Detecting Monkeypox Virus by Immunohistochemistry

ABSTRACT

Background

Mpox (formerly known as monkeypox), a zoonotic disease caused by Monkeypox virus (MPXV), has become an international outbreak since May 2022. Mpox often presents with a mild systemic illness and a characteristic vesiculopustular skin eruption. In addition to molecular testing, histopathology of cutaneous lesions usually shows distinctive findings, such as epidermal necrosis, balloon degeneration, papillary dermal edema, and focal dermal necrosis, which have proven helpful in the diagnosis of mpox. Viral cytopathic changes with areas of multinucleation, smudging of the nuclei, and intracytoplasmic inclusions have also been described. Although useful, these features are relatively nonspecific. The use of a monoclonal antibody for immunohistochemical (IHC) staining of MPXV may be a useful tool in confirming mpox infection.

Methods

Three cases of PCR‐confirmed mpox were biopsied and subjected to IHC staining with a monoclonal MPXV‐specific antibody targeting viral envelope protein A29. As controls, cell lines transduced to express other MPXV viral antigens and samples of cutaneous viral infections involving Molluscum contagiosum, Herpes simplex, Herpes zoster, or Cytomegalovirus were also subjected to IHC staining with this antibody.

Results

All three mpox patient biopsies performed on lesional skin subjected to MPXV IHC staining reliably detected viral infection in lesional skin with a diffuse cytoplasmic and focally nuclear staining pattern. No staining was seen in transduced cell lines expressing off‐target MPXV viral antigens and in lesional skin of other common viral infections listed above.

Conclusions

The monoclonal MPXV‐specific antibody may be used as an adjunct tool to confirm mpox infection.

1. Introduction

Mpox (formerly known as monkeypox) is a zoonotic disease caused by Monkeypox virus (MPXV), which belongs to the Orthopoxvirus genus of the Poxviridae family. Historically, mpox has been largely confined to Central and West Africa. However, since May 2022, it has emerged as a dynamic and multinational outbreak, affecting over 95 000 individuals and recording a total of 171 deaths in 116 countries [1]. Although animal–human transmission was considered the major source of infection, the current outbreak is transmitted primarily through skin‐to‐skin contact [2]. Polymerase chain reaction (PCR) of skin lesions from the patient is considered the gold standard for mpox diagnosis and MPXV detection. However, examination of skin lesions by histopathology and immunohistochemistry (IHC) have also been proven helpful in diagnosing mpox. Prior reports have utilized polyclonal antibodies against Vaccinia virus, without a defined viral antigenic target [3]. This can lead to nonspecific staining, especially in areas of necrosis, which may be mistakenly interpreted for true positive staining [4]. Here, we report the first use of a MPXV‐specific monoclonal antibody for IHC detection of the MPXV in lesional skin of three immunocompromised patients with PCR‐confirmed mpox infection.

2. Materials and Methods

Under an institutional IRB, three cases of suspected mpox cutaneous lesions were subjected to IHC detection utilizing the MPXV A29 antibody (see below). Staining patterns were qualitatively evaluated by dermatopathologists LM and KB. Skin lesional specimens from Molluscum contagiosum (n = 1), Herpes simplex virus‐1 (HSV‐1; n = 5), Herpes zoster virus (VZV; n = 5), and Cytomegalovirus (CMV; n = 3) infections were also stained with the antibody. For specificity analysis, HEK293 cells transfected with various commercially obtained MPXV proteins (A29L, A30L, A35R, B16R, D6L, H3L, I1L; Raybiotech, Peachtree Corners, GA) as well as nontransfected HEK293 cells were additionally stained with the antibody.

