Distinct Patterns of Antiamyloid-β Antibodies in Typical and Atypical Alzheimer Disease | Dementia and Cognitive Impairment | JAMA Neurology | JAMA Network
[Skip to Navigation]
Figure 1. Amyloid load is similar in patients with typical Alzheimer disease (AD) and patients with posterior cortical atrophy with AD (PCA-AD). Scatterplots show the radiolabeled carbon 11–Pittsburgh Compound B (PiB)-standard uptake value ratio global index in normal control individuals, patients with AD, and patients with PCA-AD.

Figure 1. Amyloid load is similar in patients with typical Alzheimer disease (AD) and patients with posterior cortical atrophy with AD (PCA-AD). Scatterplots show the radiolabeled carbon 11–Pittsburgh Compound B (PiB)-standard uptake value ratio global index in normal control individuals, patients with AD, and patients with PCA-AD.

Figure 2. Serum total antiamyloid-β antibodies in patients with typical Alzheimer disease (AD), and patients with atypical posterior cortical atrophy and AD (PCA-AD) and control individuals. A, Enzyme-linked immunosorbent assay relative optical densities (ODs) of IgM and IgG antibodies. B, Enzyme-linked immunosorbent assay relative ODs of IgG1, IgG2, IgG3, and IgG4 antibodies.

Figure 2. Serum total antiamyloid-β antibodies in patients with typical Alzheimer disease (AD), and patients with atypical posterior cortical atrophy and AD (PCA-AD) and control individuals. A, Enzyme-linked immunosorbent assay relative optical densities (ODs) of IgM and IgG antibodies. B, Enzyme-linked immunosorbent assay relative ODs of IgG1, IgG2, IgG3, and IgG4 antibodies.

Table. Demographic and Clinical Data of Studied Groups
Table. Demographic and Clinical Data of Studied Groups
Original Contribution
Sep 2012

Distinct Patterns of Antiamyloid-β Antibodies in Typical and Atypical Alzheimer Disease

Author Affiliations

Author Affiliations: INSERM, UMRS 938, St-Antoine Hospital, Paris, (Drs Dorothée, Moukari, and Aucouturier); UPMC Univ-Paris6, Paris (Drs Dorothée, Moukari, and Aucouturier); CEA, DSV, I2BM, Service Hospitalier Frédéric Joliot, Orsay, (Drs Bottlaender and Maroy); CEA, DSV, I2BM, NeuroSpin, F-91191 Gif-sur-Yvette Cedex (Dr Bottlaender); CRICM, UPMC Univ Paris 6, Pitié-Salpêtrière Hospital, Paris (Drs de Souza, Corlier, Colliot, Chupin, Lehéricy, Dubois, and Sarazin); Alzheimer Institute, Research and Resource Memory Centre; and Centre de Référence de Démences Rares, Centre de référence maladie d’Alzheimer jeune, AP-HP, Pitié-Salpêtrière Hospital, Paris (Drs de Souza, Corlier, Dubois, and Sarazin); CNRS, UMR 7225, Pitié-Salpêtrière Hospital,Paris (Drs Colliot, Chupin, and Lehéricy); Department of Metabolic Biochemistry, AP-HP, Pitié-Salpêtrière Hospital, Paris (Dr Lamari); and Department of Neuroradiology, Center for Neuroimaging Research–CENIR, Pitié-Salpêtrière Hospital, Paris (Dr Lehéricy), France.

Arch Neurol. 2012;69(9):1181-1185. doi:10.1001/archneurol.2012.604

Objective To compare serum antiamyloid-β (Aβ) antibodies in typical and atypical Alzheimer disease (AD).

Design Preliminary observations.

Subjects Thirteen patients with AD, 8 patients with posterior cortical atrophy with evidence of AD (PCA-AD) pathophysiological process by both cerebrospinal fluid (CSF) biomarkers and amyloid imaging, and 12 age-matched control individuals.

Interventions The class and subclass levels of serum anti-Aβ antibodies were measured using an oligomer-based enzyme-linked immunosorbent assay. This method allowed measuring both free antibodies and, after acidic treatment, the total fraction that includes all antibodies complexed with circulating Aβ40/42 and any cross-reacting antigen.

