Alpha-synuclein expression patterns in the colonic submucosal plexus of the aging Fischer 344 rat: Implications for biopsies in aging and neurodegenerative disorders?

Abstract

Background

This experiment assessed normative expression patterns of alpha-synuclein (SYNC), including ganglionic remodeling and development of SYNC pathologies, in the submucosal plexus (SMP) of the colon during healthy aging. The observations address age-associated changes in bowel function and are relevant to evaluations of SMP-containing colonic biopsies for SYNC or synucleinopathies associated with aging and peripheral neurodegenerative diseases.

Methods

Colonic submucosal whole mounts from groups of virgin male Fischer 344 rats (n ≥ 8 per group) at 4, 8, 16 and 24 months of age were processed immunohistochemically for SYNC and the pan-neuronal marker HuC/D. Additionally, macrophages immunoreactive for MHCII were examined. Stereological protocols were used to generate unbiased estimates of neuron density, neurons per ganglion, neurons per ganglionic area, and neuron size.

Key Results

The protein SYNC was expressed in a subpopulation of SMP neurons, in both nucleus and cytoplasm. The general age-associated pattern across different cell counts was an increase in the number of SYNC+ neurons between 4 and 8 months of age, with progressively decreasing numbers of both SYNC+ and SYNC− neurons over the remaining lifespan. The soma size of SYNC+ neurons increased progressively with age. Aggregated SYNC occurred in the aging SMP, and macrophages with alternatively activated profiles were located adjacent to pathological SYNC deposits, consistent with ongoing phagocytosis.

Conclusions & Inferences

Changes in SYNC expression with age, including a baseline of accumulating synucleopathies in the healthy aging SMP, need to be considered when interpreting either functional disturbances or biopsies of the aging colon.

INTRODUCTION

The normative pattern of alpha-synuclein (SYNC) expression in the submucosal plexus (SMP) is poorly described, especially for the healthy aging colon. Further, much of the information is indirect: in tissue obtained from partial colectomies addressing disease (carcinoma; adenoma; prolapse), Bottner and colleagues¹ described native SYNC throughout the human SMP, regardless of age (i.e., 15 to 83 years of age). In the Fischer 344 rat model of aging, the expression of SYNC has not been evaluated specifically for colonic SMP, but SYNC is expressed in a subpopulation of myenteric neurons in the stomach and small intestines of adults², and aggregated SYNC accumulates in the myenteric ganglia and tunica muscularis of aged rats³. This aggregated SYNC in the aged rat smooth muscle is likely representative of the age-related dystrophies that appear to be a normal part of the healthy aging process of the GI tract and, as such, may generalize to the SMP. Given the recognized age-related disorders of the colon related to secretion and motility^4-6, a fuller characterization of SYNC expression in the SMP would clearly have utility.

An additional need stems from the fact that colonic biopsy material is being screened for alpha-synuclein^1,7-14. If the presence of SYNC in colonic biopsies is going to be fully interpretable, a better understanding of the normative expression of the protein in the colon SMP of health aging subjects sampled over the lifespan is needed, especially taking into account progressive degenerative changes that are a normal part of aging and that could present as false-positive signs. The goals of the present study, therefore, were to a) describe the pattern of expression of native SYNC within the colonic SMP, b) ascertain if age-related changes occur in the SYNC+ population of neurons in the colonic SMP, and c) determine if aggregated SYNC is present in the colonic SMP in healthy aging.

MATERIALS AND METHODS

Subjects

Virgin Male Fischer 344 (F344; n = 67) rats were purchased from the National Institute on Aging colony maintained by Taconic Farms (Germantown, NY). The animals were housed (n = 2/cage) in polypropylene cages containing Alpha-dri bedding (Cincinnati Lab Supply, Cincinnati, OH) in a room kept at 20-22 °C on a 12:12 hour light:dark schedule. Group housing, nylabones and polycarbonate tunnels (Bio Serv, Frenchtown, NJ) provided environmental enrichment to minimize stress¹⁵. Solid chow (NIH-31M; Zeigler Feed, Gardners, PA) and filtered tap water were available ad libitum. Conditions in our AAALAC-approved colony approximated the housing, husbandry, and barrier conditions recommended by the National Institutes on Aging¹⁵, but did not provide a specific pathogen-free environment. Each animal's health status was checked on a daily basis, and the colony room was monitored for known rat pathogens by the Purdue University veterinary surveillance program through the use of sentinels that were rotated through the colony room every three months, followed by examination for parasites and serum testing for antiviral antibodies. All procedures were conducted in accordance with the National Institute of Health Guide for the Care and Use of Laboratory Animals (8^th ed., The National Academic Press, Washington, D.C.), and were approved by the Purdue University Animal Care and Use Committee.

