Long-term activity drives dendritic branch elaboration of a C. elegans sensory neuron

Neuronal activity often leads to alterations in gene expression and cellular architecture. The nematode Caenorhabditis elegans, owing to its compact translucent nervous system, is a powerful system in which to study conserved aspects of the development and plasticity of neuronal morphology. Here we focus on one sensory neuron in the worm, termed URX, which senses oxygen and signals tonically proportional to environmental oxygen. Previous studies have reported that URX has variable branched endings at its dendritic sensory tip. By controlling oxygen levels and analyzing mutants, we found that these branched endings grow over time as a consequence of neuronal activity. Furthermore, we observed that the branches contain microtubules, but do not appear to harbor the guanylyl cyclase GCY-35, a central component of the oxygen sensory transduction pathway. Interestingly, we found that although URX dendritic tips grow branches in response to long-term activity, the degree of branch elaboration does not correlate with oxygen sensitivity at the cellular or the behavioral level. Given the strengths of C. elegans as a model organism, URX may serve as a potent system for uncovering genes and mechanisms involved in activity-dependent morphological changes in neurons.

ABSTRACT 23 Neuronal activity often leads to alterations in gene expression and cellular architecture. 24 The nematode Caenorhabditis elegans, owing to its compact translucent nervous 25 system, is a powerful system in which to study conserved aspects of the development 26 and plasticity of neuronal morphology. Here we focus on one sensory neuron in the 27 worm, termed URX, which senses oxygen and signals tonically proportional to 28 environmental oxygen. Previous studies have reported that URX has variable branched 29 endings at its dendritic sensory tip. By controlling oxygen levels and analyzing mutants, 30 we found that these branched endings grow over time as a consequence of neuronal 31 activity. Furthermore, we observed that the branches contain microtubules, but do not  45 The nervous system often displays morphological plasticity in response to 46 prolonged input or activity. These activity-dependent changes in neuron shape allow 47 animals to interact more adeptly with their environment. For instance, the growth and 48 pruning of specific synapses as well as axon and dendritic branches allow neural 49 circuits to alter synaptic weighting during forms of learning and homeostatic plasticity [1, 50 2]. Interneurons also adjust the number and shape of their minute dendritic spines to 51 filter input differently in neuronal networks [3][4][5]. In the sensory system, photoreceptor 52 outer segment length has been shown to change in response to different light levels [6]. 53 Thus, although the gross structure of the adult nervous system often remains static, 54 many neurons change shape at subtle spatial and temporal scales. 55 The transparency, genetic tractability, and compact nervous system of the 56 nematode C. elegans make the worm an excellent system to study genes that underlie 57 how neurons achieve and adjust their shape. Many aspects of neuronal morphology 58 have been examined in C. elegans, such as axonal and dendritic establishment [7,8], 59 dendritic tiling [9], synapse specification [10], and sensory cilia morphogenesis and 60 maintenance [11,12]. The worm has also been used to study how neurons alter their 61 shape in response to changes in environment, such as the reshaping of the ciliated  113  114  115  116  117  118  119  120  121  122  123  124  125  126  127  128  129  130  131  132  133  134  135  136  137  138  139  140  141  142  143  144  145  146  147  148  149  150  151  152  153  154  155  156  157  158  159  160  161  162  163  164  165  166  167  168 required for the development and maintenance of sensory cilia in C. elegans have 66 conserved roles across species [17,18]. 67 Most sensory neurons in the head of C. elegans are bilaterally symmetric and 68 have a cell body that projects a single dendrite to the tip of the nose, where the sensory 69 transduction machinery is often localized [19]. Here we focus on one of these sensory   81 We report here that continuous exposure to surface level oxygen causes the 82 URX neuron to steadily grow elaborate branches at its dendritic sensory ending over the 83 course of adulthood. Branch elaboration depends on oxygen levels because cultivating 84 worms in low oxygen (1% O 2 ) prevented growth of these complex-shaped dendritic tips. 85 We also find that the oxygen sensory pathway is necessary for this growth, suggesting 86 that branch elaboration is due to neuronal activity. The components of the oxygen 87 sensing pathway normally localize to a position at the end of the dendrite just beneath 169  170  171  172  173  174  175  176  177  178  179  180  181  182  183  184  185  186  187  188  189  190  191  192  193  194  195  196  197  198  199  200  201  202  203  204  205  206  207  208  209  210  211  212  213  214  215  216  217  218  219  220  221  222  223  224   88   the surface of the nose of the worm, where they are thought to assemble This dendritic process ends just beneath the skin, where environmental oxygen   225  226  227  228  229  230  231  232  233  234  235  236  237  238  239  240  241  242  243  244  245  246  247  248  249  250  251  252  253  254  255  256  257  258  259  260  261  262  263  264  265  266  267  268  269  270  271  272  273  274  275  276  277  278  279  280  may diffuse a short distance to bind the molecular receptor

