Immunohistological detection of Chlamydia pneumoniae in the Alzheimer's disease brain
© Hammond et al. 2010
Received: 1 April 2010
Accepted: 23 September 2010
Published: 23 September 2010
Skip to main content
© Hammond et al. 2010
Received: 1 April 2010
Accepted: 23 September 2010
Published: 23 September 2010
Sporadic late-onset Alzheimer's disease (AD) appears to evolve from an interplay between genetic and environmental factors. One environmental factor that continues to be of great interest is that of Chlamydia pneumoniae infection and its association with late-onset disease. Detection of this organism in clinical and autopsy samples has proved challenging using a variety of molecular and histological techniques. Our current investigation utilized immunohistochemistry with a battery of commercially available anti-C. pneumoniae antibodies to determine whether C. pneumoniae was present in areas typically associated with AD neuropathology from 5 AD and 5 non-AD control brains.
Immunoreactivity for C. pneumoniae antigens was observed both intracellularly in neurons, neuroglia, endothelial cells, and peri-endothelial cells, and extracellularly in the frontal and temporal cortices of the AD brain with multiple C. pneumoniae-specific antibodies. This immunoreactivity was seen in regions of amyloid deposition as revealed by immunolabeling with two different anti-beta amyloid antibodies. Thioflavin S staining, overlaid with C. pneumoniae immunolabeling, demonstrated no direct co-localization of the organism and amyloid plaques. Further, the specificity of C. pneumoniae labeling of AD brain sections was demonstrated using C. pneumoniae antibodies pre-absorbed against amyloid β 1-40 and 1-42 peptides.
Anti-C. pneumoniae antibodies, obtained commercially, identified both typical intracellular and atypical extracellular C. pneumoniae antigens in frontal and temporal cortices of the AD brain. C. pneumoniae, amyloid deposits, and neurofibrillary tangles were present in the same regions of the brain in apposition to one another. Although additional studies are required to conclusively characterize the nature of Chlamydial immunoreactivity in the AD brain, these results further implicate C. pneumoniae infection with the pathogenesis of Alzheimer's disease.
Alzheimer's disease (AD) is a progressive neurological disease that affects millions of older individuals. Distinctive pathological hallmarks associated with the disease include tau accumulations forming neuropil threads (NTs) and neurofibrillary tangles (NFTs), and deposits of extracellular amyloid comprising neuritic senile plaques (NSPs). In general, there are two distinct forms of Alzheimer's disease, familial AD and sporadic late-onset AD. The early onset form of the disease, known commonly as familial AD, is caused by dysregulation of many processes due to genetic mutations that lead to the aforementioned pathology. For example, mutations in presenilins 1 and 2 genes and the gene responsible for amyloid-β protein precursor (AβPP) result in an increased accumulation of beta-amyloid (Aβ) in the brain. In late-onset sporadic AD, similar pathological accumulations occur, although without the gene mutations seen in familial AD (for review see Duyckaerts 2009) .
Most investigations have focused on the extracellular, deposited forms of amyloid in the AD brain. The extracellular accumulations of amyloid in the brain are composed principally of amyloid β 1-40 and 1-42 and form neuritic senile plaques (NSP) [1, 2]. However, intracellular accumulations of amyloid, which may occur prior to extracellular deposition, also have been demonstrated [3–6]. Further, Aβ-derived diffusible ligands (ADDLs), soluble forms of amyloid, have been postulated to be a toxic form of amyloid at the synapses and are not found in typical neuropathological or histopathologic accumulations of amyloid . Since late-onset AD lacks the same mutations seen in familial AD, determination of the cause of amyloid pathology in late-onset AD remains poorly understood.
The interplay between normal processes and environmental factors, both independently and in concert with other genetic factors, may lead to late-onset AD. In particular, infections as environmental factors may have an impact on the delicate amyloid and tau balance in the brain and lead to the pathology seen in AD. A number of infectious agents have been associated with late-onset AD [8–12]. Our focus has been on the obligate, intracellular bacterium Chlamydia pneumoniae, which has been demonstrated to be highly prevalent in the AD brain [8, 12], as well as associated with other systemic and neurological diseases [13, 14]including atherosclerosis [15, 16], stroke , encephalitis , and multiple sclerosis .