Fixation Protocol and Whole Mount Preparation

At designated ages, rats were weighed and then killed with a lethal dose of sodium pentobarbital (180 mg/kg. i.p) and perfused through the left ventricle of the heart with 200 ml of 0.01M PBS followed by 400 ml of Zamboni's fixative. The entire large intestine was removed, and the length of the organ was determined by measuring from the cecal-colonic junction to the anal sphincter. Based on the criteria of Hebel and Stromberg¹⁶, 8.0 cm of large intestine distal to the pelvic brim was collected and defined as the colon for the present study. The mucosa and muscularis externa were dissected away from the colon tissue, leaving an intact whole mount of submucosa (SM) containing the submucosal plexus (SMP). The specimen was then divided into smaller whole mounts and further fixed in Zamboni's fixative overnight.

Immunohistochemistry

Permanent single labeling

For single labeling of alpha-synuclein (SYNC), whole mounts of the SM were rinsed in PBS, and then incubated for 30 min in a hydrogen peroxide:methanol (1:4) solution to quench endogenous peroxidase. Following several more PBS rinses, whole mounts were soaked for 72 h in PBS containing 5% normal horse serum, 2% bovine serum albumin, 0.5% Triton X-100, and 0.08% Na azide. Free floating whole mounts were then incubated for 24 h with a mouse antiserum raised against SYNC (1:2500; 610787; BD Bioscience, San Jose, CA) diluted with PBS containing 2% normal horse serum, 2% bovine serum albumin, 0.3% Triton X-100, and 0.08% Na azide. Specimens were then rinsed in PBS and incubated for 1 h with biotinylated anti-mouse IgG, rat absorbed, raised in horse (1:500; BA-2001; Vector Laboratories, Inc., Burlingame, CA), followed by incubation for 1 h with an avidin-biotin-horseradish peroxidase complex (Vectastain Elite ABC Kit, Standard; PK-6100; Vector Laboratories, Inc.). Horseradish peroxidase was then reacted for 5 min with the Vector SG Peroxidase Substrate (SK-4700; Vector Laboratories, Inc; working solution mixed according to the manufacturer's instructions), and finally rinsed and stored in cold dH₂O. Other than the final rinses, every step was done at room temperature with gentle agitation on a shaker table. Stained whole mounts were mounted on gelatin-coated slides, air-dried overnight, dehydrated in an ascending series of alcohols, cleared in two xylene steps, and coverslipped with Cytoseal XYL (Richard-Allen Scientific, Kalamazoo, MI). Specificity was determined by omitting the primary antibody.

Permanent double-labeling

Double staining using antibodies to the panneuronal marker HuC/D¹⁷ (1:5000; A-21271; BD Transduction, San Jose, CA) and, in different specimens, either SYNC (1:2500; 610787; BD Bioscience), neuronal nitric oxide synthase (nNOS; 1:8000; SC-648; Santa Cruz Biotechnologies, Santa Cruz, CA), the calcium binding proteins calbindin-D-28K (CB; 1:8000; CB-38; Swant, Bellinoza, Switzerland) and calretinin (CR; 1:4000; VP-RM11; Vector Laboratories, Inc.), or the macrophage marker major histocompatibility complex class II¹⁸ (MHCII; 1:2000; MCA46R; AbD Serotec, Oxford, UK), was done sequentially using the staining protocol described above with the chromagen diaminobenzidine (DAB) reacted with H₂O₂ for 3 min, prior to the reaction with the Vector SG chromagen. Depending upon the host species of the primary, either biotinylated anti-mouse IgG (1:500; BA-2001; Vector Laboratories, Inc.) or biotinylated anti-rabbit IgG (1:500; 111-065-144; Jackson Immuno Research, West Grove, PA) was used as the secondary (cf., Phillips and Powley¹⁹ for a detailed description of our sequential double-labeling protocol). Specificity was determined by omitting the relevant primary antibody.

Fluorescent labeling

To determine colocalization of SYNC with either nNOS, CB, or CR, SM whole mounts were soaked for 4 d in PBS containing 5% normal goat serum, 2% bovine serum albumin, 0.5% Triton X-100, and 0.08% Na azide, followed by incubation for 48 h in a cocktail consisting of the mouse primary SYNC (1:2500; 610787; BD Bioscience) and one of the following three rabbit antibodies: nNOS (1:2500; SC-648; Santa Cruz Biotechnologies), CB (1:8000; CB-38; Swant), or CR (1:4000; VP-RM11; Vector Laboratories, Inc.). Whole mounts were then rinsed in PBS and incubated for 2 h in a cocktail consisting of goat anti-mouse ALEXA Flour 488 (1:500; A11029; Molecular Probes, Eugene, OR) and goat anti-rabbit ALEXA Flour 594 (1:500; A11037; Molecular Probes), diluted with PBS. To quench lipofuscin autofluorescence present in the aged tissue, all fluorescently labeled whole mounts, regardless of age, were soaked for 90 min in cupric sulfate^3,20. Finally, labeled whole mounts were rinsed in PBS, mounted on gelatin-coated slides, air dried overnight, dehydrated in alcohol, cleared in xylene, and cover-slipped with DePeX (13515; Electron Microscopy Sciences, Harfield, PA). Negative controls were run by omitting the relevant primary antibodies.

