GTC
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE ADVANCED SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Genes to Cells (2007) 12, 1011-1022. doi:10.1111/j.1365-2443.2007.01107.x
© 2007 Blackwell Publishing or its licensors
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Hattori, Y.
Right arrow Articles by Uemura, T.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Hattori, Y.
Right arrow Articles by Uemura, T.

Selective expression of Knot/Collier, a transcriptional regulator of the EBF/Olf-1 family, endows the Drosophila sensory system with neuronal class-specific elaborated dendritic patterns

 Yukako Hattoria, Kaoru Sugimuraa,b and Tadashi Uemura*

Graduate School of Biostudies, Kyoto University, Kyoto 606-8501, Japan


   Abstract
Top
 Abstract
Introduction
 Results
 Discussion
 Experimental procedures
 References
 
Dendritic tree morphology is a hallmark of cellular diversity in the nervous system, and Drosophila dendritic arborization (da) neurons provide an excellent model system to study its molecular basis. The da neurons are classified into four classes I–IV in the order of increasing branching complexity. A transcriptional regulator of the early B-cell factor (EBF)/olfactory 1 (Olf-1) family, Knot (Kn)/Collier (Col) is expressed selectively in class IV neurons, which generate the most expansive and complicated dendritic trees in the four classes. Loss of kn function in class IV neurons greatly reduced the number of their dendritic branches. Conversely mis-expression of kn in classes I and II produced supernumerary higher-order branches, whereas class III-specific short and straight terminal branches was hardly formed by kn mis-expression. Neither kn loss of function nor mis-expression were associated with dramatic alterations in the expression patterns of two other transcriptional regulators, Abrupt (Ab) and Cut (Ct), which play important roles in shaping dendritic trees with distinct class specificity from Kn. In contrast, Kn was necessary and sufficient to drive expression of a gene that encodes a class IV-specific channel protein. Collectively, all of our results suggest that Kn exerts its cell-autonomous function to control the formation, and possibly the function, of class IV-like elaborated dendritic arbors.


   Introduction
Top
 Abstract
 Introduction
Results
 Discussion
 Experimental procedures
 References
 
Neurons develop distinctive dendritic morphologies to receive and process synaptic or sensory inputs (reviewed in Scott & Luo 2001; Whitford et al. 2002; Grueber et al. 2005; Landgraf & Evers 2005). Shapes of dendrites are highly variable from one neuronal type to another, and it has been suggested that this diversity supports the differential processing of information in each type of neurons (Häusser & Mel 2003; Jan & Jan 2003; London & Häusser 2005). Therefore development of class-specific dendritic patterns is an essential process to realize the proper function of the nervous system. Recent studies on dendritic arborization (da) neurons in the Drosophila peripheral nervous system (PNS) have contributed much to shed light on the genetic programs underlying dendritic diversity, as described below.

The embryonic PNS organizes a stereotyped pattern of identified sensory neurons; and da neurons constitute a subfamily of multidendritic (md) neurons (Fig. 1A; Bodmer & Jan 1987; Jan & Jan 1993; Campos-Ortega & Hartenstein 1997). The da neurons grow 2D dendrites underneath the epidermis and on the musculature during late embryonic, larval and pupal stages (Bodmer & Jan 1987; Williams & Truman 2004). The 15 da neurons identified in each abdominal hemisegment are classified into four categories, classes I–IV, in the order of increasing territory size and/or branching complexity (Fig. 1A; Grueber et al. 2002). Class I neurons are characterized by formation of simple comb-like small dendritic trees (Fig. 1B); and class II ones, by simple, but larger arbors. Class III neurons produce short straight branches designated as "spikes" (Fig. 1C,D). In contrast, class IV neurons develop far more complicated and expansive arbors (Fig. 1E; Grueber et al. 2002; Sugimura et al. 2003). Taking advantage of class-specific transgenic GFP markers and in vivo time-lapse recordings, we and others clarified the essential roles of transcriptional factors at post-mitotic stages in generating morphologically distinct dendritic trees (Grueber et al. 2003a; Sugimura et al. 2004). Expression of the BTB-zinc finger protein Abrupt (Ab) is highly restricted to class I, and it is necessary and sufficient to endow da neurons with the class-I-specific dendritic pattern (green in Fig. 1I; Li et al. 2004; Sugimura et al. 2004). The homeodomain protein Cut (Ct), on the other hand, is expressed to different degrees in classes II–IV; and these different levels regulate dendrite growth as well as class-specific terminal branching, in particular class III-specific spike formation (yellow and orange in Fig. 1I; Grueber et al. 2003a). More recently, it has been shown that a number of other transcriptional factors are required for the proper dendritogenesis of da neurons (Kim et al. 2006; Parrish et al. 2006).


Figure 1
View larger version (61K):
[in this window]
[in a new window]

  		Figure 1  Distinct morphological classes of da neurons and selective knot (kn) expression in class IV. (A) A diagram of positions of multidendritic (md) neurons in an abdominal hemisegment of the Drosophila embryonic and larval PNS. Diamonds represent individual dendritic arborization (da) neurons; and triangles, other types of md neurons. Besides all da neurons, only two bipolar dendrite neurons (dbd and vbd) and dmd1 are indicated for simplicity. d, l, v’ and v represent dorsal, lateral, ventral prime and ventral clusters, respectively. Each class of the da neurons is differently colored. Black lines represent fascicles of axons that extend from the neuronal cell bodies. In this and all subsequent figures, anterior is to the left and dorsal is at the top. Adapted from Sugimura et al. (2004) with permission from Elsevier. (B–E) Dendritic morphologies of a class I neuron (ddaE; B), a class III neuron (ddaA; C, D) and the class IV neuron (ddaC; E) in the dorsal cluster at wandering 3rd instar stage. Genotypes were elav-GAL4C155 UAS-mCD8::GFP hs-FLP/+; FRTG13 UAS-mCD8::GFP (B, C and E) and C161-GAL4 UAS-mCD8::GFP/+ (D). (F–H) knot (kn) expression was revealed by in situ hybridization at embryonic stage 14 (F) and stage 16 (G, H). d, v’ and v represent signals of class IV da neurons in individual clusters (see A). Outside the nervous system, kn was expressed in a muscle (arrow in F) and in a dorsal vessel (bracket in F). (I) Dendrite morphologies of da neurons and selective or differential expression of transcription factors. Bars, 50 µm for B–E.