Various cell types found in the brain have been shown to be susceptible to infection by C. pneumoniae including endothelia, astroglia, microglia, and neurons [8, 12, 20–23]. Once inside the cell, C. pneumoniae reside in an intracellular inclusion that resists lysosomal fusion and immune recognition. C. pneumoniae developmentally cycle from the infectious elementary body (EBs) to the metabolically active reticulate body (RBs), which divide by binary fission. This obligate intracellular pathogen both interacts with and manipulates the host by gathering energy and nutrients that are required for replication, such as sphingomyelin and cholesterol [24–26]. Chlamydiae also inhibit apoptosis [20, 27–29] and release factors such as chlamydial lipopolysaccharide (LPS) and glycolipid protein into both the cell itself and into the surrounding milieu . Further propagation and spread of the organism may follow one or more pathways. The infectious progeny, EBs, may be released upon eventual cell death or by extrusion from the cell in a membrane bound package into the surrounding environment . Thus, C. pneumoniae and/or antigens derived from the organism may be localized both intracellularly and extracellularly at the site of infection.
Due to the chronic nature of AD and the complexity of C. pneumoniae infections, establishing an association with disease pathogenesis has proved difficult. Validating this association relies on a variety of detection methods for the organism. Our current study focuses on the use of immunohistochemistry (IHC) with a battery of commercially available anti-chlamydia antibodies on frontal and temporal cortical sections of human AD brains. Our data suggest this methodology provides a valuable insight into the interrelationship between infection and AD pathology.
Commercially available Chlamydia antibodies
BioDesign, Meridian Life Sciences
BioDesign, Meridian Life Sciences
BioDesign, Meridian Life Sciences
Intracellular and extracellular C. pneumoniae immunoreactivity was observed in the entorhinal cortex, the hippocampal formation, and the frontal cortex, regions of the brain that typically demonstrate AD pathology. Clear discrimination between chlamydia immunoreactivity and age-related lipofuscin accumulation within neurons was demonstrated. Serial sections of brain tissue displayed both amyloid pathology and the presence of C. pneumoniae immunoreactivity. Thioflavin S staining for fibrillar amyloid and specific antibody labeling for C. pneumoniae revealed deposition of both when performed on the same section. As some C. pneumoniae labeling was extracellular, a more atypical pattern, pre-absorption studies with Amyloid β 1-40 and 1-42 were performed. These studies revealed that C. pneumoniae antibodies were not cross-reacting with Aβ. Collectively, these data demonstrate that evidence of C. pneumoniae infection is present in brain tissues in areas of amyloid pathology, thereby suggesting that an interrelationship exists between these entities in the pathogenesis of sporadic late-onset AD.
Immunolabeling for Chlamydia may be overlooked in brain tissues, as it is different from what is observed in cellular infections in vitro. In all AD samples analyzed in this study, both typical intracellular perinuclear chlamydia immunoreactivity and atypical extracellular labeling were observed. Intracellular labeling demonstrated punctate elementary bodies and membrane bound inclusions similar to that of in vitro studies (see Figure 2E). This specific labeling was differentiated from lipofuscin by using red chromogens, either alkaline phosphatase (AP)red or AP magenta, as the substrate to denote C. pneumoniae immunoreactivity. Horseradish peroxidase labeling with 3, 3′-Diaminobenzidine (DAB), a brown chromogen, was not used as this labeling may be confused with the golden brown lipofuscin found in neurons of aged brains.
