*3.2. Dose-Dependent Inhibition by Clozapine*

The effects of varying concentrations of clozapine on the reactivation of EBV into the lytic cycle were tested. Clozapine at 2, 10, or 50 μM did not induce BZLF1 viral gene expression in the HH514-16 Burkitt lymphoma cells compared to untreated cells (Figure 3). When added with butyrate, lower concentrations of clozapine (i.e., 2 or 10 μM) inhibited BZLF1 expression by ~40%–50% compared to the level reached in cells treated with only NaB. Clozapine at 50 μM inhibited lytic reactivation by NaB by >95% compared to the reactivation seen with NaB alone.

**Figure 3.** Inhibition of Epstein–Barr virus (EBV) lytic reactivation by clozapine was dose dependent. Clozapine (2, 10, and 50 μM) was tested in the presence and absence of NaB (3 mM) for the effects on BZLF1 expression in HH514-16 cells. Values are the average of seven biological replicates. There was no significant difference between untreated cells and cells treated with only clozapine at any concentration. Differences in BZLF1 expression comparing cells treated with butyrate in the absence and presence of clozapine are marked with \* for *p*-value < 0.05 and \*\*\* for a *p*-value < 0.001.

The concentrations of clozapine that inhibited EBV lytic reactivation, up to 50 μM, did not limit cell growth or the percentage of dead cells when treated for 24 h (Figure 4). With 50 μM clozapine, cells remained >93% ± 4% viable (*n* = 6) after 48 h and 88% ± 3% viable (*n* = 3) after 72 h of treatment. Toxicity was observed within 24 h when the clozapine concentration reached 100 μM. Cell toxicity with 100 μM clozapine varied widely among experiments, but the average of 12 replicates resulted in ~40% cell death. When the clozapine concentration reached 200 μM, nearly all of the cells were dead after 24 h in all experiments.

**Figure 4.** Cells remained viable when treated with 50 μM clozapine for 24 h. Clozapine was tested at concentrations from 2–200 μM in the presence and absence of NaB (3 mM) for the effects on the viability of Burkitt lymphoma cells. Data from four or more biological replicates were averaged, and error bars represent the standard deviation. Conditions are not marked unless significantly different than untreated cells. Differences with a *p*-value < 0.001 are denoted with \*\*\*.

#### *3.3. Clozapine Decreased EBV Lytic Induction by dAzaC and TPA*

Like butyrate, 5-aza-2'-deoxycytidine (dAzaC) also induces lytic gene expression in HH514-16 cells [3]. Sold under the drug name Decitabine, dAzaC is a DNA methyltransferase inhibitor that is thought to activate EBV by a different mechanism than butyrate [1]. dAzaC (10 μM) was not as potent an activator of BZLF1 expression (~40-fold) as butyrate in HH514-16 cells, but activated the expression of BZLF1 significantly compared to untreated cells (Figure 5). The addition of clozapine (50 μM) at the same time as dAzaC resulted in a 60% decrease in BZLF1 expression compared to dAzaC alone. Clozapine decreased EBV lytic reactivation stimulated by two different lytic inducing agents, but the effectiveness varied. This may have been due to the different mechanisms used by the inducing agents and the shorter length of exposure time required for dAzaC to induce the EBV lytic cycle [12].

**Figure 5.** Clozapine decreased EBV lytic BZLF1 expression induced by 5-aza-2'-deoxycytidine (dAzaC). BZLF1 expression was measured in HH514-16 cells after treatment for 24 h with dAzaC (10 μM) alone or combined with clozapine (50 μM) and compared to untreated cells. The average of six biological replicates was plotted as a percentage of BZLF1 expression induced by dAzaC. *p*-value < 0.05 is denoted with \*, *p*-value < 0.001 with \*\*\*.

To determine the effectiveness of clozapine as an inhibitor in a separate EBV<sup>+</sup> cell line, lytic reactivation was tested in Raji cells—a Burkitt lymphoma cell line with a different genetic background than HH514-16 cells. The lytic cycle was induced in the Raji cells by the addition of TPA (20 ng/mL) and detected by the expression of the EBV BRLF1 mRNA (Figure 6). When the cells were treated with clozapine (50 μM) and TPA for 24 h, the induction of BRLF1 expression was blocked. Raji cells treated with clozapine alone showed a 60% decrease, though not statistically significant, in basal levels of BRLF1 expression compared to untreated cells. These results provide evidence that clozapine inhibits EBV lytic reactivation by different classes of inducing agents and in different cell lines.

**Figure 6.** Clozapine inhibited EBV lytic BRLF1 expression in Raji cells. BRLF1 expression was measured in Raji cells after treatment with TPA (20 ng/mL) for 24 h in the absence and presence of clozapine (50 μM). The average BRLF1 expression of three or more biological replicates for each of the treated conditions was compared to untreated cells. Treated conditions are marked not significantly different (n.s.) or different with *p*-value < 0.01 (\*\*) compared to untreated.

