Main

Monkeypox virus (MPXV) is a zoonotic Orthopoxvirus1. A global outbreak of MPXV has been ongoing since May 20222, this outbreak being characterized by atypical clinical findings and a disproportionate incidence in men who have sex with men3,4,5.

The first suspected case in France was a male patient who presented at an HIV and sexual health centre on 19 May 2022 with vesiculopustular lesions on the face and the genital organs, and no reported travel history. At clinical examination, the patient presented with one lesion on the lip, another on the nipple and two lesions on the genitalia. MPXV infection at all three sites was confirmed by real-time quantitative polymerase chain reaction assays at the National Reference Center for Orthopoxviruses. The PCR assay, which was adapted from published assays for MPXV, comprised initial screening for the genus Orthopoxvirus6, followed by a second specific real-time quantitative polymerase chain reaction for MPXV7. The viral strain MPXV/France/IRBA2211/2022 was isolated on Vero cells (online Methods). Using transmission electron microscopy (TEM) performed on infected Vero cells, we observed classical pox-like intracellular virions (Fig. 1a) approximately 250 nm × 125 nm in size, with an hourglass-shaped core. The sequences of the genome from both the primary sample and the isolate were identical. We examined the phylogeny of MPXV/France/IRBA2211/2022 by comparison with a dataset of 62 publicly available complete genome sequences and found that it clusters within the newly proposed B.1 lineage in clade IIb8 (formerly designated as the ‘West African’ clade9) (Fig. 1b).

Fig. 1: MPXV/France/IRBA2211/2022 imaging and phylogeny.
figure 1

a, Transmission electron microscope view of MPXV virions (MPXV/France/IRBA2211i/2022) within cells captured at 48 h post infection at higher (left) and lower (right) magnification. More than 20 MPXV-infected cells were imaged at different magnifications, and a representative section is presented here. b, The maximum-likelihood phylogenetic tree of MPXV clade IIb contains sequences from previous endemic circulation in Nigeria and later outbreaks in other countries, including the global outbreak in 2022 (subclade B.1). The tree is rooted according to the full phylogeny of MPXV and clade support values are shown as percentages above the corresponding branches. The tree tips are coloured by country and sequences from this study marked with strain names. MPXV/France/IRBA2211/2022 corresponds to the sequence from the primary sample, while MPXV/France/IRBA2211i/2022 corresponds to the isolated virus. The sequences of both isolates were identical. The E353K label marks a mutation in the F13L homologue occurring in all sequences of the B.1 subclade and 1 subclade A sequence (MT903341).

Source data

Human MPXV infection leads to a wide spectrum of clinical presentation and disease severity, including pauci- or asymptomatic infection as reported within the 2022 outbreak10. The case fatality ratio of MPXV is far less than for smallpox and varies from 10% for clade I (formerly Congo Basin clade), to 3.6% for clades IIa and IIb. Human MPXV infection often begins with a combination of non-specific symptoms such as fever, chills, asthenia, lymph node swelling, back pain and muscle aches. This prodromal phase is followed by a classical rash phase in which papules turn into vesicles, then pustules, then scabs. The number of skin lesions may range from a few to thousands. The 2022 outbreak of MPXV infection is characterized by atypical features including short incubation phase11, often mild clinical presentation, and localized sometimes painful rashes and lesions in genital and perianal sites3,4,5. Importantly, severe disease can occur in immunocompromised patients, pregnant women and children.

No specific treatment is currently approved for MPXV infection. However, several therapies developed to treat smallpox (tecovirimat (ST-246), cidofovir, brincidofovir) have shown broad spectrum activity both in vitro and in animal models against multiple orthopoxviruses, including MPXV, variola virus and rabbitpox12,13,14,15. Brincidofovir treatment of 3 MPXV-infected patients between 2018 and 2021 was complicated by elevated liver enzymes16, but no safety signals have been identified in reports of compassionate use of tecovirimat, making it a potential therapy for patients in the current outbreak. Tecovirimat has been authorized in Europe by the European Medicine Agency under ‘exceptional circumstances’, although its efficacy has not been shown on the current circulating strain17. Tecovirimat has been shown to target vaccinia virus (VACV) F13L gene product VP37 protein, which is required for extracellular virus particle formation. VP37 is highly conserved in the Orthopoxvirus genus. Interestingly, all 2022 MPXV genomes harbour the E353K substitution in the VP37 protein, this substitution being absent from the most recent common ancestor in clade IIb (Fig. 1b), but was previously identified in a single 2018-outbreak genome (MT903341).