MPXV IHC staining was performed using a commercially available antibody from Sino Biological (Wayne, PA), Clone 0006 from 1:1000 dilution (0.6 μg/mL) to 1:100 000 dilution (0.006 μg/mL). Staining was performed on the Leica Bond III auto‐staining system using Bond Polymer Refine Detection (DS9800; Deer Park, IL). The antigen retrieval used was Epitope Retrieval 2 (AR9640) for 30 min and 30‐min primary incubation time. All IHC photomicrographs were stained using 1:1000 primary antibody dilution.

3. Results

Case #1 involves a 54‐year‐old male with a history of HIV/AIDS with inconsistent adherence to antiretroviral therapy, Kaposi sarcoma, syphilis, recurrent perianal HSV infection, and verruca vulgaris. He presented with large keratotic verrucous plaques on the right nasal dorsum, fingers (Figure 1A,C), and perianal areas, as well as crusted, scalloped plaque on the left cheek (Figure 1B) for a period of 4 weeks. At the time the patient was evaluated, he had an absolute CD4 T cell count of 24/μL and a viral load of 35 000 copies/mL. The biopsy from the left cheek revealed ballooning degeneration, papillary dermal edema, and necrosis in the epidermis, dermis, and follicular epithelium in association with a mixed inflammatory infiltrate (Figure 2A,B). The findings were highly suspicious for viral infection. IHC stains for HSV‐1 and HSV‐2 were negative. Additional infectious stains for bacteria, fungi, and acid‐fast bacteria were also negative. Given the suspicion for possible MPXV infection, an experimental protocol for evaluation of in situ viral infection was pursued utilizing a commercially available antibody targeting the A29 protein antigen of MPXV (Sino Biological, Clone 0006 with 0.6 μg/mL of antibody). Positive staining was detected predominantly in the areas of infected, viable epidermis, with particularly strong reactivity in the follicular epithelium (Figure 2C). A diffuse cytoplasmic staining pattern was seen in keratinocytes of the follicular epithelium with few cells showing nuclear staining. Notably, no discrete staining is seen in necrotic tissue or other cell types. The patient's MPXV infection was subsequently confirmed using a PCR‐based assay. He began treatment with tecovirimat with resolution of cutaneous lesions within 4 weeks.

FIGURE 1.

Case #1 with a crusted plaque on the right nasal dorsum (A), an ulcerated plaque on the preauricular cheek (B), and large keratotic verrucous plaques on the fingers (C).

FIGURE 2.

A punch biopsy from the plaque on the left cheek lesion of Case #1 revealed ulceration, epidermal and dermal necrosis, ballooning degeneration, papillary dermal edema, and necrosis of the follicular epithelium. An MPXV immunohistochemical stain revealed a diffuse cytoplasmic staining in the follicular epithelium (A) H&E, 40×; (B) H&E, 200×; (C) MPXV IHC, 10×.

Case #2 is a 54‐year‐old male with a history of renal transplant on an immunosuppressive medication regimen consisting of cyclosporine, mycophenolate mofetil, and prednisone. The patient initially presented to an outpatient dermatology clinic with an indurated growth on the left neck (Figure 3A). The biopsy done at this visit showed ulceration with focal epidermal and dermal necrosis on hematoxylin and eosin (H&E) (Figure 4A) and tissue culture done at that time did not grow any organism. Over the next 3 weeks, the lesion continued to grow and became more painful despite multiple courses of broad‐spectrum oral and intravenous antibiotics (Figure 3B,C). Importantly, the patient also developed a new lesion on the left lateral wrist while he remained on broad‐spectrum antibiotics. Two additional biopsies were done on the lesion on the neck, and one biopsy was done on the new lesion on the left wrist. All three biopsies were suggestive of an infectious process but IHC stains for HSV‐1, HSV‐2, and VZV and special stains for organisms (PAS, GMS, Fite, Gram, and Warthin Starry) were negative. At this time, Karius testing of the patient's blood revealed 1648 copies of MPXV and PCR swabs were obtained to confirm the diagnosis. The patient was started on oral tecovirimat and intralesional injection of cidofovir with resolution of his cutaneous lesions. The MPXV antibody for immunohistochemistry was performed on previously collected biopsies and showed diffuse cytoplasmic staining within infected epidermal and follicular keratinocytes with focal nuclear staining in all specimens (Figure 4B,C).

