Disulfiram: A Repurposed Drug in Preclinical and Clinical Development for the Treatment of Infectious Diseases

Abstract

This article reviews preclinical and clinical studies on the repurposed use of disulfiram (Antabuse) as an antimicrobial agent. Preclinical research covered on the alcohol sobriety aid includes uses as an anti-MRSA agent, a carbapenamase inhibitor, antifungal drug for candidiasis, and treatment for parasitic diseases due to protozoa (e.g., giardiasis, leishmaniasis, malaria) and helminthes (e.g., schistosomiasis, trichuriasis). Past, current, and pending clinical studies on disulfiram as a post-Lyme disease syndrome (PTLDS) therapy, an HIV latency reversal agent, and intervention for COVID-19 infections are also reviewed..

1. INTRODUCTION

Disulfiram (DSF) is a thiuram disulfide drug whose medicinal use as an alcohol sobriety aid deters alcohol consumption by eliciting physical discomforts (e.g., headache, nausea, hypotension) [1]. Discovery of a potential utility in medicine can be traced to a 1937 report that described rubber plant workers exposed to thiuram vulcanization agents as “involuntary total abstainers” of alcohol use due to its toxic effects [2]. A decade later, intolerance to alcohol was observed during the evaluation of DSF as an oral anthelmintic drug that led to its assessment as a sobriety aid in human trials [3, 4]. In 1951, the U.S. Food and Drug Administration (FDA) approved the oral use of DSF in human medicine to treat alcohol use disorder (AUD) under the common trade name Antabuse^® [4].

The toxic effects precipitated by alcohol use during DSF therapy are manifested by the accumulation of acetaldehyde resulting from the inactivation of aldehyde dehydrogenases (ALDHs) [5]. The principal inhibitors of ALDH are believed to be compounds derived from DSF metabolism [6–8]. Upon ingestion, DSF hydrolyzes to its initial metabolite diethyldithiocarbamate (DDTC) in an acidic stomach, which may decompose to carbon disulfide (CDS) and diethylamine (DEA) (Fig. 1) [6]. Formation of DDTC outside the stomach can also result from the reactions of DSF with endogenous thiols (e.g., glutathione), cysteine-containing enzymes (e.g. ALDH), and trace copper (i.e., Cu[DDTC]₂). Hepatic metabolism of DDTC further culminates in the formation of other DSF metabolites (e.g., 1, 2) that are collectively believed to contribute to the inhibition of ALDH [6–8].

Fig. (1).

Disulfiram undergoes extensive metabolism that results in the initial formation of DDTC, which can complex with endogenous metals or be metabolized further in the body. The physical discomforts experienced by patients that consume alcohol while receiving disulfiram therapy are believed to be due to the collective inhibition of ALDHs by metabolites 1 and 2, which culminates in the toxic accumulation of acetaldehyde [6–8].

As a pleiotropic drug with sulfhydryl-modifying and chelating abilities (Fig. 1), DSF is able to invoke a broad range of pharmacological effects to enable repurposing in human medicine. Examples of recent repurposing studies include clinical trials as an intervention for cocaine dependence [9], small cell lung cancer [10], recurrent glioblastoma [11], and latent HIV-1 infections (Table 1) [12]. To this end, numerous reports have described research on the repositioned use of DSF to treat infectious diseases. This article summarizes the findings of these reports in the following sections arranged according to pathogen type (i.e., bacteria, fungi, etc.)

Table 1.

Clinical trials on the use of disulfiram in antimicrobial chemotherapy.

NCT No.	Indication	Phase	N	Primary Outpoint Measure(s)	Statusa
NCT03891667	Lyme Disease	I, II	24	Fatigue Severity Scale (FSS) and Quality of Life Enjoyment and Satisfaction Questionnaire over a 10 week interval.	Recruiting (est. completion date 3/22)
NCT01944371	HIV	I, II	30	Fold change cell-associated HIV RNA in total CD4 T-Cells with ART and a 3 day course of 500, 1000, or 2000 mg DSF.	Completed (5/14)
NCT01286259	HIV	–	16	Impact of 2 week course of 500 mg DSF with ART on infectious units per million cells; number of patients with adverse events; fold change in mean viremia compared to baseline; and number of patients with detectable plasma HIV RNA.	Completed (5/14)
NCT00878306	HIV	I	40	Determine the effects of DSF on the pharmacokinetics of efavirenz, atazanavir, and ritonavir.	Completed (4/13)
NCT03198559	HIV	I, II	2	Determine the change of baseline plasma HIV RNA levels of DSF with vorinostat.	Terminated (risks to participants were greater than originally anticipated)
NCT04485130	CoVID-19	II	60	Measure of immunologic impact by fold-change in plasma pro-inflammatory cytokines with 5 days of disulfiram	Recruiting (est. completion date 4/22)
NCT04594343	CoVID-19	II	140	Time to clinical improvement based on WHO score	Completed (9/21)

^astatus date: December 1, 2021.