Sampling Protocols

To address the sampling complications that result from (a) organ growth, (b) differential patterns of growth of different tissues within an organ, and (c) inhomogeneous shrinking and stretching associated with tissue preparation^21-28, we employed multiple indices of neuronal number, density and distribution.

Neuron and macrophage density

Unbiased estimates of neuron and macrophage density (or number per unit area) were generated with stereology using the colon SM whole mounts single-stained for SYNC or double stained for SYNC and MHCII, respectively. Stereo Investigator (Version 10; MicroBrightField, Inc., Williston, VT) was used for systematic random sampling of selection sites in the whole mounts. Briefly, the contour of each whole mount was traced, the specimen's area was determined, and then sites within each contour were chosen using the fractionator probe. A square grid (0.5 mm × 0.5 mm = 0.25 mm²) was projected onto each site and every cell that fell within the grid was counted. In this way, the entire width and length of the whole mount was sampled. Observations were done at a magnification of 250x by the same experimenter who was blind to the age of the rat from which the whole mount originated.

“Dilution” of enteric neuronal density is thought to be a normal artifact of age as the organ grows and as the tissue area of the large intestine increases with age, so both “uncorrected” and “corrected” estimates of neuronal counts were calculated. In earlier whole mount surveys of the myenteric plexus, normalizing areal density counts to the youngest age examined using a correction factor derived from the difference in organ area between the youngest age and the age being compared has proven to be reasonably accurate^21,23,24,29. The correction factor's accuracy is likely aided by the uniform nature of the innervation of the smooth muscle wall by the myenteric plexus. In the present study, though, we did not know how stretch or age-related growth would affect the areal density estimates of neurons in the colonic SMP. Thus, whereas in the myenteric plexus, stretch and growth effects on areal neuron density appear to be somewhat uniform with age, it was not so clear that this would be the case for the SMP, especially taking into account the diffuse placement of ganglia throughout the SMP and their affinity for blood vessels. Therefore, in the present study, we report both the uncorrected and corrected areal estimates along with several other measures to more completely capture the patterns of change that occur over the lifetime of the aging SMP. Rather than relying on any single measure, the trend seen in multiple measures is more likely to reflect the actual pattern of age-related change in the SMP.

Neurons per ganglion and neurons per ganglionic area

The boundary of a ganglion stained for HuC/D in the rat colonic SMP was easily delineated, making it practical to take additional quantitative measures of the plexus. Specifically, counts of neurons/ganglion and neurons/ganglionic area, two measures considered to be stretch-independent and therefore especially important and provisionally valid indices for aging studies estimating neuron numbers^22,30, were made. Similar to the process described above for estimating the density of SYNC stained neurons, Stereo Investigator was used to generate an unbiased selection of sites in each SM whole mount--stained for both HuC/D and SYNC−-from which ganglia were chosen for analysis. Quantification of ganglia was done by the same experimenter who was blind to the age of the rat from which the whole mount originated.

A ganglion was defined as consisting of two or more neurons³¹ with a distance between the neurons not exceeding one neuron diameter. Neuron diameter, estimated in a pilot study measuring 2074 neurons in colon SM whole mounts from 8-9 month-old-rats, was 22.2 μm. Once a ganglion was identified, every HuC/D+ neuron within the ganglion was marked, and those neurons that were SYNC+ were designated with a second marker. Finally, the outline of the ganglion's perimeter was traced. Since the area of a ganglion is determined by the amount of neuropil and the sizes of the neuronal somata, both of which could vary, as well as by the neuron number, both “neurons per ganglion” and “neurons per ganglionic area” were determined.

Estimates of neuron size

Using Stereo Investigator, measurement of the size (i.e., cross sectional area of the soma) of randomly chosen SYNC+ neurons was made using the whole mounts single-labeled for SYNC. A microscopic image was projected to a monitor, and the somata were outlined. The investigator making the measurements was blind to the age of the rat from which the tissue originated. The results are expressed as mean percentages of the total number of SYNC+ neurons using a bin width of 50 μm². The mean percentages for each interval were obtained from the individual percentages of each rat²⁹.

Nitrergic and cholinergic subpopulations and their colocalization with SYNC

Whole mounts of the large intestine SM double stained for HuC/D and either nNOS, CB, or CR were used to count and then to calculate the proportion of total neurons in the colon SMP that were either nitrergic (i.e., nNOS+) or cholinergic (i.e., CB+ or CR+). Stereo Investigator was used for systematic random sampling to provide unbiased selection of sites in the whole mounts. A square grid (0.5 mm × 0.5 mm = 0.25 mm²) was projected onto each site and every neuron that fell within the square grid was marked. A second marker was then placed on the double-stained neurons. Observations were done at a magnification of 250x by the same experimenter who was blind to the age of the rat from which the whole mount originated.

