Commentary: E4orf1 as a key modulator in oncogenesis and of metabolism in Adenovirus infection

1 Introduction

We read with interest the nice article by Hernández-Magaña et al, 2023 (1), which refers to E4orf1, a protein encoded by most human adenoviruses (HAdVs), as oncogenic and generalizes this potential for all HAdV serotypes without supporting evidence. The E4orf1 differs substantially among HAdVs in their nucleotide and protein sequences, structural properties, and biological functions. Importantly, experimental tumorigenicity, defined as tumor formation in susceptible hosts under controlled experimental conditions following deliberate manipulation, should be distinguished from human oncogenesis, which is a naturally occurring disease process in humans involving uncontrolled cell proliferation and clonal evolution within complex tissue and immune contexts (2, 3). Therefore, these claims need inspection as outlined below.

2 Commentary and discussion

2.1 HAdVs are not reported to cause cancer in humans

The authors implied that certain HAdVs are oncogenic because they induce tumors in primary cells and animal models (1). This statement does not clearly distinguish between tumor formation in experimental models and true oncogenesis in humans. The formation of tumors is not the same as oncogenesis or causing cancer, which is characterized by metastasis (4, 5). Also, the animal models used in these experiments were non-native hosts mainly with impaired or undeveloped immune systems which make them permissive and susceptible to tumor formation that does not reflect the biology of the native hosts.

Among the 110 HAdV serotypes, only a few serotypes, from species A (HAdV-12, -18, -31), B (HAdV-3, -7), C (HAdV-2, -5), and D (HAdV-9, -10), exhibit oncogenic potential, exclusively in vitro or in non-native models. For example, HAdV-A12, readily induced tumors in newborn Syrian hamsters, natal multimammate mice, C3H/Bi mice and Fisher rats (6–8), while HAdV-A18, B3, B7, C2 and C5 showed varying degrees of transformation or tumor induction (9–12). Notably, E4orf1 was essential for tumor formation by HAdV-D9 in 24- to 48 hour- old newborn Wistar/Furth rats (13). However, these models such as nude mice, newborn or thymectomized hamsters have underdeveloped or immunocompromised immune systems (6, 9, 14–16). The tumorigenesis observed in these models does not constitute evidence of oncogenesis in humans, the native host (6). The thymus-mediated immunity is crucial for tumor surveillance. Immunocompetent hamsters did not develop tumors when challenged with HAdV-2-transformed cells, in contrast to the high incidence of tumor formation observed in newborn and neonatally thymectomized hamsters (16). In natural hosts, effective immune responses prevent tumor development by oncogenic viruses (17). Unlike carcinogenic viruses such as Human papillomavirus, Hepatitis C virus and Epstein–Barr virus (18), HAdVs have not been reported to cause cancer in humans, nor, have the large-scale tumor studies detected adenoviral DNA or gene expression in human tumors (19, 20).

2.2 HAdV-D36 is not reported to cause tumors

The authors’ statements, especially in the abstract, may give the impression that HAdV-D36 belongs to the group of tumorigenic adenoviruses such as HAdV-C2, C5, D9, and A12 (1). HAdV-D36 is not tumorigenic in animals or humans.

HAdV-D36, has been extensively studied for its metabolic effects in various host models. Over three decades of research show that HAdV-D36 enhances glucose uptake, reduces cholesterol and triglyceride levels, and promotes adipogenesis (21). Experimentally infected chicken, mice, rats and non-human primates exhibited increased adiposity and weight gain, without tumor formation or oncogenesis (22–25). HAdV-D36 infection accelerated cell proliferation without inducing tumors or immortalization, in embryonated chicken eggs, a non-native, immunologically immature model (26).