2. BACTERIA

2.1. Gram-Positive

Investigations on DSF as a treatment for Gram-positive bacterial infections has largely focused on Staphylococcus aureus, the most common isolate of skin, bone, and joint infections. Susceptibility of methicillin-resistant S. aureus (MRSA) to DSF has been detected as low as minimum inhibitory concentration (MIC) 1.33 μg/mL [13]. In 2017, an MIC₉₀ of 16 μg/mL was reported for a thirty-strain S. aureus panel that included VISA (intermediate vancomycin-resistant) and VRSA (vancomycin-resistant) variants [14]. Pharmacodynamic studies revealed DSF to be bacteriostatic at treatments up to 4 × MIC [14] and bactericidal at 10 × MIC [15]. At 5 × MIC, a prolonged post-antibiotic effect of 5 hours was observed, while an intracellular killing effect was detected in macrophages at 10 × MIC [15]. In vivo assessment further established that a 50 mg/kg intraperitoneal dose of DSF reduced the S. aureus burden in a neutropenic murine thigh infection model [15].

Table 2 shows the reported MICs of DSF for fifteen additional Gram-positive species. Susceptible bacteria with MICs of ≤4 μg/mL include Bacillus anthracis (anthrax), Bacillus cereus, Rhodococcus erythropolis, and Staphylococcus epidermidis. Moderate to weak susceptibilities (MICs 16–32 μg/mL) were observed for isolates of vancomycin-resistant Enterococcus faecium (VRE), Listeria monocytogenes, Streptococcus mutans, Streptococcus pneumoniae, and Streptococcus pyogenes. Conversely, the DDTC byproduct of DSF metabolism was an ineffective inhibitor of Gram-positive bacteria growth at 32–64 μg/mL, with the exception of B. anthracis.

Table 2.

Reported MIC (μg/mL) ranges for disulfiram (DSF) and its initial metabolite DDTC.

Speciesa,b	Isolates	DSF	DDTC	Refs.
Acinetobacter baumannii ‡	1	>32	>32	[16]
Bacillus anthracis †	4	≤0.5	1–4	[16]
Bacillus cereus †	1	4	32	[16]
Bordetella bronchiseptica ‡	1	>32	>32	[16]
Borrelia burgdorferi ‡	1	0.18–0.3	—	[28, 30]
Brucella neotomae ‡	1	16	>32	[16]
Burkholderia cepacia ‡	1	>32	>32	[16]
Burkholderia multivorans ‡	1	32	>32	[16]
Citrobacter freundii ‡	1	>32	>32	[16]
Corynebacterium striatum †	1	8	>32	[16]
Enterobacter cloacae ‡	1	>32	>32	[16]
Enterococcus faecalis †	1	32	64	[16]
Enterococcus faecium (VRE) †	2	16–32	32–64	[14, 15]
Escherichia coli ‡	2	>32	>32	[14, 15]
Francisella tularensis ‡	1	2	8	[27]
Helicobacter pylori	12	1–10	—	[31]
Klebsiella pneumoniae ‡	2	>32	>32	[14, 15]
Listeria monocytogenes †	1	32	>32	[16]
Micrococcus luteus †	1	8	32	[16]
Mycobacterium abscess (NTM)	1	32	—	[39]
Mycobacterium avium (NTM)	2	25	>100	[38]
Mycobacterium bovis	1	3.13	6.25	[38]
Mycobacterium chelonae (NTM)	1	32	—	[39]
Mycobacterium fortuitum (NTM)	1	32	—	[39]
Mycobacterium smegmatis (NTM)	1	25	>100	[38]
Mycobacterium tuberculosis	20	0.78–1.56	1.56–3.13	[38]
Mycobacterium tuberculosis MDR/XDR	22	0.78–1.56	1.56–6.25	[38]
Mycoplasma hominis	3	0.19	—	[24]
Mycoplasma pneumoniae	3	0.19	—	[24]
Pasteurella multocida ‡	1	>32	>32	[16]
Proteus mirabilis ‡	1	>32	>32	[16]
Pseudomonas aeruginosa ‡	2	>32	>32	[14, 15]
Rhodococcus erythropolis †	1	4	>32	[16]
Salmonella typhi ‡	1	>32	>32	[16]
Shigella dysenteriae ‡	1	>32	>32	[16]
Staphylococcus aureus (VSSA) †	10	8–32	16–>32	[14]
Staphylococcus aureus (VISA) †	10	4–32	16–64	[14]
Staphylococcus aureus (VRSA) †	10	4–32	8–64	[14]
Staphylococcus epidermidis †	2	1–4	32–>64	[16]
Streptococcus agalactiae †	1	32	>32	[16]
Streptococcus gallolyticus †	1	>32	>32	[16]
Streptococcus mutans †	1	16	—	[22]
Streptococcus pneumoniae †	1	32	>32	[16]
Streptococcus pyogenes †	1	16	32	[16]
Vibrio cholerae ‡	1	32	>32	[16]
Yersinia enterocolitica ‡	1	32	>32	[16]
Yersinia pestis ‡	1	8	16	[16]
Yersinia pseudotuberculosis ‡	1	>32	>32	[16]

^aGram-positive (†), and Gram-negative (‡).

