did not change (Fig.  1a). Serum Aβ ranged between 25 and 479  pg/ml with age. in the founder animals at 4  weeks of age and transgene copy number positively correlated with serum AB42 lev- Behavior changes among AP + PS1, PS and fischer wild els in founder animals (r  =  .54). Serum Aβ and Aβ type rats 40 42 levels at 8 weeks of age were also determined in founder Barnes maze animals (Fig. 1b). Due to the segregation of multiple inte- There were no statistical differences in performance gration sites in subsequent generations of transgenic between male and female rats, thus the data were col- animals, founder animals with one copy transgene were lapsed over sex for all of the behavioral analyses. used for breeding. Founders with one PS1 transgene copy and relatively high serum Aβ levels were considered in Acquisition Overall, APP  +  PS1 rats made signifi - order to determine the ultimate founder animal. Founder cantly more errors compared to WT rats (p = 0.04) dur- 964 had one of the highest serum Aβ (196  pg/ml) lev- ing acquisition training, but there were no group differ - els as well as Aβ /Aβ levels (0.28) at 8  weeks of age. ences in latency (p > 0.05; Fig. 4a). More specifically, there 42 40 Furthermore, the F1 generation from founder 964 had were no statistical differences in performance among the the highest serum Aβ levels and had a relatively large- groups on acquisition days 1 and 2 (ps  >  0.05), however, sized litter. Thus, founder 964 was used as the transgenic on day 3, APP + PS1 rats made significantly more errors founder for generation of APP + PS1 transgenic rats. compared to both APP and WT rats (p < 0.001) and had Northern blot analysis of various tissues from the significantly longer latencies compared to WT rats only founder APP  +  PS1 transgenic animals showed that (p = 0.02; Fig. 4d). Overall, these results suggest that the brain, heart, kidney, liver, and lung had 38, 8, 19, 15, and APP + PS1 rats had a larger learning deficit in the Barnes 15 fold more PS1 mRNA (P  <  0.01), respectively, com- maze task than the APP rats. pared to the APP21 rat line. Serum Aβ and Aβ lev- 40 42 els at 8 weeks were positively correlated with PS1 mRNA Retention After the retention interval, APP  +  PS1 rats levels in APP  +  PS1 founders (Table  1). Statistical anal- made numerically more errors (p = 0.1) and had signifi - ysis of the PS1 expression of founder animals was done cantly longer latencies (p = 0.01) than both the APP and by grouping the animals based on their 8 wk serum Aβ WT rats (Fig. 4b), implying a larger memory deficit in the levels. The groups comprised of rats with Aβ42 serum APP  +  PS1 rats. These differences in memory retention levels of less than 100  pg/ml (<100), between 100 and may be due to the larger learning deficit in APP  + PS1 ani- 149 pg/ml (100–149), between 150 and 200 pg/ml (150– mals during acquisition training. 200), and greater than 200 pg/ml (>200), as well as APP21 rats that did not have the PS1 transgene. PS1 expression Reversal Overall, APP  +  PS1 rats made significantly in tissues increased significantly (p  < 0.05) as serum Aβ more errors (p = 0.04) and had significantly longer laten - levels increased (Fig. 2; Additional file  1: Table 1) with the cies (p  =  0.03) compared to both the WT and APP rats exception of the liver, in which PS1 levels were not sig- during reversal training (Fig. 4c). More specifically, there nificantly different. were no significant differences on days 1 and 2 of reversal Agca et al. BMC Neurosci (2016) 17:46 Page 5 of 10 Fig. 1 a Southern blot analysis of the genomic DNA of the founder animals used to determine the copy number of PS1 transgene. The copy num- ber of the PS1 transgene was determined by the number of hybridized genomic DNA segments in the autoradiogram and is shown in parenthesis above the rat numbers. M represents the molecular weight (MW ) markers. b Serum Aβ /Aβ levels of APP + PS1 double transgenic founder 42 40 animals and APP21 rats at 8 weeks of age. The Aβ /Aβ ratio for rat 988 could not be determined because its Aβ level was lower than the detec- 42 40 42 tion limit of the test. Founder 964 was selected based on having one PS1 transgene and high Aβ /Aβ levels as well as for having a large litter with 42 40 high Aβ /Aβ levels 42 40 Table 1 Correlation coefficients of  serum Aβ and  Aβ Histopathology of brain 42 40 levels to PS1 mRNA levels of the organs The hippocampal pyramidal cell layer of Cornu Ammonis (CA) 1 and CA2 regions of the APP transgenic rats con- PS1 mRNA Serum Aβ Serum Aβ 42 40 tained many necrotic neurons and neurofibrillary tan - Brain 0.49 0.36 gles (Fig.  5a, d, and g). Mineralization was also evident Kidney 0.48 0.30 as indicated by the rough cell membrane observed in Heart 0.60 0.34 some of the neurons in the hippocampus and the cortex Liver 0.45 0.22 in both the APP and the APP +  PS1 rats. In addition to Lung 0.54 0.31 neurons containing flame-shaped neurofibrillary tangles, hippocampus also contained remnants of neurofibrillary tangles that were embedded in the neuropil (ghost tan- gles) in the APP rats (Fig.  5d, g). Neurofibrillary tangles training (ps > 0.05), but on day 3, APP + PS1 rats made were present in viable cells that had nuclei as well as neu- significantly more errors (p  <  0.01) and had significantly rons with no visible nuclei (Fig.  5). The CA3 region also longer latencies (p  =  0.04) than both the APP and WT contained necrotic neurons and neurofibrillary tangles in rats (Fig.  4e). These results suggest deficits in reversal both the APP and APP + PS1 rats (Fig. 5j, k). learning in the APP + PS1 rats. Additional file  3: Table 3 Both the APP and APP  +  PS1 rats had fewer pyrami- contains the dataset of the Barnes Maze. dal neurons in the CA3 region compared to WT, but Agca et al. BMC Neurosci (2016) 17:46 Page 6 of 10 central and segmental chromatolysis, showing eccentric displacement of the nucleus and loss of Nissl substance with the exception of the margins of the cell body (Fig. 5h, k). Both the APP and the APP  +  PS1 brain cortex con- tained neurons with various degrees of degeneration that ranged from neurons with condensed Nissl substance to dark necrotic neurons and ghost tangles (Fig. 5p, q). The brain cortex of the APP  +  PS1 animals had larger areas with dark stained neurons, which indicated more severe neuronal necrosis (Fig. 5q). Discussion The production of transgenic rats using lentiviral vec - Fig. 2 PS1 expression in founder animals and APP controls as deter- tors is an efficient method for gene transfer. However, the mined by Northern blot analysis. The expression of PS1 was analyzed major drawback of lentiviral vectors is the integration of by combining animals with similar serum Aβ levels and statistical analysis conducted for each tissue. The figure legend indicates the the transgene into multiple sites of the genome. Multiple range of serum Aβ level (pg/ml) of the founder animals or the APP copies of the transgene result in high levels of expression, control rats. Different letters above each bar indicates significantly dif- but segregation in subsequent generations drastically ferent (p < 0.05) expression decreases transgene expression. In this study, the founder animal was selected among the animals that had only one copy of the transgene and those that also had the greatest the APP  +  PS1 animals had more severe neuronal loss serum Aβ levels and Aβ /Aβ ratio. 