The percentage of SYNC+ neurons co-reactive for nNOS, CB, or CR was determined using the quantitative sampling strategy described and validated by Burnstock and colleagues^32,33 for fluorescently labeled neurons. Briefly, the SM whole mounts stained for SYNC and either nNOS, CB or CR were systematically scanned until a SYNC+ neuron was identified, and then co-localization (or its absence) with the second marker (i.e., either nNOS, CB, or CR) was determined.

Brightfield and Fluorescent Imaging, Image Post-processing, Statistical Analysis, and Graphs

Whole mounts were examined using a Leica DM microscope (Leica Microsystems, Inc., Buffalo Grove, IL), and images were captured using a Spot Flex camera (Diagnostic Instruments, Sterling Heights, MI) controlled using Spot Software (V4.7 Advanced Plus; Diagnostic Instruments). Focus stacking³⁴ was occasionally used to maximize the depth of field of images taken from thick whole mounts of the SM. Helicon Focus Pro X64 (Version 5.1.23; HeliconSoft Ltd., Kharkov, Ukraine) generated the all-in-focus images. Photoshop CS6 (Adobe Systems, San Jose, CA) was used to: (1) organize the layout of the figures; (2) adjust color, hue, brightness, contrast, and sharpness of images; and (3) apply text and scale bars. The goal of these adjustments was to generate final images which accurately conveyed the neural innervations of the GI tract as viewed under the microscope. GraphPad Prism (Version 6.01; GraphPad Software, San Diego, CA) was used for statistical analysis of the data and the generation of graphs. One-way analysis of variance (ANOVA) with Tukey tests was used to compare the means of more than two ages. A Student's t-test was used to compare means when only two ages were sampled. For all statistical tests, a level of probability of 0.05 or less was considered significant.

RESULTS

Growth, Age and Tissue Specimens

At the four ages sampled (4, 8, 16 and 24 months), changes occurred in body and organ size, dictating the use of the multiple indices of neuron number. There was a significant change in body weight with age (ANOVA; p < 0.0001), as the three oldest ages were significantly heavier compared to 4-months (Tukey; p values < 0.05); however, weight loss occurred at 24 months of age, as this age was significantly lighter compared to 16-months (Tukey; p < 0.05). The mean (± SEM) body weights for the four ages were 354 ± 5, 454 ± 10, 501 ± 12, and 435 ± 10 grams at 4, 8, 16, and 24 months of age, respectively. Similarly, there was a significant increase in large intestinal length with age (ANOVA; p < 0.0001), with the large intestines of 8, 16, and 24 month old rats being significantly longer compared to 4-months (Tukey; p values < 0.05). The mean (± SEM) large intestinal lengths for the four ages were 20 ± 0.3, 22 ± 0.3, 23 ± 0.4, and 24 ± 0.4 cm at 4, 8, 16, and 24 months of age, respectively.

Neuron Density

Whole mounts of the colon SM from F344 rats were stained for SYNC and sampled to determine neuron density (or number per unit area). Eight rats per age at 4, 8, 16, and 24 months of age were sampled. A subset of submucosal neurons was SYNC+, and these neurons expressed the protein in the nucleus and cytoplasm; Fig. 1A. An average of 37 sites was sampled per SM whole mount. Areal density counts, uncorrected for age-related differences in tissue area, revealed a significant change in SYNC expression with age (ANOVA; p < 0.01). Specifically, there was a 39% decrease in SYNC+ neurons in the colons of 16 and 24 month old rats compared to 4-months (Tukey; p values < 0.05); Fig. 2A. Areal density counts corrected for age-related changes in tissue area^21,23,24 similarly reflected a significant change in SYNC expression in SMP neurons with age (ANOVA; p < 0.001). For corrected counts, however, there was a significant developmental increase in the number of SYNC+ neurons between 4 and 8 months of age (Tukey; p < 0.05), followed by a significant decrease in SYNC+ neurons between 8 and 24 months of age (Tukey; p < 0.05); Fig. 2B.

Figure 1.

Alpha-synuclein (SYNC) staining in the colon submucosal plexus (SMP). A: Neurons immunoreactive for SYNC expressed the protein in both the nucleus and the cytoplasm of the somata (black; Vector SG), with an unstained nucleolus. B: Neurons HuC/D+ (brown; DAB) and SYNC+ (black; Vector SG) had brown cytoplasm with black nuclei with unstained nucleoli; neurons that were HuC/D+ but SYNC− had brown cytoplasm with unstained nuclei (e.g., the cell in the upper right corner). C: Varicosities immunoreactive for SYNC (black; Vector SG) occurred throughout the SMP ganglia. Scale bar = 10 μm in C (applies to A-C).

Figure 2.