118
To study the effect of oxygen level on branch elaboration, we visualized URX 119 neurons using cytoplasmic GFP driven by the gcy-32 promoter. This gcy-32 reporter is 120 robustly expressed in URX, AQR, and PQR neurons [31], but because neither AQR nor 121 PQR send processes to the nose, we could clearly visualize the dendritic endings of 122 URX in these strains. We characterized the dendritic morphology of URX in three 123 independently-derived transgenic strains to control for artifacts caused by variation in 124 GFP expression [32]. 125 We imaged individual worms repeatedly across each day of adulthood and 126 observed that URX dendritic branches in worms maintained at high oxygen levels 127 continued to grow in length and complexity as the animal aged ( Figure 1C). By day four 128 of adulthood, the difference between the dendritic branches of worms grown in 21% and 129 those grown in 1% was pronounced, so we chose this particular age to quantify 130 differences between conditions. The morphological variability of the dendritic tips was 131 difficult to describe; however, we found that we could unambiguously classify dendritic 132 tips with elaborate branches as "complex", and those without as "simple". Specifically, if the dendritic tip had at least one secondary branch longer than 5 µm, we classified it as 134 complex; otherwise the dendritic tip was classified as simple. In worms grown in 21% 135 oxygen, the vast majority of dendritic tips were complex in each of the three transgenic 136 reporter lines (complex = 96.3%, 97.1%, and 93.7%), while worms grown in 1% oxygen 137 had mostly simple dendritic tips (complex = 8.2%, 14.7%, 18.4%) ( Figure 1D). These 138 results show that oxygen drives growth of elaborate branches at the end of the URX 139 sensory dendrite, and that growth continues as long as the worm remains exposed to maintained for the next two days at either high or low oxygen, at which point we again 150 quantified the total length of the branches (Fig 1E). We found that while the dendritic 151 branches in worms kept in high oxygen (21%) continued to grow over the two days, the 152 branches in worms moved to low oxygen (1%) neither grew nor reduced, but rather  The oxygen sensing pathway is necessary for dendritic branch growth in URX 157 URX is a sensory neuron for oxygen, which suggests the possibility that branch 158 growth at the dendritic tip is caused by prolonged sensory activity. In URX, molecular 159 oxygen is coordinated at the dendritic tip by a heterodimer of the membrane-tethered 2B). This strongly suggests that oxygen sensation drives branch elaboration at the URX 167 dendritic tip. 168 We also considered the alternative hypothesis that branch growth is repressed 169 by the hypoxia pathway in low oxygen (1%) and thus is revealed at high oxygen (21%).

170
To test this hypothesis, we examined mutants lacking the prolyl hydroxylase EGL-9,  and that the lack of branch growth in low oxygen conditions is due to decreased activity 179 of URX, not repression by the hypoxia pathway.
180 Intracellular calcium influx is sufficient for dendritic branch growth 181 Both cGMP and calcium levels increase in URX as a result of oxygen sensation.

389
The following strains were used:  953  954  955  956  957  958  959  960  961  962  963  964  965  966  967  968  969  970  971  972  973  974  975  976  977  978  979  980  981  982  983  984  985  986  987  988  989  990  991  992  993  994  995  996  997  998  999  1000  1001  1002  1003  1004  1005  1006  1007  Dendrites were scored as complex if they had at least one secondary branch that 422 extended ≥5 µm from the primary dendritic stalk ending, and were scored as simple 423 otherwise. We found that in some worms one of the URX neurons would be simple 424 while the other would be complex, and a small number of worms expressed the Pgcy-32::GFP transgene in only one neuron of the URX pair. Therefore we scored each 426 dendrite individually, giving an n of 2 or 1 per animal. Image analysis was performed 427 using ImageJ (NIH). For Figure 1D, dendritic branch length was quantified using the 428 segmented line tool in ImageJ. Pictures were analyzed in pairs to ensure consistent 429 start and end points for branch measurements but blinded to age and condition. To image day one adults, we picked L4 animals expressing the YC2.60 Ca 2+ sensor 24 460 h before imaging. To image day four adults, we picked L4s five days before the assay.