^bAbbreviations: Extensively drug-resistant (XDR); Multidrug-resistant (MDR); Nontuberculosis mycobacteria (NTM); Vancomycin-resistant Enterococcus (VRE); Vancomycin-susceptible S. aureus (VSSA); Vancomycin-intermediate resistant S. aureus (VISA); Vancomycin-resistant S. aureus (VRSA).

The growth inhibition observed for B. anthracis was partly attributed to the reactivity of DSF with intracellular thiols and chelation of metal micronutrients by metabolite DDTC [16]. To this end, research on the antibacterial activity of metal-binding substances alone [17] and in combination with antimicrobial transition metals [18, 19] have been subjects of prior reviews. Copper (II) complexes are among the most potent metallo-bactericides, and the antimicrobial action of DSF has been attributed to the formation of Cu[DDTC]₂ by absorption of endogenous copper [20]. Accordingly, studies on the synergistic potential of DSF-copper combinations have been conducted on MRSA [21], Streptococcus mutans [22], Mycobacterium tuberculosis [23], and Mycoplasma hominis [24].

Combination antibiotic testing has further revealed the possible application of DSF as an antibiotic adjuvant for multidrug-resistant MRSA infections [14, 16, 25]. Isobologram (checkerboard) studies demonstrated that DSF lowers the MIC of vancomycin in VISA and VRSA to levels observed for VSSA (i.e., MICs ≤ 2 μg/mL) [14, 16]. Synergistic activity was similarly observed with fosfomycin in a fosB⁺ MRSA strain, suggesting that the resistance-mediating enzyme FosB and/or its substrate bacillithiol [26] are cellular targets of DSF.

2.2. Gram-Negative

Most Gram-negative bacteria are nonsusceptible to DSF at concentrations ≤ 32 μg/mL (Table 2) with notable exceptions of zoonotic pathogens Francisella tularensis [27], Yersinia pestis [16], and Borrelia burgdorferi [28–30]. DSF also demonstrated growth-inhibiting properties against Helicobacter pylori at a MIC of 1 μg/mL [31]. This data is supported by the observation of a reduced relative abundance of Helicobacteriaceae in a mouse cancer model when DSF was combined with copper; however, the effects may have been associated with microbiota alterations [32].

Although the activity spectrum excludes most Gram-negative species of clinical relevance, DSF may have clinical utility as an inhibitor of β-lactam antibiotic inactivating enzymes (i.e., β-lactamases) in the bacteria. In March 2020, Chen and coworkers reported that DSF is an inhibitor of Gram-negative metallo-β-lactamases NDM-1, IMP-1, and ImiS with respective IC₅₀ values of 0.13 ± 0.05, 2.9 ± 0.8, and 3.6 ± 0.7 μM [33]. Computer-aided molecular docking studies revealed possible coordination of DSF with Zn (II) ions and thiol-disulfide exchange with Cys₂₀₈ in the active site of NDM-1. Antibacterial testing further corroborated the mechanistic studies on DSF as a carbapenamase inhibitor. The addition of 16 μg/mL DSF lowered the MICs of imipenem from 64 – 128 to 8 – 32 μg/mL in NDM-1 clones of E. coli, K. pneumoniae, and P. aeruginosa. In the same study, the addition of chelated metabolite Cu[DDTC]₂ also lowered the MICs of imipenem in NDM-1 producing bacteria to 4 – 16 μg/mL.

2.3. Borrelia

Lyme disease, a spirochete infection by the Borrelia species, is the most common tick-borne disease in the U.S. and Europe. Given their ability to travel from vertebrate host to tick vector, Borrelia bacteria possess survival strategies to adapt to changing and hostile environments. Current antibiotic therapies (e.g., doxycycline) for early Lyme disease have been shown to be less effective against stationary phase spirochetes [34]. Following treatment, patients can experience chronic illness (e.g., fatigue, arthralgia, memory loss) that may persist for over 6 months in the form of post-treatment Lyme disease syndrome (PTLDS). Although the precise cause remains unknown, autoimmunity against microbial antigenic debris and the presence of undetected persistent Borrelia bacteria are believed to be implicated [30]. Identifying new therapies that effectively eradicate both the actively growing and persistent dormant spirochetes is thought to be the key to the prevention or treatment of post-treatment borreliosis disorders.