Results Anti-Aβ IgG were restricted to the IgG1 and IgG3 subclasses. Their total levels were strikingly lower and more homogeneous in patients with PCA compared with both typical AD and controls, while biomarkers of amyloid deposition (CSF Aβ42 and positron emission tomography amyloid imaging) were similar in patients with AD and patients with PCA.

Conclusions Serum anti-Aβ IgG1 and IgG3 antibodies differ between distinct forms of AD. Its significance is discussed for possible implications as immune effectors in the specific pathophysiology of AD variants.

Brain amyloid-β peptide (Aβ) deposition is a major feature of Alzheimer disease (AD). IgG antibodies directed to Aβ naturally occur in elderly persons, and their serum and cerebrospinal fluid (CSF) levels may be significantly altered in the course of AD.1-4 While anti-Aβ antibodies may help to control the development of amyloid plaques,2 the clinical significance of their serum levels remains unclear.

We, and others, recently showed that posterior cortical atrophy (PCA) may be defined as an atypical focal form of AD (PCA-AD).5-7 While AD and PCA-AD display clearly distinct clinical presentations, they are similar for biomarkers of amyloid deposition, such as CSF Aβ42 levels and Pittsburgh Compound B (PiB) binding patterns.5-7 In the present study, we compared the pattern of serum anti-Aβ antibodies in typical AD and PCA-AD.


Eight patients with PCA-AD were enrolled according to the following diagnostic criteria: (1) gradual progression of cognitive impairment beginning with visual complaints; (2) presentation with visuospatial deficits with intact primary visual function; (3) features suggestive of Balint syndrome associated or not with Gerstmann syndrome; (4) proportionally less episodic memory impairment; (5) relatively preserved insight; (6) no parkinsonian signs; (7) glucose hypometabolism on fluorodeoxyglucose F18–positron emission tomographic (PET) examination; (8) cortical atrophy on magnetic resonance imaging with a predominance in the posterior cortical region; (9) an AD profile of CSF biomarkers (see the “CSF Biomarker Analysis” subsection); and (10) a global cortical index of radiolabeled carbon 11 (11C)–PiB uptake on PiB-PET > 1.6 (see Table note b for details).8

Thirteen patients with typical AD (Clinical Dementia Rating [CDR] ≥0.5) matched to patients with PCA-AD for age, disease duration, and Mini-Mental State Examination (MMSE) score were selected according to the New Research Criteria.7 Twelve healthy elderly controls were recruited according to the following criteria: (1) MMSE score ≥27/30 and CDR = 0; (2) no history of neurological or psychiatric disorders; and (3) no memory complaint or cognitive deficit.

Standard protocol approvals, registrations, and patient consents

The study was conducted by the Institut National de la Santé Et et la Recherche Médicale (INSERM; grant ANR-07-LVIE-002-01) and was approved by the Ethics Committee of Pitié-Salpêtrière Hospital. All the subjects provided written informed consent before participating. The controls underwent the same procedure as did the patients with AD and PCA-AD except for lumbar puncture, which was not proposed for ethical reasons.


PET imaging with 11C-labeled PiB was performed in all the subjects except 4 controls. The method was the same as previously described5 and summarized in table legend c.

Csf biomarker analysis

Levels of total Tau (T-Tau), T181-phosphorylated Tau (P-Tau), and Aβ42 were measured by enzyme-linked immunosorbent assay (ELISA) according to the manufacturer's (Innogenetics) instructions in CSF samples from all the patients except 3 patients with AD. We calculated derived ratios from single biomarkers, including T-Tau/Aβ42 and P-Tau/Aβ42. We also calculated the Aβ:Tau ratio, defined by the formula Aβ42/[240 + (1.18 × T-tau)], where a score below 0.8 is suggestive of AD.8


Serum samples were diluted at 1:50, and levels of anti-Aβ were analyzed by indirect ELISA using oligomeric Aβ-coated plates, with and without prior dissociation of immune complexes at pH 3.5, as described by Britschgi et al.3 IgG, IgM, and IgG subclasses were revealed using specific anti-γ and anti-μ secondary antibodies (Jackson ImmunoResearch) and monoclonal antibodies NL16, GOM2, ZG4, and RJ4 (University of Birmingham). Anti-Aβ monoclonal antibody BAM-10 diluted to 0.5 μg/mL was used as an internal reference in each ELISA plate, and results were expressed as ratios of the samples' optical densities to that yielded by BAM-10.


