Tags

Question refers to this news report, https://www.newscientist.com/article/2191814-we-may-finally-know-what-causes-alzheimers-and-how-to-stop-it/

The jury’s still out on

  • Whether one or more chronic infections explain all of Alzheimer’s pathophysiology.
  • Which infection/s are or could be most relevant to Alzheimer’s.

While several epidemiological studies have suggested an association of various bacterial and viral infections with Alzheimer’s (below from 1), most were cross-sectional, hospital-based studies with small numbers which study exposure and outcome at the same time – precisely the kind of studies that lack statistical power and thus preclude firm conclusions, something that requires large, case-control, population-based studies which compare diagnosed cases with controls either looking back at exposure, outcome in hand or looking forward to outcome, exposure in hand, studies that are expensive and long-term.

Meantime billions of dollars have been spent, futilely as it turns out, exploring the amyloid hypothesis in increasingly more complicated genetically engineered mouse models and in testing drug candidates against amyloid plaques. A pronounced and ongoing mismatch between the kind of research actually being funded and the kind of research necessary to establish first principles about its etiology (causation) hinders improved understanding of Alzheimer’s. Important to keep this context in mind while assessing the role of Porphyromonas gingivalis in Alzheimer’s. In short,

  • Porphyromonas gingivalis and periodontitis (gum disease): Association? Yes. Causation? Questionable.
  • There may be some kind of association between P. gingivalis and Alzheimer’s; Causation is as yet a stretch too far.

Porphyromonas gingivalis And Periodontitis (Gum Disease): Association? Yes. Causation? Questionable.

P. gingivalis is known to be associated with periodontitis (gum disease) but it’s not clear it causes it. Along with Tannerella forsythia and Treponema denticola, P. gingivalis (2, 3, 4, 5).

  • Is among the bacteria fairly reliably represented in periodontitis samples.
  • Is routinely identified in sub-gingival plaque.
  • Can be pathogenic in experimental animal-based disease models.
  • Can be grown in culture.

These factors and especially that last convenience helped establish the narrative that imposed a gum disease causative role on P. gingivalis (5) except (6)

  • Its presence doesn’t seem to be essential for development of human periodontal disease.
  • It’s present in gums of a substantial proportion of healthy individuals with no sign of periodontitis.
  • Colonization of germ-free mice with P. gingivalis doesn’t induce periodontitis (7).

In other words, not only a complicated association that’s not yet fully understood (8, 9, 10, below from 4), P. gingivalis also doesn’t pass the smell test of necessary and sufficient as it pertains to a causative role in periodontitis.

There May Be Some Kind Of Association Between P. gingivalis And Alzheimer’s; Causation Is As Yet A Stretch Too Far

Some studies in recent years have suggested a link between P. gingivalis and Alzheimer’s. Individuals with mild cognitive decline or Alzheimer’s were reported to have considerably elevated levels of antibodies against P. gingivalis prior to diagnosis, such an inflammatory response being interpreted as a risk factor for Alzheimer’s (11, 12, 13).

Things prevailed at this observational level until January 2019 when a peer-reviewed published study made the bald and bold claim of having found a causative role for P. gingivalis in Alzheimer’s (14). This study follows the recent trend of generating data from both human samples and mouse models to stitch together a plausible narrative that would likely pass muster with a relatively high profile journal such as Science Advances. Mouse model data alone won’t cut it any more, at least not for ambitious studies such as this one. Even that counts for progress these days.

