*2.2. Ricin Enzymatic Activity*

The introduction of a cell-free system utilizing a lysate from rabbit reticulocytes [38] helped to clarify that ricin inhibited the peptide chain elongation (Figure 1) [27]. The two polypeptides showed di fferent properties: The A chain possessed the toxic activity, while the B chain was a galactose-specific lectin binding the cell surface [26]. Treating the toxin with reducing agents resulted in more activity in inhibiting cell-free protein synthesis [39]. Firstly, the target of the toxic action was identified as the ribosome (Figure 1), then as the 60 S subunit of eukaryotic ribosome [40], which became unreactive toward elongation factors [41]. The toxin was found to prevent the binding between elongation factors and ribosomes avoiding the subsequent elongation-factor-dependent GTPase activity [41,42]. The A-chain molecule was very active on its substrate and it was calculated that one molecule can inactivate 2000 ribosomes/min, with a K m of 0.1–0.2 mM [43].

In addition to ricin, several other plant proteins have been identified to possess a similar protein synthesis inhibiting action. Most of them had a single polypeptide chain similar to the A chain of ricin. They were called Ribosome-Inactivating Proteins (RIPs) (reviewed in [44,45]).

The already supposed enzymatic nature of ricin A chain was finally demonstrated in 1987 by Endo et al. who discovered that ricin A-chain cleaved the N-glycosidic bond of an adenine residue, A4324 in rat 28 S RNA, from the ribose of a highly conserved ribosomal RNA single-stranded loop involved in the binding of elongation factors (Figure 1). The toxin did not directly break the RNA chain, but the depurinated RNA was susceptible to hydrolysis [46,47]. Consequently, ricin activity was identified as an rRNA N-glycosidase (EC 3.2.2.22).

Following this, it was demonstrated that the enzymatic activity of RIPs was broader than previously described. All tested RIPs were able to release adenine from DNA, in addition to rRNA, and some of them were also able to act on other polynucleotide substrates, releasing adenine from the sugar phosphate backbone of polynucleotide substrates (Figure 1) [48,49]. For this reason, the name of adenine polynucleotide glycosylase was proposed for RIPs. Thus, the ability of acting on various substrates and extensively depurinating some of them, suggested that the protein synthesis inhibition could be only one of the ways of RIP-mediated cell killing. Ricin was shown to be able to release adenine from rRNA, DNA (chromatin and naked), and also poly(ADP-ribosyl)ated poly(ADP-ribose) polymerase, an enzyme involved in DNA repair [48,50]. Furthermore, it was observed that many RIPs were able to cleave more than one adenine: ricin was able to detach few adenines from the DNA (tens), but some single-chain RIPs were able to detach even thousands of them. The hypothesis that ricin could act directly on DNA in cellular models was strengthened by the evidence that damage to nuclear DNA, consistent with the enzymatic activity (adenine release) on DNA in cell-free systems, was concomitant with protein synthesis inhibition and preceded apoptosis [51].

#### *2.3. Ricin Cellular Uptake, Routing, and Toxicity*

Starting from the mid-1970s, several research groups focused on ricin binding and internalization studies, demonstrating that the interaction of ricin with the cell started from the binding of the B chain to galactosyl residues on the cell surface, allowing access to the endosomal compartment [52]. Ricin binds to both glycolipids and glycoproteins with terminal galactose. Since ricin binds to a variety of different molecules, it seems to be internalized by different endocytic pathways, as well as by using different pathways to reach the Golgi apparatus to intoxicate the cell. In HeLa cells, about 10<sup>7</sup> binding sites were found for ricin, but only small amount of the bound toxin reached the Golgi network and participated in cell intoxication [52].