42 42 40 in the CA3 region (Fig.  5b). The hippocampal pyramidal Serum Aβ and Aβ levels were positively correlated 42 40 cell layer in the APP + PS1 rats had many neurons with with PS1 mRNA expression, indicating that the addition Fig. 3 Serum Aβ (A, D), Aβ (B, E) and Aβ /Aβ (C, F) levels of 30 day old F1 and 50 day old F2 generations. Different letters above each bar 42 40 42 40 indicate significantly different levels of Aβ and Aβ /Aβ 42 42 40 Agca et al. BMC Neurosci (2016) 17:46 Page 7 of 10 Fig. 4 Overall performance of Barnes maze Acquisition Training (A) Retention (B), and Reversal Training (C). Daily performance of Acquisition Train- ing (D), and Reversal Training (E). Data is expressed as group mean ± SEM. Different letters above each bar indicates significantly different (p < 0.05) expression of the PS1 transgene contributed to the production of Histologic examination of the brains showed that both Aβ and Aβ . Furthermore, the greater correlation both the APP and the APP  +  PS1 animals exhibited 42 40 efficiencies for Aβ and PS1 mRNA expression indicate pathological abnormalities typically associated with AD. preferential cleavage of Aβ by the PS1 enzyme. Trans- Interestingly, the APP rats had large hippocampal sec- fection using PS1 resulted in an approximately 1.6 fold tions containing neurons with neurofibrillary tangles. increase in Aβ /Aβ levels [16] in cell cultures. Simi- The hippocampus of the APP  +  PS1 rats also contained 42 40 larly, the APP  +  PS1 transgenic animals had approxi- neurofibrillary tangles, but these were more sporadic mately a twofold Aβ /Aβ increase compared to the compared to what was observed in the APP rats. Interest- 42 40 APP rats. The level of PS1 mRNA expression varied by ingly, neuronal loss was much greater in the CA3 region tissue with the highest PS1 expression in the brain and of the APP  +  PS1 rats. Another difference between the kidney, while the heart and liver had lower expression. hippocampus in APP and the APP  +  PS1 animals was Tissue specific expression was also observed in the APP that neurons in the CA1 and the CA2 regions of the transgenic rats [1]. APP + PS1 animals had chromatolysis. Chromatolysis is Agca et al. BMC Neurosci (2016) 17:46 Page 8 of 10 Agca et al. BMC Neurosci (2016) 17:46 Page 9 of 10 Fig. 5 Hematoxylin and Eosin staining of APP, APP + PS1, and WT rat brains. 4× magnification of a hippocampus section from an APP (a), APP + PS1 (b), and WT rat (c). CA1, CA2, CA3 regions of the hippocampus and dentate gyrus (DG) are labeled in c. Panels d–r show 60X magnifica- tion of the brain and each arrow style represents similar pathology throughout the figure. d APP CA1 region with necrotic and dark colored pyrami- dal neurons (black arrowhead), ghost tangles (white arrow), and normal looking neurons (white arrow head). e APP + PS1 CA1 showing degeneration with condensed Nissl substance (grey arrow) and chromatolysis (grey arrow head). f WT CA1. g APP CA2 with neurofibrillary tangles in viable neurons with visible nuclei (black arrow) and non-viable neurons (black arrow with white outline) and ghost tangles. h APP + PS1 CA2 showing chromatolysis in neurons. i WT CA2. j APP CA3 with neurofibrillary tangles and condensed Nissl substance. k APP + PS1 CA3 with neurofibrillary tangles. l WT CA3. m APP DG granule cells with condensed Nissl substance and necrotic cells. n APP + PS1 DG with necrotic cells and neurofibrillary tangles. o WT DG. p, q APP and APP + PS1 cortex neurons with condensed Nissl substance, neurofibrillary tangles, necrotic neurons, and ghost tangles. r WT cortex caused by axonal injury in motor neurons [18] and it has between the APP  +  PS1, APP, and WT rats in learning been associated with neurotoxicity and AD [5, 15, 22], and retention in the task may be related to the differences but it is not reported as widely as the presence of neurofi - in the pathology of the hippocampus. brillary tangles or Aβ plaques. A previously reported rat model for AD, the TgF344- Conclusions AD model [6], was found to produce plaques at The addition of the PS1 transgene to the APP transgenic 15 months. The difference in plaque production between rat genome affected serum Aβ and Aβ /Aβ levels, 42 42 40 the TgF344-AD and the APP + PS1 rats could be due to brain histopathology, as well as, the cognitive behavior the fact that the PS1 transgene in the TgF344-AD had Δ of the double transgenic rats. Although the APP rat brain exon 9 mutation, as opposed to the L166P mutation form had many histopathogical abnormalities that are often of the PS1 transgene that was used in these APP  +  PS1 seen in AD brains, their memory did not show significant rats. On the other hand, the behavioral characteriza- impairment during the behavioral testing. In contrast, tion of the APP + PS1 rats in this current study revealed the APP  +  PS1 rats showed significantly reduced mem - learning and memory deficits similar to what has been ory retention compared to the APP and the WT rats, in found in TgF344-AD rats. In addition, previous research addition to, neuronal loss and necrosis in the hippocam- has found that an AD mouse model that did not develop pus and brain cortex. plaques even as late as 30 months old also showed mem- Additional files ory deficits [9]. Consequently, the presence of Aβ plaques may not always correlate with AD severity. For exam- Additional file 1: Table 1. contains dataset for Northern blot analysis ple, Shankar et  al. [21] showed that Aβ dimers are syn- (see Fig. 2). PS1 mRNA level, serum Aβ42 and tissue are reported in the aptotoxic and disrupted memory, whereas insoluble Aβ table. plaque cores from Alzheimer’s diseased cortex did not Additional file 2: Table 2. contains dataset for serum Aβ , Aβ and 42 40 impair long-term potentiation. Furthermore, the number Aβ /Aβ ratio of F1 and F2 generation transgenic rats (see Fig. 3). 42 40 of Aβ plaques in the brain has not been found to corre- Additional file 3: Table 3. contains dataset for Barnes maze (see Fig. 4). The table includes acquisition day 1, 2 and 3 latency and errors and their late with the severity of cognitive impairments in humans averages, retention latency and errors and reversal days 1, 2 and 3 latency or in APP and/or PS transgenic mice [3]. and errors and their averages. Wild type rats outperformed both the APP and the APP  +  PS1 rats during the initial learning phase of the Barnes maze test. Additionally, retention of the learned Abbreviations AD: alzheimer’s disease; Aβ: amyloid-beta; APP: amyloid precursor protein; CA: task was poorer in the APP + PS1 rats relative to both the cornu Ammonis; eGFP: enhanced green fluorescence protein; IP: intraperito - APP and the WT rats. As stated earlier, one of the major neal; pLV: lentivial vector; PMSG: pregnant mares’ serum gonadotropins; PS1: differences between the APP  + PS1 and the APP rats was presenilin 1; SSC: saline sodium citrate; SDS: sodium dodecyl sulfate; Ubi-C: ubiquitin-C; WT: wild type; WRE: woodchuck hepatitis virus post-transcrip- that the