Two distinct extracellular patterns of chlamydia immunoreactivity were observed: one, a punctate pattern signifying the elementary body form of the bacteria, which can be extruded from infected cells into the surrounding milieu , and two, an amorphous foci pattern most likely indicating secreted chlamydial factors such as lipopolysaccharide [30, 31]. These patterns in the cerebrum will require further study although similar profiles of Chlamydia labeling in situ have been demonstrated in a different organ . Furthermore, our data demonstrated that C. pneumoniae extracellular immunoreactivity was not reflective of cross-reactivity with extracellular Amyloid β 1-40 or 1-42. However, C. pneumoniae extracellular organism and related antigens may interact with extracellular proteins and lipids in the brain. Although not always in direct overlap with amyloid plaque deposits, chlamydial antigens may interact with soluble oligomeric forms of amyloid, such as ADDLs, that are less likely to be found in mature plaques due to their soluble nature . These intriguing findings and their implications require further understanding of the possible relationship between amyloid and chlamydia in the same cortical regions of the brain. This relationship will vary with each individual AD patient. Each AD patient has different levels of pathology and may have corresponding variability in extent and distribution of C. pneumoniae infection in the cerebrum. Following further studies into this variability, the relationship between pathology and infection can be more thoroughly evaluated.
Although C. pneumoniae is principally a respiratory pathogen, infection of the brain has been shown following intranasal and lung infection [33, 34]. In this regard, monocytes infected with C. pneumoniae in the lungs may spread the infection via the peripheral circulation to the brain through the blood brain barrier or circumventricular organs [22, 23]. Alternatively, a more direct and insidious route of infection may follow the olfactory pathways. As such, the infection becomes established in the olfactory nasal neuroepithelia, progresses to the olfactory bulbs, and eventually infects brain structures such as the entorhinal cortex and hippocampus. The olfactory structures, the entorhinal cortex, and the hippocampal formation are the most vulnerable and the earliest regions affected in the onset of AD [35, 36]. Our current study highlights C. pneumoniae detection in the frontal and temporal cortices, including the entorhinal cortex and the hippocampal formation. Thus, infection of these regions in the brain may have great impact on the development of AD pathology.
Previous studies have demonstrated C. pneumoniae in both human and animal olfactory bulbs [8, 33, 34]. In both cell culture and animal studies, C. pneumoniae has been shown to infect nasal neuroepithelial cells . In the animal studies, infection appeared to spread centrifugally from the vulnerable neuronal cells in the olfactory bulb into the brain [33, 34, 37]. Further, Chlamydia isolated from AD brains in a prior study was shown to have more commonalities with Chlamydia respiratory strains than with Chlamydia strains from atherosclerosis with the suggestion that the organism itself may have a tropism for specific cell types in the CNS . Upon consideration of these data and our current data, a rationale for the selective vulnerability of specific brain regions to infection and resultant pathology emerges.
Notably, histopathological amyloid plaques and tangles are used to define the stage of AD, but the correlation with the pathology of the disease and the clinical manifestations of the disease are not always clear . Some individuals who have massive pathology have little to no symptoms. On the other hand, some symptomatic individuals may show little pathology upon post-mortem histopathological examination. As such, there are many variations in the amount and type of damage evident in AD . The variability of correlation between the symptomology and histopathology suggests other events and/or ingredients may be missing in the pathobiology of AD.
The response in the brain to infection may determine the extent of pathology and symptomatology that may arise. In this regard, C. pneumoniae infection characteristically promotes an inflammatory response whereby cytokines such as IL-1β and TNF-α are secreted and may initiate cellular damage . These cytokine responses to infection parallel similar responses to amyloid accumulation . Additionally, another cellular response to C. pneumoniae infection in culture is the production and processing of amyloid peptides. Labeling of infected cells in culture for Amyloid β 1-42 often reveals intracellular immunoreactivity at early post-infection times (unpublished observations CJH, DMA, CSL, BJB). Interestingly, our earlier study demonstrated intracellular and extracellular amyloid deposits in the brains of non-transgenic BALB/c mice following intranasal C. pneumoniae infection . Our current study showed a similar relationship in which we demonstrated both amyloid and chlamydia immunoreactivity in the temporal cortex of the AD brain.
