Tag Archives: lyme arthritis

Lyme Arthritis and Inflammation: Shut It Off!

A summary of 3 Lyme arthritis studies and how our immune system can fail to shut off when there’s an infection.

by Mayla Hsu, Ph.D., Director of Research and Science, GLA

Arthritis is one of the most common symptoms of Lyme disease, commonly presenting as swelling and pain in the joints. Borrelia burgdorferi bacteria, which cause Lyme disease, migrate to the joints and create the arthritic symptoms. However, growing evidence implicates not only the bacteria, but the immune system itself as playing a key role in the disease. It’s an example of how something that should protect us can also be harmful.

Our immune system has evolved to help us get rid of pathogens like B. burgdorferi. But if it’s working like a finely tuned machine, the immune system should turn off when the job is done. Instead, research has shown multiple ways that the immune system may be failing to stop the inflammatory response after infection, thus prolonging symptoms that can be very debilitating. To study this in Lyme arthritis, synovial fluid surrounding the joints can be collected from patients. It is then analyzed for the presence of immune cells and cytokines, the chemical messengers produced by cells to help kill pathogens.

Some Lyme arthritis patients have symptoms that do not improve after antibiotic therapy, known as antibiotic-refractory Lyme arthritis. Synovial fluid from these patients has been previously studied for the presence of regulatory T cells (Tregs). These immune cells, which are a subset of specialized T cells, were counted because they act as an “off switch” for inflammation. It was suspected that one pathway to long-term Lyme arthritis may be through insufficient or malfunctioning Tregs.

In the synovial fluid of antibiotic-refractory arthritis patients, an average of 5% of activated T cells were Tregs, as compared to antibiotic-responsive arthritis patients, who had 12%. Those with fewer Tregs were less responsive to anti-rheumatic medications, their arthritis took longer to resolve, and a number of them required synovectomies, or surgical removal of inflamed joint tissues, to resolve their symptoms.

A limitation of this early study was that the amount of Tregs found in the patients before and during early B. burgdorferi infection could not be analyzed, since the patients were only identified after Lyme arthritis was well underway. Might higher pre-existing Tregs be associated with quicker recovery from arthritis? If so, the number of these cells could be used as a possible prognostic marker to help with treatment decisions.

One way to determine this would be to assess Treg cells before and after experimentally infecting animals with B. burgdorferi, since such an experiment could not be done in humans. A new study in mice used an engineered mouse strain called C57BL/6 DEREG to address this question. In these mice, Treg cells can be depleted when animals are administered minute, nontoxic doses of diphtheria toxin. This was done either before or after infecting mice, and results were compared to nondepleted mice.

The researchers found that depletion of Tregs before infection with B. burgdorferi caused earlier tibiotarsal joint swelling (day 10 after infection) than in non-depleted mice (day 16 after infection). Additionally, the Treg-depleted mice had significantly more joint swelling than control mice, and in fact, had a second wave of swelling that peaked at day 22 after infection.

The contribution of Tregs after B. burgdorferi infection was studied by depleting mice one to three weeks post-infection, and then comparing joint swelling with nondepleted mice. Although these Treg-depleted mice did have increased joint swelling compared to non-depleted animals, the difference in swelling did not achieve statistical significance and was less than in mice depleted before infection.

When the joints were examined for evidence of pathology, all of the mice whose Tregs were depleted prior to infection had lymphocyte infiltration into the joints and surrounding soft tissues, indicating the presence of immune cells homing to a site of inflammation. However, only a single mouse in the nondepleted group had a mild degree of immune cell infiltration on the surface of the joint. For mice whose Tregs were depleted one to three weeks after infection, there were no inflammatory pathological changes observed in the joint tissues.

Together, the mouse studies suggest that joint swelling and pathology were more dependent on the amount of Tregs before, rather than after infection by B. burgdorferi. But Treg function, not just abundance, may also be important. In other words, what do Tregs actually do to reduce arthritis?

Experiments that assess Treg function often focus on their ability to regulate, or suppress the proliferation of other immune cells. Tregs are also studied for their inhibition of inflammatory cytokine production. In one study of a limited number of Lyme patients, T cells, of which Tregs were a subset, were collectively cultured from the synovial fluid of either antibiotic-responsive or antibiotic-refractory arthritis patients. Tregs from the refractory patients were less able to suppress the production of inflammatory cytokines like interferon gamma (IFNγ) and tumor necrosis factor alpha (TNFα) than those from the antibiotic-responsive patients.

These findings suggest that both the lower amount of Tregs, and the loss of their suppressive functions, may be why some patients have antibiotic-refractory arthritis. The study of Treg depletion in mice showed this tendency too. More studies are needed to explain how other cytokines may be involved in promoting an infection environment that ultimately, resolves or continues inflammation.