In 2006, high-throughput screening of an FDA-approved drug library identified DSF as a potent growth inhibitor of stationary-phase B. burgdorferi persisters. The BacTiter-Glo^™ viability assay was used to determine that 96 hours of 1.25 μM DSF treatment resulted in a 99.8% reduction in luminescence [28]. Subsequent testing confirmed a MIC and MBC of 0.625 (0.18 μg/mL) and 1.25 μM (0.38 μg/mL), respectively, after 72 hours treatment with DSF. In a separate study, the same research group found comparable inhibition with 5 μm (1.48 μg/ml) DSF against the log and stationary-phase B. burgdorferi by SYBR Green-I/propidium iodide detection while doxycycline only showed inhibition of log-phase spirochetes [29]. In vivo assessment in a borreliosis mouse model further demonstrated that intraperitoneal DSF conferred reductions in cytokine levels (e.g., IL-1β, IL-10, TNF-α, IFN-γ), antibody titers (e.g., IgG, IgM), and overall B. burgdorferi burden (i.e., heart, ear, bladder) in the 21 days post-treatment group.

In 2020, a different research group reported a MIC of 0.3 μg/mL for DSF against B. burgdorferi persisters using SYBR Green-I/propidium iodide to detect for viability [30]. Measurement of B. burgdorferi viability (%) following 7-day 50 μM treatment revealed greater survival for doxycycline (69.7%) and amoxicillin (46%), but not cefuroxime (35.4%), compared to DSF (39.4%). In a similar study using 5 μg/mL drug combinations, only a moderate decrease in survival was observed when DSF was combined with doxycycline, amoxicillin, and cefuroxime.

In 2019, the first clinical trial on the use of DSF for the treatment of PTLDS was registered in the U.S. [35] (Table 1). The 14-week randomized, double-blind, placebo-controlled pilot study will evaluate the side effects, tolerability, and clinical effectiveness of DSF in twenty-four PTLDS patients between 18 and 65 years of age. The participants will receive either 4 or 8 weeks of DSF titrated up from 250 to 500 mg after week one according to tolerance [35]. The patients will be evaluated at weeks 10 and 14 by two primary and three secondary outcomes. The first primary outcome is using a scale (FSS) to measure severity of fatigue in different situations on a weekly basis for ten weeks. A 16-item questionnaire (Q-LES-Q-SF) will be utilized to measure the improvement in quality of life enjoyment and satisfaction over a 10-week interval for the second primary outcome. Using a health survey (SF-36), a general symptom questionnaire (GSQ-30) every two weeks, and a self-report system (PROMIS-29), the secondary outcomes measure changes in disability, multiple symptom burden, worsening of symptoms, pain intensity, and social functioning over ten weeks.

Registration of the clinical trial came in the same year that a case report was published on the experience of patients who received DSF treatment for chronic relapsing borreliosis [36]. In two of the patients, DSF eliminated the need for retreatment with antibiotics during the observation period. The third patient was symptom-free for six months and retreated with DSF. During DSF treatment, each of the patients reported adverse effects from treatment ranging from fatigue to psychotic episodes. In response to receiving notification of severe and persistent toxic events in patients administered DSF for chronic Lyme disease, The French Federation against Tick Borne Diseases (FFMVT) distributed a self-reported survey to their members who suffer from post-treatment Lyme disease syndrome (PTLDS) [37]. Pain and fatigue benefits were reported in 44% of the 16 patients that responded. A higher sensitivity to DSF-induced toxic or side effects (e.g., cognitive issues, pain, and fatigue) was demonstrated in PTLDS (13/16) than alcohol dependence treatment (1/15,000).

2.4. Mycobacteria

Multiple groups have investigated the repurposing potential of DSF for treating mycobacterial infections [38, 39]. M. tuberculosis is reportedly more susceptible than nontuberculosis mycobacteria (NTM) to DSF; however, second-line antituberculosis agents, amikacin and moxifloxacin, were shown to exhibit synergy with DSF against NTM species M. abscessus and M. fortuitum [39]. In these bacteria, DSF displayed bactericidal activity at 10 × MIC and intracellular killing in infected macrophages at 5 × MIC. In vivo assessment in a murine bacteremia model demonstrated that 50 mg/kg/day oral DSF reduced the M. fortuitum kidney burden after 15 days, although amikacin was more effective.

Prior to this report, Horita and coworkers published a comprehensive study on the antitubercular activity of DSF and DDTC against multidrug- and extensively drug-resistant M. tuberculosis (MDR/XDR-TB) [38]. Table 2 shows the reported MIC ranges for forty-two susceptible and drug-resistant TB isolates that corresponded to MIC₉₀s of 1.56 and 3.13 μg/mL, respectively. Intracellular killing effects of M. tuberculosis in human monocytes were also observed for DSF and DDTC; however, isoniazid and rifampicin were more effective overall. In vivo studies further established that oral DSF invokes bactericidal effects in serum and reduced pulmonary and spleen CFUs in a mouse model of chronic tuberculosis following a 4-week regimen of 80 to 160 mg/kg/day.

