Figure 1.
The Dementin Family of proteins.
Dendrogram (A) and schematic (B) showing the sequence similarity and structural organization of Dementin and related proteins and their Unigene codes. “CC” represents domains predicted to form coiled-coil structures, and “TM” represents regions predicted to form transmembrane domains. The percent similarity of the domains conserved between TMCC2 and Dementin are indicated by brackets. See Fig. S1 for alignments. (C) Structure of the Drosophila dementin gene from flybase.org showing the locations of the enhancer trap (dmtn-Gal4) and the lethal mutation (dmtn1) used in this study; coding exons are in filled boxes and non-coding exons in open boxes, the intron indicated by a dashed line is not to scale.
Table 1.
Identities of dementin-related genes in human, mouse, D. melanogaster, C. elegans and C. intestinalis.
Figure 2.
Expression pattern of Dementin in the brain.
(A) Schematic of the adult optic lobe. (B) Immunoblot of preparations from the heads of young adult flies which have RNAi-mediated suppression of Dementin in both glia and neurons (repo- and elav-Gal4), glia alone (repo-Gal4), neurons alone (elav-Gal4), or using dmtn-Gal4. Each lane was loaded with the equivalent of 3.5 heads prepared from pools of 20–30 heads each. (C and D) Expression of dmtn-Gal4 driven GFPnls (green) in the optic lobes of white pupae. (C) GFP was detected in the cell body (inset 1) and axons (inset 2) of retinal neurons detected by antibody mAb24b10 against Chaoptin (magenta). GFP was also detected in retinal axons in brains stained for Repo (inset 3, arrowhead). GFP co-localized with the glial transcription factor Repo detected using antibody 8D12 (magenta). See Figs. S4A and S4B for sections of the optic lobes of larvae, animated in movies S1 and S2. (E and F) Expression of dmtn-Gal4 in the adult optic lobe. (E) Detection of GFPnls (green) in glia and neurons of the adult optic lobe in a homozygous dmtn-Gal4 fly stained with anti-Repo (magenta). Panel 1, oblique optical section; panel 2, vertical optical section. Panels 3 and 4, examples of dmtn-Gal4-driven cell-surface targeted mCD8::GFP (green) using MARCM in neurons of adult optic lobes stained with anti-Chaoptin (magenta). La, lamina; Lo, lobula complex; Me, medulla; Re, retina; cg, cortex glia; lg, lamina glia; sg, surface glia; xg, chiasm glia; ln, lamina neurons, lon lobula neurons; mn, medulla neurons. See Fig. S2C for an image of the adult central brain.
Figure 3.
Genetic interaction of Dementin with human APP.
(A) eclosion rates and wing expansion phenotypes. The image is of a fly expressing human APP in neurons, showing unexpanded wings. % M, percent of flies eclosing that were male. (B) Immunoblot and densitometric quantification prepared from the heads of male flies expressing human APP (UAS-APP), which also have RNAi for Dementin (UAS-RNAi), or which ectopically express Dementin (UAS-dmtn) driven by elav-Gal4. Each sample was prepared from 30 heads and the equivalent of 3 heads was loaded per lane. Replicate immunoblots are shown in Fig. S2G. Densitometry results are expressed as a proportion of the result obtained from flies only expressing APP, corrected by loading control. APP was detected with antibody A5137 targeted against the C-terminus of APP. l.c., loading control. Data are from 10 to 12 independent crosses.
Figure 4.
Alteration of APPL metabolism both by RNAi for Dementin and by mutation of dementin.
(A) Western blot of heads from 10-day-old flies with RNAi for Dementin driven in neurons shows the appearance of a novel ∼50 kDa APPL-derived C-terminal fragment (APPLf). c1 and c2, control parent lines: UAS-Dcr2;+;UAS-RNAi/TM6B and elav-Gal4;+;+; RNAi, progeny with RNAi for Dementin: Dcr2/elav-Gal4; +; UAS-RNAi/+. Each lane was loaded with equivalent of 3 heads per lane from preparations made with 20 heads each. (B) Western blot of heads from dmtn1 flies: dmtn1 also induced production of the ∼50 kDa APPLf. Each lane is loaded with the equivalent of 4 heads from preparations made from 20–40 fly heads collected on the day of their eclosion. All flies were 9 generations isogenic. APPL was detected in all blots using antibody dR-14, the APPLf was also detected in each case with antibody dC-21 (not shown). See Fig. S2E for replicates. (C) Western blot for Dementin from the same samples showing a truncated Dementin peptide. l.c., loading control. (D) Densitometric quantification of (B) showing that dmtn1 does not affect total levels of APPL or the ratio of holoAPPL to sAPPL, but does induce the production of APPLf (p<0.0001). Values for total APPL and the ratio of sAPPL to sAPPL+hAPPL are expressed as relative to that obtained with control w1118 flies (left-hand scale); the percent of total APPL that was APPLf is shown on the right-hand scale. (E) Wings of homozygous dmtn1 flies (left), but not heterozygous controls (right, dmtn1/TM3 Ser) fail to expand after eclosion. Both images were captured 3 h after eclosion. Error bars represent the SEM.