And, why some people have lower Tregs to begin with is also not yet understood – whether it is genetically determined or occurs in response to infection. But what is obvious is that the interplay between B. burgdorferi and the host immune response is complicated, depending on switches to turn inflammation on and off. More research will help us understand this regulation, and how host protection can possibly turn into harm.

Related Blogs:

Lyme Arthritis: The Antibody Connection
Research POV: Lyme Arthritis and Peptidoglycan
Possible Clue to Lyme Arthritis Found in People’s Inflamed Joints

Lyme Arthritis: The Antibody Connection

by Mayla Hsu, Ph.D., Director of Research and Science, GLA

Lyme disease, caused by Borrelia burgdorferi bacteria, often leads to arthritis, with up to 60 percent of late-stage patients suffering from painful joints and swelling. Symptoms can persist for months and even years. Its suspected that a protein called arthritis-related protein (Arp), which is located on the surface of the Lyme bacteria causes these symptoms. Mice experimentally infected with B. burgdorferi develop arthritis symptoms that are resolved when the animals begin to make anti-Arp antibodies.

But an enduring medical mystery has been why these anti-Arp antibodies coincide with resolving Lyme arthritis but not any other aspect of Lyme disease. One reason could be that other proteins, also on the bacterial surface, may be protecting Arp and the bacteria from antibodies. And if those shielding proteins constantly adapt, they could be even more effective at protecting Arp.

It turns out that this shielding mechanism might be possible. Recent data reported by Abdul Lone and Troy Bankhead at Washington State University indicates that another bacterial surface protein known as  VlsE, may protect the bacteria from anti-Arp antibodies and act as a shield to prevent the immune system from fighting Lyme.

The mechanism of how VlsE defends the Lyme bacteria from Arp antibodies is an interesting story. Earlier work by this research team tested a mutant B. burgdorferi that lacks VlsE protein. This mutant was tested for its ability to infect and grow in immunodeficient mice. Growing the bacteria in mice (in vivo) is more informative than growing the bacteria in a test tube (in vitro), because a mouse has the complexity of an animal with multiple organ systems, and the bacteria disseminate through the body that has different cell types, a circulatory system and other features not found in vitro. Immunodeficient mice lack a functioning immune system, so growing bacteria in them allows observation of how the bacteria replicate unimpeded, in a complex animal system. 

In severe combined immunodeficient (SCID) mice, the VlsE-deficient mouse was capable of infection, showing no growth deficiency. However, if the animals were treated with anti-Arp antibodies, the animals were able to clear the infection.  This suggests that anti-Arp antibodies can stick to Arp on the bacteria and promote their destruction. In contrast, normal bacteria that have VlsE on their surface could not be cleared by anti-Arp antibodies in SCID mice. The contrast in bacterial clearance in the mice shows that VlsE somehow affects the clearance of bacteria by anti-Arp antibodies.

Next, the investigative team used microscopy to observe whether anti-Arp antibodies can stick to B. burgdorferi spirochetes when VlsE is also present on the bacterial surface. They found that although anti-Arp antibodies can bind to VlsE-negative bacteria, they cannot do so in VlsE-positive bacteria. This result suggests that VlsE may actually be a physical barrier that prevents anti-Arp antibodies from sticking to their target, the Arp protein. Precisely how VlsE does this is an interesting question. For example, are Arp and VlsE close together on the bacterial surface? Is VlsE larger or does it block only the specific epitopes, or subdomains that are bound by Arp antibodies?

Moreover, there were limits on VlsE-mediated protection of B. burgdorferi proteins. Control experiments indicated that VlsE may not block antibodies other than those targeting Arp. This was shown by using a blood serum (antisera) purified from mice infected with an Arp-negative, VlsE-negative bacteria. These antisera would lack antibodies against Arp and VlsE, but have antibodies against other B. burgdorferi proteins. Mice were pretreated with the Arp-negative, VlsE-negative antisera and then challenged with normal bacteria which had both Arp and VlsE. The normal bacteria could not infect, which showed that the presence of VlsE could not protect the bacteria from antibodies that target non-Arp bacterial antigens. This suggests that VlsE may only protect Arp, and not other Borrelia antigens.

Work by other investigators has shown that VlsE undergoes extensive antigen variation. The antibodies elicited by VlsE also change during disease course. So, the work of Lone and Bankhead raises questions about whether specific VlsE variants protect Arp, and if there is a correlation with disease stage and the onset of arthritis.

These findings are also important because they further highlight how, in addition to other immune dysfunctions, the bacteria have evolved multiple ways to specifically evade the humoral, or antibody-mediated immune response. Not only does the VlsE protein change during bacterial infection, providing a mechanism of immune evasion and persistence for the bacteria, but it may also protect other bacterial proteins from being targeted. Further studies will clarify how VlsE shields Arp.