3. FUNGI

Fungal infections due to yeast (e.g., Candida, Cryptococcus) and molds (e.g., Aspergillus) can cause serious diseases, particularly in immunocompromised patients (e.g., cancer, HIV, transplant). As a colonizer of the human intestinal tract, Candida albicans is implicated in a large number of superficial and systemic mycoses. The opportunistic pathogen is considered susceptible to first-line echinocandins (e.g., caspofungin) and azoles (e.g., fluconazole); however, drug resistance occurs in other Candida species, including C. glabrata [40]. The most comprehensive study on the antifungal potential of DSF reported an MIC₉₀ value of 4 μg/mL for both C. albicans and C. glabrata [41]. The MICs were further found to be equivalent to the minimum fungicidal concentrations (MFCs), indicating DSF to be a fungicidal agent (Table 3).

Table 3.

Reported MIC and MFC ranges for disulfiram.

Organism	Isolates	MICa	MFCa	Refs.
Aspergillus fumigatus	2	8	8	[41]
Candida albicans	4	4–8	2–8	[41, 42]
Candida glabrata	3	4	4	[41]
Candida krusei	1	2	2	[41]
Candida tropicalis	1	16	16	[41]
Candida parapsilosis	1	16	16	[41]
Cryptococcus neoformans	4	1–6.25	1–6.25	[41, 42]
Histoplasma capsulatum	1	32	32	[41]
Pythium insidiosum	27	8–32	—	[45]

^aμg/mL.

Other publications examining the antifungal properties of DSF have reported MIC data for A. fumigatus [41, 42], C. albicans [42–44], Cryptococcus neoformans [42], and Pythium insidiosum [45] (Table 3). In addition, DSF was found to lower the MIC of fluconazole in C. albicans [44] and sensitized Saccharomyces cerevisiae expressing the C. albicans Cdr1p efflux pump to fluconazole, miconazole, and nystatin [46]. Mechanistic studies have further revealed that DSF impairs C. albicans hyphal induction [47] and spore generation in C. neoformans [42].

4. VIRUSES

4.1. HIV

Infections with human immunodeficiency virus (HIV) require lifelong antiretroviral therapy (ART) due to reservoirs that the virus establishes in latently infected CD4⁺ T-cells [48]. The earliest ART studies related to DSF evaluated the sodium form of its primary metabolite DDTC (Imuthiol^®). Prior to the arrival of highly active antiretroviral therapy (HAART), a 1988 report described the clinical effects of oral Na[DDTC] versus placebo in HIV patients [49]. Improvement in clinical status was noted for 38 patients treated Na[DDTC], with none progressing to AIDS compared to 3 of 39 for the placebo group after 16 weeks. The Na[DDTC] cohort also showed moderately improved CD4+ cell counts but no change in HIV transcription rate. The authors attributed the clinical response to Na[DDTC] to increased T-cell maturation and differentiation, HIV metalloenzyme inhibition, and/or anti-oxidant protective effect against inflammation.

Additional ART studies on Na[DDTC] were conducted, including a clinical trial from 1987 to 1989 with an enrollment size of 389 to compare a once-weekly 400 mg/m² oral dose to placebo [50]. Patients receiving zidovudine, which was approved by the FDA in 1987, also participated in the trial. The primary objectives were to determine if Na[DDTC] prevented AIDS-defining opportunistic infections (OIs) or the occurrence of Kaposi sarcoma. Following the 24-week study, the authors concluded that Na[DDTC] reduced the incidence of OIs in symptomatic HIV patients, but no statistical difference could be established for Kaposi sarcoma.

After the introduction of HAART, it was realized that an HIV cure would require reversal of the latent viral reservoirs [51]. In 2011, Xing et al. demonstrated that DSF and its DDTC metabolite were capable of reactivating latent HIV-1 in a primary CD4⁺ T-cell mouse model without increasing global T-cell activation and associated host toxicities [52]. A mechanism proposed for the reactivation was the depletion of the PTEN tumor suppressor, which mediates Akt signaling pathway and downstream release of active, positive transcription factor b for viral production [53]. However, this mechanism has been investigated in other ex vivo experiments, and the mechanism of induced RNA transcription remains uncertain [54, 55].

In clinical studies, use as an HIV latency reversal agent was evaluated in an open-label, single-arm pilot study whereby 16 virally undetectable (< 50 copies/mL) patients received 500 mg of DSF daily with their ART for 14 days [56]. The study failed to demonstrate a significant effect on the latent reservoirs, but it did find a transient, post-dose viremia with great inter-subject variability that suggested the pharmacokinetics and pharmacodynamics (PK-PD) of DSF may have had an impact. Moreover, a dose-response relationship was demonstrated in a 30 patient phase II dose-escalation study, wherein subjects receiving three days of 2000 mg DSF had a higher detectable HIV RNA concentration than that of baseline at 30 days, which was not seen in patients receiving 500 mg or 1000 mg. All dosing regimens demonstrated an increase in cell-associated HIV RNA with a 2.1 (95% CI 1.3–2.2, p<0.001), 2.5 (95% CI 1.9–3.3; p<0.001), and 2.1 (95% CI 1.5–3.1; p<0.003) fold increase in cell-associated HIV RNA after receiving three days of 500 mg, 1000 mg, and 2000 mg DSF, respectively [56]. Conversely, the median baseline plasma RNA was numerically higher in the 500 mg group (5.7 copies/mL) than the 1000 mg (2.6 copies/mL; p = 0.32) and 2000 mg (0.87 copies/mL; p = 0.27) groups, which may have confounded DSF efficacy [12].