 
In this study, we focused on Knot (Kn)/Collier (Col), a member of the early B-cell factor (EBF)/olfactory 1 (Olf-1) family, which is expressed selectively in class IV da neurons (arrowheads in Fig. 1G,H; Crozatier et al. 1996; Orgogozo et al. 2004; Orgogozo & Grueber 2005). We took two complementary approaches to address the potential role of Kn in shaping class-specific dendritic arbors: one was to find kn loss-of-function phenotypes in class IV neurons and the other was to study effects of kn mis-expression on dendrite morphogenesis of classes I–III neurons. The results of both approaches showed that Kn cell-autonomously endowed class IV neurons with their characteristic elaborated patterns of dendritic branching. Then we studied whether there was a cross-regulatory mechanism at the level of gene expression and/or dendrite morphogenesis between Kn, Ab and Ct. Finally, we searched for target genes of these transcription factors.


   Results
Top
 Abstract
 Introduction
 Results
Discussion
 Experimental procedures
 References
 
Loss of kn function in single class IV da neurons downsized their dendritic arbors and made them less elaborate

In addition to being expressed in class IV da neurons, kn is also expressed in other tissues such as DA3 muscle (Fig. 1F); and it was reported that DA3 muscle was deformed in kn mutant embryos (Crozatier & Vincent 1999). To exclude a possible secondary effect of the muscle phenotype on dendrite development, we generated da neurons homozygous for a strong knKN4 mutation (Nestoras et al. 1997) and selectively labeled those mutant neurons in otherwise heterozygous mature larvae with the mosaic analysis with a repressible cell marker (Fig. 2; MARCM; Lee & Luo 1999).


Figure 2
View larger version (76K):
[in this window]
[in a new window]

  		Figure 2  Loss of kn function made dendritic arbors of class IV neurons less elaborate and downsized them. Wild-type (A) and knKN4 MARCM clones (B–F) of class IV ddaC (A–D), class I ddaD (E) and class III v’pda (F). Class IV ddaC (arrows in the insets of A and B) and class I ddaE (arrowheads in insets of A and B) were verified by co-staining for the clone (green in A–C) and all sensory neurons (magenta in A–C). Cell bodies and dendrite of all da neurons in the dorsal cluster were traced with different colors (C'). (D) Tracings of two mutant ddaC clones. Numbers in A–D are those of branch terminals. Clone genotypes were elav-GAL4C155 UAS-mCD8::GFP hs-FLP/+; FRTG13 UAS-mCD8::GFP (A) and elav-GAL4C155 UAS-mCD8::GFP hs-FLP/+; FRTG13 knKN4 UAS-mCD8::GFP (B–F). Bars, 50 µm for A–F.

 
The kn mutation directed class IV neurons to form far less complicated and less expansive arbors than elaborated by the wild-type neuron (compare Fig. 2A with Fig. 2B–D; n = 8 for kn– clones). The number of branch terminals of the wild-type dorsal class IV neuron, ddaC, is over 500 in the mature larvae (Grueber et al. 2003a); whereas that of the kn mutant neuron was below 250 and variable from one neuron to another (see numbers in Fig. 2A–D). This variability is reminiscent of the phenotypes of ct clones of ddaC (Grueber et al. 2003a) and could be due to perdurance of the wild-type protein, as previously discussed (Lee & Luo 1999). Importantly, individual kn mutant clones appeared to be single-cell clones, which means that those clones were generated by mitotic recombination just before final cell divisions of kn/+ progenitor cells. This finding strongly suggests that Kn was required for formation of elaborate dendritic arbors of class IV and acted in a cell-autonomous manner. The da neurons of other classes did not require Kn for shaping their dendritic arbors properly, as shown by the normal-looking dendritic trees of class I (n = 23; Fig. 2E), class II (n = 22; data not shown) and class III neurons (n = 40; Fig. 2F).

Ectopic expression of kn extensively increased the number of higher-order branches in class I neurons

To realize the characteristic dendrite morphology of classes I–III neurons, is it essential to silence kn in these classes? We addressed this question by mis-expressing kn in those classes from late embryos onwards and throughout larval stages (Figs 3–5). Class I neurons, which had been expressing kn ectopically (kn+ neurons) till the mature 3rd instar stage, generated supernumerary branch terminals and displayed bushy appearances with complete penetrance (compare Fig. 3A,B; n > 30 examined). The total number of terminals had been already increased extensively at the 2nd instar stage, with statistical significance (Fig. 3C vs. Figs 3D and 4D,E); and this phenotypic change was detected even at the 1st instar stage (Fig. 3E,F) and in the embryo (data not shown).