  • Figures 1, 2 and 3 examined 29 each postmortem tissue microarrays specifically from the middle temporal gyrus (presumably because it’s a known Alzheimer’s degeneration hotspot) from sex- and age-matched dementia-free controls or Alzheimer’s cases for evidence of P.gingivalis presence.
    • They found considerably higher levels of P. gingivalis-specific enzymes called gingipains in Alzheimer’s brain tissues compared to sex- and age-matched control brain tissues.
  • Figure 4 shows data from oral and cerebrospinal fluids from 10 living subjects with probable Alzheimer’s.
    • They report presence of P.gingivalis DNA in these fluids based on quantitative PCR.
    • However, P.gingivalis DNA levels reported in a given patient showed no correlation between oral versus cerebrospinal fluids.
  • In figures 5 and 6, they cultured a human bone marrow-derived neuronal cell line, SH-SY5Y – Wikipedia, with purified P. gingivalis gingipains and showed (along with parallel experiments in mice),
    • Gingipains are toxic to these cells.
    • Small molecule gingipain inhibitors blocked this toxicity (and in brains of treated mice as well, figure 6 C and D).
    • Gingipains could use the neuronal cells’ tau protein as a substrate and hydrolyze it.
    • Neuronal gingipain levels correlated with cellular tau protein levels.
  • In figures 7 and 8, they infected wild type mice (i.e., standard inbred lab mouse strain that hadn’t been genetically modified one way or another) with P. gingivalis to establish a periodontal infection and claim it led to brain lesions that recapitulate Alzheimer’s lesions – specifically, amyloid beta plaques and neurofibrillary tangles.

Plausible narrative except for one key weakness and liberal use of questionable statistics throughout – need to get a bit into the weeds to highlight the pronounced weaknesses of the study.

  • Figures 1, 2, 3 and 4 are the linchpins of this paper because they were the only data generated from actual Alzheimer patients and appropriate controls so if they turn out to be not reproducible, the rest don’t matter.
    • Proprietary polyclonal rabbit antibodies were generated specifically for this study and used in figures 1, 2, 3 and 4 to show brain tissue from Alzheimer’s but not from controls stained positive for P.gingivalis-specific gingipains. Investigators unconnected to this team need to independently confirm these results using these reagents since this is the critical experiment for this narrative.
    • Without this data, the rest of the experiments using the human cell line and the mouse model cannot sustain the narrative by themselves. The rest of the paper is essentially a lot of work trying to square this circle.
  • Figure 4 as well as other figures in the supplementary data only report PCR data on P.gingivalis DNA from Alzheimer’s patients. Why didn’t they show data from equal number of sex- and age-matched controls the way they did in figures 1, 2, and 3?
  • Questionable use of statistics.
    • For example, figure 1 shows the results of brain tissue stains from n of 99, 104, 84 or 89 and yet they state they examined tissue arrays from only 29 each postmortem control or Alzheimer’s patients.
    • These numbers don’t match meaning they are reporting multiple tissue sections from individual postmortem patient samples as independent data points. However, 29 patient samples are 29 patient samples, no matter how many sections they chose to stain from a given patient’s sample.
    • On top of that, they commit the even more flagrant statistical lapse of using the non-parametric Mann-Whitney test to report statistically significant differences between Alzheimer and control brain tissues. To resort to a non-parametric test even after artificially inflating sample numbers is a clear sign of the extent of contrivance that was necessary to arrive at statistical significance.
    • Hello, anonymous reviewers of this paper, sleep-walked your way through peer-review, eh?
  • Reporting mean plus/- standard error of the mean (figures 3, 6, 7, 8) instead of mean plus/- standard deviation is sleazy because it artificially minimizes intra-group variations in order to artificially accentuate between-group differences. Data that need to be thus massaged (cooked) are a lot noisier than the authors are willing to admit.

The causative role of P.gingivalis in periodontitis itself is currently far from assured. Its role in Alzheimer’s is even more tenuous, this paper notwithstanding. After all, claiming causation on the basis of 29 postmortem Alzheimer’s brain tissues and a pair of proprietary antibody sera does sound like a stretch too far, doesn’t it? One swallow does not a summer make.