APP  +  PS1 rats had neuronal loss in the CA3 tional regulatory element. region of hippocampus (Fig.  5b shows an APP  +  PS1 rat with very few neurons in the CA3 region). The CA1 Authors’ contributions CA was responsible for genotypic and phenotypic characterization of trans- region recodes information from the CA3 region and sets genic rats, statistical analysis of data, colony management, and writing the up associatively learned backprojections to the neocortex paper. JJL generated lentiviral vectors and performed ELISA. DK and TRS have to allow subsequent retrieval of information to the neo- conducted rat behavioral assessment and analysis. YA and AWC generated transgenic rats. YA designed the research, and contributed to the analysis of cortex [13]. The hippocampal CA3 region and the den - the data. All authors read and approved the final manuscript. tate gyrus also play an important role in the encoding of new spatial information and processing the geometry Author details Department of Veterinary Pathobiology, College of Veterinary Medicine, Uni- of the environment [12], which are both crucial to the versity of Missouri, 1600 East Rollins Street, Room W191, Columbia, MO 65211, successful completion of the task. Thus, the differences 2 USA. Department of Psychological Sciences, University of Missouri, Columbia, Agca et al. BMC Neurosci (2016) 17:46 Page 10 of 10 MO 65211, USA. Yerkes National Primate Research Center, Emory University, 9. Gandy S, Simon AJ, Steele JW, Lublin AL, Lah JJ, Walker LC, Levey AI, Krafft Atlanta, GA 30329, USA. Department of Neurology, Center for Neurodegen- GA, Levy E, Checler F, Glabe C, Bilker WB, Abel T, Schmeidler J, Ehrlich erative Disease, Emory University, Atlanta, GA 30322, USA. ME. Days to criterion as an indicator of toxicity associated with human Alzheimer amyloid-beta oligomers. Ann Neurol. 2010;68(2):220–30. Acknowledgements 10. Jarrett JT, Berger EP, Lansbury PT Jr. The C-terminus of the beta protein is This study was funded by a grant from Alzheimer Research Consortium and critical in amyloidogenesis. Ann N Y Acad Sci. 1993;695:144–8. the University of Missouri startup funds. The authors would also like to thank 11. Kandimalla RJ, Wani WY, Binukumar BK, Gill KD. siRNA against presenilin 1 Dr. Marcia Hart for helping assess brain histopathology. (PS1) down regulates amyloid β42 production in IMR-32 cells. J Biomed Sci. 2012;19:2. Competing interests 12. Kesner RP. Behavioral functions of the CA3 subregion of the hippocam- The authors declare that they have no competing interests. pus. Learn Mem. 2007;14(11):771–81. 13. Kesner RP, Rolls ET. A computational theory of hippocampal function, Availability of data and materials and tests of the theory: new developments. Neurosci Biobehav Rev. Datasets are submitted as supplementary material. 2015;48:92–147. 14. Kim J, Onstead L, Randle S, Price R, Smithson L, Zwizinski C, Dickson Ethics approval and consent to participate DW, Golde T, McGowan E. Abeta40 inhibits amyloid deposition in vivo. J All animal studies were approved by University of Missouri’s Animal Care and Neurosci. 2007;27(3):627–33. Use Committee (Protocol Number 7995). 15. Kotariya NT, Bikashvili TZ, Zhvaniya MG. Chkhikvishvili TsG. Ultrastruc- ture of hippocampal field CA1 in rats after status epilepticus induced Received: 21 April 2016 Accepted: 27 June 2016 by systemic administration of kainic acid. Neurosci Behav Physiol. 2010;40(2):127–30. 16. Li N, Liu K, Qiu Y, Ren Z, Dai R, Deng Y, Qing H. Eec ff t of Presenilin Muta- tions on APP Cleavage; insights into the pathogenesis of FAD. Front Aging Neurosci. 2016;8:51. 17. Lois C, Hong EJ, Pease S, Brown EJ, Baltimore D. Germline transmission References and tissue-specific expression of transgenes delivered by lentiviral vec- 1. Agca C, Fritz JJ, Walker LC, Levey AI, Chan AW, Lah JJ, Agca Y. Develop- tors. Science. 2002;295:868–72. ment of transgenic rats producing human beta-amyloid precursor 18. McIlwain DL, Hoke VB. The role of the cytoskeleton in cell body enlarge- protein as a model for Alzheimer’s disease: transgene and endogenous ment, increased nuclear eccentricity and chromatolysis in axotomized APP genes are regulated tissue-specifically. BMC Neurosci. 2008;9:28. spinal motor neurons. BMC Neurosci. 2005;6:19. 2. Alzheimer’s Association. 2015 Alzheimer’s disease facts and figures. 19. Masters CL, Selkoe DJ. Biochemistry of amyloid beta-protein and Alzheimers Dement 2015; 11(3):332–84. amyloid deposits in Alzheimer disease. Cold Spring Harb Perspect Med. 3. Benilova I, Karran E, De Strooper B. The toxic Aβ oligomer and Alzheimer’s 2012;2(6):a006262. disease: an emperor in need of clothes. Nat Neurosci. 2012;15(3):349–57. 20. Rosen RF, Fritz JJ, Dooyema J, Cintron AF, Hamaguchi T, Lah JJ, LeVine 4. Burdick D, Soreghan B, Kwon M, Kosmoski J, Knauer M, Henschen A, H 3rd, Jucker M, Walker LC. Exogenous seeding of cerebral β-amyloid Yates J, Cotman C, Glabe C. Assembly and aggregation properties of deposition in βAPP-transgenic rats. J Neurochem. 2012;120(5):660–6. synthetic Alzheimer’s A4/beta amyloid peptide analogs. J Biol Chem. 21. Shankar GM, Li S, Mehta TH, Garcia-Munoz A, Shepardson NE, Smith 1992;267(1):546–54. I, Brett FM, Farrell MA, Rowan MJ, Lemere CA, Regan CM, Walsh DM, 5. Cataldo AM, Hamilton DJ, Nixon RA. Lysosomal abnormalities in degener- Sabatini BL, Selkoe DJ. Amyloid-beta protein dimers isolated directly ating neurons link neuronal compromise to senile plaque development from Alzheimer’s brains impair synaptic plasticity and memory. Nat Med. in Alzheimer disease. Brain Res. 1994;640(1–2):68–80. 2008;14(8):837–42. 6. Cohen RM, Rezai-Zadeh K, Weitz TM, Rentsendorj A, Gate D, Spivak I, Bho- 22. Sil S, Goswami AR, Dutta G, Ghosh T. Eec ff ts of naproxen on immune lat Y, Vasilevko V, Glabe CG, Breunig JJ, Rakic P, Davtyan H, Agadjanyan MG, responses in a colchicine-induced rat model of Alzheimer’s disease. Kepe V, Barrio JR, Bannykh S, Szekely CA, Pechnick RN, Town T. A trans- NeuroImmunoModulation. 2014;21(6):304–21. genic Alzheimer rat with plaques, tau pathology, behavioral impairment, 23. Walsh DM, Selkoe DJ. A beta oligomers—a decade of discovery. J Neuro- oligomeric aβ, and frank neuronal loss. J Neurosci. 2013;33(15):6245–56. chem. 2007;101(5):1172–84. 7. De Strooper B. Aph-1, Pen-2, and Nicastrin with Presenilin generate an active gamma-Secretase complex. Neuron. 2003;38(1):9–12. 8. Deng Y, Tarassishin L, Kallhoff V, Peethumnongsin E, Wu L, Li YM, Zheng H. J Neurosci. 2006;26:3845–54. Submit your next manuscript to BioMed Central and we will help you at every step: • We accept pre-submission inquiries • Our selector tool helps you to find the most relevant journal • We provide round the clock customer support • Convenient online submission • Thorough peer review • Inclusion in PubMed and all major indexing services • Maximum visibility for your research Submit your manuscript at www.biomedcentral.com/submit http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png BMC Neuroscience Springer Journals

Presenilin 1 transgene addition to amyloid precursor protein overexpressing transgenic rats increases amyloid beta 42 levels and results in loss of memory retention

Download PDF
10 pages

Loading next page...