Figure 3
View larger version (59K):
[in this window]
[in a new window]

  		Figure 3  kn mis-expression increased the number of branch terminals in class I neurons. (A–F) Images of control class I da neurons ddaD and ddaE (A, C and E) and the neurons that ectopically expressed kn (B, D and F). Stages of larvae are wandering 3rd instar (A, B), 54–56 h AEL (C, D) and 28–31 h AEL (E, F). Over 20 pairs of neurons of each genotype were observed at each stage. (G, H) Two examples of time-lapse recordings of kn+ ddaE (G) and ddaD (H). Third- and higher-order branches (see Fig. 4) sprang out from pre-existing branches throughout larval development (arrows and arrowheads). The series of images were taken at 31–34, 53–56 and 72–75 h AEL. (I) In addition to branches of the two class I neurons (green arrows in the inset), branches of class IV ddaC (magenta arrow) were also visualized and traced in magenta. The number of ddaC neurons traced was 4. Genotypes of the control and kn+ larvae were +; GAL42-21 UAS-mCD8::GFP/+ and UAS-kn::HA/+; GAL42-21 UAS-mCD8::GFP/+, respectively. Bars, 50 µm for A–D and 10 µm for E–H.

 

Figure 5
View larger version (50K):
[in this window]
[in a new window]

  		Figure 5  Effects of kn mis-expression on dendrite morphology of classes II and III neurons. Control neurons (A, A'C) and kn+ neurons (B, B' D) in wandering 3rd instar larvae. Cell bodies of class II ldaA are indicated in A and B, and their branches were traced in A and B. Brackets in A and B indicate branches of class III ldaB. Branches of control class III ddaA were covered with spikes (arrowheads in C), whereas branches of kn+ ddaA were almost naked (arrowheads in D) with only a small number of longer and curved protrusions (yellow arrow in D). Branches of class II ldaA are indicated by the blue arrow in D. Genotypes of the control and kn+ larvae were +; C161-GAL4 UAS-mCD8::GFP/+ and UAS-kn::HA/+; C161-GAL4 UAS-mCD8::GFP/+, respectively. Bar, 50 µm for A–D.

 
To analyze how the mis-expression made the dendritic arbors much more complicated, we quantified the order of branching complexity in the control and the kn+ class I neurons (Fig. 4A,B'). The mis-expression did not abrogate the comb-like shape of the class I dendritic tree, which was formed by 1st- and 2nd-order branches. The increase in the total number of branch terminals was primarily due to supernumerary 3rd- and higher-order branches, and the contributions of the 1st- and 2nd-order ones, if any, appeared to be inconsiderable. Those additional higher-order branches tended to sprout from lower-order branches at variable angles, and some of them were curved. The wild-type class I neuron becomes almost inactive in branching and stabilizes its overall dendritic shape by mid-larval stages; in contrast, the wild-type class IV continues branch formation throughout larval development (Sugimura et al. 2003). Our time-lapse recordings showed that higher-order branches of the kn+ class I neuron were generated continuously in larvae (Fig. 3G,H), suggesting that they were reminiscent of those of class IV in terms of their morphology and the temporal profile of the appearance.


Figure 4
View larger version (41K):
[in this window]
[in a new window]

  		Figure 4  Quantitative analysis of class I neurons (ddaD and ddaE) that mis-expressed kn alone or both kn and ab. (A–C and A'–C') Images of ddaD and ddaE at 54–56 h AEL (A–C) and their tracings (A'–C'). Control neurons (A, A'), kn+ neurons (B, B') and neurons that co-overexpressed kn and ab (kn+ ab+; C, C'). (A'–C') We focused on dorsal branches that extended from cell bodies to the most dorsal tips and defined them as 1st-order branches (dark blue). Second-order branches (green) are those that sprouted from the 1st-, 3rd- and higher-order branches are traced in magenta. (D–F) The number of branch terminals of ddaD (D) and ddaE (E), and length distributions of 3rd- and higher-order branches of ddaE (F) at 54–56 h AEL. (D, E) Numbers of neurons analyzed are indicated above individual bars. Asterisks represent significant differences between the control and each mis-expression (Student's t-test, P < 0.0001). (F) Distinct length distributions of kn+ ddaE and kn+ ab+ ddaE. The total numbers of branches analyzed are indicated above the bars. For each genotype, data were collected on four neurons. Genotypes of the control and kn+ larvae were as described in the legend of Fig. 3. The genotype of the kn+ ab+ larvae was UAS-kn::HA/+; GAL2-21 UAS-mCD8::GFP/UAS-abL. Bar, 50 µm.

 
We examined whether the formation of the kn+-induced higher-order branches could be suppressed or not by increasing the expression level of ab, which characterizes class I-specific simple dendritic trees. The expression level of Ab in class I was further increased by introducing ab transgene together with kn (Fig. 4C,C'). Under this condition, kn+ab+ ddaE reduced the length of 3rd- and higher-order branches (Fig. 4F). This result showed that Kn and Ab interfered with each other's function. The field size of each dendritic tree can be quantified by delineating a polygon connecting dendritic tips (Grueber et al. 2002). In contrast to the striking increase in the number of branch terminals, the size of the dendritic arbor of kn+ class I neurons did not expand under our experimental condition employed.

kn mis-expression did not endow class I neurons with the class IV-like inhibitory dendro–dendritic interaction

In addition to elaborated and expansive dendrite, class IV neurons are also characterized by their mutual avoidance of both iso- and heteroneuronal dendritic branches; and this allows complete, but minimal overlapping, innervation of the body wall (designated as "tiling"; Wässle et al. 1981; Grueber et al. 2002, 2003b; Sugimura et al. 2003). On the other hand, class I neurons do not show such an inhibitory dendro–dendritic interaction (Grueber et al. 2003b; Sugimura et al. 2003). We explored whether branches of kn+ class I neurons displayed the class IV-like repulsive behaviors or not.