Further evaluation is required to specifically demonstrate how amyloid and C. pneumoniae, both intracellular and extracellular, are interrelated. In addition, future studies are required to further characterize the atypical extracellular Chlamydial immunolabeling profiles. Others have demonstrated evidence for an association of infection and amyloid in AD . HSV1 viral DNA was shown to specifically associate with AD amyloid plaques . Previous hypotheses have even suggested that amyloid in the AD brain may act as an antiviral agent  or an entrapping agent for infection . Intriguingly, a recent study suggests that amyloid has anti-microbial properties, and may arise in response to brain infection in AD . As 2 of 5 non-AD cases in our current study showed occasional chlamydia immunoreactivity and diffuse amyloid deposition, future analysis must also include mild cognitive impairment cases, as well as non-AD cases, as infection may be a prodromal event leading to eventual AD pathology.
The exact consequences of infection that correlate best with the stage and subtype of Alzheimer's disease require ongoing investigation. This is true especially with regard to the hallmarks of the disease such as amyloid plaques and tau tangles. Alzheimer's disease is manifest with multifactorial aspects of pathology and with potentially multiple associated environmental factors, including infectious agents [8, 10–12]. Further, our Chlamydia pneumoniae data suggest that the areas of the brain involved with olfaction are important to study as infection in these regions, as well as in the olfactory neuroepithelia and olfactory bulbs, may be a precursor to the pathology associated with AD. Identification and localization of infectious agents, such as Chlamydia pneumoniae, in specifically vulnerable areas and cell types in the brain is paramount. This determination may be a missing link in the current strategy of associating symptomatology to disease pathogenesis in sporadic late-onset Alzheimer's disease.
Post-mortem tissue samples from various brain regions (frontal and temporal cortices) of patients with and without AD were obtained through Dr. William Hill of the Medical College of Georgia (Augusta, GA), from the MCP-Hahnemann School of Medicine Department of Pathology, currently Drexel University College of Medicine (Philadelphia, PA), and the Alzheimer's Research Center of the Health Partners Research Foundation at Regions Hospital (St Paul, MN) under approved protocols at each procurement site. Five AD and 5 non-AD age-matched male and female archival brains were examined. All cases were confirmed as AD or non-AD by neuropathological examination at the source using standard diagnostic criteria (NINDS/CERAD) .
To detect chlamydia in the brain, IHC was performed on serial sections from formalin-fixed, paraffin-embedded Alzheimer brain tissue. Parallel IHC was performed using antibodies for AD amyloid plaque pathology, specifically anti-amyloid 1-42 antibodies. Amyloid plaques were best visualized when antigen retrieval was not performed whereas the optimal labeling for chlamydia antigens was achieved following antigen retrieval. Optimum immunoreactivity occurred at 37°C in a humidified chamber for both amyloid and chlamydial antigens. To visualize amyloid deposits, 3, 3' Diaminobenzidine (DAB) (golden brown color) was used. Red chromogens such as alkaline phosphatase (AP) red and AP new magenta were found to best visualize both intracellular and extracellular C. pneumoniae, as nothing in a typical brain section should appear red. For counterstaining, Mayer's hematoxylin was chosen as it only stains the nucleus leaving the cytoplasm of the cell clear allowing easier visualization of intracellular chlamydia EB and inclusion immunolabeling. Controls were used to rule out non-specific labeling. Our controls included: no primary or secondary antibodies, primary with no corresponding secondaries, no primary but HRP conjugated secondary and/or AP conjugated secondary alone and in combination, and anti-human IgG primary with appropriate HRP and/or AP secondaries alone and in combination, all of which were reacted with appropriate substrates DAB and/or AP red and/or AP magenta, alone and in combination. Controls for fluorescent work also included no primary, secondary, or stain to allow for evaluation of autofluorescence of the brain tissues.