Bibliography

1. Sochocka, Marta, Katarzyna Zwolinska, and Jerzy Leszek. “The infectious etiology of Alzheimer’s disease.” Current neuropharmacology 15.7 (2017): 996-1009. The Infectious Etiology of Alzheimer’s Disease

2. Arweiler, Nicole B., and Lutz Netuschil. “The oral microbiota.” Microbiota of the Human Body. Springer, Cham, 2016. 45-60.

3. Curtis, Mike. “An introduction to microbial dysbiosis.” The Human Microbiota and Chronic Disease: Dysbiosis as a Cause of Human Pathology (2016): 37.

4. Nobbs, Angela H., David Dymock, and Howard F. Jenkinson. “Microbiota of the Mouth: A Blessing or a Curse?.” The Human Microbiota (2013): 135-166.

5. Socransky, S. S., et al. “Microbial complexes in subgingival plaque.” Journal of clinical periodontology 25.2 (1998): 134-144.

6. Nibali, Luigi. “The periodontal diseases: microbial diseases or diseases of the host response?.” The Human Microbiota and Chronic Disease: Dysbiosis as a Cause of Human Pathology (2016): 217.

7. Hajishengallis, George, et al. “Low-abundance biofilm species orchestrates inflammatory periodontal disease through the commensal microbiota and complement.” Cell host & microbe 10.5 (2011): 497-506. Low-Abundance Biofilm Species Orchestrates Inflammatory Periodontal Disease through the Commensal Microbiota and Complement

8. Griffen, Ann L., et al. “Porphyromonas gingivalis strain variability and periodontitis.” Journal of clinical microbiology 37.12 (1999): 4028-4033. https://jcm.asm.org/content/jcm/37/12/4028.full.pdf

9. Darveau, R. P., G. Hajishengallis, and M. A. Curtis. “Porphyromonas gingivalis as a potential community activist for disease.” Journal of dental research 91.9 (2012): 816-820. Porphyromonas gingivalis as a Potential Community Activist for Disease

10. Jiao, Y., M. Hasegawa, and N. Inohara. “The role of oral pathobionts in dysbiosis during periodontitis development.” Journal of dental research 93.6 (2014): 539-546. https://www.researchgate.net/profile/Naohiro_Inohara/publication/260949569_The_Role_of_Oral_Pathobionts_in_Dysbiosis_during_Periodontitis_Development/links/5523f4dd0cf2caf11bfbd68c.pdf

11. Kamer, Angela R., et al. “TNF-α and antibodies to periodontal bacteria discriminate between Alzheimer’s disease patients and normal subjects.” Journal of neuroimmunology 216.1-2 (2009): 92-97. TNF-α and antibodies to periodontal bacteria discriminate between Alzheimer’s disease patients and normal subjects

12. Noble, James M., et al. “Periodontitis is associated with cognitive impairment among older adults: analysis of NHANES-III.” Journal of Neurology, Neurosurgery & Psychiatry 80.11 (2009): 1206-1211. https://pdfs.semanticscholar.org/3f3d/658d3249180a27b646f1ce841b0f4f9ad0c1.pdf

13. Stein, Pamela Sparks, et al. “Serum antibodies to periodontal pathogens are a risk factor for Alzheimer’s disease.” Alzheimer’s & Dementia 8.3 (2012): 196-203. Serum antibodies to periodontal pathogens are a risk factor for Alzheimer’s disease

14. Dominy, Stephen S., et al. “Porphyromonas gingivalis in Alzheimer’s disease brains: Evidence for disease causation and treatment with small-molecule inhibitors.” Science advances 5.1 (2019): eaau3333. Porphyromonas gingivalis in Alzheimer’s disease brains: Evidence for disease causation and treatment with small-molecule inhibitors

https://www.quora.com/What-is-the-relationship-between-Alzheimers-and-the-bacteria-that-causes-gum-disease-Porphyromonas-gingivalis-Does-it-cause-Alzheimer-s-sometimes-or-strongly-contribute-to-its-progression/answer/Tirumalai-Kamala