In Vitro Models for Emerging Infectious Disease Detection and Host–Pathogen Interaction Studies

Abstract

Many emerging and re-emerging infectious diseases have been observed over the last few decades around the globe due to population growth, international travel, environmental changes, and microbial adaptation and evolution, despite advances in the medical field. The spread of these diseases is related to complex interactions between pathogens and their hosts. Accordingly, this review summarises current knowledge on infection development and discusses methods used for detection and modeling. Recent studies have revealed the limitations of two-dimensional models and increasingly rely on 3D systems, including spheroids, organoids, and organ-on-a-chip systems, that offer more realistic tissue environments, allowing researchers to more effectively study host–pathogen interactions. Overall, the integration of complementary approaches and the development of 3D models are crucial for enhancing diagnosis, developing new therapeutic approaches, and strengthening control strategies of emerging outbreaks.

1. Introduction

Throughout antiquity, humanity has documented severe infectious diseases that have emerged and re-emerged over the globe. More recently, in January 2023, 64 distinct infectious diseases were documented worldwide, reported across 235 countries and regions [1]. Over the last two decades alone, several major epidemics have recurred within short intervals, such as Severe Acute Respiratory Syndrome (SARS) in 2003, the H1N1 influenza pandemic in 2009, Middle East Respiratory Syndrome (MERS) in 2012, the Ebola virus epidemic in West Africa between 2013 and 2016, and the Zika virus outbreak in 2015 and the COVID-19 pandemic that began in 2019 [2]. In recent years, the world has seen the re-emergence of several infectious diseases, such as monkeypox, dengue fever, measles, cholera, and invasive group A streptococcus infection [1]. These epidemics remind us that new and recurrent infections remain a serious public health problem worldwide.

Many factors are involved in the emergence of these diseases, including population growth, international travel and trade, human activity, environmental and ecological impact, industrialisation, deficiencies in public health systems, and pathogen adaptation and resistance [2]. The emergence of novel infectious diseases could pose severe global health challenges, as observed over the last 500 million years [3].

Multimodal surveillance systems involving several disciplines have been established to protect human health as well as animals and ecosystems [4]. Nevertheless, due to the limited integration of crucial biological components such as host–pathogen interactions, pathogen reservoirs, and environmental factors, these programs continue to face some difficulties in predicting emerging and/or re-emerging outbreaks.

Studies increasingly indicate that infectious diseases are attributed to complex host–pathogen interactions, involving dynamic bidirectional exchanges, some of which remain poorly understood [5]. Indeed, genetic diversity, immunosuppression, immune system disorders, malnutrition, ageing, and environmental influences are the main factors that can make individuals more susceptible to emerging infections [2]. Pathogens can also evolve rapidly, adapting their gene expressions to novel environments, colonising diverse ecological niches, and mitigating the effects of both biotic and abiotic stresses [6].

In view of the above, this review aims to highlight the interactions between human pathogens and their hosts, summarise current knowledge on pathogen infection development, and emphasise the methods used for detection and modeling.

2. Host–Pathogen Interactions

The study of infectious diseases depends on understanding how pathogens interact with their hosts. These interactions elucidate the mechanism of infection and host immune response. Generally, infection develops through four main stages: (1) host invasion and colonisation, (2) evasion of immune defense, (3) intracellular replication and spread, and (4) immune recognition and elimination of the pathogen [7] (Figure 1). In addition, pathogens can produce toxins and other virulence factors that influence disease severity. The resulting host response generates measurable biomarkers, which can serve as useful indicators for disease progression and clinical diagnosis.

2.1. Host Invasion by Overcoming Primary Barriers

The invasion of a host by a pathogen starts by overcoming the primary host barriers. This process is often mediated by specialised protein secretion systems that inject effector proteins into host cells. These proteins facilitate pathogen adhesion, internalisation, survival, and multiplication. Likewise, some pathogens could exploit sphingolipids to facilitate their adhesion on host cells and facilitate intracellular survival [8].

To date, six bacterial secretion systems (Type I–IV) have been described [9]. For example, Salmonella enterica uses a type III secretion system to inject proteins into the host cytosol, resulting in a vacuole formation that protects bacteria and facilitates their spread [10]. In contrast, Chlamydia enters host cells by endocytosis and forms intracellular inclusions that mimic host membranes [11]. These structures interfere with vesicular transport, allowing the pathogen to obtain the nutrients and complete its development cycle [11]. Although bacterial systems have been extensively studied, similar fundamental processes are observed in viral and eucaryotic pathogens, such as protozoan and fungal pathogens. For example, viral infections initiate through specific interactions between viral attachment receptors, such as cell adhesion molecules, sialic acid, and phosphatidylserine receptors on host cells, which enable the virus to surmount the plasma membrane barrier, facilitate entry into host cells, and access the intracellular machinery required for replication [12]. Eukaryotic pathogens, including parasites and fungi, also display specialized invasion strategies. Plasmodium spp. invade host hepatocytes and, after liver stage development, merozoites engage specific erythrocyte receptors and active entry processes involving parasite adhesins and motor complexes to penetrate red blood cells and form a parasitophorous vacuole [13]. Similarly, Candida albicans adheres to epithelial surfaces through adhesins such as Als3 and Ssa1 and invades host tissues by combined induced endocytosis and active hyphal penetration facilitated by hydrolytic enzymes [14].