 
/lp/springer-journals/presenilin-1-transgene-addition-to-amyloid-precursor-protein-yZdq9GWoJJ

References (21)

Publisher
Springer Journals
Copyright
Copyright © 2016 by The Author(s)
Subject
Biomedicine; Neurosciences; Neurobiology; Animal Models
eISSN
1471-2202
DOI
10.1186/s12868-016-0281-8
pmid
27388605
Publisher site
See Article on Publisher Site

Abstract

Background: We previously reported the production of transgenic rats (APP21 line) that over-express human amyloid precursor protein (APP) containing Swedish and Indiana mutations. In order to generate a better model for Alzheimer’s disease (AD), the APP21 rat line was used to generate double transgenic line that over-expressed Preseni- lin 1 (PS1) with L166P mutation in addition to APP transgene (APP + PS1 line). Results: Thirty-two double transgenic founders were generated and the ultimate transgenic founder was selected based on PS1 transgene copy number and level of amyloid-beta (Aβ) peptide. The APP + PS1 double transgenic rats had 38 times more PS1 in brains compared to APP rats. Behavioral assessment using Barnes maze showed that APP + PS1 rats exhibited a larger learning and memory deficit than APP21 rats. Double transgenic rats also produced more Aβ . Histological examination of the brains showed that the APP21 rat line displayed neurofibrillary tangles and in contrast, the APP + PS1 line showed chromatolysis in hippocampal neurons and neuronal loss in CA3 region of hippocampus. Conclusions: Due to the separate segregation of APP and PS1 transgenes in APP + PS1 double transgenic rats, this transgenic line may be a valuable model for studying the effects of various levels of APP and PS1 transgenes on vari- ous aspects of brain pathologies associated with the AD phenotype. Background Amyloid peptides are produced via the cleavage of APP Alzheimer’s disease (AD) is one of the most devastat- by ϒ-secretase, a protease complex of four proteins that ing and costly diseases that affect approximately 12 % of includes PS1 or PS2, which are both homologous pro- the population over the age of 65 [2]. Alzheimer’s dis- teins that contain the catalytic site of the enzyme [7]. ease is associated with the over-production and reduced ϒ-secretase cleaves the β-amyloid protein into two major clearance of amyloid-beta (Aβ) peptides [19]. Cleavage forms of Aβ polypeptides, Aβ and Aβ . The relative 40 42 of amyloid-β precursor protein (APP) by the preseni- amount of the longer form of Aβ, Aβ , is particularly lin (PS) 1 enzyme results in the release of Aβ peptides, critical for AD progression because it is more prone to which aggregate and form Aβ plaques in the brain. These aggregation than the shorter Aβ peptide [4, 10, 23]. Aβ plaques and neurofibrillary tangles are the two pri - In addition, the ratio of Aβ /Aβ correlates with the 42 40 mary pathological manifestations of AD in the brain. load of Aβ plaques in the brain [8, 14]. Furthermore, the reduction of PS1 mRNA using siRNA against PS1 reduced Aβ in cultured cells [11] which suggests that *Correspondence: agcay@missouri.edu 1 lower levels of PS1 lead to decreased production of Aβ . Department of Veterinary Pathobiology, College of Veterinary Medicine, University of Missouri, 1600 East Rollins Street, Room W191, Columbia, We previously generated rats that overexpress human MO 65211, USA APP containing the Swedish and Indiana mutations [1]. Full list of author information is available at the end of the article © 2016 The Author(s). This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/ publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Agca et al. BMC Neurosci (2016) 17:46 Page 2 of 10 The APP21 line expressed high levels of APP, but failed gonadotropins (PMSG) followed by IP injections of to produce Aβ plaques in the brain. However, intrahip- 15  IU human chorionic gonadotropin 52  h after PMSG. pocampal seeding of dilute Alzheimer brain extracts con- Twenty-four hours after mating, rats were euthanized taining aggregated Aβ led to the production of plaques using CO and zygotes were collected from the oviducts. in APP21 rats [20]. This finding indicates that plaque Morphologically normal zygotes having two pronuclei formation depends on environmental challenges, condi- and a sperm tail were used for lentiviral vector injections tions which laboratory rats are not exposed to routinely. into perivitelline space as described earlier [1]. Lentiviral In order to increase the pathogeny of overexpressed APP, vector-injected zygotes were then transferred into the we generated double transgenic rats from the APP21 line oviducts of 8–10 week-old Sprague–Dawley pseudopreg- that over-express both the PS1 and the APP transgenes. nant recipient rats. Our hypothesis was that the addition of a PS1 transgene would increase more toxic Aβ polypeptide and thus, Genomic DNA isolation and polymerase chain reaction would produce a more suitable model for examining AD. Genomic DNA from tail-snips was isolated using the Wiz- ard Genomic DNA Purification Kit (Promega, Madison, Methods WI). Screening of transgenic rats was conducted by PCR Animals, construct design, and preparation of vesicular using primers annealing the Ubi-C promoter (Forward; stomatitis virus glycoprotein pseudotyped lentivirus TGTCCGCTAAATTCTGGCCGTT) and PS1 transgene The homozygous APP21-transgenic rat line [1 ], con- (Reverse; AGCATTCAGAATTGAGTGCAGGGC). The taining human APP with the Swedish double missense PCR reactions were carried out using 50  ng genomic mutation (K595M/N596L) and Indiana single missense DNA, 100  ng of each primer and 1.5 U Biolase taq (Bio- mutation (V642F), were used for the generation of APP- line, Randolph, MA), and amplified PCR products were PS1 double transgenic rats. Inbred Fischer 344 rats were size separated through a 1 % agarose gel stained with eth- used as the background strain for the APP21 rat line. The idium bromide for visualization. human PS1 cDNA coding region corresponding to 285- 1688 bases of PS1 variant 1 (GenBank Accession num- ELISA measurement of Aβ and Aβ 40 42 ber NM_000021.3)  with L166P mutation was cloned Serum samples were analyzed for Aβ and Aβ using 40 42 into the Lentivial vector (pLV) containing ubiquitin-C commercial ELISA kits (Genetics Company, Schlieren, (Ubi-C) and enhanced green fluorescence protein (eGFP; Switzerland). Serum samples were diluted in assay buffer pLVU-eGFP) cassette [17] in place of the eGFP coding and processed according to the manufacturer’s recom- sequence. The new vector was designated as pLVU-PS1. mended protocols. Briefly, samples and standards were The Ubi-C promoter, which is upstream of PS1, controlled incubated in capture wells overnight at 4  °C with either PS1 transcription. The pLVU-PS1 was a self-inactivating a biotinylated Aβ or an Aβ -specific antibody. After 40 42 vector, composed of the woodchuck hepatitis virus post- several rinses, the enzyme-conjugated detection rea- transcriptional regulatory element (WRE) to increase gent was added to the wells for 30  min. After additional transcription levels and minimize position effects. Addi - rinses, wells were incubated with the chromogen solution tionally, a human immunodeficiency virus-1-flap element for 30 min at room temperature and were shielded from was inserted between the 5′ long terminal repeat and the light. After the addition of the stop solution, the wells internal promoter, which increased the titer. The prepara - were read for absorption at 450  nm and the Aβ concen- tion of vesicular stomatitis virus glycoprotein