Serum IgG and IgM levels were measured by the immunoturbidimetric method (Abbott Labs). Serum IgG subclass levels were measured using a competitive ELISA with specific monoclonal antibodies.9

Statistical analyses

All the comparisons of antibody levels were performed using the Mann-Whitney nonparametric test for comparison of means between 2 groups and the Wilcoxon test (GraphPad Prism Software, Inc) for paired AD and PCA-AD cases.


No statistical difference was observed between the AD and PCA-AD groups concerning Aβ42, T-Tau, P-Tau levels, and Aβ:Tau ratios. The mean PiB indices were identical in the PCA-AD and AD groups concerning all the analyzed regions of interest (Figure 1 and Table).


Acidic pretreatment of serum samples allowed the evaluation of the total fractions of anti-Aβ antibodies after dissociation of immune complexes, while analyses of untreated sera measured the free antibody fractions. As shown in Figure 2, total serum anti-Aβ antibodies essentially belonged to the IgM class and IgG1 and IgG3 subclasses. This total fraction may include all antibodies that are initially complexed with circulating Aβ40/42 as well as any other cross-reacting antigen. While IgM antibody levels were quite similar in the 3 groups, patients with PCA-AD displayed significantly lower and more homogeneous concentrations of IgG antibodies than patients with AD and healthy controls (Figure 2A), which was most evident for IgG1 (Figure 2B). Comparison of typical AD and PCA-AD cases paired for age, disease duration, and CDR confirmed that those with PCA-AD have lower IgG1 (P = .008) and IgG3 (P = .023) circulating anti-Aβ antibodies.

Overall serum IgG, IgM, and IgG subclass levels were similar in the PCA-AD and AD groups ( eFigure 1), which ruled out possible immunoglobulin deficiencies. Comparisons of ratios between anti-Aβ antibodies and corresponding total immunoglobulin levels for each class and subclass confirmed that anti-Aβ IgG1 and IgG3 were strongly and specifically lower in patients with PCA-AD than in patients with AD (P < .001 for both IgG1 and IgG3, eFigure 2).

Anti-Aβ antibodies in both groups of samples—total and free anti-Aβ antibodies—were found only for isotypes IgM, IgG1 and IgG3. No significant difference was found between the free antibody fractions of controls, patients with PCA-AD and patients with AD, except for free anti-Aβ IgG1, which were lower in patients with PCA-AD than in patients with AD (P = .03, data not shown).

Concerning all groups, no significant correlation was found between antibody levels and any of the tested clinical data (age, MMSE, disease duration), levels of CSF biomarkers and PiB global index. No correlation was observed between antibody levels and global PiB index nor CSF biomarkers when we pooled patients with AD and patients with PCA-AD.


This preliminary study explored anti-Aβ antibodies in carefully identified populations of patients with typical AD and patients with PCA-AD, the latter being defined as PCA syndrome associated with clear biological and imaging evidence of AD pathophysiological process. The total amount of serum anti-Aβ IgG antibodies, especially those of the IgG1 subclass, was strikingly lower in patients with PCA-AD compared with patients with typical AD and aged controls.

Serum anti-Aβ antibodies occur naturally in elderly individuals, and their role in AD remains unclear. Recent studies suggested that serum anti-Aβ antibodies may have beneficial effects on amyloid pathology and neuron toxicity.2,3 In typical patients with AD, previous results showed a striking heterogeneity of serum antibody levels, which may be decreased,3 increased10 or unchanged11 compared with healthy controls. Such divergent results may relate to a diversity of methods used for anti-Aβ antibody assessment. A 2010 report by Storace et al4 showed that the total fraction of serum anti-Aβ antibodies was higher in patients with mild cognitive impairment who progressed to AD than stable cases, suggesting that this blood marker is associated with AD progression. Using a similar method, we observed a clear difference in serum total anti-Aβ IgG antibody levels between patients with AD and patients with PCA-AD, while their amyloid pathological markers were similar.