And it is another piece of the puzzle of how B. burgdorferi may cause persistent arthritis. Previous findings implicated peptidoglycan, a different component of the bacterial cell wall. How these pieces fit together will explain disease outcomes and long-term symptoms.

Related Blogs:

Research POV: Lyme Arthritis and Peptidoglycan
Possible Clue to Lyme Arthritis Found in People’s Inflamed Joints


research pov_lyme arthritis_Peptidoglycan

Research POV: Lyme Arthritis and Peptidoglycan

Important study identifies the Borrelia burgdorferi peptidoglycan as an immunogen likely to cause Lyme arthritis in some patients


by Mayla Hsu, Ph.D., Director of Research and Science, GLA

A longstanding mystery in Lyme disease is why some antibiotic-treated patients continue to suffer long-term symptoms, while others recover. It is possible that a hyperactive immune system in some patients causes inflammation-related tissue damage. Another possibility is that the bacteria, Borrelia burgdorferi, may still be alive in compartments of the body that are not readily accessible to antibiotics, and the bacteria’s continued replication causes ongoing symptoms.

A recent article found that a component of the bacterial cell wall, peptidoglycan (PG), might be driving the persistent inflammation that causes Lyme arthritis even after antibiotic treatment. The authors, who include a GLA-funded researcher, found B. burgdorferi PG antibodies in the synovial fluid that surrounds the knees of Lyme arthritis patients, which is evidence of an ongoing immune response against bacteria or fragments of its cell wall. In contrast, they did not find significant antibodies against PGs of any other bacteria. A strength of this study is that they compared synovial fluid from Lyme arthritis patients with that of other types of non Lyme-related arthritis. Only Lyme arthritis patients had synovial fluid B. burgdorferi PG antibodies.

Blood was also collected from the same Lyme patients whose synovial fluid was studied. Here, they found that sera (serums) of Lyme arthritis patients also had PG antibodies, although the levels were much lower than in the synovial fluid. Moreover, Lyme arthritis patient sera had much higher levels of PG antibodies than any detected in control patient sera. These findings indicate a continued immune response against PG, particularly in the synovial joints, specifically against B. burgdorferi.

Another asset of this study was its analysis of patient samples both before and after antibiotic treatment. The authors designed a way to detect PG itself, not only antibodies, in synovial fluid and sera. Using this test, they could not find PG in control synovial fluid, nor in Lyme arthritis patient sera. However, it was present in 92% of Lyme arthritis synovial fluids, both before and after treatment with antibiotics; and this was true of patients who had been treated with oral as well as intravenous antibiotics. In addition, the amount of PG strongly correlated with the level of PG antibodies in the same synovial fluid samples. This finding suggests that the bacterial cell wall itself is present in the synovia of patients, and that its presence elicits antibodies against PG.

Peptidoglycan_figure for blog
Comparison of B. burgdorferi PG antibodies for serum and synovial fluid, from Lyme arthritis patients (red) and from controls (black) (healthy and non-Lyme arthritis)

Do these results mean that B. burgdorferi is actively replicating in the patients in whom PG or PG antibodies were found? Testing for bacterial DNA was done on synovial fluid and blood, both before and after antibiotic treatment. Although both compartments were often positive for bacterial DNA before therapy, almost all samples became negative after therapy. This finding suggests that PG and PG antibodies persisted even after eliminating actively replicating B. burgdorferi at these sites, but it does not rule out bacterial replication elsewhere, and leaves open the question of how bacterial cell wall components could remain long after the bacterial DNA is no longer detectable.

This study also identified proinflammatory cytokines in synovial fluid, which play an important role in mediating immune responses. In vitro, the addition of purified B. burgdorferi PG to cultured peripheral blood mononuclear cells (PBMCs) caused increased levels of inflammatory cytokines like TNFα, IL1α, and IFNγ. These and others were also significantly elevated in the synovial fluid of Lyme arthritis patients, suggesting a possible mechanistic link between PG and inflammation. Finally, injection of purified PG in mice resulted in ankle swelling and inflammatory changes that are consistent with the established mouse model for Lyme arthritis.

In sum, these results make a strong case that B. burgdorferi PG may be responsible for driving inflammation and continued symptoms in post-antibiotic Lyme arthritis. Whether it is also implicated in promoting other long-term symptoms remains to be seen. These findings open up a new avenue of inquiry that may yield fruitful insights that will help us to better understand and care for people suffering from persistent symptoms.

Learn more about GLA’s research initiatives and accomplishments:

Research Report
Published Research Findings
Current Grantees 
First Observational Study for Lyme Disease Treatment
Post-treatment Lyme: Two Million by 2020
Blog: Why Good Science is Crucial