Clinical Microbiology and Infection
April 2017
By: G.R. Thompson, J.M. Tuscano, M.Dennis, A. Singapuri, S. Libertini, R. Guadino, A. Torres, J.M.P. Delisle, J.D. Gillece, J.M. Schupp, D. M. Engelthaler, L. Leibovici
Click here to link to the original article at

Dear Sir,
To reduce the risk of infection in highly immunocompromised patients, prophylactic antibacterial, antifungal and antiviral agents are frequently prescribed and patients are routinely advised to reduce their risk of exposure by avoiding soil, plants and cut flowers due to the presence of Aspergillus and other moulds and Nocardia spp. [1, 2]. Other recommendations to limit exposures include the avoidance of water-retaining materials given their association with Pseudomonas aeruginosa, raw vegetable sprouts (Escherichia coli), undercooked eggs (Salmonella enteritidis), fresh salsa, and berries (Cyclospora etc.) among others [1]. These recommendations do not comment on the infectious risks of medical marijuana—a substance now legal in 11 countries—and used by ∼13% of the US population yearly.

Medical marijuana obtained from dispensaries does not differ in form from recreational marijuana and consists of dried material from the Cannabis plant purchased in a variety of preparations. Patients seek out marijuana (or related products) in an attempt to control nausea or vomiting due to cytotoxic chemotherapy, to control pain or neuropathy, or to stimulate appetite. Legalization has impacted the public’s perception of safety [3], and inferences on safety are implied by the ability of physicians to write prescriptions for medical marijuana.

The potential infectious risks of marijuana obtained from dispensaries has not been systematically evaluated and we believe that this unduly places patients (unknowingly) at risk for acquisition of severe infections. Using both 16S/internal transcribed spacer community analysis and metagenomics approaches we sought to define the potential infectious risks of marijuana.

Twenty cannabis samples from different dispensaries in northern California were collected by Steep Hill Laboratories (Berkeley, CA) and catalogued, and their DNA was extracted using the PowerSoil DNA Isolation Kit (MoBio, Carlsbad, CA). The fungal communities (assessed by internal transcribed spacer gene sequence analysis) and the bacterial communities (from 16S sRNA gene) were analysed using published methods [4, 5, 6]. More specific whole metagenome analyses were used to better understand the presence of distinct species within the microbiome [6].

Internal transcribed spacer analysis resulted in ∼4000 total fungal taxonomic classifications. Of the 20 fungal genera observed at >0.1% frequency, many were genera that contain opportunistic pathogens (e.g. Cryptococcus, Mucor, Aspergillus) (Appendix 1). Cross-amplification of the 16S primers with plant chloroplast and mitochondria DNA resulted in limited bacterial 16S amplification, as previously reported in microbiome assessments of plant material [7]. However, samples yielded enough 16S data for reliable bacterial community analysis (>5000 operational taxonomic units) in at least one sample (MJ150-008).

Species identification using GOTTCHA [8] on the whole metagenome data for this sample returned multiple pathogens that would probably be attributed to hospital-acquired infection rather than to exposure to marijuana obtained from a dispensary: E. coli, Klebsiella pneumoniae, Pseudomonas aeruginosa, Pseudomonas fluorescens and Pseudomonas putida, Acinetobacter baumannii and Stenotrophomonas maltophilia (Appendix 2) [9]. Although limited in coverage, WMGS reads from the samples analysed mapped to the fungal genomes targeted: sample 138-011 read (n=534) mapping supports the presence of Alternaria alternata and Cladosporium sphaerospermum. Whole metagenome sequence reads from sample MJ150-008 (n=658) indicated the presence of Aspergillus fumigatus, Cryptococcus laurentii and Mucor circinelloides, all well-known causes of invasive fungal infections in immunocompromised hosts.

We found numerous Gram-negative bacilli and fungal pathogens contaminating medical marijuana. These pathogens potentially pose a grave risk to our patients, particularly the immunosuppressed. Although our report is limited to genomic analyses, a few previous studies have cultured multiple fungi (Aspergillus and Penicillium spp.) and bacteria (Klebsiella, Enterobacter, Salmonella and Bacillus) [10] confirming that viable organisms can be recovered from cannabis.

Click here to continue reading the full article at