Mean ± SEM density of SYNC+ neurons in the aging colon SMP. A: For uncorrected areal counts there was a significant decrease in the density of SYNC+ neurons at 16 and 24 months of age. B: For areal density counts of SYNC+ neurons corrected for age-related changes in tissue area there was a significant increase in SYNC+ neurons at 8 months of age followed by a linear decrease in SYNC+ neurons through 24 months of age. An “a” above an error bar indicates a significant difference compared to 4 months of age, and “b” indicates a significant difference compared to 8 months of age.

Neurons per Ganglion, Neurons per Ganglionic Area, and Solo Neurons

Adjacent whole mounts of the SM--taken from the same 32 rats used to generate areal density estimates--were double labeled with the pan-neuronal antibody HuC/D, visualized with DAB, plus immunohistochemistry for SYNC, visualized with Vector SG. This labeling protocol resulted in the cytoplasm of all SMP neurons staining brown, and the nuclei of those neurons that were SYNC+ staining black; Fig. 1B. Alpha-synuclein-positive fibers and varicosities were also observed throughout the SMP ganglia; Fig. 1C. The permanent-staining double-labeling protocol allowed for extensive quantification of the SMP ganglia. A total of 470, 402, 342, and 350 ganglia, sampled at 4, 8, 16, and 24 months of age, respectively, was used to determine the number of HuC/D+ neurons per ganglion, SYNC+ neurons per ganglion, HuC/D+ neurons per ganglionic area, and SYNC+ neurons per ganglionic area.

Age had a significant effect on the number of HuC/D+ neurons per ganglion (ANOVA; p < 0.0001), with significantly fewer HuC/D+ neurons per ganglion at 16 and 24 months of age compared to both 4- and 8-month-old rats (Tukey; p values < 0.05); Fig. 3A. There was a similar change in SYNC+ neurons with age (ANOVA; p < 0.01), with significantly fewer SYNC+ neurons per ganglion at 16 and 24 months of age compared to 8-months (Tukey; p values < 0.05); Fig. 3B. Interestingly, 90% of the 1564 SMP ganglia surveyed contained at least one SYNC+ neuron, and 97% of the 757 ganglia, which consisted of four or more HuC/D+ enteric neurons, contained at least one SYNC+ neuron.

Figure 3.

Quantitative comparisons of the age-related changes in the density of HuC/D+ and SYNC+ neurons expressed as per ganglion and per ganglionic area. A,B: There was a significant age-related decrease in the number of HuC/D+ and SYNC+ neurons/ganglion. C,D: The number of HuC/D+ neurons per ganglionic area was significantly decreased with age, whereas no significant differences were noted in the SYNC+ population of neurons. An “a” above an error bar indicates a significant difference compared to 4 months of age, and “b” indicates a significant difference compared to 8 months of age.

Ganglionic area was significantly different across age overall (ANOVA; p < 0.01), but no individual differences between the group means were revealed by Tukey post-hoc tests (p values > 0.05). The mean (± SEM) ganglionic area was 1824 ± 89, 1817 ± 61, 1520 ± 60, and 1521 ± 110 μm² at 4, 8, 16, and 24 months of age, respectively. There was a significant decrease in HuC/D+ neurons per ganglionic area with age (ANOVA; p < 0.05), with 8% fewer HuC/D+ neurons at 24 months of age compared to 4-months (Tukey; p < 0.05); Fig. 3C. There was no significant effect of age on the number of SYNC+ neurons per ganglionic area (ANOVA; p = 0.08); Fig. 3D.

Because there is an age-related decrease in the number of HuC/D+ neurons per ganglion, we hypothesized that if this decrease reflected neuron death with age, there might subsequently be more solo neurons (i.e., a neuron that was not part of a ganglion) present in the colon SMP, as the ganglia degenerate with age. To test this prediction, a total of 350, 346, 349, and 356 sites were sampled at 4, 8, 16, and 24 months of age, respectively. A marker was placed on each solo HuC/D+ neuron occurring within a square grid (0.5 mm × 0.5 mm = 0.25 mm²), and then a second marker was assigned if the neuron was also immunoreactive for SYNC. Areal density counts of solo neurons were corrected for age-related differences in tissue area. There was a significant effect of age on the population of solo HuC/D+ neurons (ANOVA; p < 0.0001), but not SYNC+ neurons (ANOVA; p = 0.06). Specifically, there were significantly more solo HuC/D+ neurons at 16 and 24 months of age compared to 4- and 8-months (Tukey; p values < 0.05); Fig. 4A. There was a gradual linear increase in the density of solo neurons expressing SYNC with age; Fig. 4B.

Figure 4.

Solo neurons, that is individual neurons that were not part of any ganglion, were scattered throughout the colon SMP. A: There was a significant increase in the density of solo HuC/D+ neurons at 16 and 24 months of age. B: The density of solo neurons immunoreactive for SYNC also increased with age. An “a” above an error bar indicates a significant difference compared to 4 months of age, and “b” indicates a significant difference compared to 8 months of age.

