It’s a curious fact that when you remove the spleens of old mice, they live longer.1
Like, by a lot. Mice that were splenectomized at 2 years old (which is elderly for a mouse) had a median lifespan 19% longer than control mice, and a whopping 34% longer than sham-operated mice.
This doesn’t mean you should go get your spleen removed. You need your spleen; people who get splenectomies or who have inadequate spleen function are at greater risk for infection, because the spleen is full of macrophages which destroy bacteria and memory B cells which produce antibodies. And abdominal surgery in otherwise healthy people is almost never a good risk-benefit tradeoff.
But it’s intriguing. If removing aged spleens can extend life, is there something that the aged spleen is producing that’s bad for you?
Well, what’s different about young and old spleens?
Aged spleens are smaller than young spleens.2 They have 1/8 of the volume of white pulp (the lymphoid tissue, full of lymphocytes) but 27 times as much volume of red pulp (which is mostly connective tissue, blood cells, and specialized macrophages which eat damaged blood cells.)
And aged spleens have fewer of every type of cell except mast cells, which are much more abundant.
Could mast cells be the culprit?
Let’s look at another angle: the Tabula Muris Senis, a database which compares single-cell gene expression of mice at various ages. Which cell types are more common in aged spleens than young spleens? Which genes are expressed more in aged than young spleens?
There’s significantly more erythroblasts after 24 months
There’s significantly more granulocytes (which include mast cells) after 24 months
There’s significantly more megakaryocyte-erythroid progenitor cells after 24 months
There’s significantly more plasma B cells at 30 months
The genes whose expression rose the most between 3-18 months and 24-30 months are:
Beta-s, found mostly in erythroblasts and megakaryocyte-erythroid progenitor cells;
S100a8 and S100a9, the two subunits of calprotectin, an inflammatory marker found at its highest levels in granulocytes and also in erythroblasts and megakaryocyte-erythroid progenitor cells;
Camp, the gene for cathelicidin, mostly found in granulocytes and used to attack bacteria;
Ngp, or neutrophilic granule protein, mostly found in granulocytes
So this is consistent with a picture where granulocytes and granulocyte-specific genes rise in abundance with age.
So what are granulocytes/mast cells doing in the spleen?
Mast cells release lots of pro-inflammatory signals when activated by an IgE antibody. They are responsible for allergic responses. Usually they hang out near blood vessels and the mucosa of the intestine and lung.3
The spleen, in other words, is not typically rich in mast cells. But mast cells accumulate in the spleen and other organs when some problem is going on in the body that increases mast cell production or activity.
Mast cells accumulate in the spleen in humans who died of anaphylactic shock, compared to other causes of death. This makes sense given the role of mast cells in anaphylaxis.4
When a mouse gets a parasitic infection (which activates mast cells), during recovery the senescent mast cells migrate to the spleen.5 The mast cells aren’t proliferating in the spleen; it seems the spleen is where mast cells go to die.
Mast cells infiltrate the spleen and other tissues during disorders that increase the number of mast cells (mastocytosis). They have been found in the spleen as a result of urticaria pigmentosa, an autoimmune skin disease caused by abnormal proliferation of mast cells.6
The bottom line is, mast cells are rare or absent in the spleen unless something is going on to activate or produce mast cells — mastocytosis, allergy, or parasitic infection.
Is “normal” aging, or common age-related disease, one of those things that makes mast cells get frisky?
And do mast cells play a causal role in age-related diseases?
The answer looks like yes, though not obviously a bigger role than other inflammatory cells and signals.
Mast cell density and degranulation (spilling their toxic contents) is associated with more advanced fibrosis in non-alcoholic fatty liver disease, one of the most common diseases of aging. In fact, mast cells can cause fibrosis when you add them to liver tissue.7
Mast cells are more common around the mesenteric lymph vessels in old than young rats (27% higher density), and much more likely to be activated in old rats (403% more activated mast cells.) Activated mast cells weaken the pumping action of the neighboring lymph vessels, reducing the flow of lymph which clears out damaged cells and misfolded proteins from tissues. Lymphatic drainage grows weaker with age in general, and the loss of the ability to clear out damaged cells and proteins has been implicated in many age-related diseases.8
Mast cells are more common, larger, and secrete more histamine and heparin, in aged mouse skin than in young mouse skin. Signaling downstream of heparin can degrade collagen, and it’s collagen degradation that’s responsible for wrinkles and other age-related changes in skin.9 Mast cells also become more common with age in human skin.10
Mast cells also accumulate in atherosclerotic plaques.11
It’s not clear in many of these cases whether mast cells are a cause or effect (or both) of age-related chronic inflammation and tissue damage. But the association seems to exist.
Let’s go back and look into more detail at splenectomy and spleen transplants to understand exactly what kind of effect they have on aging.
Transplanting young mice with spleen cells from old mice with an autoimmune kidney disease called membranous glomerulonephritis can give the young mice the same kidney disease, suggesting that the old spleen cells carried or produced the autoantibodies that attack kidney tissue. This effect is probably not caused directly by mast cells, as IgG antibodies (including the autoantibodies that cause membranous glomerulonephritis) are produced by B cells.
Tumors injected into mice grow faster when the recipient mice are old than young. Tumors injected into young mice grow faster when they are mixed with living spleen cells from old mice — but not dead, irradiated spleen cells from old mice. This tumor-enhancing effect of old spleen cells appears to come from T cells. When mice have their thymuses removed, which reduces the pool of T cells in the body, their aged spleen cells no longer accelerate tumor growth. 12 While total spleen T cell counts decline with age, perhaps some subset of them proliferates or changes in a way that promotes tumor growth.