We addressed this question by examining whether the branches of kn+ class I neurons avoided those of nearby class IV or not. This assumption was based on a previous study that when supernumerary class IV neurons are generated in close vicinity, those duplicated neighboring neurons divide up the field that is normally covered by a single cell (Moore et al. 2002). In our mis-expression experiment in class I, branches of class IV ddaC were weakly labeled in the background of strong signals of class I branches, in a small portion of dorsal clusters in mature larvae (Fig. 3I). We showed that branches of kn+ ddaD and ddaE (class I) and those of ddaC (class IV) extensively overlapped with each other (Fig. 3I; n = 4), suggesting that kn mis-expression did not confer on class I, a capacity for recognition of class IV dendrites to avoid cross over. Taken together, kn mis-expression transformed dendritic trees of class I into those of class IV-like in terms of branching complexity and continuous higher-order branch formation, but not in terms of the field size or the avoidance behavior.

Effects of kn mis-expression on classes II and III neurons

In response to mis-expression of kn, not only class I but also class II neurons acquired exuberant branches. By using a subset marker, we could trace relatively easily the dorsal branches of one of the class II neurons, ldaA (Fig. 5A,A',B,B'). It appeared that kn+ ldaA developed far more complicated dendritic arbors than the control branches in terms of both number and total length. On the other hand, kn mis-expression greatly decreased the number of terminal structures that distinguish class III (compare arrowheads in Fig. 5C,D). Short straight "spikes" typically cover long branches of the wild-type class III dendrite; in contrast, branches of kn+ ddaA were almost naked with only a small number of longer and curved protrusions (yellow arrow in Fig. 5D).

Neither kn loss of function nor its mis-expression was associated with dramatic alteration of ab or ct expression pattern in the embryo

Next we addressed whether the dendritic malformations that were described above were associated with alterations of expression patterns of two other transcriptional regulators, Ab and Ct, because formation of dendritic trees of the four classes are differentially affected by loss of function or mis-expression of either ab or ct (Grueber et al. 2003a; Li et al. 2004; Sugimura et al. 2004). Furthermore, some of the reported phenotypes of ab and ct have similarities to, though they are not identical to, kn phenotypes described in this study. For example, mis-expression of ab in class IV simplified its dendritic arbor, as did the loss of ct or kn function (Fig. 2). Loss of ab function in class I resulted in production of supernumerary branches, as did mis-expression of ct or kn (Figs 3 and 4). Moreover, loss of ct function in class III eliminated spikes and so did ab or kn mis-expression (Fig. 5).

It could be that loss of kn function resulted in ectopic expression of ab or loss of ct expression in class IV, which then caused the dendritic simplification; however, this did not seem to be the case (Fig. 6). In kn mutant embryos, the Ab level remained undetected in class IV (brackets of Fig. 6A–F), and Ct expression did not look much altered (data not shown). We also showed that kn mis-expression did not down-regulate Ab in class I in the embryo (arrowheads of Fig. 6G–I); likewise, it did not reduce the level of Ct in class III, either (arrows in Fig. 6J–O). Our results thus suggest the possibility that there may be no strong cross-regulation at the level of embryonic gene expression between kn and ab or ct.


Figure 6
View larger version (57K):
[in this window]
[in a new window]

  		Figure 6  Neither a kn mutation nor kn mis-expression dramatically altered expression patterns of Ab and Ct. (A–F) The wild-type (A–C) and knKN4 homozygous embryos (D–F) were stained for the pan-neuronal marker Elav (A and D, green in C and F) and for Ab (B and E, magenta in C and F). Signals of class I neurons (ddaD and ddaE) are indicated with arrowheads throughout this figure. In the kn mutant, Ab immunoreactivity was hardly detected in classes II–IV neurons as in the wild-type (brackets in B, C, E and F). (G–I) kn+ embryo, in which kn was mis-expressed in all neurons (elav-GAL4C155 UAS-mCD8::GFP/+ or Y; UAS-kn::HA/+), was stained for GFP (G, green in I) and for Ab (H, magenta in I). kn mis-expression did not decrease the level or the number of Abrupt-positive nuclei in the cluster. kn+ embryos of this genotype did not hatch; nevertheless, their dendritic trees looked exuberant compared with those of the control embryo (data not shown). (J–O) Control embryo (elav-GAL4C155 UAS-mCD8::GFP/+ or Y; J–L) and kn+ embryo (elav-GAL4C155 UAS-mCD8::GFP/+ or Y; UAS-kn::HA/+; M–O) were stained for GFP (J and M, green in L and O) and for Ct (K and N, magenta in L and O). kn mis-expression did not alter the normal pattern of Ct staining in the cluster. Arrows point to class III neurons that strongly expressed Ct. The one next to ddaD is ddaF, and the more ventrally located one is ddaA (Grueber et al. 2003a). All of the embryos in this figure were at stage 16. Bar, 10 µm for A–O.

 
Promoter activity of a class IV-specific gene pickpocket1 (ppk1) was regulated by Kn and Ab in positive and negative ways, respectively

As one way to search for target genes of the transcription factors, we conducted a microarray analysis and compared a gene expression profile of wild-type embryos and that of embryos that mis-expressed ab in all neurons (see details in Experimental procedures). ppk1 was included in a group of genes whose expression level was reduced by fourfold in ab+ embryos (Supplementary Table S1). ppk1 is expressed selectively in class IV, encodes a member of the degenerin/epithelial sodium channel, and coordinates larval locomotion (Adams et al. 1998; Ainsley et al. 2003).