pseudotyped tration in the samples was calculated based on standard Lentivirus was as described by Lois et  al. [17]. All animal curves. studies were performed in accordance with the University of Missouri’s Animal Care and Use Committee guidelines Southern blot analysis and the Institute for Laboratory Animal Research Guide Southern blot analysis was conducted to determine the for the Care and Use of Laboratory Animals. The rats were copy number of the transgene in founder rats. Genomic housed in conventional cages at 20–25  °C in a controlled DNA was digested with EcoRI, which cut the junction of lighting environment and provided free access to water the PS1 and the WRE transgene. The digestion products and standard pellet rodent chow. were size-separated through a 0.8 % agarose gel and trans- ferred to a Genescreen plus membrane (Perkin Elmer, Zygote collection, microinjection of lentiviral vector, Wellesley, MA) overnight. The 524  bp PS1 probe tem - and embryo transfer plate was prepared by amplification of pLVU-PS1 using Homozygous APP21-transgenic female rats (28–30  day- forward (AGGTCCACTTCGTATGCTGGTTGA) and old) were super ovulated by administering intraperi- reverse (TGATGGAGATTGGAAGAGTGGCA) prim- toneal (IP) injections of 15  IU pregnant mares’ serum ers. The P labeled probe was generated using the probe Agca et al. BMC Neurosci (2016) 17:46 Page 3 of 10 template, Ready-To-Go DNA Labeling Beads (Amersham lumins were hung above the platform served to brightly Biosciences, Piscataway, NJ) and [α- P]-dCTP (Perkin light the maze in order to create a potentially aversive Elmer, Wellesley, MA). The membranes were prehybrid - environment to help motivate the rats to escape from the ized in 10  % dextran sulfate, 6  ×  saline sodium sitrate brightly lit, open surface in favor of the dark environment (SSC), 1  % sodium dodecyl sulfate (SDS) for 2  h and of the escape box. One light hung 68.5 on side of the hybridized using the P labeled probe overnight before maze while the other hung 137 cm from the other side. exposure to BioMax MS-1 Autoradiography Film (Kodak, Each rat was assigned an escape hole number under Rochester, NY). which the escape box was placed on each test trial; assigned hole numbers were alternated across rats to Northern blot analysis eliminate odor cues for consecutively tested rats. The Total RNA was isolated using Trisure (Bioline, CA) from escape box location remained constant for any individual founder APP21 + PS1 and APP21 transgenic rats. Brain, rat across test trials. Behavioral testing consisted of two heart, kidney, liver, and lung tissues were used in North- pretraining trials on the first day, and 6 evaluation trials ern blot analysis. Total RNA was size-separated through (2 trials/day) over a period of 3  days. Each day, the ani- 1  % agarose gel before transferring to the Genescreen mals were transferred in their home cage from their col- plus membrane overnight. The membrane was prehy - ony room to the testing room 30 min prior to the start of bridized in 10  % dextran sulfate, 5 ×  SSPE, 50  % forma- testing. On the pretraining trial, the rat was preexposed mide, 5  ×  Denhardt’s, 1  % SDS at 42  °C for 6  h prior to to the goal box (with the goal box placed in the same hybridization with the PS1 probe overnight. After expo- hole that would be used during subsequent training) by sure to BioMax MS-1 Autoradiography Film, membranes placing the rat in the goal box for 90  s. The rats’ escape were stripped to remove labeled cDNA probe by boil- from the goal box was prevented by covering the open- ing for 1 h in 1 %SDS and 0.1× SSC. The intensity of the ing to the maze surface with a grey, opaque start box (a bands was determined using Kodak 1D v 3.6.3 software 23  cm  ×  23  cm box). A training trial began by placing (New Haven, CT). the rat under the start box positioned in the center of the platform. After 30 s, the box was lifted and the rat had a Behavioral measures maximum of 5 min (300 s) to locate and enter the escape Behavioral assessment was conducted using APP  +  PS1 box. Latency (time it took for the rat to find and enter the (n =  12), APP (n =  11), and non-transgenic Fischer 344 escape box) and total errors (nose-pokes into non-escape (n  =  12) rats when their average age was approximately holes) were recorded. If the rat did not enter the escape 10  months. Prior to behavior studies, each rat was han- box within 5 min, it was gently guided there by the exper- dled at least 3 times to reduce stress due to handling dur- imenter’s hand. After 30  s, the rat was removed from ing the behavioral testing. the escape box and returned to its home cage. Rats were allowed to rest in their home cage in the testing room for Barnes maze 30  min before starting their second daily trial. The plat - The maze consisted of a grey circular platform 122 cm in form and escape box were cleaned after every trial with a diameter. The platform was surrounded by a wall that was 20 % ethanol solution. After the third day, testing abated 30.5 cm in height. The maze was elevated 83.8 cm above for 3  days, after which retention was evaluated for two the floor by a stand. Twenty holes measuring 10.2 cm in additional trials (1 day), conducted exactly like those just diameter were evenly spaced around the perimeter. A described. The following day, the rats were given reversal rectangular grey escape box (28 cm in length ×  12.7  cm training in which the rats were given training exactly like wide × 7.6 cm high at the area closest to the maze taper- those of original acquisition, except that the escape box ing to 16.5  cm high) could be placed beneath any hole. was located in a new location (in the opposite quadrant). The escape box included an entry ramp that provided Reversal training continued for a total of 3 days. easy entry access for the rat. Black curtains were hung around the maze and above the maze walls to surround Histopathology the apparatus and ensure that rats could only use the Nine of the female rats [3 of each: APP + PS1, APP, and visual cues provided in the maze, rather than the distal wild type (WT)] were euthanized at 18–19 months of age cues within the testing room. Proximal cues were more and their brains were collected. Tissues were fixed in 10 % likely to remain constant, within subjects and across sub- formalin for 48 h prior to paraffin embedding. Hematox - jects, during the course of training. Four visual cues con- ylin and eosin staining was used to determine histological sisting of various shapes (triangle, square, circle, cross) changes. The stained slides were analyzed using a Nikon were placed at evenly spaced intervals on the inside of the Eclipse E600 microscope (Melville, NY) and an Olympus maze walls. Two 86-W, 120-V floodlights producing 1690 DP72 Camera (Center Valley, PA). Agca et al. BMC Neurosci (2016) 17:46 Page 4 of 10 Statistical analysis Subsequent generations of APP + PS1 double transgenic Statistical analysis for PS1 mRNA expression, serum Aβ rat line level comparisons, and behavioral analyses were per- The APP21 line has been previously bred as a homozy - formed using general linear models in SAS (SAS 9.3, SAS gous transgenic line. However, homozygous APP  +  PS1 Institute Inc., Cary, NC, USA). For the Barnes maze, the rats failed to produce pups. Thus, the APP  +  PS1 dou- average of each animal’s daily score was used in the anal- ble transgenic line was retained as homozygous for APP ysis of overall performance during acquisition and rever- transgene and