FIGURE 3.

Case #2 with an enlarging ulcerated nodule on the left lateral anterior neck at the initial visit (A), at the time of biopsy (B), and after a two‐week follow‐up (C).

FIGURE 4.

A punch biopsy from the edge of the ulcer of Case #2 revealed ulceration with focal epidermal and dermal necrosis. MPXV immunohistochemical stain showed diffuse cytoplasmic and nuclear staining in the infected keratinocytes in the epidermis and follicular epithelium with rare nuclear staining. (A) H&E, 100×; (B) MPXV IHC, 200×; (C) MPXV IHC, 200×.

Case #3 involves a 63‐year‐old male with a history of HIV/AIDS and diffuse large B cell lymphoma in remission after recent chimeric antigen receptor T cell (CAR‐T) therapy. The patient presented with a nonhealing perianal ulcer of 5 days' duration (Figure 5) and a new onset pustular eruption over the face, trunk, and extremities. Punch biopsy of the perianal ulcer showed full‐thickness epidermal necrosis with a mixed dermal inflammatory infiltrate (Figure 6A,B). Use of the MPXV antibody for IHC demonstrated diffuse cytoplasmic staining in non‐necrotic, infected keratinocytes (Figure 6C,D). Some cells displayed a more granular staining in the cytoplasm. A viral PCR swab from the perianal ulcer was positive for non‐Variola Orthopoxvirus and negative for HSV‐1, HSV‐2, and VZV DNA. A separate swab from a pustular lesion on the back was positive for VZV. Given these concomitant viral infections, the patient was initiated on intravenous acyclovir and oral tecovirimat treatment. At 3 weeks posttreatment, all lesions had resolved.

FIGURE 5.

Case #3 with a perianal ulcer.

FIGURE 6.

A punch biopsy of the ulcer from Case #3 showed full‐thickness epidermal necrosis with mixed inflammation. MPXV immunohistochemical stain demonstrated diffuse cytoplasmic staining in the infected keratinocytes in the epidermis and follicular epithelium with rare nuclear staining. (A) H&E, 40×, (B) H&E, 200×, (C) MPXV IHC, 200×, and (D) MPXV IHC, 200×.

4. Discussion

Here, we report three cases of patients with mpox, who presented with distinct clinical characteristics, histopathological findings, and striking immunoreactivity with a monoclonal MPXV‐specific antibody. The age of the patients ranged from 54 to 63 years with a mean age of 57. All three patients were immunocompromised and developed atypical skin lesions. All lesions were rapidly progressing clinically to large plaques and nodules, representing an escalation in severity from the usual vesiculopapular eruptions that are seen in immunocompetent patients. In line with other descriptive case studies, our cohort of immunocompromised patients demonstrated more severe cutaneous disease and a protracted course compared to immunocompetent patients [5, 6, 7]. The diagnosis of mpox was confirmed in all three cases via viral PCR swabs and by Karius testing in Case #2. All three patients were treated with antiviral medication and demonstrated immediate clinical improvement at 2 months follow up.

The clinical course of the current mpox outbreak has been relatively well characterized. The incubation period ranges from 5 to 21 days. Initial symptoms include fever, headache, and asthenia. Most patients develop skin lesions a few days after the onset of initial symptoms [4]. These lesions are typically papules and/or pseudopustules with central necrotic umbilications, which eventually become hemorrhagic crusts. In our series, all three patients were severely immunocompromised as they either had HIV infection or malignancy or were iatrogenically immunosuppressed with medications. Their lesions were larger and more atypical compared with typical mpox lesions seen in immunocompetent individuals. This observation is in line with published findings of mpox in this population, who have been found to have more prolonged illness, larger lesions, and higher rates of secondary bacterial infections and genital ulcers [8].