4.2. Coronavirus

In 2018, Lin and coworkers reported on the inhibitory activity of DSF on coronavirus (CoV) papain-like cysteine proteases [57]. Their research showed that papain-like proteases (PL^pro) from MERS-CoV and SARS-CoV were inhibited by DSF in an allosteric and competitive (or mixed) manner at IC₅₀s 22.7 ± 0.5 and 14.2 ± 0.5 μM, respectively. Mechanistic studies further indicated that the inhibition of SARS-CoV PL^pro likely involves DSF modification of Cys₂₇₁ as the DDTC adduct. Moreover, synergistic inhibition of MERS-CoV PL^pro was observed with 6-thioguanine and mycophenolic acid, suggesting potential utility as combination treatments for COVID.

In June 2020, Jin et al. conducted high-throughput screening of a 10,000 member chemical library to identify inhibitors of recombinant SARS-CoV-2 M^pro [58]. Dose-response curves of their six lead compounds revealed that DSF (IC₅₀ 9.35 μM) was more effective than shikonin and PX-12, but not ebselen, tideglusib, and carmofur. In March 2021, Chen et al. established that DSF catalyzes the disassociation of Zn²⁺ from the replication and transcription complex (RTC) of SARS-CoV-2 while impairing viral ATPase (IC₅₀ 0.4 μM) and exoribonuclease (IC₅₀ 2.7 μM) activity [59]. Combination in vitro studies further demonstrated that DSF and remdesivir confer synergistic inhibition of SARS-CoV-2 replication by concomitant disruption of viral polypeptide proteolysis and RNA transcription/translation.

On the heels of these reports, two phase II clinical trials (NCT04485130, NCT04594343) were registered to evaluate the clinical response of SARS-CoV-2 positive individuals treated with DSF. Clinical trial NCT04485130 aims to assess four outcomes in a randomized (2:1) placebo-controlled, double-blind study with COVID-19 patients presenting mild or moderate/severe symptoms [60]. The two cohort study will evaluate the clinical response to a five consecutive day course of either 1000 mg/day or 2000 mg/day capsules of DSF or placebo containing cellulose. The primary trial outcome measures will compare the immunologic impact by the fold-change in plasma levels of pro-inflammatory cytokines (e.g., IL-6, IL-1β) at days 5, 15 and 31. Secondary outcome measures will further evaluate fold-change in: 1) SARS-CoV-2 virus copies per million cells; 2) safety and tolerability in accordance to the Common Terminology Criteria for Adverse Events (CTCAE) v4.0; and 3) COVID-19 symptoms in accordance to a 5-point adapted somatic symptom severity score (SSS-8). As of December 1, 2021, the trial was in the recruitment phase, with an estimated study completion date of April 30, 2022.

In September 2021, clinical trial NCT04594343 completed a randomized (1:1) placebo-controlled, double-blind study of 140 patients with moderate COVID-19 [61]. Patients received a 2-week, daily course of either placebo or 500 mg DSF by oral administration or a nasogastric enteral tube if in mechanical ventilation. The primary outcome measure was evaluated for time to clinical improvement from baseline enrollment with an improvement in WHO score of ≥1 point. Secondary outcome measures included the mean number of days on supplemental oxygen, time to hospital discharge, and 28-day mortality rate. As of December 1, 2021, the clinical trial results have not been reported to our knowledge.

Although clinical trial data is not yet available, retrospective analysis of COVID-19 patients prescribed DSF for alcohol use disorder (AUD) has provided initial insight on its potential utility in treatment. In October 2021, Fillmore et al. published a study that compared COVID-19 disease severity in patients prescribed DSF to untreated patients [62]. Clinical records obtained between February 2020 and February 2021 from the U.S. Veterans Affairs (VA) COVID-19 Shared Data Resource and the VA Corporate Data Warehouse (CDW) were used in the retrospective cohort study. The database identified 167,327 of 944,127 patients that received a positive SARS-CoV-2 test during this period. Among them were 188 patients with a pharmacy record for DSF. Table 4 presents the findings of lower severe outcome incidences for SARS-CoV-2 patients prescribed DSF for AUD. The authors concluded that DSF may have contributed to the reduced incidence and severity of COVID-19 infections in patients prescribed the treatment for AUD.

Table 4.

Clinical outcome comparison of COVID-19 patients by disulfiram treatment status [62].

Patient Status	Untreated (Incidenceb)	Treated (Incidenceb)	P-Value
screened for SARS-CoV-2	941,894	2,223	–
positive SARS-CoV-2 testa	167,139	188	–
ICU admissiona	7,403 (0.044)	11 (0.059)	0.44
mechanical ventilationa	959 (0.006)	1 (0.005)	1.00
deatha	5,009 (0.030)	0 (0.000)	0.03
all severe outcomesa	13,371 (0.080)	12 (0.063)	0.45

^apatients with confirmed SARS-CoV-2 infection by a positive PCR or antigen test.