To verify if ppk1 expression was negatively regulated by Ab, we employed a ppk1–EGFP transgenic line that has a fusion gene of a ppk1 cis-element with the EGFP coding sequence and expresses EGFP selectively in class IV neurons (Fig. 7A,C,E–G; Grueber et al. 2003b). The ab mis-expression suppressed the ppk1–EGFP expression in all of over 25 class IV neurons observed (Fig. 7B). In contrast to ab, kn was necessary for ppk1–EGFP expression in class IV (Fig. 7C,D). In kn homozygous embryos, class IV neurons were still born as shown by our Ct staining (data not shown); however, ppk1–EGFP signals of class IV were very faint or hardly detected in all segments in seven mutant embryos examined. Furthermore kn was sufficient to drive ppk1–EGFP expression, as shown by the fact that ppk1–EGFP signals were detected in 8 out of 29 kn+ ddaD and in 16 out of 28 kn+ ddaE (arrowheads in Fig. 7E–J). Thus, it is likely that ppk1 is one of the target genes of Kn.


Figure 7
View larger version (67K):
[in this window]
[in a new window]

  		Figure 7  ppk1 expression was under the control of Ab and Kn. (A, B) Control wandering 3rd instar larva (A; ppk-EGFP, UAS-abS/+) and a larva that mis-expressed ab in all da neurons (B; 109(2)80/+; ppk-EGFP UAS-abS/+). ppk-EGFP expression was abolished by ab mis-expression. (C, D) Control embryo (ppk-EGFP/ppk-EGFP) and kn mutant embryo (knKN4/knKN4; ppk-EGFP/ppk-EGFP) at 18–20 h AEL. v and v' indicate class IV neurons in each cluster (see Fig. 1A). ppk-EGFP signals of class IV were hardly visible (D). Images of C and D were acquired and processed with identical parameters. Bar, 25 µm. (E–J) Control larva (E-G) (GAL2-21, UAS-myr-RFP/ppk-EGFP) and a larva that mis-expressed kn in class I da neurons (H–J) (UAS-kn::HA/+; GAL2-21 UAS-myr-RFP/ppk-EGFP) at wandering 3rd instar stage. Images of RFP (E and H), those of EGFP (F and I), and merged images (G and J) are shown. Bar, 50 µm.

 

   Discussion
Top
 Abstract
 Introduction
 Results
 Discussion
Experimental procedures
 References
 
To address mechanisms underlying the distinctive modes of dendrite development of da neurons, we studied the roles of kn, which is selectively expressed in the class IV neuron. The kn mis-expression in classes I and II neurons produced supernumerary higher-order branches. One possible interpretation of these results would be that Kn makes dendritic trees complex by promoting branching in a broad range of neuronal types. However, this hypothesis would be difficult to reconcile with our finding that kn mis-expression substantially decreased the number of terminal protrusions of class III, ‘spikes,’ and made the dendritic tree smooth at least in one of class III neurons ddaA. The easiest interpretation of these results, together with the results of MARCM analysis, would be that selective expression of kn in the class IV neuron plays a pivotal role in forming its characteristic dendritic arbors and that development of less complex arbors of classes I–III and spike formation in class III depends on the absence of Kn.

Although our manipulations of kn mis-expression caused severe and reproducible phenotypes, they did not necessarily provide evidence for almost complete morphological transformation from classes I–III into class IV neurons. The partial alteration of the arbor patterns might be due to a late onset and/or a low level of kn transgene expression obtained by using the available post-mitotic drivers. Another non-mutually exclusive possibility would be that Kn is not sufficient enough to confer all the features of class IV dendrite, including the expansive territorial field and the avoidance behavior between branches, on non-class IV neurons, in particular on class I that maintained endogenous expression of Ab. This speculation is analogous to a previous proposal about how Kn affects muscle development (Crozatier & Vincent 1999). Kn is expressed in founders of DA3 muscle, and thin muscle fibers are formed at the position of DA3 in kn mutant embryos; but mis-expression of kn does not transform other muscles into DA3-like ones. A working hypothesis of DA3 formation postulates the combined activity of Kn and at least one other muscle-identity factor. In the context of normal development of da neurons, Kn-independent mechanisms may work in class IV. Such mechanisms might be under the control of other transcriptional factors, Ct and Spineless (Ss), which have been shown to contribute to class IV dendrite morphogenesis. This and previous studies suggest that there may be no strong cross-regulation at the level of embryonic gene expression between kn, ct and ss, and that each transcriptional factor may start exerting its cell-autonomous function to control shaping the dendrite by modulating target genes (Grueber et al. 2003a; Li et al. 2004; Sugimura et al. 2004; Kim et al. 2006). This hypothesis could be tested, for example, by examining if kn mis-expression may better transform classes I to IV-like in ab mutant class I da neurons.

We searched for target genes of Kn, and our results suggested that ppk1 may be one of the candidates. However, it was reported that ppk1 mutant larvae show a defect in locomotion, but not in neuronal morphology (Ainsley et al. 2003), suggesting the possibility that target genes of Kn may include a group of genes controlling physiological functions of class IV neurons and others responsible for the dendritic morphogenesis. When the two class-selective transcription factors Kn and Ab were examined at the level of dendritic morphology as a final read-out, we found that they could interfere with each other's activity upon over-expression. Therefore target genes of Kn may be partially overlapped with those of Ab, and the interference could be at least partially due to competition between the two for cis regulatory elements of the same target gene. Actions of Kn and Ab on ppk1 expression were consistent with this hypothesis.