hemizygous for PS1 transgene. sal training. Comparisons of daily performance were Serum Aβ and Aβ levels were determined in 40 42 analyzed separately and differences were considered sig - F1 generation pups at 30  days of age and in F2 gen- nificant at p < 0.05 for all analyses. eration pups at 50  days of age (Fig.  3; Additional file  2: Table 2). The Aβ levels increased with age in both APP Results and APP  +  PS1 lines and both F1 and F2 generation APP21‑PS1 double transgenic founders APP +  PS1 rats had significantly greater (p  <  0.05) Aβ Forty-five rats were born from the lentiviral vector which levels compared to APP rats. Serum Aβ /Aβ levels 42 40 contained PS1 transgene-injected embryos and 32 of were approximately two times greater in the APP +  PS1 them were PCR positive for the PS1 transgene. Southern rats compared to the APP rats for both F1 and F2 gen- blot results showed that transgene copy number ranged erations (p  <  0.001), but serum Aβ levels were similar between 1 and 5 for each PS1 transgenic founder rat between PS1  +  APP and APP rats and did not change (Fig.  1a). Serum Aβ ranged between 25 and 479  pg/ml with age. in the founder animals at 4  weeks of age and transgene copy number positively correlated with serum AB42 lev- Behavior changes among AP + PS1, PS and fischer wild els in founder animals (r  =  .54). Serum Aβ and Aβ type rats 40 42 levels at 8 weeks of age were also determined in founder Barnes maze animals (Fig. 1b). Due to the segregation of multiple inte- There were no statistical differences in performance gration sites in subsequent generations of transgenic between male and female rats, thus the data were col- animals, founder animals with one copy transgene were lapsed over sex for all of the behavioral analyses. used for breeding. Founders with one PS1 transgene copy and relatively high serum Aβ levels were considered in Acquisition Overall, APP  +  PS1 rats made signifi - order to determine the ultimate founder animal. Founder cantly more errors compared to WT rats (p = 0.04) dur- 964 had one of the highest serum Aβ (196  pg/ml) lev- ing acquisition training, but there were no group differ - els as well as Aβ /Aβ levels (0.28) at 8  weeks of age. ences in latency (p > 0.05; Fig. 4a). More specifically, there 42 40 Furthermore, the F1 generation from founder 964 had were no statistical differences in performance among the the highest serum Aβ levels and had a relatively large- groups on acquisition days 1 and 2 (ps  >  0.05), however, sized litter. Thus, founder 964 was used as the transgenic on day 3, APP + PS1 rats made significantly more errors founder for generation of APP + PS1 transgenic rats. compared to both APP and WT rats (p < 0.001) and had Northern blot analysis of various tissues from the significantly longer latencies compared to WT rats only founder APP  +  PS1 transgenic animals showed that (p = 0.02; Fig. 4d). Overall, these results suggest that the brain, heart, kidney, liver, and lung had 38, 8, 19, 15, and APP + PS1 rats had a larger learning deficit in the Barnes 15 fold more PS1 mRNA (P  <  0.01), respectively, com- maze task than the APP rats. pared to the APP21 rat line. Serum Aβ and Aβ lev- 40 42 els at 8 weeks were positively correlated with PS1 mRNA Retention After the retention interval, APP  +  PS1 rats levels in APP  +  PS1 founders (Table  1). Statistical anal- made numerically more errors (p = 0.1) and had signifi - ysis of the PS1 expression of founder animals was done cantly longer latencies (p = 0.01) than both the APP and by grouping the animals based on their 8 wk serum Aβ WT rats (Fig. 4b), implying a larger memory deficit in the levels. The groups comprised of rats with Aβ42 serum APP  +  PS1 rats. These differences in memory retention levels of less than 100  pg/ml (<100), between 100 and may be due to the larger learning deficit in APP  + PS1 ani- 149 pg/ml (100–149), between 150 and 200 pg/ml (150– mals during acquisition training. 200), and greater than 200 pg/ml (>200), as well as APP21 rats that did not have the PS1 transgene. PS1 expression Reversal Overall, APP  +  PS1 rats made significantly in tissues increased significantly (p  < 0.05) as serum Aβ more errors (p = 0.04) and had significantly longer laten - levels increased (Fig. 2; Additional file  1: Table 1) with the cies (p  =  0.03) compared to both the WT and APP rats exception of the liver, in which PS1 levels were not sig- during reversal training (Fig. 4c). More specifically, there nificantly different. were no significant differences on days 1 and 2 of reversal Agca et al. BMC Neurosci (2016) 17:46 Page 5 of 10 Fig. 1 a Southern blot analysis of the genomic DNA of the founder animals used to determine the copy number of PS1 transgene. The copy num- ber of the PS1 transgene was determined by the number of hybridized genomic DNA segments in the autoradiogram and is shown in parenthesis above the rat numbers. M represents the molecular weight (MW ) markers. b Serum Aβ /Aβ levels of APP + PS1 double transgenic founder 42 40 animals and APP21 rats at 8 weeks of age. The Aβ /Aβ ratio for rat 988 could not be determined because its Aβ level was lower than the detec- 42 40 42 tion limit of the test. Founder 964 was selected based on having one PS1 transgene and high Aβ /Aβ levels as well as for having a large litter with 42 40 high Aβ /Aβ levels 42 40 Table 1 Correlation coefficients of  serum Aβ and  Aβ Histopathology of brain 42 40 levels to PS1 mRNA levels of the organs The hippocampal pyramidal cell layer of Cornu Ammonis (CA) 1 and CA2 regions of the APP transgenic rats con- PS1 mRNA Serum Aβ Serum Aβ 42 40 tained many necrotic neurons and neurofibrillary tan - Brain 0.49 0.36 gles (Fig.  5a, d, and g). Mineralization was also evident Kidney 0.48 0.30 as indicated by the rough cell membrane observed in Heart 0.60 0.34 some of the neurons in the hippocampus and the cortex Liver 0.45 0.22 in both the APP and the APP +  PS1 rats. In addition to Lung 0.54 0.31 neurons containing flame-shaped neurofibrillary tangles, hippocampus also contained remnants of neurofibrillary tangles that were embedded in the neuropil (ghost tan- gles) in the APP rats (Fig.  5d, g). Neurofibrillary tangles training (ps > 0.05), but on day 3, APP + PS1 rats made were present in viable cells that had nuclei as well as neu- significantly more errors (p  <  0.01) and had significantly rons with no visible nuclei (Fig.  5). The CA3 region also longer latencies (p  =  0.04) than both the APP and WT contained necrotic neurons and neurofibrillary tangles in rats (Fig.  4e). These results suggest deficits in reversal both the APP and APP + PS1 rats (Fig. 5j, k). learning in the APP + PS1 rats. Additional file  3: Table 3 Both the APP and APP  +  PS1 rats had fewer pyrami- contains the dataset of the Barnes Maze. dal neurons in the CA3 region compared to WT, but Agca et al. BMC Neurosci (2016) 17:46 Page 6 of 10 central and segmental chromatolysis, showing eccentric displacement of the nucleus and loss of Nissl substance with the exception of the margins of the cell body (Fig. 5h, k). Both the APP and the APP  +  PS1 brain cortex con- tained neurons with various degrees of degeneration that ranged from neurons with condensed Nissl substance to dark necrotic neurons and ghost tangles (Fig. 5p, q). The brain cortex of the APP  +  PS1 animals had larger areas with dark stained neurons, which indicated more severe neuronal necrosis (Fig. 5q). Discussion The production of transgenic rats using lentiviral vec - Fig. 2 PS1 expression in founder animals and APP controls as deter- tors is an efficient method for gene transfer. However, the mined by Northern blot analysis. The expression of PS1 was analyzed major drawback of lentiviral vectors is the integration of by combining animals with similar serum Aβ levels and statistical analysis conducted for each tissue. The figure legend indicates the the transgene into multiple sites of the genome. Multiple range of serum Aβ level (pg/ml) of the founder animals or the APP copies of the transgene result in high levels of expression, control rats. Different letters above each bar indicates significantly dif- but segregation in subsequent generations drastically ferent (p < 0.05) expression decreases transgene expression. In this study, the founder animal was selected among the animals that had only one copy of the transgene and those that also had the greatest the APP  +  PS1 animals had more severe neuronal loss serum Aβ levels and Aβ /Aβ ratio. 