Indeed, a subset of T cells in aged mice seem to suppress the ability of other spleen cells to produce antibodies against bacterial toxins. Cultivation of both old and young spleen cells together has a smaller antibody response than young spleen cells alone. This immunosuppressive effect was abolished when the old spleen cells were treated with a compound that selectively kills T cells.13
T cells that suppress antibody production are now known as regulatory T cells, or Tregs, and they increase with age in both mice and humans, particularly in the spleen. Tregs are believed to block the immune response to cancers.
Splenectomy also has some specific immunological benefits.
It provides protection against brain damage due to stroke in aged mice14 if they are splenectomized prior to or immediately after the stroke, but not if you wait more than three days.15 Immediate splenectomy reduces inflammation due to stroke and improves functional recovery. This points to aged spleens producing some kind of pro-inflammatory substance.
Overall, though, the “active ingredient” that aged spleens produce that does bad things is still mysterious. Is it mast cells? Is it Tregs? Is it a sub-population of B cells? Unknown.
This would be a great topic for a more in-depth immunological or biomarker profiling study. How do the immune cell populations of splenectomized vs. non-splenectomized aged mice differ? What about inflammatory cytokines in blood or various tissues?
Once we understand what splenectomy is doing “right”, we might be able to work on getting that effect in isolation, without the nasty complications of actual removal of the spleen.
Albright, J. F., T. Makinodan, and Julia W. Deitchman. "Presence of life-shortening factors in spleens of aged mice of long lifespan and extension of life expectancy by splenectomy." Experimental gerontology 4.4 (1969): 267-276.
El-Naseery, Nesma I., et al. "Aging-associated immunosenescence via alterations in splenic immune cell populations in rat." Life sciences 241 (2020): 117168.
Dwyer, Daniel F., Nora A. Barrett, and K. Frank Austen. "Expression profiling of constitutive mast cells reveals a unique identity within the immune system." Nature immunology 17.7 (2016): 878-887.
Edston, Erik. "Accumulation of eosinophils, mast cells, and basophils in the spleen in anaphylactic deaths." Forensic science, medicine, and pathology 9.4 (2013): 496-500.
Friend, Daniel S., et al. "Senescent jejunal mast cells and eosinophils in the mouse preferentially translocate to the spleen and draining lymph node, respectively, during the recovery phase of helminth infection." The Journal of Immunology 165.1 (2000): 344-352.
Metcalfe, Dean D. "Mast cells and mastocytosis." Blood, The Journal of the American Society of Hematology 112.4 (2008): 946-956.
Kundu, Debjyoti, et al. "The Dynamic Interplay Between Mast Cells, Aging/Cellular Senescence, and Liver Disease." Gene Expression 20.2 (2020): 77.
Chatterjee, Victor, and Anatoliy A. Gashev. "Aging-associated shifts in functional status of mast cells located by adult and aged mesenteric lymphatic vessels." American Journal of Physiology-Heart and Circulatory Physiology (2012).
Hafez, Sara Mohamed Naguib Abdel. "Age related changes in the dermal mast cells and the associated changes in the dermal collagen and cells: A histological and electron microscopy study." Acta histochemica 121.5 (2019): 619-627.
Gunin, Andrei G., et al. "Age-related changes in proliferation, the numbers of mast cells, eosinophils, and cd45-positive cells in human dermis." Journals of Gerontology Series A: Biomedical Sciences and Medical Sciences 66.4 (2011): 385-392.
Atkinson, James B., et al. "The association of mast cells and atherosclerosis: a morphologic study of early atherosclerotic lesions in young people." Human pathology 25.2 (1994): 154-159.
Gozes, Y., and Nathan Trainin. "Enhancement of Lewis lung carcinoma in a syngeneic host by spleen cells of C57BL/6 old mice." European journal of immunology 7.3 (1977): 159-164.
Segre, Diego, and Mariangela Segre. "Humoral immunity in aged mice: II. Increased suppressor T cell activity in immunologically deficient old mice." The Journal of Immunology 116.3 (1976): 735-738.
Chauhan, Anjali, et al. "Splenectomy protects aged mice from injury after experimental stroke." Neurobiology of aging 61 (2018): 102-111.
Ran, Yuanyuan, et al. "Splenectomy fails to provide long-term protection against ischemic stroke." Aging and disease 9.3 (2018): 467.
Another class of hypotheses for what's going on here: based on my admittedly-poor understanding, the spleen mainly turns over various cell types from the blood (notably including mast cells and RBCs). Presumably the body has feedback loops to maintain cell count for these cell types, so if the spleen is removed, production of these cell types should also slow/stop. I could imagine several different stories for how that might alleviate aging problems:
- to the extent that stem cell exhaustion is a thing, slowing production of cells might allow stem cell reserves to last longer
- slowed production might slow proliferation of precancerous cells
- rapid proliferation might push cells past the critical point for senescence
- ... probably lots of other possibilities I haven't thought of
This is pretty interesting. If one had access to a hospital's database, I imagine it wouldn't be hard to do a retrospective analysis looking at people who had a splenectomy and then seeing if they ended up with lower risk for cardiovascular disease. All I know is lymphocytes are implicated in the early stages of atherosclerosis. How that fact might be connected, I have no idea, but it seems it could be.