To assess the activity of tecovirimat on the 2022 MPXV strain, we performed dose response studies on Vero cells (online Methods). Tecovirimat completely abolished MPXV replication (measured by plaque reduction assay of plaque forming units (p.f.u.)) at 100 nM, with a 50% inhibitory concentration (IC50) of 12.7 nM. We also used a VACV ANCHOR-GFP infection model, in which virus infection and propagation can be visualized in living cells, and found a comparable IC50 (6 to 8.6 nM) (Extended Data Fig. 1). Cidofovir in vitro IC50 for MPXV lineage B.1 in Vero cells was 30 μM (Fig. 2), indicating 3,000-fold lower potency than that of tecovirimat.

Fig. 2: Tecorivimat and cidofovir potency against MPXV/France/IRBA2211i/2022 in vitro.
figure 2

a,b, ST-246 and cidofovir inhibit plaque formation of the MPXV MPXV/France/IRBA2211i/2022 isolate. Vero cells were infected with MPXV/France/IRBA2211i/2022 and treated with indicated concentrations of ST-246 (a) or Cidofovir (b) for 72 h. Lysis plaque inhibition is expressed in %, normalized over control conditions. IC50 and R2 are indicated. Data are presented as mean ± s.d. Experiments were performed twice in triplicate.

Resistance mutations to tecovirimat have been described for various orthopoxviruses18,19 (Extended Data Fig. 2), including cowpox, camelpox or vaccinia viruses, but not MPXV. Disseminated vaccinia developing increasing 50% effective concentration to tecovirimat in an immunocompromised patient treated with tecovirimat and brincidofovir combination was reported in a clinical setting20. Thus, a regimen that includes adherence to treatment should be tested in human clinical trials. It will also be important to monitor the evolution of MPXV lineage B.1 in response to circulation in humans and in response to administration of any therapies, including tecorivimat.

Our results support the use of tecovirimat in the clinical response to the 2022 MPXV outbreak, in particular for immunosuppressed patients. Of note, tecovirimat is not recommended for pregnant women due to the lack of safety data. Clinical trials in humans will be needed to further establish the efficacy of tecovirimat in vivo.

Methods

Ethics

The IRBA is the national reference laboratory for orthopoxviruses, designated by the French Ministry of Health (through the ‘Arrêté du 7 mars 2017 fixant la liste des centres nationaux de référence pour la lutte contre les maladies transmissibles’) to process samples for identification and characterization of MPXV. The patient signed an informed consent form, and the sample subjected to viral genetic characterization was processed in an anonymized fashion. All work with infectious virus was performed in BSL-3 containment laboratories.

Virus detection

Test and control samples were extracted using the QIAamp DNA mini kit (Qiagen) with small modifications. Briefly, pre-extraction inactivation was performed in BSL-3 by adding 200 µl of sample to a tube containing 200 µl of Qiagen buffer AL and 20 µl of proteinase K. The tubes were incubated at 70 °C for 10 min to inactivate the virus before extraction.

The pan-Orthopoxvirus and pan-MPXV PCR assays have been previously described6,7. Each reaction consisted of 5 μl of extracted DNA and 15 μl of iTaq Universal Probes Supermix (BioRad), the latter containing 0.4 µM of each primer (Eurogentec), 0.25 µM of probe in the pan-Orthopoxvirus PCR and 0.2 µM of probe in the pan-MPXV PCR. All assays were performed on a CFX96 thermocycler (BioRad). Data and results were analysed and reported using BioRad CFX Maestro.