Thus, differences in serum anti-Aβ antibodies may hardly be explained by mere adsorption onto amyloid deposits. However, this difference may relate to differential intracerebral reactivity with molecular species that do not fix PiB, such as Aβ monomers or soluble oligomers.

Intergroup differences were evident for IgG1 and IgG3 subclasses, which were lower and more homogeneous in the patients with PCA-AD. IgG1 and IgG3 have the unique ability to bind FcγR-I receptors on monocytes/macrophages, which are actively recruited to the brain parenchyma in the course of AD.12 Thus, lower antibody levels in patients with PCA-AD might relate to more efficient recruitment of anti-Aβ-bearing monocytes to the central nervous system. On the other hand, lower antibody levels in patients with PCA-AD may result from weaker activation of anti-Aβ immune responses.

IgG1 and IgG3 are the most potent IgG subclasses in activating the complement classical pathway through C1q binding. Because complement fragments and receptors, including C1q and CR1, are involved in neuroprotection, neuroinflammation, or both, different rates of IgG1 and IgG3 antibody transport to the brain might contribute to pathological differences between focal and classical forms of AD.13

Because PCA seems to remain a focal posterior disease even at the late stage of evolution,14 it provides a model for studying in vivo factors that might influence AD progression and the interactions between amyloid and Tau pathology. Our results point to new targets for investigating this crucial field of research by showing different implications of anti-Aβ immune effectors between typical AD and PCA-AD.

In conclusion, our observations support the idea that systemic immunological responses, especially the anti-Aβ IgG1 and IgG3 antibodies, could contribute to shaping clinical presentations of AD. The above hypotheses require studies on a larger series of patients and further investigations into the specific pathophysiology of typical and atypical forms of AD, with particular focus on immunological features. Such studies would allow better understanding of the pathophysiology of the disease and could lead to the development of innovative immunotherapy approaches in AD.

Back to top
Article Information

Correspondence: Marie Sarazin, MD, PhD, Department of Neurology, Alzheimer Institute and Reseach and Resource Memory Centre, Pavilion J. Lhermitte, Pitié-Salpêtrière Hospital, 47 bd de l’hôpital, 75013 Paris, France (marie.sarazin@psl.aphp.fr) or Pierre Aucouturier, PhD, INSERM, UMRS 938, St-Antoine Hospital, 184 rue du Faubourg St-Antoine, 75012 Paris, France (pierre.aucouturier@inserm.fr).

Accepted for Publication: March 7, 2012.

Published Online: June 18, 2012. doi:10.1001/archneurol.2012.604

Author Contributions: Drs Dorothée, Moukari, de Souza, Maroy, Dubois, Sarazin, and Aucouturier had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Drs Sarazin and Aucouturier contributed equally to the manuscript. Study concept and design: Dorothée, Bottlaender, Maroy, Dubois, Sarazin, and Aucouturier. Acquisition of data: Bottlaender, Moukari, de Souza, Corlier, and Sarazin. Analysis and interpretation of data: Dorothée, Bottlaender, Moukari, de Souza, Maroy, Colliot, Chupin, Lamari, Lehéricy, Dubois, Sarazin, and Aucouturier. Drafting of the manuscript: Bottlaender, Moukari, Maroy, Dubois, Sarazin, and Aucouturier. Critical revision of the manuscript for important intellectual content: Dorothée, de Souza, Maroy, Corlier, Colliot, Chupin, Lamari, Lehéricy, Dubois, Sarazin, and Aucouturier. Statistical analysis: Moukari, de Souza, Maroy, Colliot, Lamari, Dubois, and Aucouturier. Obtained funding: Sarazin. Administrative, technical, and material support: Maroy, Corlier, Chupin, Dubois, Sarazin, and Aucouturier. Study supervision: Dorothée, Maroy, Dubois, Sarazin, and Aucouturier.