Secretion systems function as sophisticated protein injection machines that play central roles in host–pathogen interactions. Type II secretion systems (T2SS) are commonly used by Gram-negative bacteria, such as Escherichia coli and Vibrio cholerae, to export folded toxins and hydrolytic enzymes across the outer membrane, contributing to epithelial barrier disruption and disease progression [15].

In contrast, Type III secretion systems (T3SS) act as molecular syringes that inject effector proteins directly into the host cell cytosol. This system is utilized by several invasive pathogens, including Salmonella enterica and Yersinia pestis, to subvert host cytoskeletal dynamics and modulate immune signaling. For instance, Salmonella enterica employs a T3SS to induce the formation of a membrane-bound vacuole that protects bacteria and facilitates intracellular survival and dissemination. Similarly, Chlamydia species enter host cells via endocytosis and form intracellular inclusions that mimic host membranes, thereby hijacking vesicular trafficking pathways to acquire nutrients and complete their developmental cycle [16].

Other pathogens rely on Type IV secretion systems (T4SS), which are evolutionarily related to bacterial conjugation machinery and are capable of translocating proteins or nucleoprotein complexes into host cells. Legionella pneumophila and Coxiella burnetii use T4SSs to deliver large repertoires of effector proteins that remodel host endocytic and lysosomal pathways, enabling intracellular replication within specialized vacuoles [17].

Finally, Type VI secretion systems (T6SS), structurally related to contractile bacteriophage tails, are employed by pathogens such as Pseudomonas aeruginosa to inject toxic effectors into neighboring bacterial or eukaryotic cells. This system contributes both to interbacterial competition and to host colonization by eliminating competing microbiota and modulating host defenses [18].

Together, these diverse secretion systems highlight the multiplicity of strategies used by pathogenic bacteria to breach host barriers, manipulate cellular processes, and establish infection.

2.2. Evasion of Host Defenses by Pathogens

Pathogens must evade host immune defenses to successfully establish an infection. Therefore, many bacteria, such as Clostridium and Bacillus have developed strategies to survive in extreme conditions through the production of spores. For example, Spores of Bacillus cereus promote biofilm formation and increase the production of secondary metabolites, like catalase and superoxide dismutase, known for their protective role against oxidative stress and other host immune responses [19]. Direct immunomodulation is employed by other pathogens, such as Bordetella pertussis, which secretes multiple virulence factors, including pertussis toxin (PTX), interfering with both innate and adaptive immune responses. This disorder enables long-term persistence and increases pathogenicity. Beyond immune evasion, PTX also affects host physiology by altering pancreatic beta-cell function and disrupting glucose regulation [20]. Similarly, viruses have developed sophisticated immune evasion mechanisms like modifying immunodominant epitopes, interfering with the cellular immune response, and interfering with the effector functions of the immune system by changing cytokine expression or blocking apoptosis [21]. Eukaryotic pathogens also employ complex strategies to evade host immunity, often involving antigenic variation and immune modulation. For instance, African trypanosomes evade the immune responses through the intermittent switching of their major variant surface glycoprotein, associated with tolerance to high levels of interferon-gammaand avoiding complement-mediated destruction [22].

2.3. Pathogen Replication Within the Host

Pathogens must successfully replicate within their hosts in order to establish the infection and ensure the transmission to others. Chlamydia species provide a typical model of this adaptive strategy. The infectious elementary body enters host cells and subsequently transforms into a metabolically active reticulate body, which facilitates intracellular replication and produces effectors modulating host responses. While the elementary body primarily initiates and disseminates infection, reticulate bodies replicate within specialized vacuoles; chlamydial inclusion bodies [8]. Due to these inclusions, a close contact is established with the host cell organelles, particularly the Golgi apparatus, in order to acquire essential nutrients required for bacterial multiplication and proliferation. After many rounds of divisions, reticulate bodies convert into elementary bodies, which are released during host cell lysis, thereby facilitating the dissemination of infection [13]. On the other hand, viruses are entirely dependent on their host cells for the production of progeny viruses [23]. Viral replication within host cells is driven by genome-specific replication strategies that exploit host transcriptional, translational, and membrane-associated machineries, leading to the formation of specialized replication compartments that coordinate viral genome synthesis, protein production, and virion assembly while actively modulating host antiviral responses [24].

2.4. Host Immune Control and Pathogen Elimination

The host triggers immune defenses to mitigate pathogen invasion and colonization. Firstly, the innate immune system intervenes via pattern recognition receptors (PRRs) of various immune cells. These receptors recognise conserved pathogen-associated molecular patterns (PAMPs), including ligands detected by endosomal and cytosolic receptors, involved in the assembly of inflammasomes. Once recognised, PRRs activate protein complexes such as NLRP1, NLRP3, AIM2, ASC, and caspase-1 that stimulate the maturation and release of pro-inflammatory cytokines, including IL-1β and IL-18. The host response can control infection (e.g., by Bacillus cereus) by inducing pyroptosis, a form of programmed cell death associated with inflammation [24,25].