Figure 5.
The effect of Dementin on survival and optic lobe development.
(A) dmtn1 is a pupal lethal allele. Bars represent the percentage of dmtn1 flies that remain alive at various stages of development. (B) The survival rate of adult dmtn1 flies and controls; all flies were 9 generations or more isogenic. (C and D) dmtn1 flies have an underdeveloped medulla. Retinal axons projections into the optic lobe of heterozygous (C) and homozygous (D, n>26) dmtn1 flies expressing GFPnls under control of repo-Gal4 (green) were stained with mAb24b10 against Chaoptin (magenta). (E) RNAi for Dementin driven in glia with repo-Gal4 and with dmtn-Gal4, but not in neurons with elav-Gal4 is developmentally lethal. For a Western blot of extracts prepared from the heads of these flies see Figure 2B. (F and G) Optic lobes stained with antibody mAb24b10 for retinal axons show that RNAi for Dementin driven in glia with repo-Gal4 (F, n = 8), but not in neurons with elav-Gal4 (G, n = 12), recapitulates the medulla defects seen in dmtn1 flies. La, lamina; Me, medulla.
Figure 6.
Rescue of dmtn1 flies by wild-type Dementin.
The optic lobe defect of dmtn1 flies is fully rescued by expression of Dementin in neurons, and partly by expression in glia. (A) Eclosion rate relative to the expected Mendelian ratio of homozygous dmtn1 flies where wild-type Dementin expression was driven from a UAS-dmtn construct in either glia (repo-Gal4), neurons (elav-Gal4), or in both. (B) Western blot showing expression of Dementin in the same flies. Each lane was loaded with the equivalent of 3 heads prepared from pools of 20–30 heads. l.c., loading control. Corresponding data for Dementin in w1118 and dmtn1 flies is shown in Fig. 4C. (C–F) Representative images of retinal axon projections stained with mAb24b10 in the optic lobe of homozygous dmtn1 flies not rescued by wild-type Dementin (C, n>26), rescued by wild-type dementin specifically in glia (D, n = 11), neurons (E, n = 20) or in both glia and neurons (F, n = 12). The arrow indicates incorrectly targeted retinal axons in dmtn1 flies rescued only in glia. La, lamina; Me, medulla.
Figure 7.
AD-like pathology and neurodegeneration in rescued dmtn1 flies. dmtn1 flies whose neurons express only Dmtn1 showed mislocalization of Futsch/MAP1B (A, enlarged in A′), whereas flies whose glia express only Dmtn1 (B) did not; neither did dmtn1 flies expressing wild-type Dementin in both neurons and glia (C).
Quantification of these data is shown in panel D. All flies whose neurons express only Dmtn1 (L) showed patchy staining for Bruchpilot in the lobula (arrows), whereas flies whose glia express only Dmtn1 did not (M); neither did flies expressing wild-type Dementin in both neurons and glia (N). dmtn1 flies whose neurons express only Dmtn1 further developed vacuoles (E, enlarged in E′), whereas dmtn1 flies expressing wild-type Dementin in either neurons only, or in both neurons and glia did not (F, G, quantified in K). (H) Western blot and quantification of Futsch/MAP1B in rescued flies 1 day after eclosion and 10 days later showing that dmtn1 flies neurons express only Dmtn1, have an age-related decline in Futsch levels, see also Fig. S2E; error bars are the SEM. This was accompanied by an age-related increase in APPLf levels in dmtn1 flies expressing wild-type Dementin in glia only but not in those expressing wild-type Dementin in neurons only (I and J). dmtn1 flies whose neurons express only Dmtn1 (O) showed deterioration of synaptic staining whereas flies whose glia express only Dmtn1 did not (P), neither did flies expressing wild-type Dementin in both neurons and glia (Q). (R) dmtn1 flies whose neurons express only Dmtn1 (n = 118) had shorter lifespans than those who glia express only Dmtn1 (n = 138), or with wild-type Dementin in both neurons and glia (n = 54). Images are representative immunostained paraffin sections from 10–20 rescued dmtn1 flies in each group fixed on the day after eclosion indicated and stained as indicated.