Financial Disclosure: None reported.

Funding/Support: This study was supported in part by the Institut National de la Santé et de la Recherche Médicale (INSERM), Agence Nationale de la Recherche (ANR, grant ANR-07-LVIE-002-01), International Foundation of Research on Alzheimer's Disease, Fondation Nationale de Gérontologie, and MEDIPAR.

Role of the Sponsor: No funders had a role in the design and conduct of the study; in the collection, analysis, and interpretation of the data; or in the preparation, review, or approval of the manuscript.

Additional Contributions: Gisèle Jollivet provided technical assistance in IgG subclass measurements.

Du Y, Dodel R, Hampel H,  et al.  Reduced levels of amyloid beta-peptide antibody in Alzheimer disease.  Neurology. 2001;57(5):801-80511552007PubMedGoogle ScholarCrossref
Kellner A, Matschke J, Bernreuther C, Moch H, Ferrer I, Glatzel M. Autoantibodies against beta-amyloid are common in Alzheimer's disease and help control plaque burden.  Ann Neurol. 2009;65(1):24-3119194878PubMedGoogle Scholar
Britschgi M, Olin CE, Johns HT,  et al.  Neuroprotective natural antibodies to assemblies of amyloidogenic peptides decrease with normal aging and advancing Alzheimer's disease.  Proc Natl Acad Sci U S A. 2009;106(29):12145-1215019581601PubMedGoogle Scholar
Storace D, Cammarata S, Borghi R,  et al.  Elevation of β-amyloid 1-42 autoantibodies in the blood of amnestic patients with mild cognitive impairment.  Arch Neurol. 2010;67(7):867-87220625095PubMedGoogle Scholar
de Souza LC, Corlier F, Habert MO,  et al.  Similar amyloid-β burden in posterior cortical atrophy and Alzheimer's disease.  Brain. 2011;134(pt 7):2036-204321705422PubMedGoogle Scholar
Rosenbloom MH, Alkalay A, Agarwal N,  et al.  Distinct clinical and metabolic deficits in PCA and AD are not related to amyloid distribution.  Neurology. 2011;76(21):1789-179621525424PubMedGoogle Scholar
Dubois B, Feldman HH, Jacova C,  et al.  Revising the definition of Alzheimer's disease: a new lexicon.  Lancet Neurol. 2010;9(11):1118-112720934914PubMedGoogle Scholar
Koric L, Felician O, Ceccaldi M. Use of CSF biomarkers in the diagnosis of Alzheimer's disease in clinical practice [in French].  Rev Neurol (Paris). 2011;167(6-7):474-48421420704PubMedGoogle Scholar
Aucouturier P, Mounir S, Preud’homme JL. Distribution of IgG subclass levels in normal adult sera as determined by a competitive enzyme immunoassay using monoclonal antibodies.  Diagn Immunol. 1985;3(4):191-1964085174PubMedGoogle Scholar
Nath A, Hall E, Tuzova M,  et al.  Autoantibodies to amyloid beta-peptide (Aβ) are increased in Alzheimer's disease patients and Aβ antibodies can enhance Aβ neurotoxicity: implications for disease pathogenesis and vaccine development.  Neuromolecular Med. 2003;3(1):29-3912665674PubMedGoogle Scholar
Hyman BT, Smith C, Buldyrev I,  et al.  Autoantibodies to amyloid-beta and Alzheimer's disease.  Ann Neurol. 2001;49(6):808-81011409436PubMedGoogle Scholar
Rezai-Zadeh K, Gate D, Gowing G, Town T. How to get from here to there: macrophage recruitment in Alzheimer's disease.  Curr Alzheimer Res. 2011;8(2):156-16321345166PubMedGoogle Scholar
Veerhuis R, Nielsen HM, Tenner AJ. Complement in the brain.  Mol Immunol. 2011;48(14):1592-160321546088PubMedGoogle Scholar
Kas A, de Souza LC, Samri D,  et al.  Neural correlates of cognitive impairment in posterior cortical atrophy.  Brain. 2011;134(pt 5):1464-147821478188PubMedGoogle Scholar