42 42 40 in the CA3 region (Fig.  5b). The hippocampal pyramidal Serum Aβ and Aβ levels were positively correlated 42 40 cell layer in the APP + PS1 rats had many neurons with with PS1 mRNA expression, indicating that the addition Fig. 3 Serum Aβ (A, D), Aβ (B, E) and Aβ /Aβ (C, F) levels of 30 day old F1 and 50 day old F2 generations. Different letters above each bar 42 40 42 40 indicate significantly different levels of Aβ and Aβ /Aβ 42 42 40 Agca et al. BMC Neurosci (2016) 17:46 Page 7 of 10 Fig. 4 Overall performance of Barnes maze Acquisition Training (A) Retention (B), and Reversal Training (C). Daily performance of Acquisition Train- ing (D), and Reversal Training (E). Data is expressed as group mean ± SEM. Different letters above each bar indicates significantly different (p < 0.05) expression of the PS1 transgene contributed to the production of Histologic examination of the brains showed that both Aβ and Aβ . Furthermore, the greater correlation both the APP and the APP  +  PS1 animals exhibited 42 40 efficiencies for Aβ and PS1 mRNA expression indicate pathological abnormalities typically associated with AD. preferential cleavage of Aβ by the PS1 enzyme. Trans- Interestingly, the APP rats had large hippocampal sec- fection using PS1 resulted in an approximately 1.6 fold tions containing neurons with neurofibrillary tangles. increase in Aβ /Aβ levels [16] in cell cultures. Simi- The hippocampus of the APP  +  PS1 rats also contained 42 40 larly, the APP  +  PS1 transgenic animals had approxi- neurofibrillary tangles, but these were more sporadic mately a twofold Aβ /Aβ increase compared to the compared to what was observed in the APP rats. Interest- 42 40 APP rats. The level of PS1 mRNA expression varied by ingly, neuronal loss was much greater in the CA3 region tissue with the highest PS1 expression in the brain and of the APP  +  PS1 rats. Another difference between the kidney, while the heart and liver had lower expression. hippocampus in APP and the APP  +  PS1 animals was Tissue specific expression was also observed in the APP that neurons in the CA1 and the CA2 regions of the transgenic rats [1]. APP + PS1 animals had chromatolysis. Chromatolysis is Agca et al. BMC Neurosci (2016) 17:46 Page 8 of 10 Agca et al. BMC Neurosci (2016) 17:46 Page 9 of 10 Fig. 5 Hematoxylin and Eosin staining of APP, APP + PS1, and WT rat brains. 4× magnification of a hippocampus section from an APP (a), APP + PS1 (b), and WT rat (c). CA1, CA2, CA3 regions of the hippocampus and dentate gyrus (DG) are labeled in c. Panels d–r show 60X magnifica- tion of the brain and each arrow style represents similar pathology throughout the figure. d APP CA1 region with necrotic and dark colored pyrami- dal neurons (black arrowhead), ghost tangles (white arrow), and normal looking neurons (white arrow head). e APP + PS1 CA1 showing degeneration with condensed Nissl substance (grey arrow) and chromatolysis (grey arrow head). f WT CA1. g APP CA2 with neurofibrillary tangles in viable neurons with visible nuclei (black arrow) and non-viable neurons (black arrow with white outline) and ghost tangles. h APP + PS1 CA2 showing chromatolysis in neurons. i WT CA2. j APP CA3 with neurofibrillary tangles and condensed Nissl substance. k APP + PS1 CA3 with neurofibrillary tangles. l WT CA3. m APP DG granule cells with condensed Nissl substance and necrotic cells. n APP + PS1 DG with necrotic cells and neurofibrillary tangles. o WT DG. p, q APP and APP + PS1 cortex neurons with condensed Nissl substance, neurofibrillary tangles, necrotic neurons, and ghost tangles. r WT cortex caused by axonal injury in motor neurons [18] and it has between the APP  +  PS1, APP, and WT rats in learning been associated with neurotoxicity and AD [5, 15, 22], and retention in the task may be related to the differences but it is not reported as widely as the presence of neurofi - in the pathology of the hippocampus. brillary tangles or Aβ plaques. A previously reported rat model for AD, the TgF344- Conclusions AD model [6], was found to produce plaques at The addition of the PS1 transgene to the APP transgenic 15 months. The difference in plaque production between rat genome affected serum Aβ and Aβ /Aβ levels, 42 42 40 the TgF344-AD and the APP + PS1 rats could be due to brain histopathology, as well as, the cognitive behavior the fact that the PS1 transgene in the TgF344-AD had Δ of the double transgenic rats. Although the APP rat brain exon 9 mutation, as opposed to the L166P mutation form had many histopathogical abnormalities that are often of the PS1 transgene that was used in these APP  +  PS1 seen in AD brains, their memory did not show significant rats. On the other hand, the behavioral characteriza- impairment during the behavioral testing. In contrast, tion of the APP + PS1 rats in this current study revealed the APP  +  PS1 rats showed significantly reduced mem - learning and memory deficits similar to what has been ory retention compared to the APP and the WT rats, in found in TgF344-AD rats. In addition, previous research addition to, neuronal loss and necrosis in the hippocam- has found that an AD mouse model that did not develop pus and brain cortex. plaques even as late as 30 months old also showed mem- Additional files ory deficits [9]. Consequently, the presence of Aβ plaques may not always correlate with AD severity. For exam- Additional file 1: Table 1. contains dataset for Northern blot analysis ple, Shankar et  al. [21] showed that Aβ dimers are syn- (see Fig. 2). PS1 mRNA level, serum Aβ42 and tissue are reported in the aptotoxic and disrupted memory, whereas insoluble Aβ table. plaque cores from Alzheimer’s diseased cortex did not Additional file 2: Table 2. contains dataset for serum Aβ , Aβ and 42 40 impair long-term potentiation. Furthermore, the number Aβ /Aβ ratio of F1 and F2 generation transgenic rats (see Fig. 3). 