The biopsies of all three cases showed well‐documented histopathologic features as previously reported, including epidermal and dermal necrosis, ballooning degeneration, and papillary dermal edema. Viral cytopathic changes were observed with areas of multinucleation, smudging of the nuclei, and intracytoplasmic inclusions [4]. These features are relatively nonspecific and make identifying the types of viral infection difficult. Previous attempts to detect MPXV in lesional skin have been made with the use of a polyclonal antibody‐targeting Vaccinia virus, which are not specific to MPXV and have been shown to stain necrotic keratinocytes and sebocytes, and may not necessarily reflect true viral infection [3, 4]. We have now demonstrated the use of a monoclonal MPXV‐specific antibody that can reliably detect viral infection in lesional skin. Positive staining is recorded as diffuse cytoplasmic staining in keratinocytes within the infected, viable epidermis and/or follicular epithelium, with focal nuclear staining. No immunoreactivity was seen in the dermal fibroblasts or endothelial cells. Prior studies utilizing IHC approaches to detect MPXV have found staining in a diffuse cytoplasmic pattern, reflecting where virion replication and assembly takes place and is evident in all cases examined in our cohort. MPXV may also create “mini‐nuclei” in infected cells due to reorganization of rough endoplasmic reticulum in an enclosed structure, which may explain rare instances of apparent nuclear staining seen in one of our cases [9]. In Cases 1 and 2, there was a particularly strong reactivity involving the follicular epithelium, which has been observed in mpox and other cutaneous viral infections [10, 11].

Similar to other orthopoxviruses, MPXV exists in two distinct infectious forms, the intracellular mature virus and the extracellular enveloped virus, each carrying different surface glycoproteins and utilizing different mechanisms to infect cells [12]. The monoclonal antibody we utilized in this study targets the A29 heparin sulfate binding protein that mediates virion entry into the host cell [13]. The A29 protein is found in the mature virus and is homologous to the A27 protein found in Vaccinia virus (VACV) and the A30 protein in Variola virus [14]. While this antibody has been shown to cross‐react with VACV A27 protein under reducing conditions (per manufacturer data), it has not been shown to do so in IHC applications. To determine the antibody's specificity, it was also used to stain HEK293 cells transfected with individual MPXV proteins, including A29L, A30L, A35R, B16R, D6L, H3L, and I1L. Only the target A29L transfectants were stained with the antibody, while stainings for the other off‐target MPXV proteins were negative (data not shown). Further testing showed a lack of antibody reactivity in the lesional skin of patients with Molluscum contagiosum, HSV‐1, VZV, and CMV (data not shown). Nevertheless, the MPXV anti‐A29 antibody is notably more specific than the commonly used polyclonal anti‐Vaccinia virus antibody, making it a valuable tool for confirming the presence of MPXV infection. This study is limited by the low sample size, the lack of testing on other Orthopoxviruses, and the likely inability of the antibody to distinguish between clades of MPXV.

It is the current standard of care to confirm all suspected mpox cases through molecular testing. In addition to PCR, histopathologic and immunohistochemical study of skin lesions can serve as valuable confirmatory tool, especially in cases where mpox was not on the initial differential diagnosis or in instances where multiple viral infections (e.g., Herpes simplex, Herpes zoster, Molluscum contagiosum, etc.), bacterial infections (e.g., chancroid, lymphomogranuloma venereum, etc.), or inflammatory conditions (e.g., genital aphthous ulcers) are under diagnostic consideration. Furthermore, although molecular testing may be a more sensitive method of viral detection, potential false‐negative results from inadequate sample collection may be overcome with adjunctive confirmatory assays, such as IHC [15, 16].

Ethics Statement

This study was conducted in accordance with an approved institutional IRB protocol (20‐02021524; Weill Cornell Medical College).

Conflicts of Interest

The authors declare no conflicts of interest.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.