Each member of the early B-cell factor (EBF)/olfactory 1 (Olf-1) family possesses a helix-loop-helix domain necessary for its dimerization and a distinct DNA-binding domain (reviewed by Hagman & Lukin 2006). In neuronal development, roles of EBF members have been studied in a couple of systems: EBF members are responsible for switching gene expression in post-mitotic cells that leave the subventricular zone on the way to the mantle in vertebrate CNS development (Garel et al. 1999; Garcia-Dominguez et al. 2003). In Caenorhabditis elegans, high-level expression of Unc-3 occurs as ventral cord motor neurons begin to project axons to target tissues; and unc-3 mutant worms show severe defasiculation defects (Prasad et al. 1998). It would be intriguing to address whether these EBF members also play pivotal roles in dendrite morphogenesis of defined subsets of neurons in those model animals.

Genetic studies on several putative transcriptional factors of other families have shown that transcriptional regulation is a common mechanism to generate morphological diversity of dendrites of several neuronal types (Moore et al. 2002; Komiyama et al. 2003; Vrieseling & Arber 2006). Therefore, it is likely that class-specific profiles of gene expression controlled by these factors are responsible for distinctive neuronal morphogenesis. Although it is still challenging to identify target genes of these transcriptional regulators, the da neuron is an appropriate model system to pursue how morphological and physiological characteristics are coordinately regulated by the transcriptional mechanism and to eventually elucidate how different types of neurons play unique roles in neural circuits.


   Experimental procedures
Top
 Abstract
 Introduction
 Results
 Discussion
 Experimental procedures
References
 
Molecular biology

A kn cDNA clone, RE03728, was purchased from Invitrogen. For expression in Drosophila, the coding sequence of kn was amplified using a pair of primers and cloned into pUAST carrying three copies of the HA tag (Brand & Perrimon 1993; Niwa et al. 2002). The pUAST plasmids were microinjected into w1118 embryos by using standard techniques to generate transgenic strains (Spradling & Rubin 1982). To study kn expression patterns, we used RE03728 as a template to prepare a probe for in situ hybridization (Lehmann & Tautz 1994).

Drosophila strains

To drive kn expression in class I da neurons and in all post-mitotic neurons, we used GAL42-21 UAS-mCD8::GFP (Grueber et al. 2003a) and elav-GAL4c155 UAS-mCD8::GFP hsFLP (Lee & Luo 1999), respectively. Compared to elav-GAL4c155, which initiates transgene expression at an early post-mitotic phase (stage 13 or earlier), GAL42-21 started driving transgene expression at stage 15 or later stages. For mis-expression in classes II and III, C161-GAL4 mCD8::GFP was used (Williams & Truman 2004).

knKN4 is a strong embryonic lethal allele (Nestoras et al. 1997). We mated the following pairs: females of elav-GAL4c155 UAS-mCD8::GFP hsFLP; tub-Gal80 FRTG13 were mated to males of FRTG13 UAS-mCD8::GFP/FRTG13 UAS-mCD8::GFP to generate control clones, and were mated to males of knKN4 FRTG13 UAS-mCD8::GFP/CyO to generate kn mutant clones. A protocol for FLP induction by heat shocks was previously described (Grueber et al. 2002). Other strains used were UAS-abL and UAS-abS (Sugimura et al. 2004), 109(2)80-GAL4 (Gao et al. 1999), ppk-EGFP (Grueber et al. 2003b) and UAS-myr-mRFP (Bloomington Stock Center, Bloomington, Indiana). All fly embryos and larvae were grown at 25 °C.

Image collection of dendritic trees and quantification

Imaging, single cell labeling, time-lapse recordings and quantitative analysis of the images were essentially done as described (Grueber et al. 2002; Sugimura et al. 2003, 2004). Because dendrites extend on approximately 2D planes underneath the epidermis, Z-series files were projected into 2D images, which were then used for quantification of dendritic morphologies. Total length of dendritic branches was measured by using IMAGEJ. Data were presented as means ± SD.

Immunochemistry

Whole-mount embryos and dissected larvae were stained according to standard protocols with the following primary antibodies: rabbit anti-Ab (Hu et al. 1995), mouse anti-Ct (Blochlinger et al. 1990; 2B10 from Developmental Studies Hybridoma Bank [DSHB] at the University of Iowa), mouse 22C10 (Fujita et al. 1982; DSHB), rat anti-Elav (7E8A10; DSHB), mouse anti-GFP (Santa Cruz Biotechnology, Santa Cruz, CA), rabbit anti-GFP (Molecular Probes, Eugene, OR) and rat anti-CD8 (CALTAG Laboratories, Burlingame, CA).

Microarray analysis

The genotype of an ab-overexpressing animal and that of a control were elav-GAL43E1/UAS-abL, UAS-abS and elav-GAL43E1/+, respectively. Total RNA was extracted from 130 embryos at 18.5–19.5 h AEL by using RNeasy mini kit (QIAGEN). Gene Chip Drosophila Genome 2.0 array (Affymetrix) was used, and the detailed experimental procedure was carried out as previously described (De Gregorio et al. 2001), except for that we started with 5 µg of total RNA instead of poly-A RNA. Quality of the analysis was confirmed according to the manufacturer's manual. Data were analyzed by using GENECHIP Operating Software (GCOS) 1.0 (Affymetrix).


   Acknowledgements
 
We thank Jym Mohler, Yuh Nung Jan, Wes Grueber, Akinao Nose, Darren Williams, the Bloomington Stock Center and Drosophila Genetic Resource Center (Kyoto) for Drosophila stocks; and Yo-ichi Nabeshima for giving us the opportunity to perform the microarray analysis in his facility. We are also grateful to Developmental Studies Hybridoma Bank for antibodies and Adrian Moore for communicating prior to publication. This work was supported by grants from the programs Grants-in-Aid for Scientific Research on Priority Areas-Molecular Brain Science of the MEXT of Japan (17024025 to T. U.). K.S. was supported by a JSPS Research Fellowship for Young Scientists.


