Figure 8 Supplemental Figure 1

Supplemental Figure 2

Supplemental Figure 3

Supplemental Figure 4

1

SUPPLEMENTAL MATERIAL

Supplemental Figure 1. CaM-KII labeling at individual Club endings. In order to

illustrate the variability in the amount of CaM-KII labeling within the population of CEs, 8

bit images of individual CEs (n=43) were thresholded and total and central areas of each

CE were identified following procedures identical to those used for calculations of the

periphery-center index (see methods). (A) Histogram shows the total amount of CaM-KII

labeling detected at individual CEs, expressed as total CaM-KII pixels above threshold

per total CE pixel area. (B) Histogram shows the amount of CaM-KII labeling detected at

the center of individual CEs, expressed as CaM-KII pixels above threshold at the center

per center CE pixel area. (C) Graph illustrates the relationship between center and total

CaM-KII labeling at individual CEs. Note the strong correlation between these two

measurements.

Supplemental Figure 2. Sequence alignment of the auto-inhibitory region of αCaM-

KII with its fish ortholog and the Cx36 binding sites to αCaM-KII with its orthologs.

(A) The mouse αCaM-KII and teleost (zebrafish) regulatory regions shows high levels of

sequence identity. (B) Sequence alignment between Cx36 cytoplasmic loop (CL) and

carboxyl-terminal (CT) αCaM-KII binding sites (Alev et al., 2008) and its fish homologs

show high degrees of similarity. The Cx36CL binding site, similar to the pseudotarget

binding site of αCaM-KII, shows similarity to amino acids surrounding core residues for

binding. The Cx36CT binding site, highly similar to the pseudosubstrate binding site of

αCaM-KII, reveals a conserved motif and high degree of identity with perch Cx35 and

Cx34.7.

2

Supplemental Figure 3. Surgical isolation of hindbrain tissue containing the

Mauthner cells. (A) Mid- saggital view of a goldfish brain showing its different regions:

tectum, cerebellum, vagal Lobe and the hindbrain region in which the M-cells are

located. Gray area represents the portion to be surgically isolated for further material

isolation. (B) Hindbrain region isolated from A. After isolation from the brain, the

cerebellum is removed, leaving a portion of the hindbrain containing the intact M-cells

and CEs, along with other reticulospinal and vestibulospinal neurons. Four to five of

these hindbrain sections were used in each sample for protein quantification.

Supplemental Figure 4. In-silico predictions and alignments of the mammalian Cx36

phosphorylation sites by αCaM-KII with its homologs perch Cx35 and Cx34.7. (A)

CaM-K/CaM-KII/CaM-KIIa (α) phosphorylation sites prediction for Cx36, Cx35 and

Cx34.7 were identified using the Group-based Prediction System Version 2.1 (GPS

v2.1). Four putative αCaM-KII phosphorylation sites were identified previously for Cx36

(Alev et al., 2008) using the group-based phosphorylation scoring method (GPS). Scores

for predicted αCaM-KII phosphorylation sites for Cx36 and their fish counterpart were

higher than the cutoff values (highest threshold) for a high prediction performance. The

prediction of serine 293 as a phosphorylation target site in Cx36 required a lower

threshold (medium), which still offers high accuracy and sensitivity. (B) Sequence

alignment of the regions of Cx36 cytoplasmic loop and carboxyl-terminal containing

phosphorylation sites for αCaM-KII and its fish homologs perch Cx35 and Cx34.7,

demonstrate a high degree of homology. Putative αCaM-KII phosphorylation sites are

outlined in red. Completely conserved amino acids are indicated with “I”, conserved

substitutions with “:”, and semi-conserved substitutions with “.”.