Figure 1.
Mutant α-Actinin-4 Sediments Abnormally
(A) Pattern of sedimentation of α-actinin-4 in a 10%–40% sucrose gradient. Results shown are for wild-type α-actinin-4, K228E α-actinin-4, and K228E α-actinin-4 after addition of excess cold (wild-type) α-actinin. A fraction of the mutant α-actinin-4 sedimented at least as quickly than the highest molecular weight marker, catalase, which has a sedimentation coefficient of 11.3. This was observed with all of the other mutants tested as well (data not shown), but never with labeled wild-type α-actinin-4. Addition of cold α-actinin did not alter the sedimentation pattern seen with the mutant form of α-actinin-4.
(B) Results illustrated graphically. Units are arbitrary.
Figure 2.
Mutant α-Actinin-4 Behavior in Cells
(A) Mutant and wild-type α-actinin-4 show different localization and dynamics when expressed in a conditionally immortalized differentiated mouse podocyte cell line. Differentiated podocytes were injected in the nucleus with equal concentrations of expression plasmid for GFP fusions of mutant and wild-type actinins. At 2–4 h after injections, cells were imaged and both phase and fluorescence images recorded as described in the Materials and Methods. To illustrate changes in distribution of the fluorescence signal, three fluorescence images each 1 min apart were overlaid as red, green, and blue panes. Areas of fluorescence that were the same in all panes show as white, while dynamic areas are indicated by the color. The top panel indicates the initial phase and overlain dynamic fluorescence images of wild-type α-actinin-4, while the bottom two panels illustrate characteristic results for mutants K228E and T232I at 3 min time intervals. (See Videos S1–S3.)
(B) Transfections in podocytes derived from mutant and wild-type mice. When transfected into conditionally immortalized podocytes of all three α-actinin-4 genotypes (+/+, K228E/+, or K228E/K228E), wild-type GFP–α-actinin-4 shows diffuse cytoskeletal localization. Mutant GFP–α-actinin-4 shows a similar alteration in localization when expressed in these three cells types.
Figure 3.
(A) Targeting construct. As we have described elsewhere (Kos et al. 2003), targeting initially resulted in a “knockout” allele, due to disruption of normal transcription, presumably by the intronic loxP-flanked neomycin resistance cassette. After breeding to Cre transgenic mice, this neomycin cassette was excised, as illustrated.
(B) Northern blot analysis using kidney total RNA illustrates that the expression of the Actn4 transcript in K228E/K228E is similar to the expression in wild-type mice.
(C) RT-PCR and sequencing of the relevant portion of Actn4 exon 8 confirms that the transcript in mice homozygous for the targeted allele contains the desired point mutation (top, wild-type; bottom, targeted).
(D) Western blot showing markedly decreased expression of α-actinin-4 protein in K228E/K228E homozygous mice and moderately decreased expression in heterozygotes. Shown are blots using protein from cultured fibroblasts. Results were similar using protein extracted from lung, brain, liver, and kidney (data not shown). β-actin control is shown below.
(E) Western blot showing expression of α-actinin-4 in lymphocytes from a human K228E/+ heterozygote (Kaplan et al. 2000) and three wild-type controls (two related, one unrelated). β-actin control is shown below.
Figure 4.
α-Actinin-4 Synthesis and Degradation
(A and B) Synthesis of α-actinin-4 by wild-type and K228E/K228E fibroblasts. The rate of increase in the accumulation of mutant and wild-type α-actinin-4 is similar, as indicated by the superimposed shapes of the synthesis curves.
(C and D) Pulse–chase experiments showing degradation of α-actinin-4 in K228E/K228E cells. Half-life of wild-type α-actinin-4 is greater than 30 h. Half-life of mutant α-actinin-4 is approximately 15 h.
(E) Half-life of K228E mutant α-actinin-4 is restored to near-normal levels by the addition of lactacystin. Shown is labeled α-actinin-4 levels, expressed as a percentage of α-actinin-4 at time 0 and at 16 h and in the presence of 2.5 μM lactacystin in DMSO or in DMSO alone.
Figure 5.
Electron micrographs from Actn4 wild-type (A) and Actn4K228E/K228E mice (B–D). As shown, Actn4K228E/K228E mice were found to have abnormalities that were typically focal, with some areas of podocyte foot process effacement (B), as well as areas that appeared essentially normal (C). Bottom image ([D] using tannic acid counterstaining) illustrates electron-dense deposits observed in several podocyte cell bodies in Actn4K228E/K228E mice. No such deposits were observed in wild-type or heterozygous mice.
Figure 6.
Biochemical Characteristics of Mutant Mice
(A) Average BUN and creatinine levels in Actn4K228E/– (n = 12), Actn4+/+ (n = 8), and Actn4K228E/K228E (n = 12) mice at the time of sacrifice. Differences were not statistically significant. Error bars show standard deviation.
(B) Summary of proteinuria in wild-type, Actn4+/–, Actn4K228E/–, Actn4–/–, and Actn4K228E/K228E mice, measured by albumin dipstick. Distribution of measurements are illustrated graphically for each genotype.
Figure 7.
(A) IF studies of glomerular protein expression in Actn4+/+, Actn4K228E/+ , and Actn4K228E/K228E mice. As indicated, expression of α-actinin-4, ZO-1, and nephrin is shown, as is a merged image of α-actinin-4 and ZO-1 expression.
(B) Glomerular expression of α-actinin-4 in normal human kidney and in an individual heterozygous for a K228E mutation.