Archival, paraffin-embedded, human brain tissues were deparaffinized through xylenes and graded alcohols. Antigen retrieval was performed in The Retriever (Electron Microscopy Science, Fort Washington, PA) according to manufacturer's directions using 1 × Citra Antigen retrieval buffer (BioGenex, San Ramen, CA). The sections were rinsed with filtered water 3 × 5 min and treated with Alkaline Phosphatase/Horseradish Peroxidase Block (BioFX Laboratories, Owings Mills, MD) for 30 min at room temperature (RT). The tissues were rinsed with filtered water, PBS 3 × 5 min, and blocked in 1% Fetal Bovine Serum (FBS)/PBS for 5 min at RT. Sections were incubated in a humidified chamber with anti-chlamydia primary antibodies for 90 min at 37°C. Anti-chlamydial antibodies, genus and species specific, were purchased from commercial sources and used at dilutions shown in Table 1. Following the incubation, the sections were rinsed with PBS 3 × 5 min and blocked with 1% FBS/PBS block for 5 min at RT. The sections were incubated in a humidified chamber with appropriate anti-mouse or anti-rabbit Alkaline Phosphatase conjugated secondary antibody for 60 min at 37°C (BioFX Laboratories, Owings Mills, MD or Zymed- Invitrogen Corporation, Carlsbad, CA). After 3 × 5 min water rinses, the sections were reacted with BioFX AP-New Magenta IHC Substrate (BioFX Laboratories- SurModics, Owings Mills, MD) 40 min at RT or they were reacted with AP red (Zymed- Invitrogen Corporation, Carlsbad, CA) according to the manufacturer's directions. The sections were rinsed with water and counterstained with Mayer's hematoxylin (Electron Microscopy Science, Fort Washington, PA). The slides were aqueous mounted using Crystal Mount (Biomeda, Thermo Fisher Scientific, Pittsburgh, PA) and then permanently mounted using Permount (Thermo Fisher Scientific, Pittsburgh, PA). Sections were viewed on a Nikon Eclipse E800 or a Nikon Eclipse 90i microscope and captured using: the spot RT (Diagnostic Instruments, Starling Heights, MI) with the Image Pro Plus Phase 3 Imaging software (Media Cybernetics, Silver Spring, MD), the Nikon DS-Fi1 camera using the NIS-Elements Advanced Research version 3.0 software (Nikon) or the Nikon DS-Ri1 camera using the NIS-Elements Advanced Research version 3.0 software (Nikon).Depending on the available tissue, the IHC was performed on 2-3 serial sections for each antibody for each section of the brain that was available. Out of the battery of antibodies, 5 monoclonal and 1 polyclonal antibodies were more routinely used (Table 1). In order to be considered a positive sample, the sample must have been positive by IHC with at least 2 C. pneumoniae species-specific monoclonal antibodies in 2 of the 3 serial sections of each brain section. All chlamydia antibodies used on the tissue sections were also used on astrocyte, epithelial, monocyte, and/or neuronal cell lines that were uninfected or that had been infected with AR39 Chlamydia pneumoniae (ATCC, Manassas, VA) as antibody specificity controls.
Archival, paraffin-embedded, human brain tissues were deparaffinized through xylenes and graded alcohols. The sections were rinsed with filtered water and phosphate buffered saline (PBS) 3 × 5 min. Endogenous peroxidase was quenched using 3% H2O2 or Alkaline Phosphatase/Horseradish Peroxidase Block (BioFX Laboratories, Owings Mills, MD) for 30 min at RT. The tissues were rinsed with sterile filtered water, PBS 3 × 5 min, and blocked in 1% (FBS)/PBS for 5 min at RT. The primary antibody, mouse anti-human β-Amyloid (4G8, Signet, Covance, Cambridge, MA) or rabbit anti-Amyloid peptide β cleavage site 42 (Sigma, St Louis, MO), were incubated in a humidified chamber for 90 min at 37°C. Following the incubation, the sections were rinsed with PBS 3 × 5 min and blocked in 1% FBS/PBS block for 5 min at RT. The sections were incubated with anti-rabbit horseradish peroxidase conjugated (HRP) secondary antibody (Zymed, Invitrogen Corporation, Carlsbad, CA) in a humidified chamber for 60 min at 37°C. After a water rinse and PBS washes, 3 × 5 min, the sections were reacted with 3, 3′-Diaminobenzidine (DAB) (Sigma Aldrich, St Louis MI) for 10-20 min at RT. The sections were rinsed with water and counterstained with Mayer's hematoxylin (Electron Microscopy Science, Fort Washington, PA). Slides were dehydrated through graded alcohols and xylenes followed by cover slipping using Permount (Thermo Fisher Scientific, Pittsburgh, PA). Sections were viewed and captured as noted previously.