^bincidence proportion in patients with confirmed SARS-CoV-2 infection.

5. PROTOZOA

5.1. Giardia lamblia

Multiple groups have investigated DSF as a potential treatment for giardiasis, the most common waterborne diarrheal disease in the world [63]. The primary pathogenic species G. lamblia is reported to have DSF susceptibility at minimum lethal concentration 1.23 ± 0.32 μM [64]. In vivo evaluation showed a modest cure rate of 21% for mice receiving a 100 mg/day gavage of DSF suspension in water compared to a 100% of mice dosed with 20 mg/day metronidazole. Despite the lower cure rate, a significant reduction in intestinal trophozoite levels was observed, and a higher cure rate was believed attainable with an optimized dosing vehicle. Following this publication, two labs reported pharmacological studies on carbamate kinase [65] and triosephosphate isomerase [66] as cellular targets of DSF in G. lamblia. Enzyme inhibition was attributed to DSF modification of a cysteine residue with respective IC₅₀ values of 0.64 and 6.6 μM.

5.2. Cryptosporidium parvum

In 2019, Sarwono and coworkers reported on DSF as a prospective treatment for cryptosporidiosis [67], a fecal-oral transmitted watery diarrheal disease that can escalate to a chronic and fatal infection in immunocompromised patients. Their study screened a 1,570-member off-patent drug library for IMPDH inhibition of purine biosynthesis in C. parvum and identified DSF as a lead compound. Preliminary in vivo assessment showed that 1000 mg/day DSF for 4 weeks cleared 48% of the intestinal burden in C. parvum-infected mice.

5.3. Trichomonas vaginalis

In 1998, Bouma et al. reported on the susceptibility of sexually transmitted T. vaginalis to DSF and metabolite DDTC [68]. Higher susceptibility levels were detected for DSF (MIC 0.1 – 0.7 μM) than with DDTC (MIC 0.3 – 9 μM) under aerobic conditions, and no cross-resistance was observed with DSF in metronidazole-resistant strains. Conversely, anaerobic evaluation was found to have a significant effect on susceptibility with MICs measuring over 100 μM for DSF. In 2013, a second lab similarly reported higher T. vaginalis viability when comparing DSF efficacy under aerobic and anaerobic conditions [69].

5.4. Plasmodium

Both DSF and DDTC have been reported to block erythrocyte parasitemia by P. falciparum, the primary species implicated in mosquito-borne malaria [70]. Treatments of 0.1 to 100 μg/mL were found to reduce parasitemia by 74–100% after 96 hours. Dose- and time-dependent inhibition of glucose utilization and lactate production were also observed in DSF-treated cultures. In a separate study, DSF was shown to potentiate the in vivo efficacy of chloroquine [71]. Decreased parasitemia and prolonged survival were both observed for the DSF cohort of chloroquine-treated mice infected with P. vinckei petteri. The authors rationalized that DSF oxidation of reduced glutathione used in heme detoxification conferred an increase in Plasmodium sensitivity to chloroquine.

5.5. Leishmania

Sandfly-borne Leishmania infections manifest cutaneous and mucosal diseases that are often not responsive to treatment [72]. As a member of the World Health Organizations Neglected Tropical Disease list, leishmaniasis affects millions of people in impoverished regions where low cost, nontoxic therapies with oral efficacy have been sought. Between 2009 and 2015, multiple groups reported studies that included evaluation of DSF as an antileishmanial agent. Researchers at the University of Pittsburgh screened chemical libraries in an L. major promastigote growth inhibition assay that identified DSF as a lead compound with an EC₅₀ of 0.5 μM [73]. Assessment of DSF in a footpad inoculation model of L. major cutaneous infection demonstrated its in vivo efficacy with up to 50% reduction in footpad thickness compared to 80–85% reduction for amphotericin B.

In 2014, Peniche and coworkers also identified DSF as an effective inhibitor of ex vivo cultured L. major with an EC₅₀ of 0.06 μM [74]. Their follow up article in 2015 reported an EC₅₀ of 0.062 μM for L. donovani and test results for DSF-metal salt combinations [75]. Among the salts evaluated, ZnCl₂ and ZnSO₄ were the most effective potentiators of DSF growth inhibition in L. major and L. donovani, respectively. The addition of the zinc salts further increased the in vitro therapeutic index values as an indicator of lower cytotoxicity for the combinations.

5.6. Trypanosoma cruzi

T. cruzi is the etiology of insect-borne Chagas’ disease, a parasitic infection that can cause chronic heart disease if untreated. In 1995, researchers at Vanderbilt University reported that DSF and DDTC were more effective than first-line drug benznidazole as inhibitors of T. cruzi epigmatigotes growth [76]. In their follow up article, DSF and DDTC were shown to have better antitrypanosomal activity against smooth muscle-infecting amastigotes than benznidazole, though not for the infective trypomastigote form [77].