In vitro, HAdV-D36-infected cells increase glucose uptake through an insulin-independent mechanism involving activation of the PI3K/Akt pathway (27). This involves increased cAMP and Akt activity, leading to GLUT4 translocation to the cell membrane (28). HAdV-D36 E4orf1 plays a key role by interacting with Dlg1 (drosophila Disc large protein), activating Ras, and stimulating PI3K/GLUT4 signaling independent of insulin (29). While uncontrolled PI3K/Akt activation contributes to cancers, its mere activation does not imply oncogenesis, as PI3K/Akt activation is essential for cellular functions such as glucose homeostasis, lipid metabolism, protein synthesis, cardiomyocyte survival, pancreatic β-cell function, central nervous system regulation and immune development (30–32). The biological results of PI3K/Akt activation are highly context -dependent such as cell type, intensity and duration of signaling. PI3K/Akt activity can promote glucose uptake and metabolic regulation or, in other settings, support cell survival, proliferation, migration, or therapy resistance (33, 34). Its activation by HAdV-D36, is needed to probably support metabolic functions like adipogenesis, not oncogenic transformation.

2.3 HAdV-D36 early region 4 open reading frame 1 is not reported to cause oncogenesis

The authors referred to E4orf1 as a “key modulator of oncogenesis” (1). This generalization overlooks its functional variability across adenovirus serotypes. While E4orf1 is present in most HAdVs, its DNA and protein sequences differ significantly, leading to distinct structural and functional outcomes (35). Moreover, adenoviruses such as HAdV-F40 and F41, which lack E4orf1, have shown cellular transformation via left terminal genomic regions (36).

Although complete functions of individual E4orf1 proteins among HAdV serotypes remain unclear, their biological roles should be considered in light of serotype-specific variations. HAdV-D36 E4orf1 is not linked to cancer; instead, it shows promise as a therapeutic candidate for treating type 2 diabetes mellitus (37). It activates the Ras/PI3K pathway via interaction with the Dlg1 protein leading to increased expression and activation of Ras, phosphorylated AKT (p-AKT), GLUT4, GLUT1, and adiponectin (37). While Ras/Akt signaling plays a central role in metabolic regulation, Ras activation alone is not sufficient to induce oncogenesis. Ras overexpression in mice did not cause tumors or cancer but improved glycemic control (38). In long-term studies, HAdV-D36 E4orf1 improved glycemic control without tumor formation in mice (39). In silico studies showed that E4orf1 proteins are structurally different and can be grouped into three distinct forms, referred to as proteotypes, each defined by a unique amino acid sequence and potentially distinct function. For species D, three proteotypes were identified. Notably, HAdV-D9 and HAdV-D10 share proteotype #3 and are closely related phylogenetically, whereas HAdV-D36 belongs to proteotype #1 and is phylogenetically distant from these tumorigenic serotypes. (Table 1) (35, 40).

Table 1. Nucleotide and amino acid comparison of HAdV-D36-derived E4orf1 with E4orf1 from potentially tumorigenic adenovirus serotypes and their corresponding proteotypes.

N/A, Not applicable.

MYC proteins are transcription factors that orchestrate gene programs governing cellular growth and metabolism. Their aberrant activation is a hallmark of many cancers, where they drive metabolic reprogramming by promoting pathways that support rapid cell growth and proliferation, including aerobic glycolysis, glutamine utilization, and lipid biosynthesis (41, 42). The claim that HAdV-D36 E4orf1 activates MYC, is unsupported, as their cited reference #28 in (1) involves HAdV-C5, not HAdV-D36. This raises concerns about the serotype-specificity of E4orf1. Such generalizations overlook key DNA, protein, and functional differences among E4orf1 variants, which show notable codon variability between HAdV-D36 and most other tumorigenic serotypes (Table 1). Sequence comparisons suggest that E4orf1 undergoes natural selection and plays a pivotal role in adenoviral evolution (43, 44).

In summary, over the past three decades, we have been studying HAdV-D36 and its genes in cell and animal models and have not observed any oncogenic potential. Nevertheless, due diligence and continuous vigilance remain essential, including safety evaluation and long-term surveillance as E4orf1 is considered for translational or therapeutic applications. We respectfully submit that there is no evidence indicating HAdVs cause cancer in humans. Furthermore, the E4orf1 gene sequence and functions vary among adenovirus serotypes and should not be generalized. Finally, there is no evidence that the E4orf1 of HAdV-D36 is tumorigenic or oncogenic.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Footnotes

Edited by: Amir E. Mehana, Suez Canal University, Egypt

Reviewed by: Alvaro Cerda, University of La Frontera, Chile

Christian Garcia-Sepulveda, Universidad Autónoma de San Luis Potosí, Mexico