SYNC Neuron Size

Neuron soma size was based on a sample of 517, 574, 346, and 404 neurons at 4, 8, 16, and 24 months of age, respectively. Neurons were sampled from the 32 SM whole mounts (8 per age) that were single-stained for SYNC and used to estimate neuron areal density. A significant age-related increase in mean SYNC+ neuron size occurred in the SMP (ANOVA; p = 0.03) with mean neuron size progressively increasing with age and becoming significantly larger by 24 months of age compared to 4-months (Tukey; p < 0.05); Fig. 5A. The increase in mean SYNC+ neuron size at 24 months of age was a result of a decrease in the percentage of smaller SYNC+ neurons ranging between 200 to 400 μm², with a corresponding increase in the percentage of SYNC+ neurons greater than 450 μm²; Fig. 5B.

Figure 5.

Age-related changes in SYNC+ neuron size in the SMP. A: The mean ± SEM size or cross-sectional area of the somata of SYNC+ neurons increased with age. B: The mean percentages of SYNC+ neurons were obtained from the single percentages of each rat, and neuron size is graphed in 50 μm bins. The increase in mean neuron size at 24 months of age is a result of a decrease in the percentage of smaller neurons ranging between 200 to 400 μm², with a concurrent increase in the percentage of neurons greater than 450 μm². An “a” above an error bar indicates a significant difference compared to 4 months of age.

Nitrergic and Cholinergic Subpopulations and their Colocalization with SYNC

The colons from eight rats, two per age, at 8, 12, 16, and 24 months of age were divided into three SM whole mounts and stained with the pan-neuronal marker HuC/D, visualized with DAB, followed by staining for nNOS, CB, or CR, respectively, visualized with Vector SG. An average of 33 sites per whole mount was sampled for nNOS, CB, and CR, respectively, for each of the four ages. There was no difference between the four ages for the three phenotypes (ANOVA; p values > 0.05), so the counts for each of the three phenotypes were collapsed across age. There was a significant difference between the phenotypes (ANOVA; p < 0.0001), with the proportion of neurons immunoreactive for CB and CR significantly greater compared to nNOS (Tukey; p values < 0.05). The mean (± SEM) proportion of the HuC/D+ neurons co-reactive for the three phenotypes was 31 ± 2.4, 67 ± 2.2, and 69 ± 2.3 percent for nNOS, CB, and CR, respectively; Fig. 6.

Figure 6.

Both nitrergic and cholinergic neurons were present in the SMP. The mean percentage of the HuC/D+ neurons (brown; DAB) co-reactive for the different phenotypes were: A) 31% neuronal nitric oxide synthase (nNOS; black; Vector SG), B) 67% calbindin (CB; black; Vector SG), and C) 69% calretinin (CR; black; Vector SG). [In each case where a nitrergic or cholinergic marker stained positive with Vector SG, the black chromagen masked out the brown DAB staining for HuC/D.] Scale bar = 10 μm in C (applies to A-C).

For the evaluation of colocalization with SYNC, the colons from an additional 14 rats, seven per age, at 10 and 25 months of age, were divided into three SM whole mounts and fluorescently stained for SYNC followed by staining for nNOS, CB, or CR, respectively. Fifty SYNC+ neurons per whole mount were identified and colocalization with nNOS, CB, or CR was determined. This resulted in a total of 350 SYNC+ neurons per age per double-labeling protocol used to determine the percentage of SYNC+ neurons co-reactive for nNOS, CB, or CR; Fig 7. One whole mount double labeled for SYNC and CR at 25 months of age was damaged during the dissection processes, resulting in only 300 neurons sampled for that particular age and double-labeling protocol. Unpaired t-tests determined that there was no difference between the two ages for the three phenotypes (Unpaired t-test; p values > 0.05). At 10 and 25 months of age, 53 ± 2 and 50 ± 3 percent of the SYNC+ neurons were co-reactive for nNOS, 67 ± 5 and 71 ± 4 percent were co-reactive for CB, and 73 ± 5 and 69 ± 5 percent were co-reactive for CR.

Figure 7.

Alpha-synuclein-positive neurons in the SMP were co-reactive for nNOS, CB, and CR. There were no age-related differences in the proportion of the three phenotypes. A: At 10 and 25 months of age, 53 ± 2 and 50 ± 3 percent, respectively, of the SYNC+ neurons were co-reactive for nNOS. B: At 10 and 25 months of age, 67 ± 5 and 71 ± 4 percent, respectively, of the SYNC+ neurons were co-reactive for CB. C: At 10 and 25 months of age, 73 ± 5 and 69 ± 5 percent, respectively, of the SYNC+ neurons were co-reactive for CR. Scale bar = 20 μm in C (applies to A-C).

