Finally, a transgenic light chain amyloidosis mouse model

Ward and colleagues report the first transgenic light chain amyloidosis mouse model in this issue of Blood and demonstrate that it can be used to test pharmaceutical candidates.¹ 

Light chain amyloidosis results from the misfolding and aggregation of an immunoglobulin light chain usually produced by clonal plasma cells in the bone marrow. Hence, light chain amyloidosis is both a cancer and an amyloid disease and is the most rapidly fatal of the systemic amyloid diseases. Patients with light chain amyloidosis are usually treated with chemotherapy agents to eradicate the plasma cell clone. However, the toxicity of these drugs in the background of proteotoxicity caused by the process of light chain amyloid fibril formation (amyloidogenesis) often limits how much of the chemotherapeutics can be given.^2-5  Several experts have hypothesized that if amyloidogenic light chain secretion and/or light chain amyloidogenesis could be blocked, the associated organ toxicity would be ameliorated, enabling more aggressive and effective chemotherapy regimens to be used.

While transgenic cell and murine models are now available for nearly every human amyloid disease, these have proven elusive for light chain amyloidosis despite significant effort on the part of several laboratories. Light chain amyloidosis mouse models have been hard to generate probably because of the severe cytotoxicity and organ toxicity associated with the process of light chain amyloidogenesis—leading to embryonic lethality. Here, Ward et al report the long-awaited first transgenic murine model of light chain amyloidosis.¹ 

The amyloidogenic light chain levels in the 3 lines generated are comparable with nonamyloidogenic light chain levels found in healthy human adults. Because the efficiency and rate of amyloid formation is dependent on the concentration of amyloidogenic light chain, the low plasma concentration of amyloidogenic light chain minimizes amyloidogenesis, which is probably why these mouse lines could be generated. It is envisioned that these mice will be very useful for evaluating proteostasis regulator candidates that selectively lower amyloidogenic light chain secretion without altering proteome secretion in general, including antibody secretion.⁶  Moreover, Ward et al showed that all 3 mouse lines produce amyloid in the lumen of the gastric glands of the stomach. The acidic environment of the stomach probably partially unfolds the destabilized light chain, which then forms a conformational intermediate that mis-assembles, leading to a dysplastic stomach epithelium and dilated glands filled with light chain amyloid. Approximately 20% of the transgenic mice exhibited a neurodegenerative phenotype reflected by a gait disturbance and limb clenching when the mice were picked up by the tail, and these mice demonstrated impaired inclined treadmill performance.

Ward and colleagues beautifully demonstrated that these mice could be used to assess the efficacy of anti–light chain amyloid drug candidates.¹  Transgenic mice 3 to 6 months of age were treated with doxycycline in the drinking water. After 7 months of treatment, 23% of the mice had stomach amyloid detected by Congo red versus 69% of the untreated group. While the mechanism of doxycycline action merits further investigation, what is clear is that this murine model is useful for testing antiamyloid agents.

Like almost all “first transgenic disease models,” this is not the ultimate murine model in that amyloidogenic light chain expression is low and amyloidogenicity appears to require the acidity of the gastric gland to occur. However, this model appears to be superior to the nontransgenic mouse models. These include a model introduced by Pepys and colleagues wherein they repeatedly injected human light chains from light chain amyloidosis patients into mice to observe amyloidogenesis of the human protein.⁷  This probably could be done in immune-compromised mice for longer-term experiments to avoid the immune system response if sufficient amounts of amyloidogenic light chains could be procured, which is probably best done by recombinant expression. Ward and coworkers also previously published a nontransgenic light chain amyloidosis model wherein plasmacytoma cells stably transfected with an amyloidogenic light chain are injected into mice, allowing short-term siRNA proof-of-principle experiments to be conducted.⁸  It seems that the next generation of mouse models would include those where the production of the amyloidogenic light chain could be turned on at relatively high levels after the mice reach adulthood—simulating what occurs in human light chain amyloidosis.