Virus isolation, titration and production

The swabs used for sample collection were discharged in universal transport media. After homogenization, the samples were diluted 1:100 in Dulbecco’s modified Eagle medium (DMEM) containing 0.4% fetal bovine serum. To inhibit bacterial and fungal contamination, antibiotics and antifungals were added (gentamycin: 2.5 µg ml−1, Mycostatin: 10 U ml−1, penicillin: 100 U ml−1, streptomycin: 100 µg ml−1). Following clarification at 700 g for 10 min, we inoculated 50 μl of the samples onto monolayers of Vero cells (African green monkey kidney, ATCC CCL-81) in 96-well plates for titration (8 replicates per dilution) and isolation. The plates were incubated at 37 °C under 5% CO2 and observed daily for cytopathic effects (Extended Data Fig. 3). The viral titre was calculated according to the Reed and Muench method and expressed in terms of median tissue culture infectious dose per ml21.

The supernatant from a well corresponding to the sample collected from the nipple lesion, in which cytopathic effect was observed, was used to inoculate a T25 flask. At day 3, upon observation of 80% cytopathic effect, the flask was frozen/thawed 3 times and then clarified by centrifugation at 1,200 g for 10 min (passage 2). Passage 2 stock was titred using plaque forming assays in 24-well plates and sequenced. The titre was 4.1 × 106 p.f.u. per ml.

TEM

Two days after infection, Vero cells seeded in a 6-well plate were fixed with 2.5% (v/v) glutaraldehyde in sodium cacodylate buffer (0.1 M; pH 7.4; 10 mM CaCl2, 10 mM MgCl2 and 2% glucose) for 4 d at 4 °C. After washing samples with a mixture of saccharose (0.2 M) and sodium cacodylate (0.1 M), cells were postfixed with 1% (v/v) osmium tetroxide in cacodylate buffer for 1 h at room temperature. Samples were stained with 2% (v/v) uranyl acetate for 1 h at 4 °C, followed by gradual dehydration with increasing ethanol concentrations. Samples were embedded in Epon LX112 resin (Ladd Research Industries) in embedding capsules and polymerized for 24 h at 60 °C. Embedded samples were cut in ultrathin 80 nm sections with a UC6 ultramicrotome (Leica), sections placed onto 300-mesh copper grids, stained with 2% uranyl acetate and lead citrate, and examined under a Philips CM10 TEM microscope (operating at 100 kV, and equipped with a Denka LaB6 cathode and a CCD Erlanghsen 1000 Gatan camera). No filtering procedures were applied to the images. More than 20 MPXV-infected cells were imaged at different magnifications.

Sequencing

Viral DNA was extracted from the clinical sample (a swab from a lesion near the nipple) and from the passage 2 stock of the isolated virus using the Macherey Nagel pathogen kit according to the manufacturer’s instructions after DNase treatment. Libraries were prepared in parallel for Oxford Nanopore Technologies and Illumina. We used the ligation sequencing kit (SQK-LSK109) after whole-genome amplification (Cytiva kit, Thermo Fisher), following manufacturer’s instructions.

Data were filtered with kraken/bracken (v2.1.2/v2.6.2) to identify reads matching the chordopoxvirinae family. We mapped reads to monkeypox virus strain Israel_2018 (MN648051.1) with minimap2 v2.24. From this read mapping, a consensus sequence was extracted using nanopolish (v0.13.2) and bcftools (v1.14). We obtained an average coverage depth of 620X (53,632 mapped reads out of 130,176).

Libraries from the clinical isolate and passaged virus were also prepared for Illumina sequencing using the NEBNext UltraII DNA kit (New England Biolabs) after shearing with an M220 Covaris device (Covaris) following the manufacturer’s recommendations. Non-amplified (‘native’) as well as amplified extracts were processed for the clinical sample. After size selection with AMPure beads (Beckman), sequencing was carried out using a NextSeq 550 instrument (Illumina) with an HT 2 × 150 bp cartridge.