6. HELMINTHES

Evaluation of DSF as an anthelmintic drug can be traced to the 1940’s after copper-chelating substances were found to induce lethal inhibition of respiration in intestinal worms [3, 78]. Decades later, discovery as an inhibitor of copper-dependent dopamine-β-hydroxylase in norepinephrine biosynthesis led to the evaluation of DSF as a modulator of neurotransmission in Schistosoma mansoni, a water-borne parasitic blood fluke [79]. Elevated dopamine and decreased norepinephrine to an undetectable level were observed in S. mansoni isolated from infected mice treated with oral DSF. Additional studies revealed that DSF attenuated egg production and adult worm maturation, increased lactic acid levels, and disrupted phenol oxidase-mediated eggshell formation [79].

Similarly, DSF was found to impair Trichuris muris (whipworm) egg formation and reduce the infection burden in mice [80]. Infected mice receiving oral DSF had a dose-dependent decrease in worm counts following a 2.5–7.5 mg/kg daily regimen for 26 days. Eggs isolated from the feces of these mice were reportedly malformed and contained abnormal embryos. A follow-up study demonstrated that mice receiving oral gavage of malformed eggs from DSF-treated female whipworms did not develop an infection, confirming them to be nonviable. The authors proposed that an analogous phenol oxidase used in S. mansoni for egg production as a drug target of DSF in T. muris.

7. SEPSIS

Sepsis is a life-threatening immune response to an infection involving proinflammatory cytokine (e.g., IL-1β, IL-18) released from myeloid and epithelial cells upon gasdermin D (GSDMD) pore formation and pyroptosis [81]. As the final mediator of cytokine release, inhibitors of GSDMD function may have clinical value in the treatment of sepsis syndrome [82]. In a 2020 report, Hu et al. screened 3,752 compounds in a GSDMD-induced liposomal leakage assay and identified DSF as a lead compound [83]. Evaluation in macrophages revealed DSF blocked pyroptosis with IC₅₀ values of 7.7 ± 0.3 μM and 10.3 ± 0.5 μM for canonical human and non-canonical mouse inflammasome pathways, respectively. Mechanistic studies demonstrated that DSF did not affect caspase processing of GSDMD to its active form (GSDMD-NT), and GSDMD pore-forming activity was directly inhibited by DSF. Mass spectroscopic analyses also revealed that DSF undergoes thiol-disulfide exchange reaction with the Cys₁₉₁ residue in human GSDMD and the modification impaired pore formation. In vivo evaluation further showed DSF suppressed LPS-induced inflammatory immune response (i.e., IL-1β, TNF, and IL-6 serum levels), which was attributed to inhibition of GSDMD pore formation using Gsdmd^−/− mice.

CONCLUSION

Nearly seventy years after its approval as an alcohol sobriety aid, DSF continues to be evaluated for use in human medicine. Recent efforts to develop DSF as a repurposed drug to treat infectious diseases have largely centered on applications as an antimicrobial adjuvant. Preclinical investigations revealed that DSF is synergistic with first-line antimicrobials to treat infections due to MRSA (vancomycin), mycobacteria (amikacin, moxifloxacin), C. albicans (fluconazole), and P. falciparum (chloroquine). Similarly, DSF exhibits promise as a counteragent of drug-resistance mechanisms present in imipenem-resistant bacteria (carbapenamases) and azole-resistant yeast (efflux pumps).

Prospective uses as a standalone therapy have also been investigated, and such applications would be optimal for infections that DSF confers pathogen-killing effects at concentrations attainable in the body. A general PK-PD requirement of an antimicrobial agent is the ability to distribute to sites of infection at concentrations above the MIC. As with other hydrophobic antibiotics (e.g., azithromycin), DSF exhibits high soft tissue penetration, low serum concentrations, and biliary excretion as the primary route of elimination. These attributes suggest greater potential as a treatment for soft tissues and intestinal tract infections than for bloodstream and urinary tract infections. Alternatively, topical DSF may be beneficial for treating dermatological infections due to bacteria (e.g., MRSA), fungi (e.g., Candida), and protozoa (e.g., Leishmania). A precedent exists for the medicinal use of a DSF topical formulation in the form of a 2% emulsion with benzyl benzoate (Tenutex^®) as a treatment for skin infestations by scabies, head lice, and crab lice.

In the final analysis, research has shown DSF to be a repurposed drug candidate for treating a diversity of infections. Use as an antimicrobial adjuvant may accelerate treatment response, and the anti-inflammatory properties of DSF could further protect against complications stemming from the infection. Moreover, translational studies are expected to progress faster for infectious diseases lacking treatment options (e.g., PTLDS, COVID-19); however, the neuropathic effects described in case reports for PTLDS patients underlines the importance of clinical trial safety assessment in the patient population for whom DSF will be used.