   Footnotes
 
Communicated by: Kozo Kaibuchi

aThese authors contributed equally to this work. Back

bPresent address: Laboratory for Cell Function Dynamics, Advanced Technology Development Group, Brain Science Institute, RIKEN, Wako 351-0198, Japan. Back

* Correspondence: E-mail: tauemura{at}lif.kyoto-u.ac.jp


   References
Top
 Abstract
 Introduction
 Results
 Discussion
 Experimental procedures
 References
 
Adams, C.M., Anderson, M.G., Motto, D.G., Price, M.P., Johnson, W.A. & Welsh, M.J. (1998) Ripped pocket and pickpocket, novel Drosophila DEG/ENaC subunits expressed in early development and in mechanosensory neurons. J. Cell Biol. 140, 143–152.[Abstract/Free Full Text]

Ainsley, J.A., Pettus, J.M., Bosenko, D., Gerstein, C.E., Zinkevich, N., Anderson, M.G., Adams, C.M., Welsh, M.J. & Johnson, W.A. (2003) Enhanced locomotion caused by loss of the Drosophila DEG/ENaC protein Pickpocket1. Curr. Biol. 13, 1557–1563.[CrossRef][Medline]

Blochlinger, K., Bodmer, R., Jan, L.Y. & Jan, Y.N. (1990) Patterns of expression of cut, a protein required for external sensory organ development in wild-type and cut mutant Drosophila embryos. Genes Dev. 4, 1322–1331.[Abstract/Free Full Text]

Bodmer, R. & Jan, Y.N. (1987) Morpohological differentiation of the embryonic peripheral neurons in Drosophila. Roux's Arch. Dev. Biol. 196, 69–77.[CrossRef]

Brand, A.H. & Perrimon, N. (1993) Targeted gene expression as a means of altering cell fates and generating dominant phenotypes. Development 118, 401–415.[Abstract]

Campos-Ortega, J.A. & Hartenstein, V. (1997) The Embryonic Development of Drosophila melanogaster. Berlin: Springer.

Crozatier, M. & Vincent, A. (1999) Requirement for the Drosophila COE transcription factor Collier in formation of an embryonic muscle: transcriptional response to notch signalling. Development 126, 1495–1504.[Abstract]

Crozatier, M., Valle, D., Dubois, L., Ibnsouda, S. & Vincent, A. (1996) Collier, a novel regulator of Drosophila head development, is expressed in a single mitotic domain. Curr. Biol. 6, 707–718.[CrossRef][Medline]

De Gregorio, E., Spellman, P.T., Rubin, G.M. & Lemaitre, B. (2001) Genome-wide analysis of the Drosophila immune response by using oligonucleotide microarrays. Proc. Natl. Acad. Sci. USA 98, 12590–12595.[Abstract/Free Full Text]

Fujita, S.C., Zipursky, S.L., Benzer, S., Ferrus, A. & Shotwell, S.L. (1982) Monoclonal antibodies against the Drosophila nervous system. Proc. Natl. Acad. Sci. USA 79, 7929–7933.[Abstract/Free Full Text]

Gao, F.B., Brenman, J.E., Jan, L.Y. & Jan, Y.N. (1999) Genes regulating dendritic outgrowth, branching, and routing in Drosophila. Genes Dev. 13, 2549–2561.[Abstract/Free Full Text]

Garcia-Dominguez, M., Poquet, C., Garel, S. & Charnay, P. (2003) Ebf gene function is required for coupling neuronal differentiation and cell cycle exit. Development 130, 6013–6025.[Abstract/Free Full Text]

Garel, S., Marin, F., Grosschedl, R. & Charnay, P. (1999) Ebf1 controls early cell differentiation in the embryonic striatum. Development 126, 5285–5294.[Abstract]

Grueber, W.B., Jan, L.Y. & Jan, Y.N. (2002) Tiling of the Drosophila epidermis by multidendritic sensory neurons. Development 129, 2867–2878.[Medline]

Grueber, W.B., Jan, L.Y. & Jan, Y.N. (2003a) Different levels of the homeodomain protein cut regulate distinct dendrite branching patterns of Drosophila multidendritic neurons. Cell 112, 805–818.[CrossRef][Medline]

Grueber, W.B., Yang, C.H., Ye, B. & Jan, Y.N. (2005) The development of neuronal morphology in insects. Curr. Biol. 15, R730–R738.[CrossRef][Medline]

Grueber, W.B., Ye, B., Moore, A.W., Jan, L.Y. & Jan, Y.N. (2003b) Dendrites of distinct classes of Drosophila sensory neurons show different capacities for homotypic repulsion. Curr. Biol. 13, 618–626.[CrossRef][Medline]

Hagman, J. & Lukin, K. (2006) Transcription factors drive B cell development. Curr. Opin. Immunol. 18, 127–134.[CrossRef][Medline]

Häusser, M. & Mel, B. (2003) Dendrites: bug or feature? Curr. Opin. Neurobiol. 13, 372–383.[CrossRef][Medline]

Hu, S., Fambrough, D., Atashi, J.R., Goodman, C.S. & Crews, S.T. (1995) The Drosophila abrupt gene encodes a BTB-zinc finger regulatory protein that controls the specificity of neuromuscular connections. Genes Dev. 9, 2936–2948.[Abstract/Free Full Text]

Jan, Y.N. & Jan, L.Y. (1993) The peripheral nervous system. In: The Development of Drosophila melanogaster (eds M. Bate & A. Martinez-Arias), pp. 1207–1244. Cold Spring Harbor, New York: Cold Spring Harbor Laboratory Press.