We mapped the reads with bwa mem (v0.7.17-r1188) and obtained two datasets with marked differences between native and amplified samples. We obtained 46.2% of mapped reads for MPXV/France/IRBA2211/2022 after amplification, and only 9.5% for the native extract. We attributed the difference to the additional DNase treatment performed for amplified samples. There was no nucleotide difference between the two sequences, indicating that amplification did not alter original sequences.

We then aggregated the sequencing data from the two platforms by using the Nanopore consensus obtained from minimap2 as reference for mapping the Illumina reads. We filtered the sam file to remove unmapped reads and used Geneious R11 (Biomatters) to extract a consensus sequence (ON755039).

Phylogenetic analysis

All available MPXV sequences were retrieved from GenBank on 3 July 2022. This dataset was subset to keep 62 published and discussed sequences22 belonging to the newly proposed clade IIb. We used Nextalign as implemented in the MPXV Nextstrain23 build to align sequences, also adopting the Nextstrain masking strategy (https://github.com/nextstrain/monkeypox). The alignment was visually inspected for accuracy. A maximum-likelihood phylogeny was inferred with IQ-TREE v2.0.624, with 1,000 (ultrafast) bootstrap25 replicates for branch support estimates. The tree was rooted between MPXV strain Nigeria-SE-1971 (KJ642617) and sequences of subclade A in accordance with the full comprehensive MPXV phylogenetic tree.

F13L homologue alignment

F13L homologue sequences were extracted from the available full-genome orthopoxviruses tested for susceptibility to tecovirimat: VACV (AY243312.1), VARV (DQ441447.1), CPXV (AF482758.2), CMLV (AY009089.1), MPXV (HM172544.1), ECTV (AF012825.2)19 and RPXV (AY484669.1)26. Amino acid sequences were aligned with MAFFT v7.45027. Annotations for mutations associated with resistance to tecovirimat18 were added to the alignment plot in Geneious Prime (Biomatters).

Antiviral assays

For MPXV infections, Vero cells (ATCC CCL-81) were seeded in 6-well plates (1.5 × 106 cells per well). At 24 h post seeding, cells were infected with MPXV/FranceIRBA2211i/2022 to obtain 50–100 p.f.u. per well. After adsorption for 1 h, the inoculum was replaced with DMEM and 2% FBS containing the appropriate concentration of tecorivimat. After 1 h at 37 °C, cells were then overlaid with 1.6% carboxymethyl cellulose (VWR) diluted in DMEM and 2% FBS. The plates were incubated for 3 d at 37 °C in a 5% CO2 incubator. Monolayers were fixed and stained in 3.7% formaldehyde, 0.1% crystal violet and 1.5% methanol. The plaques were counted microscopically by eye. All experiments were performed in triplicate.

For VACV ANCHOR infections, HeLa cells (ATCC CCL-2) were grown in DMEM without phenol red (Sigma Aldrich), supplemented with 10% fetal bovine serum (Eurobio-Scientific), 1 mM sodium pyruvate (S8636, Sigma Aldrich), l-glutamine (G7513, Sigma Aldrich) and penicillin-streptomycin solution (P0781, Sigma Aldrich). Cells were plated (9,000 cells per well) in Corning Cellbind 96-well plates in complete DMEM. At 24 h post seeding, cells were treated with the test compound at the indicated concentration in triplicate. Cells were then infected with VACV ANCHOR virus (Copenhagen strain with the ANCHOR system in the TK locus) at multiplicity of infection 0.1 and incubated at 37 °C and 5% CO2. At 48 h and 72 h, cells were fixed with 4% formalin (Sigma) for 10 min at r.t., washed with PBS and incubated with PBS-Hoechst 33342 (1 mg ml−1).

Data acquisition by high-content microscopy was performed on a Thermo CellInsight CX7 HCS microscope using a compartmental analysis algorithm. Results were extracted, normalized over the vehicle-treated condition and expressed as the average of three independent wells ±s.d. IC50 were calculated using GraphPad Prism v9.

Reporting summary

Further information on research design is available in the Nature Research Reporting Summary linked to this article.