LL5beta: a regulator of postsynaptic differentiation identified in a screen for synaptically enriched transcripts at the neuromuscular junction

Abstract

In both neurons and muscle fibers, specific mRNAs are concentrated beneath and locally translated at synaptic sites. At the skeletal neuromuscular junction, all synaptic RNAs identified to date encode synaptic components. Using microarrays, we compared RNAs in synapse-rich and -free regions of muscles, thereby identifying transcripts that are enriched near synapses and that encode soluble membrane and nuclear proteins. One gene product, LL5beta, binds to both phosphoinositides and a cytoskeletal protein, filamin, one form of which is concentrated at synaptic sites. LL5beta is itself associated with the cytoplasmic face of the postsynaptic membrane; its highest levels border regions of highest acetylcholine receptor (AChR) density, which suggests a role in "corraling" AChRs. Consistent with this idea, perturbing LL5beta expression in myotubes inhibits AChR aggregation. Thus, a strategy designed to identify novel synaptic components led to identification of a protein required for assembly of the postsynaptic apparatus.

Figure 1.

Isolation of synaptic and nonsynaptic samples from muscle. (A) Diaphragm muscle stained with rBTX and viewed under a fluorescence dissecting microscope. Fibers run from top to bottom. NMJs are in a central “endplate band.” (B) Higher-magnification view of the endplate band shows that small groups of NMJs are tightly clustered. Synaptic samples for microarray analysis were obtained by microdissecting groups such as the one boxed in A and B. (C and D) Microdissected synaptic sample, double-stained with rBTX (C) and the nuclear stain DAPI (D). Bars: (A) 1 mm; (B) 200 μm; (C and D) 100 μm. (E) Southern analysis of two nonsynaptic (lanes 1 and 2) and two synaptic (lanes 3 and 4) samples after reverse transcription and PCR amplification. Gel was stained with ethidium bromide (EtBr) to verify equal loading. Blots were probed for transcripts known to be synaptically enriched (AChRα, AChE, PKARIα [PKA], and MuSK) or presumed to be uniformly distributed (cyclophilin [cycloph.] and GAPDH).

Figure 2.

Extent to which replicate sampling enhances reliable detection of synaptic transcripts. Rank orders for six transcripts known to be synaptically enriched (AChRα, AChRδ, AChRɛ, AChE, PKARIα, and MuSK) were calculated for each of nine independent experiments, with the most synaptically enriched equal to 1 and the least equal to 36,701 (see Table I). The mean rank order for all six transcripts was calculated for each combination of the 36 (9 × 8/2) possible sets of two pairs, each of the 84 (9 × 8 × 7/3 × 2) possible sets of three pairs, and so on. The two lines enveloping the individual points show the best and worst rank orders that could be obtained from a given number of pairs; central bars shows means.

Figure 3.

Verification of novel synaptic transcripts by whole mount in situ hybridization and quantitative RT-PCR. (A and B) Neonatal diaphragm was hybridized with probes for CD24 (A) or LL5β (B). RNA is concentrated at discrete sites identified as NMJs based on their number and position and by histochemical staining of adjacent muscle pieces with rBTX or a histochemical stain for AChE. Bar, 0.5 mm. (C) Quantitative RT-PCR of synaptic and extrasynaptic samples. A broadly distributed transcript (GAPDH, open squares) is present at similar levels in both samples, whereas dynein intermediate chain (Table II, No. 3; filled squares) is greater than sixfold enriched in the synaptic sample. Black and gray lines show data from synaptic and extrasynaptic samples, respectively. Thick line is the threshold value, used to calculate Ct (threshold cycle), from which differences between samples were determined.

Figure 4.

Structure and synaptic localization of LL5β. (A) Structures of LL5β and two homologues, LL5α and LL5γ. The percentage of amino acid identity of conserved domains is shown; amino termini of the three gene products are dissimilar. FHA, forkhead-associated domain; Ser, serine-rich region; COIL, coil-coil domain; PH, pleckstrin homology domain. (B–D) Longitudinal sections of adult muscle stained with mAb to LL5β (low magnification, B; high magnification, D) or control antibody of the same isotype (C) plus rBTX. LL5β is colocalized with AChRs. (E) Cross section of an adult rat NMJ showing LL5β concentrated in the depths of junctional folds and between AChR-rich areas at the crests of folds. (F) Synaptic concentration and colocalization were evident by prenatal day 4. (G) C2 myotubes were treated with agrin to induce AChR clusters, and then stained with anti-LL5β and rBTX. LL5β is concentrated at AChR-rich clusters. Bars: (B and C) 100 μm; (D) 20 μm; (E and F) 5 μm; (G) 10 μm.