MHCII+ Macrophages

Finally, the colons from an additional 13 rats, eight at 10 months of age and five at 24 months of age, were prepared as SM whole mounts stained for both SYNC, visualized with DAB, and the macrophage marker MHCII, visualized with Vector SG. The majority of the MHCII+ cells in the SMP consisted of smooth, round monocytes; however, short thread-like tendrils were frequently observed projecting from the perimeter of monocytes; Fig. 8E,F. A total of 167 sites containing 7,427 MHCII+ monocytes, with a mean (± SEM) of 44 ± 5 cells per 0.25 mm², were estimated for the whole mounts sampled at 10 months of age; whereas, a total of 138 sites containing 4,092 MHCII+ monocytes, with a mean (± SEM) of 25 ±2 cells per 0.25 mm², were determined from the whole mounts sampled at 24 months of age, reflecting a significant (Unpaired t-test; p = 0.02) reduction of 43% in the number of MHCII+ cells in the aged colon SMP.

Figure 8.

Morphological observations of the colonic SMP indicated age-related degenerative changes. A,B: Aggregated SYNC (black; Vector SG) was apparent in the SMP ganglia of aging rats. C,D: Swollen SYNC+ varicosities (black; Vector SG) were similarly present in the SMP ganglia (HuC/D; brown; DAB) of aging rats. E: Monocytes immunoreactive for major histocompatibility complex class II (MHCII; black; Vector SG) were present throughout the SMP. F: Some MHCII+ cells had thin tendrils projecting from the perimeter of the cell. G,H: A phagocytotic response was displayed by MHCII+ cells (black; Vector SG) within the vicinity of aggregated SYNC (brown; DAB). Panels A,C-H are from 24 month old rats and Panel B is from a 16 month old rat. Scale bar = 20 μm in H (applies to A-H).

Qualitative Observations of SYNC Aggregates

Protein aggregates stained for SYNC were observed within the ganglia of the colon SMP at 16 and 24 months of age (Fig. 8A,B,D), along with large caliber SYNC+ axons projecting away from the dystrophic ganglia; Fig. 8A,B. Additionally, SYNC+ swollen varicosities were observed encircling neurons within the aged ganglia; Fig. 8C. Finally, MHCII+ cells within the vicinity of aggregated SYNC were morphologically consistent with ongoing phagocytosis; Fig. 8G,H. Specifically, instead of having a smooth round morphology, the cells would have numerous long-thin tendrils that often came into close registration with the aggregated SYNC; Fig. 8H.

DISCUSSION

In light of indications that long-recognized age-related disturbances of colonic secretion and motility with aging^4-6 parallel the development of neuropathies and cell losses observed in the submucosal plexus (SMP)³⁵, and, further, in light of the possibility that alpha-synuclein (SYNC) staining in colonic biopsies may be a useful biomarker for diagnosing neurodegenerative disorders, the present study was designed to examine the pattern of expression of the protein in the enteric circuitry of the SMP over the course of presumably healthy aging. In addition to replicating the previously established loss of SMP neurons with age³⁵, the present experiment added several key findings including that: a) a subpopulation of SMP neurons is immunoreactive for SYNC, and neuronal expression of SYNC varies with age, b) neurons immunoreactive for SYNC increase in size with age, c) SYNC is present in both nitrergic and cholinergic SMP neurons, and d) aggregates of SYNC protein occur in the SMP of healthy aging subjects. Additionally, the presence of presumably alternatively activated macrophages at sites of aggregated SYNC suggests that macrophages use phagocytosis and not a classical inflammatory response to clear deposits of SYNC from the colonic SMP³⁶. The results can be discussed in terms of these observations.

SMP Losses and Reorganization with Age

Our earlier analysis of the effects of aging on the morphological features and extrinsic inputs of the SMP³⁵ complements the present experiment, and the two studies together begin to characterize how the plexus ages. Both experiments draw a similar picture of a progressive loss of neurons within the ganglia that approximates a 15-20% attrition by old age. More particularly, the present study found an 18% loss of HuC/D+ neurons per ganglion at 24 months of age; and our previous report³⁵, using Cuprolinic Blue, an alternative pan-neuronal marker based on RNA staining¹⁷, observed a 15% cell loss of colonic SMP neurons at 24 months of age.

The cause(s) of the cell loss cannot be specified from the available observations, but the analyses in the two experiments supply some useful initial information. Notably, corresponding with an overall loss of neurons (i.e., HuC/D+ neurons) with age in the present experiment, there was a significant loss of neurons in the SYNC+ subpopulation of the SMP. Furthermore, several indices measured, including the enlargement of the SYNC+ positive neurons with age, the accumulations of SYNC aggregates, and the presence of alternatively activated macrophages, are all consistent with the SMP neurons being under some physiological stress or load. In this regard, it may also be relevant that our earlier experiment³⁵ characterized dystrophic, or neuropathic, profiles of the extrinsic catecholaminergic (i.e. tyrosine hydroxylase positive, presumably predominately sympathetic) projections to the SMP ganglion neurons.