42 40 of Aβ plaques in the brain has not been found to corre- Additional file 3: Table 3. contains dataset for Barnes maze (see Fig. 4). The table includes acquisition day 1, 2 and 3 latency and errors and their late with the severity of cognitive impairments in humans averages, retention latency and errors and reversal days 1, 2 and 3 latency or in APP and/or PS transgenic mice [3]. and errors and their averages. Wild type rats outperformed both the APP and the APP  +  PS1 rats during the initial learning phase of the Barnes maze test. Additionally, retention of the learned Abbreviations AD: alzheimer’s disease; Aβ: amyloid-beta; APP: amyloid precursor protein; CA: task was poorer in the APP + PS1 rats relative to both the cornu Ammonis; eGFP: enhanced green fluorescence protein; IP: intraperito - APP and the WT rats. As stated earlier, one of the major neal; pLV: lentivial vector; PMSG: pregnant mares’ serum gonadotropins; PS1: differences between the APP  + PS1 and the APP rats was presenilin 1; SSC: saline sodium citrate; SDS: sodium dodecyl sulfate; Ubi-C: ubiquitin-C; WT: wild type; WRE: woodchuck hepatitis virus post-transcrip- that the APP  +  PS1 rats had neuronal loss in the CA3 tional regulatory element. region of hippocampus (Fig.  5b shows an APP  +  PS1 rat with very few neurons in the CA3 region). The CA1 Authors’ contributions CA was responsible for genotypic and phenotypic characterization of trans- region recodes information from the CA3 region and sets genic rats, statistical analysis of data, colony management, and writing the up associatively learned backprojections to the neocortex paper. JJL generated lentiviral vectors and performed ELISA. DK and TRS have to allow subsequent retrieval of information to the neo- conducted rat behavioral assessment and analysis. YA and AWC generated transgenic rats. YA designed the research, and contributed to the analysis of cortex [13]. The hippocampal CA3 region and the den - the data. All authors read and approved the final manuscript. tate gyrus also play an important role in the encoding of new spatial information and processing the geometry Author details Department of Veterinary Pathobiology, College of Veterinary Medicine, Uni- of the environment [12], which are both crucial to the versity of Missouri, 1600 East Rollins Street, Room W191, Columbia, MO 65211, successful completion of the task. Thus, the differences 2 USA. Department of Psychological Sciences, University of Missouri, Columbia, Agca et al. BMC Neurosci (2016) 17:46 Page 10 of 10 MO 65211, USA. Yerkes National Primate Research Center, Emory University, 9. Gandy S, Simon AJ, Steele JW, Lublin AL, Lah JJ, Walker LC, Levey AI, Krafft Atlanta, GA 30329, USA. Department of Neurology, Center for Neurodegen- GA, Levy E, Checler F, Glabe C, Bilker WB, Abel T, Schmeidler J, Ehrlich erative Disease, Emory University, Atlanta, GA 30322, USA. ME. Days to criterion as an indicator of toxicity associated with human Alzheimer amyloid-beta oligomers. Ann Neurol. 2010;68(2):220–30. Acknowledgements 10. Jarrett JT, Berger EP, Lansbury PT Jr. The C-terminus of the beta protein is This study was funded by a grant from Alzheimer Research Consortium and critical in amyloidogenesis. Ann N Y Acad Sci. 1993;695:144–8. the University of Missouri startup funds. The authors would also like to thank 11. Kandimalla RJ, Wani WY, Binukumar BK, Gill KD. siRNA against presenilin 1 Dr. Marcia Hart for helping assess brain histopathology. (PS1) down regulates amyloid β42 production in IMR-32 cells. J Biomed Sci. 2012;19:2. Competing interests 12. Kesner RP. Behavioral functions of the CA3 subregion of the hippocam- The authors declare that they have no competing interests. pus. Learn Mem. 2007;14(11):771–81. 13. Kesner RP, Rolls ET. A computational theory of hippocampal function, Availability of data and materials and tests of the theory: new developments. Neurosci Biobehav Rev. Datasets are submitted as supplementary material. 2015;48:92–147. 14. Kim J, Onstead L, Randle S, Price R, Smithson L, Zwizinski C, Dickson Ethics approval and consent to participate DW, Golde T, McGowan E. Abeta40 inhibits amyloid deposition in vivo. J All animal studies were approved by University of Missouri’s Animal Care and Neurosci. 2007;27(3):627–33. Use Committee (Protocol Number 7995). 15. Kotariya NT, Bikashvili TZ, Zhvaniya MG. Chkhikvishvili TsG. Ultrastruc- ture of hippocampal field CA1 in rats after status epilepticus induced Received: 21 April 2016 Accepted: 27 June 2016 by systemic administration of kainic acid. Neurosci Behav Physiol. 2010;40(2):127–30. 16. Li N, Liu K, Qiu Y, Ren Z, Dai R, Deng Y, Qing H. Eec ff t of Presenilin Muta- tions on APP Cleavage; insights into the pathogenesis of FAD. Front Aging Neurosci. 2016;8:51. 17. Lois C, Hong EJ, Pease S, Brown EJ, Baltimore D. Germline transmission References and tissue-specific expression of transgenes delivered by lentiviral vec- 1. Agca C, Fritz JJ, Walker LC, Levey AI, Chan AW, Lah JJ, Agca Y. Develop- tors. Science. 2002;295:868–72. ment of transgenic rats producing human beta-amyloid precursor 18. McIlwain DL, Hoke VB. The role of the cytoskeleton in cell body enlarge- protein as a model for Alzheimer’s disease: transgene and endogenous ment, increased nuclear eccentricity and chromatolysis in axotomized APP genes are regulated tissue-specifically. BMC Neurosci. 2008;9:28. spinal motor neurons. BMC Neurosci. 2005;6:19. 2. Alzheimer’s Association. 2015 Alzheimer’s disease facts and figures. 19. Masters CL, Selkoe DJ. Biochemistry of amyloid beta-protein and Alzheimers Dement 2015; 11(3):332–84. amyloid deposits in Alzheimer disease. Cold Spring Harb Perspect Med. 3. Benilova I, Karran E, De Strooper B. The toxic Aβ oligomer and Alzheimer’s 2012;2(6):a006262. disease: an emperor in need of clothes. Nat Neurosci. 2012;15(3):349–57. 20. Rosen RF, Fritz JJ, Dooyema J, Cintron AF, Hamaguchi T, Lah JJ, LeVine 4. Burdick D, Soreghan B, Kwon M, Kosmoski J, Knauer M, Henschen A, H 3rd, Jucker M, Walker LC. Exogenous seeding of cerebral β-amyloid Yates J, Cotman C, Glabe C. Assembly and aggregation properties of deposition in βAPP-transgenic rats. J Neurochem. 2012;120(5):660–6. synthetic Alzheimer’s A4/beta amyloid peptide analogs. J Biol Chem. 21. Shankar GM, Li S, Mehta TH, Garcia-Munoz A, Shepardson NE, Smith 1992;267(1):546–54. I, Brett FM, Farrell MA, Rowan MJ, Lemere CA, Regan CM, Walsh DM, 5. Cataldo AM, Hamilton DJ, Nixon RA. Lysosomal abnormalities in degener- Sabatini BL, Selkoe DJ. Amyloid-beta protein dimers isolated directly ating neurons link neuronal compromise to senile plaque development from Alzheimer’s brains impair synaptic plasticity and memory. Nat Med. in Alzheimer disease. Brain Res. 1994;640(1–2):68–80. 2008;14(8):837–42. 6. Cohen RM, Rezai-Zadeh K, Weitz TM, Rentsendorj A, Gate D, Spivak I, Bho- 22. Sil S, Goswami AR, Dutta G, Ghosh T. Eec ff ts of naproxen on immune lat Y, Vasilevko V, Glabe CG, Breunig JJ, Rakic P, Davtyan H, Agadjanyan MG, responses in a colchicine-induced rat model of Alzheimer’s disease. Kepe V, Barrio JR, Bannykh S, Szekely CA, Pechnick RN, Town T. A trans- NeuroImmunoModulation. 2014;21(6):304–21. genic Alzheimer rat with plaques, tau pathology, behavioral impairment, 23. Walsh DM, Selkoe DJ. A beta oligomers—a decade of discovery. J Neuro- oligomeric aβ, and frank neuronal loss. J Neurosci. 2013;33(15):6245–56. chem. 2007;101(5):1172–84. 7. De Strooper B. Aph-1, Pen-2, and Nicastrin with Presenilin generate an active gamma-Secretase complex. Neuron. 2003;38(1):9–12. 8. Deng Y, Tarassishin L, Kallhoff V, Peethumnongsin E, Wu L, Li YM, Zheng H. J Neurosci. 2006;26:3845–54. Submit your next manuscript to BioMed Central and we will help you at every step: • We accept pre-submission inquiries • Our selector tool helps you to find the most relevant journal • We provide round the clock customer support • Convenient online submission • Thorough peer review • Inclusion in PubMed and all major indexing services • Maximum visibility for your research Submit your manuscript at www.biomedcentral.com/submit

Journal

BMC Neuroscience – Springer Journals

Published: Jul 7, 2016

There are no references for this article.