**2. Results**

*2.1. The Longer Peptides Have Higher Affinity for RTA than the Shorter Peptides*

The sequences of peptides mimicking the last 11 amino acids of the P proteins are shown in Figure 1.


**Figure 1.** The sequence of the peptides corresponding to the C-terminal end of P proteins.

We used surface plasmon resonance (SPR) with Biacore T200 (GE Healthcare, Marlborough, MA, USA) to measure the affinity of these peptides for RTA. Because of solubility limitations, the highest concentration of peptide measured was 0.5 mM. For direct comparison of affinity with depurination activity, depurination buffer was used as the running buffer for the Biacore analysis. The dose-dependent interaction sensorgrams are shown in Figure 2a and the fitting is shown in Figure 2b. The interaction sensorgrams of peptides with RTA showed fast on and fast off characteristics, suggesting fast association and fast dissociation, and thus relatively low affinity (Figure 2b). As the peptide concentration increased, the equilibrium binding levels increased. Due to the fast association and dissociation, equilibrium data were used to calculate the dissociation constants (*K*D). The *K*D<sup>s</sup> of the peptides are shown in Table 1. Overall, the affinity of the peptides for RTA was in the high micromolar range. The affinity of RTA for the peptides was 10<sup>5</sup> times lower than its affinity for the ribosome, which is in the low nanomolar range [23]. The *K*D increased from 196 μM to 451 μM as the number of amino acids decreased from 11 (P11) to 4 (P4). Deletion of the three aspartic acids (D106D107D108) did not decrease the affinity dramatically, as the *K*D was 272 μM for P10 and around 300 μM for P9, P8, and P7. The decreased affinity was mainly due to the deletion of Asp106. The *K*D did not change much when Asp107 and Asp108 were deleted. The *K*D increased from 294 μM to 399 μM with the deletion of Met109 and to 497 μM when Gly110 was deleted, indicating the importance of Met109 and Gly110 in the interaction. Deleting Phe111 did not change the affinity appreciably. However, upon deleting Gly112, the affinity decreased dramatically, indicating that the peptide with only the last three amino acids of P proteins almost lost the ability to bind RTA.

**Figure 2.** The interaction curves (**a**) and the steady state affinity fitting (**b**) of the peptide—Ricin A chain (RTA) interaction. The *K*D was determined using Biacore T200. The untagged recombinant RTA was immobilized on a CM5 chip by amine coupling at about 2422 RU. The reference surface was activated and blocked. The peptides were passed over the surface at 6.2 μM (red), 18.5 μM (green), 55.6 μM (dark blue), 166.7 μM (magenta), and 500 μM (light blue).


**Table 1.** The affinity of peptides for RTA.

**\*** Data were obtained from the fitting shown in Figure 1 and shown as average ± SD of three to four replicates. MW: molecular weight. *KD*: the equilibrium dissociation constant.

#### *2.2. Peptides Bind to the Ribosome Binding Site of RTA*

We showed that the R189A, R193A, R234A, and R235A mutations affect the ribosome binding of RTA [29]. The order of decreased binding strength was R235A > R234A > R193A ≥ R189A [29], where the R235A mutant completely lost ribosome binding [29]. To verify the binding site of these peptides on RTA, the His-tagged RTA mutants R189A, R193A, R234A, and R235A were captured on an NTA chip and P11 and P4 were passed over the surface. As shown in Figure 3, the binding levels of P11 and P4 decreased for the Arg mutants. The relative decrease in binding levels of the Arg mutants to P11 and P4 was in the same order as the relative decrease in ribosome binding. The lowest binding was observed with R235A, followed by R234A, R193A, and R189A, indicating that the peptides bind at the ribosome binding site of RTA.

**Figure 3.** The interaction curves of P11 (**<sup>a</sup>**–**<sup>c</sup>**) and P4 (**d**–**f**) with wild-type (WT) RTA and the R193A, R235A, R189A, and R234A point mutants. The interactions were analyzed by Biacore T200 using the same conditions as in Figure 2, except N-terminally His-tagged wild-type RTA (10×His-RTA) or 10×His-tagged RTA mutants were captured on an NTA chip at around 2100 RU. Flow cell 1 (Fc1) was used as control, R235A or R234A was captured on Fc2, R193A or R189A was captured on Fc3, and wild type (WT)-RTA was captured on Fc4. The P11 or P4 were passed over the surface at a concentration of 166.7 μM. The signals were normalized to the chip density of WT-RTA.

The crystal structures of P6 with RTA have shown that Arg235 and the last aspartic acid of P proteins, Asp115, form a critical hydrogen bond. To determine if Asp115 is critical for binding RTA, a 5-mer peptide P5b (G110FGLF114) in which Asp115 was deleted was synthesized and its interaction with RTA was compared to P5 (F111GLFD115). The calculated *K*D of P5b was over 20 times higher than that of P5, indicating that the last aspartic acid (D115) is critical for the interaction with RTA (Figure 4). These results provide further evidence that the peptides interact with the ribosome binding site of RTA.

**Figure 4.** The interaction of peptides mimicking the last five amino acids (P5) of the P proteins (**<sup>a</sup>**,**<sup>c</sup>**) and the penultimate five amino acids (P5b) with RTA (**b**,**d**). The *K*D was determined by Biacore T200 using the same conditions as in Figure 2. The binding sensorgrams are shown in (**<sup>a</sup>**,**b**) and the fitting is shown in (**<sup>c</sup>**,**d**).

