4.7.2. Proteolytic Digestion

After revelation of the SDS-PAGE gel, each band related to the ricin chains was cut out with a stiletto and transferred to a 1 mL microcentrifuge tube, pre-washed twice with TA50 [50% acetonitrile with 50% water/trifluoroacetic acid (99.9:0.1)]. After, the samples were uncolored through two successive washes with 0.2 mL of a solution 100 mM of NH4HCO3/50% ACN, for 45 min at 37 ◦C. Then the samples were dehydrated by adding 100 μL of acetonitrile [99.9%, UV/HPLC spectroscopic from Vetec (Rio de Janeiro, Brazil)] for 10 min at room temperature, and dried under N2 flow at room temperature.

In parallel, aliquots were collected from the stock solution of trypsin (trypsin gold, mass spectrometry grade from Promega (Madison, Wisconsin, USA) 1 μg/μ<sup>L</sup> in 50 mM of acetic acid, and diluted to 20 μg/mL with 40 mM NH4HCO3/10% ACN. The dried gel pieces were then incubated and rehydrated in 30 μL of this trypsin solution at room temperature for 1 h. After, the digestion bu ffer (40 mM NH4HCO3/10% ACN) was added until covering completely the gel pieces. The tubes were well closed to avoid evaporation and incubated overnight at 37 ◦C. The day after the solution was transferred to a clean tube and 30 μL of TA50 added to the gel, which was submitted to ultrasound for 20 minutes. The resulting solution was transferred to a clean tube and totally dried under N2 (AP, 99.997% from Linde) flow, being re-suspended again with 20 μL of the solution of TFA 0.1% in water. This sample, containing the peptides from the trypsin digestion was sent for analysis by MALDI-TOF MS.

Table 9 shows the peptides expected for the complete trypsinization of ricin, named according to its positions in the sequences of RTA and RTB.


**Table 9.** Expected peptides from the total proteolysis of ricin with trypsin.


**Table 9.** *Cont.*

4.7.3. MALDI-TOF MS Analysis for the Identification of Ricin

The mixture of trypsin peptides extracted from each band of the SDS-PAGE gel was analyzed through MALDI-TOF MS using a saturated solution of 4-hydroxyα-cyanocinnamic acid (HCCA) saturated in TA30 as a matrix. The sample preparation consisted of mixing equal volumes of the peptides solution in 0.1% TFA/water with a saturated solution of HCCA in TA30. After, 0.5 μL of this new mixture was applied over the target plate and left to dry. The analyses were performed in reflector mode, by monitoring the presence of peaks in the spectral region corresponding to the weight between 700 and 4000 Da. All spectra were obtained by addition of 2000 laser shots randomly distributed over the whole surface of the sample. The laser energy was kept constant during the shots.

The mass spectra of the mixture of peptides were used for the identification of ricin. Firstly, the *m*/*z* list of the peaks obtained was exported for the software Biotools, from Bruker. After, through the MASCOT PMF search mechanism, the experimental results were compared with the information available in the data banks of proteins SwissProt [36] and NCBI [37].

## 4.7.4. Analyses by MALDI-TOF MS/MS

The results obtained by MALDI-TOF MS were confirmed by a second analytic spectrometric technique. For this, three peptides were chosen to have their amino acid sequences verified through MALDI-TOF MS/MS. The criteria established for the selection of the precursor ions were the intensity and the relevance of the peptide for the di fferentiation between ricin and other proteins, the absence of cysteine and methionine residues, and the possibility of the same peptide representing RTA and RTB.

The analyses were performed in the same target plate with the samples former prepared for the MALDI-TOF MS experiment. The equipment used also was the same used before. The spectra were obtained by the method known as fragmentation analysis and structural time of flight (FAST), which only works in the reflective mode. For each analysis the range of the ions selector and the number of segments were adjusted according to the mass of the precursor ion selected, avoiding interference of fragments from possible adjacent ions. The spectra were exported to the software Bruker Biotools and, with the help of the MASCOT searching mechanism, compared to the data existing in the data banks SwissProt [36] and NCBI [37].