Figure 5.

Synaptic localization of LL5β requires its PH domain. Muscles were electroporated with the indicated constructs. 4 d later, fibers were fixed, teased, and stained with rBTX. (A) GFP-LL5β colocalizes with AChRs at the NMJ. (B and C) A fusion protein lacking the carboxy-terminal PH domain (GFP-LL5β-ΔPH [B]) and GFP alone (C) are diffusely distributed. Bar, 20 μm.

Figure 6.

LL5β is associated with borders of developing AChR clusters. C2 myotubes were cultured on a laminin substrate to induce development of complex, branched AChR aggregates, and then double-stained with anti-LL5β and rBTX. (A) LL5β is colocalized with AChRs in simple plaques but concentrated at the circumference of the clusters. (B and E) Peripheral distribution of LL5β becomes more marked as clusters become perforated and then branched. Bracketed region in E is shown at higher magnification in F. (C and D) In some cases, LL5β is most concentrated at the borders of the perforations that transform plaques into annuli and then into branched aggregates. (F) Within aggregates, AChRs and LL5β have complementary distributions. (G) High magnification view of an NMJ from a P7 mouse (Fig. 4 F) showing similar complementarity in vivo. Bars: (A–E) 10 μm; (F and G) 1 μm.

Figure 7.

LL5β is required for AChR clustering. (A) COS cells were cotransfected with GFP-LL5β plus control duplex VIII siRNA (left), LL5β-RNAi#1 (center), or LL5β-RNAi#2 (right), and then examined for GFP fluoresence. RNAi#1 and #2 suppressed GFP-LL5β expression. (B and C) Myotubes cultured on laminin (Fig. 6) were transfected with the indicated RNAi vectors, or control RNAi vector, fused to form myotubes, and then stained with rBTX. Both LL5β RNAi's decreased the number of AChR aggregates, whereas control RNAi (a sequence that effectively suppressed GFP) had no effect. (C) Examples are shown in B and data from one experiment are graphed (C; 15–10× microscope fields were counted per culture). The difference between LL5β RNAi and either control is significant at P < 0.001 by t test. Similar results were obtained in two additional experiments. Western blots of myotube lysates, using anti-LL5β antibody 7E3, are shown in C (bottom). Densitometry showed that RNAi#1 and #2 decreased LL5β protein levels by 40–50% compared with mock-transfected and control RNAi-transfected cultures. Error bars show mean ± SEM. (D) GFP-transfected myotubes were stained with anti-GFP to show high transfection efficiency. Only the brightest myotubes would appear positive for endogenous GFP fluorescence. Inset shows mock-transfected myotubes stained with anti-GFP. Bar, 100 μm.

Figure 8.

Overexpression of LL5β perturbs AChR clustering. Myoblasts cultured on laminin (Fig. 6) were transfected with the indicated constructs, fused to form myotubes, and then stained with rBTX. (A) Few large AChR clusters are present in myotubes that overexpress GFP-LL5β; instead, receptors are present diffusely and in small patches (arrows). (B and C) Forced expression of GFP (B) or a GFP-LL5β fusion protein lacking the PH domain (GFP-LL5β-ΔPH; C) has no effect on the frequency of aggregation. Bar, 100 μm. (D) Percent of transfected (strongly fluorescent) or untransfected myotubes with complex AChR aggregates. 80 myotubes were analyzed per condition: 40 in each of two separate experiments. Difference between GFP-LL5β and each other condition is significant at P < 0.001. Error bars show mean ± SEM.

Figure 8.

Overexpression of LL5β perturbs AChR clustering. Myoblasts cultured on laminin (Fig. 6) were transfected with the indicated constructs, fused to form myotubes, and then stained with rBTX. (A) Few large AChR clusters are present in myotubes that overexpress GFP-LL5β; instead, receptors are present diffusely and in small patches (arrows). (B and C) Forced expression of GFP (B) or a GFP-LL5β fusion protein lacking the PH domain (GFP-LL5β-ΔPH; C) has no effect on the frequency of aggregation. Bar, 100 μm. (D) Percent of transfected (strongly fluorescent) or untransfected myotubes with complex AChR aggregates. 80 myotubes were analyzed per condition: 40 in each of two separate experiments. Difference between GFP-LL5β and each other condition is significant at P < 0.001. Error bars show mean ± SEM.