Doxycycline reduces fibril formation in a transgenic mouse model of AL amyloidosis

Jennifer Ellis Ward, Ruiyi Ren, Gianluca Toraldo, Pam Soohoo, Jian Guan, Carl O'Hara, Ravi Jasuja, Vickery Trinkaus-Randall, Ronglih Liao, Lawreen H Connors, David C Seldin, Jennifer Ellis Ward, Ruiyi Ren, Gianluca Toraldo, Pam Soohoo, Jian Guan, Carl O'Hara, Ravi Jasuja, Vickery Trinkaus-Randall, Ronglih Liao, Lawreen H Connors, David C Seldin

Abstract

Systemic AL amyloidosis results from the aggregation of an amyloidogenic immunoglobulin (Ig) light chain (LC) usually produced by a plasma cell clone in the bone marrow. AL is the most rapidly fatal of the systemic amyloidoses, as amyloid fibrils can rapidly accumulate in tissues including the heart, kidneys, autonomic or peripheral nervous systems, gastrointestinal tract, and liver. Chemotherapy is used to eradicate the cellular source of the amyloidogenic precursor. Currently, there are no therapies that target the process of LC aggregation, fibril formation, or organ damage. We developed transgenic mice expressing an amyloidogenic λ6 LC using the cytomegalovirus (CMV) promoter to circumvent the disruption of B cell development by premature expression of recombined LC. The CMV-λ6 transgenic mice develop neurologic dysfunction and Congophilic amyloid deposits in the stomach. Amyloid deposition was inhibited in vivo by the antibiotic doxycycline. In vitro studies demonstrated that doxycycline directly disrupted the formation of recombinant LC fibrils. Furthermore, treatment of ex vivo LC amyloid fibrils with doxycycline reduced the number of intact fibrils and led to the formation of large disordered aggregates. The CMV-λ6 transgenic model replicates the process of AL amyloidosis and is useful for testing the antifibril potential of orally available agents.

Figures

Figure 1
Figure 1
Schematic of the CMV-λ6 construct which is transmitted in the transgenic mouse genome. (A) The CMV-λ6 transgenic construct including the full-length immunoglobulin LC containing a leader (L), variable (V), joining (J), and constant (C) domain was expressed using the CMV promoter (pCMV) and capped with the bovine growth hormone polyadenylation sequence (BGH pA). (B) PCR genotyping of a representative litter of 10 pups (lanes 3-12) generated by crossing a transgenic line AL55 female (lane 1) and a wild-type male (lane 2) produced 50% transgene positive offspring as expected. Negative (no DNA, lane 13) and positive (plasmid DNA, lane 14) controls are present. The vertical line indicates that noncontiguous lanes from the same agarose gel image are displayed together.
Figure 2
Figure 2
Expression of the human λ6 LC protein in CMV-λ6 transgenic mice. (A-F) Immunohistochemistry demonstrating human λ LC expression (red-brown staining) in (A) stomach, (B) bladder, (C) pancreas, (D) kidney, and (E) heart. (F) Pancreas stained with human κ LC antibody (negative control) is also shown. Antibodies used were Dako nos. A0193 (λ) and A0191 (κ). (G-I) Immunoblots of human λ LC expression (top panels) in serum (G), pancreas (H), and stomach (I) tissue extracts from transgenic (Tg) and wild-type (WT) littermate mice. β-actin was used as a loading control (H-I bottom panel). The anti–human λ antibody used was Sigma-Aldrich no. L5267. The vertical line indicates that noncontiguous lanes from the same immunoblot image are shown.
Figure 3
Figure 3
Amyloid deposits in the stomach of CMV-λ6 transgenic mice. Stomach sections were stained with Congo red and counterstained with hematoxylin visualized by brightfield (A) or polarized light microscopy (B). (C-D) Demonstration of immunohistochemical staining with anti–human λ and anti–human κ LC, respectively. (E) Negative staining EM of the contents of the stomach glands in the stomach of a transgenic mouse with Congo red positive deposits. Fibril diameters are approximately 10 nm. (F) Immunoblot for human λ LC in protein extracted from stomach amyloid deposits isolated by LCM (lane 1); serum from a transgenic mouse as a control (lane 2). The vertical line indicates that noncontiguous lanes from the same immunoblot image are displayed.
Figure 4
Figure 4
Age-dependent increase in amyloid deposition in CMV-λ6 mice. Proportion of transgenic mice at various ages with amyloid deposits in stomach sections detected by Congo red staining of a single section of stomach. Groups included mice that were < 6 months (n = 19), 6-12 months (n = 18), 12-18 months (n = 54), 18-23 months (n = 29), and 24-30 months of age (n = 6).
Figure 5
Figure 5
Neurologic phenotype in CMV-λ6 mice. (A) The transgenic mice displayed less spontaneous activity compared with wild-type controls when fasted overnight (P < .018 horizontal, P < .0068 vertical, n = 4 each group). (B) Exercise capacity on an inclined treadmill was diminished in older transgenic mice compared with age-matched controls (P < .011, n = 4 each group). (C) H&E-stained section of spinal cord revealing dystrophic neurons (example denoted as *) in an old transgenic mouse with the limb clenching phenotype.
Figure 6
Figure 6
Doxycycline prevents amyloid deposition in vivo and directly interacts with LC in vitro. (A) In control CMV-λ6 transgenic mice drinking water alone, 69% (11/16) had Congo red positive deposits in the stomach. (B) Deposits were inhibited by oral administration of doxycycline in the drinking water, identified in only 23% (4/17) of the treated mice (P = .00006, χ2 analysis). (C) For a 5-day period of incubation, recombinant amyloidogenic LC in vitro form typical amyloid fibrils, as visualized by negative stain TEM, (D) whereas incubation with 250 mg/L doxycycline resulted in degraded and broken fibrils and disorganized bundles of immature fibrils (data not shown). (E) With 15 mg/L doxycycline, broken fibrils with frayed ends were observed. (F) Amyloid fibrils extracted from autopsy material formed numerous large aggregates after incubation with doxycycline. (C-E) Magnification ×42 000, bar = 100 nm; magnification ×10 000, bar = 500 nm.

Source: PubMed

3
Předplatit