The pattern of results across the two experiments, whether it signals a cause(s) of the aging-related neuropathies or not, is consistent with a diminution in trophic maintenance of the autonomic outflow controlling the colon. Conceivably, wear and tear on the enteric neurons in the ganglia of the SMP, and/or the correlated losses in enteric cell number, may diminish the trophic support of the preganglionic projections to the ganglia and produce the dystrophic sympathetic profiles we previously described³⁵. Alternatively, dysfunction and dystrophy of the sympathetic inputs could predate and, eventually, produce the involution of the SMP ganglia.

SYNC Expression Patterns and Their Implications for neurodegenerative diseases

Considerations of the implications of the present experiment need to be tempered by recognition that (a) the observations evaluate peripheral, not central, neurodegeneration, (b) the observations characterize a rat model, not human material, and (c) the analyses are based on the Fisher 344 rat, generally considered a strong model of normal or “healthy” aging¹⁵, not a model of disease-initiated neurodegeneration. With these provisos underscored, it is still worth considering some potential extrapolations of the present observations that need to be more fully addressed in future experiments. Such considerations can facilitate assessing what correlations and associations can eventually be made between peripheral and central neurodegeneration, rat and human neuropathies, and normal aging and aging-associated neuropathies.

In particular, the present observations address two issues relevant to synucleinopathies involving the gut, for example Parkinson's disease. One deals with mechanism(s) of such disorders, and the other addresses diagnosis. They can be considered in that order.

The present observations have implications for the “Braak hypothesis”^37,38, which posits that the CNS synucleopathies of Parkinson's disease result from neuronal retrograde transport of misfolded and insoluble SYNC (and perhaps pathogens that provoke misfolding) from the GI tract to the brain. Even with the caveat that the F344 model of aging used in the present experiment was not parkinsonian, our previous SYNC series^2,39, as well as the observations in the current study, delineate a neural network and a mechanism that has the interconnectivity needed for a pathway of retrograde transport as assumed in the Braak hypothesis^37,38. The observations collectively indicate that, in both the rostral GI tract^3,19,37 and in the large intestines^7,11-14 (the present experiment), aggregates of apparently insoluble SYNC accumulate with age in submucosal neurons (and terminals presumably originating from myenteric neurons), myenteric neurons, and preganglionic neurons (e.g., the swollen varicosities or terminals apparently contacting myenteric neurons). Similarly, and of particular clinical significance, Annerino and colleagues⁴⁰ have observed myenteric and submucosal plexus SYNC+ Lewy pathologies in a population of patients that had reached the frank motoric stages of Parkinson's disease. Should such synucleopathies observed in the preganglionic terminals eventually be transported centripetally from the SMP to the myenteric plexus and then to the somata of the neurons in the spinal cord and brain^41,42, the misfolded proteins will have reached the CNS and be situated to spread retrogradely within the CNS⁴³.

It should be stressed that our survey of SYNC protein in the SMP was done with tissue perfused and fixed overnight in Zamboni's fixative. Further, it should be noted that controversies exist as to how to separate soluble from insoluble SYNC efficiently and effectively⁴⁴ and as to what criteria to use for Lewy bodies in the gastrointestinal tract^45-47. Complete assessments of the accumulations of SYNC material we observed will require additional assays (phospho-SYNC antibodies; biochemical assays, etc.)^7,8,9,47. It is also worth noting that, had we used one of the digestion assays commonly employed as an antigen-retrieval step for well-fixed or paraffin-embedded protein antigens (e.g., formic acid; proteinase K pre-treatment)⁴⁴, we might have detected more SYNC protein. Conversely, with an aggressive digestion, we might have solublized some of the deposited SYNC protein. In recognition of the complications of interpreting the impact of different IHC protocols as well as in the controversies from using different criteria for Lewy pathologies, we have avoided the use of the Lewy body terminology. Though different analyses based on different protocols and different species will need to establish appropriate within-experiment baselines, the present experiment does suggest that SYNC expression and aggregates occur even in presumably healthy submucosal ganglia, accumulate in the plexus with age, and need to be considered when interpreting SYNC+ specimens.

Thus, the present results, in addition to their implications for the discussion of neural pathways by which SYNC and other misfolded proteins might spread centripetally from the gut into the CNS⁴⁸, are also relevant to the issue of using SYNC in the SMP from colonic biopsies to diagnose age-related problems or neurodegenerative disorders. To the extent that one can generalize from normal aging of the autonomic nervous system in a rat previously validated as a model of aging to the patterns of aging in the human enteric nervous system, it is evident that there is (a) extensive SYNC expression in the SMP of the colon throughout the lifespan and (b) a substantial incidence of aggregated synuclein or synucleinopathies in the aging SMP. Such a base rate of SYNC expression will clearly have to be gauged and subtracted from the SYNC+ staining patterns in biopsy material, if false positive diagnoses are to be avoided.