Firstly, it was reported that ricin entered into cytoplasm through clathrin-dependent endocytosis [53]. Afterwards, it became clear that clathrin-independent mechanisms were also involved [54]. After cell uptake, ricin is delivered to early endosomes, from where most of protein molecules are recycled back to the cell surface or delivered, via late endosomes, to lysosomes for proteolytical degradation. A small amount of non-degraded ricin is addressed within the trans-Golgi network [55]. The involvement of the Golgi complex in ricin routing was confirmed using different Golgi-disrupting agents, such as brefeldin A, monensin, etc. In fact, the pretreatment with these agents inhibited the cytotoxic effects of ricin [56]. It was demonstrated that ricin was cycled from Golgi to the endoplasmic reticulum via coatomer protein 1 (COP-1)-coated vesicles [57], although it was later proved that the COP-1-independent pathway could also be involved [58].

The complete elucidation of intracellular ricin traffic occurred when it was demonstrated that, after reaching the endoplasmic reticulum, the two ricin chains were separated, and the A chain was retro-translocated through the quality control pathway delivering misfolded proteins to cytosol (Figure 1) [59]. Recently, it has been demonstrated that cholesterol rafts are required for Golgi transport of ricin; meaning that glycosphingolipids may not be required (reviewed in [60]).

The portion of A chain that quickly refolded, thus avoiding ubiquitination and proteosomal degradation, was able to reach its intracellular target (reviewed in [61]). It was estimated that one molecule of active ricin that arrives to its substrate is enough to kill one cell [62].

The discovery that ricin, and some related toxins, may be retrogradely transported along neuronal processes (Figure 1) [63] opened a new field of research in neurobiology and this property has been exploited for the selective destruction of neuron bodies.

Different cell types have shown variable levels of sensitivity to ricin (reviewed in [14]), possibly because of the mannose receptor expression on the cell surface and endocytosis efficacy. Ricin has been shown to be one of the most toxic plant toxins on cell lines with IC50s (concentration inhibiting protein synthesis by 50%) ranging from less than 0.1 to 1 pM [26,64–66]. However, it must be taken into account that it is very difficult to make a direct comparison of the data available in the literature about ricin cytotoxicity, because of the differences in the experimental approaches and technical conditions.

The polynucleotide depurinating activity of RIPs suggests the possibility of a wider toxic action on many biological substrates, not excluding the induction of oxidative stress. This could explain the induction of more than one cell death pathway, e.g. apoptosis and necroptosis, caused by ricin and other RIPs (Figure 1) [64,67].

#### **3. Ricin Toxicity in Humans and Animals**

On one hand, ricin has been studied for bio-medical applications, exploiting the ability of the A-chain to kill target cells once linked to a monoclonal antibody, as below described in the immunotoxins chapter. On the other hand, ricin has attracted nefarious interests, with a history of military, criminal, and terroristic uses [68].

The acute toxicity of ricin is highly variable depending on the animal species and strain. The pathological effects and subsequent clinical signs of ricin intoxication depend also on the route of exposure, as this dictates the subsequent tissue distribution of the toxin. Following intravenous or intramuscular administration, lesions eventually develop in the spleen, liver, and kidney whilst the lung remains unaffected. After oral ingestion, the gastrointestinal tract is severely affected. Inhalational exposure produces effects that are mainly confined to the respiratory tract [69].

The majority of data on animal toxicity has been derived from laboratory experiments in rodents, principally rat and mouse models. Oral administration of ricin was reported to give a lethal dose (LD) for 50% of animals (LD50s) 20 to 30 mg/kg in rat and 15 to 35 mg/kg in mouse [70–72]. For intravenous, inhalation and intraperitoneal routes, toxicity is approximately 1000-fold higher than that obtained for the oral route, with LD50 values in mouse of 2 to 10 μg/kg, 3–5 μg/kg and 22 μg/kg, respectively [70,73]. The lower toxicity of ricin after oral exposure is due to the protein destruction in the lumen of the intestinal tract [74,75]. Ricin acts in a time- and concentration-dependent manner. Notably, there is a time delay of about 10 h before death occurs, even when very high doses are applied [76].