#### *2.3. Peptides Compete with Ribosomes for Binding to RTA*

To determine if peptide binding at the ribosome binding site of RTA can affect binding of RTA to ribosomes, a peptide-ribosome competition assay was conducted using the A-B-A injection capability of Biacore 8K. The molecular weight of the yeas<sup>t</sup> ribosome is about 3000 times larger than that of the peptides. The affinity of RTA for the ribosome is in the nanomolar range [23]. The affinity of peptides for RTA is in the high micromolar range (Table 1) and about 105-fold lower than the affinity for the ribosome. Since the structure of the last six amino acids of P proteins with RTA has been resolved, P6 was chosen for the competition assay. RTA was immobilized on a CM5 chip by amine coupling. The surface was first blocked by P6 at three different concentrations (125, 250, and 500 μM) for one minute. Ribosomes (20 nM) were mixed with the same three concentrations of P6 and passed over the surface for two minutes. The ribosome binding levels at the end of each injection were plotted against the concentration of P6. As shown in Figure 5, ribosome binding decreased as the concentration of P6 increased, indicating that P6 competed with ribosomes for binding to RTA in a dose-dependent manner. However, 104-fold higher concentration of P6 was needed compared to the ribosome and the inhibition did not reach 100%.

**Figure 5.** P6 competes with the ribosome for binding to RTA. The A-B-A capability of Biacore 8K was used for the competition analysis. RTA was immobilized on a CM5 chip at 4000 RU by amine coupling. The P6 was passed over the RTA at indicated concentrations for 1 min and then yeas<sup>t</sup> ribosomes (20 nM) were injected over the surface together with the same concentrations of P6 for another 2 min at a flow rate of 30 μL per min. The ribosome binding levels were determined at 5 s before the end of the injection. The surface was regenerated by three one-minute injections of 2 M NaCl and one injection of running buffer with 2% DMSO. The data are expressed as average ± SD from four replicates.

#### *2.4. Peptides Inhibit the Depurination Activity of RTA*

We showed that the peptides bind to the ribosome binding site of RTA and affect the interaction of RTA with the ribosome. Based on our ribosome depurination model, if ribosome binding of RTA is reduced then depurination of the SRL on the ribosome should be affected. To test this model, we examined the ability of the peptides to inhibit the depurination activity of RTA on yeas<sup>t</sup> and rat liver ribosomes. Both ribosomes were used since RTA depurinates rat liver ribosomes at a much higher rate than yeas<sup>t</sup> ribosomes [30]. Different concentrations of peptides were incubated with RTA first, and then ribosomes were added to start the depurination reaction. After 5 min at room temperature, the reaction was stopped and the level of ribosome depurination was measured by qRT-PCR [31]. The percent depurination was plotted against the concentration of the peptide (Figure 6). The IC50 values were obtained by fitting the inhibition curves with the Michaelis–Menten equation using the Origin Pro 9.1 software (OriginLab, Northampton, MA, USA). The inhibition of depurination of yeas<sup>t</sup> (Figure 6a) and rat liver (Figure 6b) ribosomes by P11 are shown. As peptide concentrations increased, the percentage of inhibition increased, and percent inhibition reached 90% for both ribosomes. The IC50 of P11 was 4.7 μM for yeas<sup>t</sup> ribosomes and 31 μM for rat liver ribosomes (Table 2). The IC50 values of P10 to P3 were measured using the same method for yeas<sup>t</sup> (Figure S1) and rat liver (Figure S2) ribosomes and data are shown in Table 2. The IC50 values for yeas<sup>t</sup> ribosomes increased from 7.9 μM to 15 μM when D106 was deleted and to 23 μM and 34 μM when D107 and D108 were deleted, respectively. Similarly, the IC50 values for rat ribosomes increased from 83 μM to 142 μM when D106 was deleted and to 267 μM when D107 was deleted. These results demonstrated that the D106D107D108 motif is important for inhibition of the depurination activity of RTA. The IC50 values were about 6 to 10 times higher for rat ribosomes than yeas<sup>t</sup> ribosomes, possibly because RTA depurinates rat ribosomes at a higher rate than yeas<sup>t</sup> ribosomes [30]. We could not measure IC50 for P7 to P4 for rat ribosomes because the peptide concentrations needed were prohibitive and limited by solubility. We could not detect any inhibition activity for P3 and P5b at the highest concentration measured (500 μM). The IC50 values correlated with the RTA interaction results (Table 1) and indicated that inhibition of ribosome binding by peptides with even a low affinity could lead to inhibition of the depurination activity of RTA.

**Figure 6.** Inhibition of depurination activity of RTA by P11. The depurination levels were determined by qRT-PCR. (**a**) Yeast ribosomes were used at 60 nM and RTA was used at 1.0 nM. (**b**) Rat liver ribosomes were used at 60 nM and RTA was used at 0.2 nM. Different concentrations of P11 and RTA were mixed first and the reaction was started by adding ribosomes. The reaction was incubated at the room temperature for 5 min and was stopped by adding 2×RNA extraction buffer. The RNA was purified and the depurination levels were determined by qRT-PCR. The depurination level of the reaction without toxin was set as 100%. The depurination levels were calculated and plotted as percent of no toxin control. Experiments were conducted four to six times and the data were fit with the Michaelis–Menten equation using Origin Pro 9.1. IC50: the half maximal inhibitory concentration. Imax: maximal inhibition.


**Table 2.** Inhibition of depurination activity of RTA by peptides.

**\*** The IC50 values were determined using the method shown in Figure 6 and Figures S1 and S2. Data are from the fitting results. NA: not analyzed.