Individual anti-chlamydia antibodies, mouse anti-C. pneumoniae (Table 1, #2, #3, BioDesign, Meridian Life Science, Inc., Saco, ME) or rabbit anti-chlamydia antibody (Table 1, #1, BioDesign, Meridian Life Science, Inc., Saco, ME), were mixed with >10 M excess of either Amyloid β 1-40 or Amyloid β 1-42 peptide (Sigma, St Louis, MO) in a batch method and allowed to incubate overnight at 4°C. The supernatant of each pre-absorbed antibody was then used for Chlamydia pneumoniae immunohistochemistry as described above with the exception that secondary antibodies were conjugated with HRP and reacted with DAB.
Slides immunolabeled for C. pneumoniae were stained using a modified Thioflavin S protocol http://www-medlib.med.utah.edu/WebPath/webpath.html. In brief, following Chlamydia pneumoniae primary and subsequent secondary incubations and AP-New Magenta substrate development, slides were counterstained with Mayer's hematoxylin and coversliped with aqueous mount. Areas of chlamydia immunoreactivity were identified by light microscopy. The aqueous mount was then soaked off and the section were immersed in 1% Thioflavin S for 5 min, differentiated in 70% alcohol for 5 min, and coverslipped with water. The sections were kept wet by cover slipping with water as the intensity of the Thioflavin S was diminished with use of Crystal mount. Sections were viewed and captured while still wet using a Nikon Eclipse E800 microscope and Nikon DS-Ri1 camera. Light and fluorescent (under FITC filter ) images were captured and merged using the NIS-Elements Advanced Research version 3.0 software (Nikon).
For determination of amyloid and chlamydia labeling and the relationship between the two in the sections, we evaluated the relationship with the following quantification schemes.
For percentage of chlamydia pneumoniae immunopositiyity and percentage of type of immunopositivity, we first qualitatively determined the types of label to be counted: intracellular, extracellular or a combination of the two. In representative frontal and temporal sections, the total numbers of positives across a section were counted and then the numbers of each category tallied and the percentage of the total positive immunoreactivity was determined per total area across the section. Analysis at 4, 10 and 40× magnification was performed manually and utilizing the NIS-Elements Advanced Research version 3.0 software (Nikon). Using the software, the immunoreactivity was selected and the area or cells were counted. A similar quantification scheme was used in determining the relationship of the chlamydia immunoreactivity relative to the amyloid immunoreactivity across a representative temporal region. In this case, we again qualitatively determined the type of immunoreactivity: intracellular, extracellular, or a mixture of the two for amyloid and chlamydia. We then determined the counts and areas of single label and then the label of both in apposition to get an indication of the relative relationships between the two antigens
We are grateful to Dr. William Hill, the Alzheimer's Research Center of the Health Partners Research Foundation at Regions Hospital and the autopsy service of Medical College of Pennsylvania/Hahnemann University (Drexel University) for tissues provided for this work. We are grateful to the families who donated tissues to the several sources listed above for research into Alzheimer's disease. We would like to thank Gwendolyn Harley for all her sectioning help. Finally, we would like to thank all the students past and present that helped with the work including Adia Taylor, and Jesty Abraham.
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.