461
On the day of the assay 5 -10 worms were glued to agarose pads (2% in M9 buffer, 1 462 mM CaCl 2 ), using Dermabond tissue adhesive, with their body immersed in OP50 463 washed off from a seeded plate using M9 buffer. The animals were quickly covered with 464 a PDMS microfluidic chamber and 7% O 2 pumped into the chamber for 2 min before we 465 began imaging, to allow animals to adjust to the new conditions. Neural activity was 466 recorded for 6 minutes with switches in O 2 concentration every 2 minutes.

467
Imaging was on an AZ100 microscope (Nikon) bearing a TwinCam adaptor  1121  1122  1123  1124  1125  1126  1127  1128  1129  1130  1131  1132  1133  1134  1135  1136  1137  1138  1139  1140  1141  1142  1143  1144  1145  1146  1147  1148  1149  1150  1151  1152  1153  1154  1155  1156  1157  1158  1159  1160  1161  1162  1163  1164  1165  1166  1167  1168  1169  1170  1171  1172  1173  1174  1175 1177  1178  1179  1180  1181  1182  1183  1184  1185  1186  1187  1188  1189  1190  1191  1192  1193  1194  1195  1196  1197  1198  1199  1200  1201  1202  1203  1204  1205  1206  1207  1208  1209  1210  1211  1212  1213  1214  1215  1216  1217  1218  1219  1220  1221  1222  1223  1224  1225  1226  1227  1228  1229    In all images, anterior is to the left, dorsal is upwards. B.) Examples of simple and complex dendritic ending morphologies in URX in wild-type worms. URX was visualized by expressing a Pgcy-32::GFP transgene, and worms were grown in either high (21%) or low (1%) oxygen until day four of adulthood. Note variable branched morphology of complex dendritic endings. Scale bars in images showing full neuron and inset are 10 µm and 5 µm, respectively. C.) The same individuals were imaged on day one and day four of adulthood to show the growth of branches over time in 21% oxygen. Scale bar is 5 µm. D.) Scoring of dendritic ending morphology in wild-type worms grown and maintained at either high or low oxygen until day four of adulthood. Three independently-derived strains were scored. In this and all similar figures, the number of dendrites scored per condition is shown above each bar. E.) Wild-type worms were reared in high oxygen until day two of adulthood, and then maintained in either high or low oxygen until day four of adulthood, with images taken at both time points. Branched endings in worms maintained at high oxygen showed significant growth over these two days (t = 2.5, p<0.05), while those in low oxygen neither grew nor shrank appreciably (t = 1.2, p>0.05). n = 8 for high oxygen conditions, 7 for low oxygen condition. Statistical significance determined by paired t-test. . B-C.) URX dendritic ending morphology was scored in worms of the given genotype on day four of adulthood after growth in high oxygen conditions. Components of the oxygen-sensing pathway were necessary for dendritic branch elaboration, while stabilization of HIF-1 in the egl-9(sa307) mutant had no effect on branch growth. Each bar is representative of three independently-derived transgenic strains expressing Pgcy-32::GFP to visualize URX. D.) URX dendritic ending morphology was scored in worms of the given genotype and age. Gain-of-function allele egl-19(gf) refers to the egl-19(n2368) allele. See Supplemental Figure 1 for example photos of the simple endings in the double mutants.
gcy-35 day 4 adults -21% O2 WT day 4 adults -21% O2 day one adults. To the right, bordering behavior in wild-type or npr-1(ky13) worms was quantified on day one or day four of adulthood. Opposing trends suggests that URX dendritic branch complexity does not contribute to this behavior. Each point is an assay of 40 worms. Error bars are 95% C.I., statistical significance determined by Student's t-test, *** indicates p < 0.01.