Jan, Y.N. & Jan, L.Y. (2003) The control of dendrite development. Neuron 40, 229–242.[CrossRef][Medline]

Kim, M.D., Jan, L.Y. & Jan, Y.N. (2006) The bHLH-PAS protein Spineless is necessary for the diversification of dendrite morphology of Drosophila dendritic arborization neurons. Genes Dev. 20, 2806–2819.[Abstract/Free Full Text]

Komiyama, T., Johnson, W.A., Luo, L. & Jefferis, G.S. (2003) From lineage to wiring specificity. POU domain transcription factors control precise connections of Drosophila olfactory projection neurons. Cell 112, 157–167.[CrossRef][Medline]

Landgraf, M. & Evers, J.F. (2005) Control of dendritic diversity. Curr. Opin. Cell Biol. 17, 690–696.[CrossRef][Medline]

Lee, T. & Luo, L. (1999) Mosaic analysis with a repressible cell marker for studies of gene function in neuronal morphogenesis. Neuron 22, 451–461.[CrossRef][Medline]

Lehmann, R. & Tautz, D. (1994) In situ hybridization to RNA. Methods Cell Biol. 44, 575–598.[Medline]

Li, W., Wang, F., Menut, L. & Gao, F.B. (2004) BTB/POZ-zinc finger protein abrupt suppresses dendritic branching in a neuronal subtype-specific and dosage-dependent manner. Neuron 43, 823–834.[CrossRef][Medline]

London, M. & Häusser, M. (2005) Dendritic computation. Annu. Rev. Neurosci. 28, 503–532.[CrossRef][Medline]

Moore, A.W., Jan, L.Y. & Jan, Y.N. (2002) hamlet, a binary genetic switch between single- and multiple-dendrite neuron morphology. Science 297, 1355–1358.[Abstract/Free Full Text]

Nestoras, K., Lee, H. & Mohler, J. (1997) Role of knot (kn) in wing patterning in Drosophila. Genetics 147, 1203–1212.[Abstract]

Niwa, R., Nagata-Ohashi, K., Takeichi, M., Mizuno, K. & Uemura, T. (2002) Control of actin reorganization by Slingshot, a family of phosphatases that dephosphorylate ADF/cofilin. Cell 108, 233–246.[CrossRef][Medline]

Orgogozo, V. & Grueber, W.B. (2005) FlyPNS, a database of the Drosophila embryonic and larval peripheral nervous system. BMC Dev. Biol. 5, 4.[CrossRef][Medline]

Orgogozo, V., Schweisguth, F. & Bellaiche, Y. (2004) Slit-Robo signalling prevents sensory cells from crossing the midline in Drosophila. Mech. Dev. 121, 427–436.[CrossRef][Medline]

Parrish, J.Z., Kim, M.D., Jan, L.Y. & Jan, Y.N. (2006) Genome-wide analyses identify transcription factors required for proper morphogenesis of Drosophila sensory neuron dendrites. Genes Dev. 20, 820–835.[Abstract/Free Full Text]

Prasad, B.C., Ye, B., Zackhary, R., Schrader, K., Seydoux, G. & Reed, R.R. (1998) unc-3, a gene required for axonal guidance in Caenorhabditis elegans, encodes a member of the O/E family of transcription factors. Development 125, 1561–1568.[Abstract]

Scott, E.K. & Luo, L. (2001) How do dendrites take their shape? Nat. Neurosci. 4, 359–365.[CrossRef][Medline]

Spradling, A.C. & Rubin, G.M. (1982) Transposition of cloned P elements into Drosophila germ line chromosomes. Science 218, 341–347.[Abstract/Free Full Text]

Sugimura, K., Satoh, D., Estes, P., Crews, S. & Uemura, T. (2004) Development of morphological diversity of dendrites in Drosophila by the BTB-zinc finger protein Abrupt. Neuron 43, 809–822.[CrossRef][Medline]

Sugimura, K., Yamamoto, M., Niwa, R., Satoh, D., Goto, S., Taniguchi, M., Hayashi, S. & Uemura, T. (2003) Distinct developmental modes and lesion-induced reactions of dendrites of two classes of Drosophila sensory neurons. J Neurosci. 23, 3752–3760.[Abstract/Free Full Text]

Vrieseling, E. & Arber, S. (2006) Target-induced transcriptional control of dendritic patterning and connectivity in motor neurons by the ETS gene Pea3. Cell 127, 1439–1452.[CrossRef][Medline]

Wässle, H., Peichl, L. & Boycott, B.B. (1981) Dendritic territories of cat retinal ganglion cells. Nature 292, 344–345.[Medline]

Whitford, K.L., Dijkhuizen, P., Polleux, F. & Ghosh, A. (2002) Molecular control of cortical dendrite development. Annu. Rev. Neurosci. 25, 127–149.[CrossRef][Medline]

Williams, D.W. & Truman, J.W. (2004) Mechanisms of dendritic elaboration of sensory neurons in Drosophila: insights from in vivo time lapse. J. Neurosci. 24, 1541–1550.[Abstract/Free Full Text]

Accepted: 4 June 2007




This article has been cited by other articles:


Home page
 DevelopmentHome page
M. M. Corty, B. J. Matthews, and W. B. Grueber
Molecules and mechanisms of dendrite development in Drosophila
Development, April 1, 2009; 136(7): 1049 - 1061.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Hattori, Y.
Right arrow Articles by Uemura, T.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Hattori, Y.
Right arrow Articles by Uemura, T.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE ADVANCED SEARCH TABLE OF CONTENTS