|
Fig. 1. Sequence conservation of AID/APOBEC-like deaminases in S. purpuratus. a Amino acid sequence alignments of human AID (HsAID) and the S. purpuratus SpAIDL1 and SpAIDL4a with the pfam08210 APOBEC-like N-terminal domain present in all vertebrate AID/APOBEC polynucleotide deaminases. Absolutely conserved residues including the catalytic HxE-PCxxC motif are shown in red, while conservation limited to HsAID or SpAIDL1/4a and pfam08210 are indicated in green or blue, respectively. b Schematic representation of the open reading frames of SpAIDL1, 2, 3, 4a, and 9 encoded by alleles from three unrelated individuals. Regions where the PCR primers precluded obtaining sequences are marked in dark blue, and the location and nature of polymorphisms are indicated
|
|
Fig. 2. Expression of SpAIDLs in S. purpuratus. a The expression levels of SpAIDL1, 2, 3, 4a, 9, and 18S RNAs in coelomocytes (C), the esophagus (E), the small intestine (SI), the large intestine (LI), and the gonad (G) obtained from three individual S. purpuratus (Sp146♀, Sp147♂, and Sp148♀) were measured by RT-qPCR with two technical replicates per sample. The expression levels are normalized to the amounts of 18S, and the error bars indicate the standard deviation. The insets are a magnified view. b, c Six animals were injected with either live V. diazotrophicus (Sp138, 139, and 140) or sterile sea water (Sp141, 142, and 143). The time course of SpAIDL2, SpAIDL9, and SpIL17-9 expression levels were measured by RT-qPCR with each data point being the mean of two technical replicates. The expression levels were first normalized to 18S and then to the expression at t = 0 for each transcript for each sea urchin. The horizontal bars indicate the mean
|
|
Fig. 3. Enzymatic activities of SpAIDLs. a–c Deaminase activities of each indicated protein were measured in bacterial reversion assays using either a Kanamycin resistance reporter plasmid (a) or the endogenous rpoB gene as the reporter (b, c). Note that the SpAIDL1 dataset from b is shown again in c for clarity. The horizontal bars represent the median (numerical value below each column), and p-values (for each deaminase compared to empty pASK for a, b, and compared to wildtype SpAIDL1 for c) were calculated using a Willcoxon rank sum for unpaired data. Datasets with p < 0.05 are marked with an asterisk (a: HsAID p = 0.000001 and SpAIDL1 p = 0.02087; b: HsAID p = 6.6 × 10−9, SpAIDL1 p = 0.0005, and SpAIDL2 p = 0.0014; c: SpAIDL1 RQAA p = 0.00048). The inset represents a magnified view omitting all data points of HsAID (a, b) and one datapoint of SPAIDL2 (b) for clarity. SpAIDL1 RQ and SpAIDL1 RQAA are mutants of SpAIDL1 in which either two (H86R and E88Q) or four (H86R, E88Q, C123A, and C126A) of the conserved active site residues were altered. d, e Western blot analysis of the expression of the indicated V5-tagged deaminases in bacteria induced with AHT for 3 h (+) or not (−). Note that the enzymatic activities were tested with untagged versions of the proteins. The empty pASK_V5 plasmids served as a control. The predicted sizes for V5-tagged HsAID, SpAIDL1 (and the mutants thereof), 2, 3, 4a, and 9 are 25.4, 24.0, 23.5, 22.3, 23.6, and 24.6 kD, respectively
|
|
Fig. 4. Mutation patterns generated by SpAIDL1 and LaAIDL1. Genomic DNA fragments of the rpoB gene from 24 individual RifR colonies from the E. coli-based deaminase assay (Figs. 3, 5) were amplified by PCR and sequenced. For each deaminase a minimum of 20 interpretable sequences were obtained (HsAID: 24, SpAIDL1: 24, SpAIDL1 RQAA: 21, LaAIDL1: 24, LaAIDL1 RQRQ: 20, pASK: 24). The frequencies of mutations at individual nucleotides numbered relative to the 5′-end of the PCR product (with A/Ts shown in blue) are shown as bar graphs. Colonies from bacteria containing only the empty expression vector (pASK) served as the negative control for spontaneously arising mutations conferring resistance to rifampicin
|
|
Fig. 5. Characterization of LaAIDLs in L. anatina. a Schematic view of LaAIDL1 and 2 with deaminase domains shown in light blue and the catalytic residues in red. b Normalized expression levels of LaAIDL1, 2, and LaRPL39 from the pedicle (Ped), the digestive cecum (DC), the gonad, and the gut from two individuals (La1 and La3) were measured by RT-qPCR with two technical replicates per sample. The error bars indicate the standard deviation. The insets represent magnified views. c–f Deaminase activities of LaAIDL1, LaAIDL2, and mutants of the former were measured in bacterial reversion assays using either a Kanamycin resistance reporter plasmid (c, d) or the endogenous rpoB gene as the reporter (e, f). LaAIDL1 RQ RQAA is putative catalytic mutant in which two residues in both active sites (H81R, E83Q, H298R, and E300Q) were altered. The data for HsAID and empty pASK plasmids is the same as shown in Fig. 3. The horizontal bars represent the median (value below each column), and p-values were calculated for each deaminase compared to empty pASK (c, e), and compared to wildtype LaAIDL1 for (d, f) using a Willcoxon rank sum test for unpaired data. Datasets with p < 0.05 (c: HsAID p = 0.000001 and LaAIDL1 p = 0.000053; d: LaAIDL1 RQRQ p = 0.021; e: HsAID p = 6.6 × 10−9 and LaAIDL1 p = 0.0022) are marked with asterisks. The insets represent a magnified views of the same data omitting all data points of HsAID (c, e) and one data point of LaAIDL1 RQRQ (f) for clarity. g, h Bacterial expression of V5-tagged LaAIDL1 and 2 (and mutants thereof), and HsAID in E. coli was assessed using a monoclonal anti-V5 antibody (for details see legend to Fig. 3b). Note that the enzymatic activities were tested with untagged versions of the proteins. The predicted sizes for V5-tagged HsAID, LaAID1, and 2 are 25.4, 52.4, and 23.5 kD, respectively, and smaller bands likely represent degradation products
|