#### *3.4. Metabolites of Clozapine*

Two of the major metabolites of clozapine—clozapine-N-oxide (CNO) and N-desmethylclozapine (NDMC; norclozapine) [13]—were tested to determine if the effects of clozapine on EBV lytic reactivation would be altered as the drug was metabolized. Neither clozapine metabolite by itself had any effect on basal levels of BZLF1 expression in HH514-16 Burkitt lymphoma cells (Figure 7). Cells were then treated with butyrate and CNO at 50 μM—the same concentration at which clozapine inhibited BZLF1 expression (Figures 2 and 3). The induction of BZLF1 expression by NaB was the same in the absence or presence of CNO, demonstrating no inhibitory effect by CNO (Figure 7). However, when combined with butyrate, NDMC (50 μM) decreased BZLF1 expression by 90% compared to butyrate alone. Therefore, CNO did not decrease BZLF1 expression, but clozapine and its metabolite NDMC did inhibit EBV lytic gene expression.

**Figure 7.** Desmethylclozapine, a metabolite of clozapine, inhibited EBV lytic reactivation, but clozapine-N-oxide did not. HH514-16 cells were treated with clozapine (50 μM), clozapine N-oxide (50 μM), or desmethylclozapine (50 μM) for 24 h in the presence and absence of butyrate (NaB; 3 mM). EBV lytic reactivation was measured by the expression of BZLF1. The averaged data are plotted as a percent of the BZLF1 expression induced by NaB. Data represent the average and standard deviation of five biological replicates. Changes from untreated are marked n.s for not significant, *p*-value < 0.001 is denoted with \*\*\*, and *p*-value < 0.05 with \*. The NaB and NaB+clozapine-N-oxide were not significantly different.

#### **4. Discussion**

#### *4.1. Concentrations of Clozapine in Therapeutic Use*

Clozapine, shown here to inhibit expression of EBV lytic genes, is an antipsychotic drug used to treat schizophrenia. The standard dosing for patients is 300–600 mg of clozapine per day [14]. The recommended therapeutic range for clozapine plasma levels range from 350–550 ng/mL for effective treatment. The actual concentrations vary by patient, with factors such as weight and whether the patient smokes influencing this greatly. Studies of patients taking 400 mg/day of clozapine have measured blood concentrations of 40–1911 ng/mL and 84–1088 ng/mL [15]. The maximum plasma level recommended varies between 600 and 2000 ng/mL [16]. The clozapine concentration that inhibited EBV was 50 μM (Figure 3), which is ~8–16-fold higher than plasma concentrations in patients. In experiments conducted in vitro, concentrations of clozapine up to 50 μM had no effect on the viability of HH514–16 Burkitt lymphoma cell line (Figure 5). In another study that used a modified tetrazolium assay to assess the viability of U-937 cells from a patient with histiocytic lymphoma, clozapine had no effect on cell survival after 24 h of treatment with 6250 ng/mL (19 μM) clozapine, and 80% of cells survived when exposed to 12,500 ng/mL (~40 μM) clozapine [17]. No toxicity was observed by clozapine at >50 μM in neutrophils, monocytes, or HL-60 human leukemia cells [18].

#### *4.2. Metabolites of Clozapine*

The metabolism of clozapine is catalyzed by the cytochrome P450 enzymes in the liver into two main metabolites: clozapine-n-oxide (CNO) and N-desmethylclozapine (NDMC). NDMC is found in patient plasma at concentrations similar to clozapine, while CNO is much less [13]. CNO is pharmacologically inactive, but has the potential to reverse-metabolize into its parent compound clozapine [19]. No therapeutic benefits of NDMC have been demonstrated for the treatment of schizophrenia [20], but it does have biological effects. While clozapine is an antagonist of the dopamine D2 receptor, NDMC is a partial agonist in some assays [21]. NDMC is an allosteric agonist at the muscarinic M1 receptor. The muscarinic agonist activity of NDMC can potentiate N-methyl-D-aspartate (NMDA) receptor currents [22]. NDMC has a higher affinity for 5-HT1C and 5-HT2 receptors than clozapine, while CNO is less potent [23]. NDMC is more effective than clozapine as a partial agonist of the 5-HT1C receptor [21]. At higher concentrations, clozapine and NDMC, but not CNO, also antagonize the GABAA receptor [24]. Overall, clozapine and NDMC are more biologically active than CNO, which correlates with the observed effects on EBV lytic reactivation where clozapine and NDMC inhibited EBV, but CNO did not (Figure 7).

#### *4.3. The E*ff*ects of Clozapine on Immune Cells*

One of the most potentially critical side effects of clozapine is agranulocytosis—a reduction in granule-containing white blood cells, particularly neutrophils. Due to the potential for agranulocytosis and a high risk of infection, patients taking clozapine require long-term hematology monitoring. A number of mechanisms for the clozapine-induced agranulocytosis have been explored [25,26]. In macrophages, clozapine affects adhesion, phagocytosis, and reactive oxygen species production [27]. Clozapine also alters cytokine production in macrophages [27]. Effects of clozapine on cytokine production have been observed in a number of studies on peripheral blood mononuclear cells or whole blood and in patients with schizophrenia, though reported results have varied, and even contradicted, possibly due to varying cell sources and treatment conditions [28]. Clozapine suppresses interferon-γ production in peripheral blood mononuclear cells and inhibits Th1 cell differentiation [29]. Clozapine inhibits the production of the T-bet transcription factor and enhances mRNA expression of STAT6 and GATA3 [29]. Whole-genome analysis using the T lymphocyte cell line JM-Jurkat treated with clozapine revealed changes in expression of hundreds of mRNAs and miRNAs involved in a number of cellular processes, including cellular metabolism and oxidative stress [30]. The effects of clozapine on B cells have been less well studied.