#### *4.8. Verification of the Toxic Activity of the Ricin Samples by MALDI-TOF MS*

The toxic activity of ricin present in the samples was verified by MALDI-TOF MS, following a method adapted from Schieltz et al. [46]. For this, a DNA substrate chemically synthesized with the nucleotide sequence GCGCGAGAGCGC, similar to rRNA 28S where the ricin attack occurs, was acquired from the Company Genone Biotechnologies (Rio de Janeiro, Brazil).

A solution containing 0.1 μmol/mL of nucleotides was prepared and mixed with a PBS10 bu ffer (pH 7.4). After, the sample solution was prepared with 20 mg/mL of the white powder extracted from castor bean seeds mixed with the PBS10 bu ffer solution. The reaction mixture was produced by mixing equal volumes of the two solutions. Then it was incubated at 37 ◦C, without stirring, for 24 h. Aliquots were collected and analyzed in times 0, 4 and 24 h. The matrix solution consisted of 3-hydroxypicolinic acid (3-HPA) saturated in TA50. This solution (0.5 μL) was applied over the target plate and left to dry at room temperature. At the time intervals mentioned above, 2.0 μL were collected from the supernatant of the reaction and mixed with more 18 μL of the matrix solution. From this mixture, one aliquot of 0.5 μL was deposited over the first layer, left for drying, and introduced in the target plate of the equipment.

The analysis method was in the reflective mode, with a range of *m*/*z* from 3000 to 4000, with the addition of spectra obtained after 2000 laser shots randomly distributed over the whole sample surface in the target plate.

We monitored the intensities of the signals of peaks at *m*/*z* 3,697, referring to the mass of the quasi-molecular ion of the oligonucleotide protonated [M + H]+ and in *m*/*z* 3564, related to the loss of adenine [M+H-A]<sup>+</sup>.

Solvents and reagents used for these experiments were: 3-hydroxypicolinic acid 99%, from Sigma Aldrich, TA50 (produced with acetonitrile 99,9%, UV/HPLC spectroscopic, from Vetec (Rio de Janeiro, Brazil); trifluoroacetic acid 99% from Sigma Aldrich and distilled and deionized water); PBS10 (prepared with Na2HPO4 99% from Sigma Aldrich; NaH2PO4·H2O 98% from Sigma Aldrich and NaCl, ACS reagent, from Vetec (Rio de Janeiro, Brazil).

**Author Contributions:** Conceptualization, R.S.B., A.L.S.L. and K.S.C.L.; methodology, R.S.B., C.G.M.S., A.L.S.L. K.S.C.L. and M.R.D.; software, R.S.B.; validation, R.S.B. and A.L.S.L.; formal analysis, R.S.B., A.L.S.L. and K.S.C.L.; investigation, R.S.B., C.G.M.S., A.L.S.L. and K.S.C.L.; resources, A.L.S.L., T.C.C.F., and K.K.; data curation, R.S.B.; writing—original draft preparation, R.S.B., A.L.S.L.; writing—review and editing, E.N., T.C.C.F. and

K.K.; visualization, E.N., R.S.B.; supervision, A.L.S.L. and K.S.C.L.; project administration, A.L.S.L. and K.S.C.L.; funding acquisition, A.L.S.L., T.C.C.F., and K.S.C.L.

**Funding:** This research was funded by Military Institute of Engineering, Brazilian financial agencies Conselho Nacional de Pesquisa (CNPq) (Grant No. 308225/2018-0) and Fundação de Amparo à Pesquisa do Estado do Rio de Janeiro (FAPERJ). (Grant No. E-02/202.961/2017), Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) Pró-Defesa, and the Brazilian Army. This work was also supported by Excellence project University of Hradec Králové (UHK).

**Acknowledgments:** Military Institute of Engineering, Institute of Chemical, Biological, Radiological and Nuclear (IDQBRN) Defense, Army Institute of Biology.

**Conflicts of Interest:** The authors declare no conflicts of interest.