2.5. Pathogen Toxins and Virulence Factors

Many bacterial pathogens cause disease through the production of toxins and virulence factors that directly damage host tissues and disrupt immune responses. Staphylococcus aureus virulence is largely mediated by a broad array of secreted toxins that induce host cell injury, inflammation, and immune dysregulation. These include pore-forming toxins such as α-, β-, and γ-hemolysins and leukocidins, which damage leukocytes and epithelial cells. Phenol-soluble modulins further contribute to cytolysis and pro-inflammatory signaling. Exfoliative toxins compromise epidermal integrity, while superantigens trigger massive, non-specific T-cell activation. Collectively, these toxins promote tissue destruction, neutrophil killing, and systemic inflammatory responses, contributing to disease severity across a wide range of infections [26,27].

The pathogenicity of Bacillus cereus relies on a diverse repertoire of toxins and tissue-damaging enzymes involved in both diarrheal and emetic syndromes. Diarrheal disease is primarily mediated by the pore-forming enterotoxins hemolysin BL (HBL), nonhemolytic enterotoxin (NHE), and cytotoxin K (CytK), which damage intestinal epithelial cells by forming membrane pores, leading to microvilli destruction and osmotic lysis. In contrast, the emetic form is caused by cereulide, a heat- and acid-stable cyclic dodecadepsipeptide encoded on a megaplasmid. Cereulide functions as a potassium ionophore and induces vomiting through gut–brain signaling involving serotonin 5-HT3 receptors. Additional hemolysins and metalloproteases, including InhA1 and NprA, contribute to cytotoxicity, immune evasion, and tissue barrier degradation. Membrane-active phospholipases further exacerbate host cell damage, enhancing disease severity and supporting the opportunistic pathogenic potential of B. cereus [28].

In Clostridium perfringens, virulence is primarily toxin-driven and depends on the coordinated action of membrane-damaging exotoxins. The alpha-toxin (CPA), a phospholipase C, is the major virulence determinant in human infections and plays a central role in tissue pathology. By hydrolyzing phosphatidylcholine and sphingomyelin, CPA induces cell lysis, hemolysis, endothelial injury, and myonecrosis [29]. In intestinal disease, C. perfringens enterotoxin (CPE) is the key pathogenic factor. CPE binds claudin receptors on epithelial cells, forms membrane pores, disrupts tight junctions, and triggers calcium-dependent cell death, leading to epithelial barrier breakdown and secretory diarrhea [30,31]. Additional toxins, such as beta- and epsilon-toxins, may further aggravate intestinal and systemic pathology. Accessory factors, including perfringolysin O, act synergistically with CPA to amplify membrane damage and inflammatory responses [30].

The virulence of Pseudomonas aeruginosa is driven by a complex and tightly regulated arsenal of toxins and persistence factors. Exotoxin A inhibits host protein synthesis, while Type III secretion system effectors (ExoU, ExoS, and ExoT) disrupt host cell membranes, cytoskeletal organization, and signaling pathways. Proteases, elastases, lipases, and phospholipases degrade host tissues and immune components. Redox-active metabolites such as pyocyanin induce oxidative stress and impair mucociliary clearance, while extracellular DNA release promotes biofilm formation [32,33]. Biofilms composed of alginate, Psl, and Pel exopolysaccharides provide protection from antibiotics and host defenses. These virulence traits are coordinately regulated by quorum-sensing networks, enabling P. aeruginosa to establish severe acute and chronic infections [34].

Pathogenic Escherichia coli strains employ distinct toxin repertoires depending on their pathotype. Enteropathogenic E. coli uses Type III secretion system effectors to disrupt the cytoskeleton and form attaching-and-effacing lesions. Enterohemorrhagic E. coli produces Shiga toxins that damage endothelial cells and can lead to hemolytic uremic syndrome. Enterotoxigenic E. coli secretes heat-labile and heat-stable enterotoxins that alter intracellular signaling pathways and induce secretory diarrhea. Enteroinvasive and enteroaggregative strains rely on invasion factors, adhesins, and cytotoxins to promote epithelial damage, inflammation, and biofilm formation. Across all pathotypes, these virulence factors collectively drive intestinal injury and disease severity [35,36].

During infection with Vibrio cholerae, pathogenicity depends on the coordinated action of multiple toxins and colonization factors. Cholera toxin (CTX) is the primary virulence determinant and is responsible for the profuse secretory diarrhea characteristic of cholera. CTX production is closely linked to the toxin-coregulated pilus, which is essential for intestinal colonization and acquisition of the CTX-encoding bacteriophage. Accessory toxins, including the multifunctional autoprocessing repeats-in-toxin, hemolysin A, zonula occludens toxin, and accessory cholera enterotoxin, further contribute to epithelial damage and barrier dysfunction. Additional factors such as flagella, outer membrane proteins, secreted proteases, sialidase, and adhesins promote intestinal persistence, immune evasion, and biofilm formation, collectively supporting efficient colonization and disease progression [37].

2.6. Infection Biomarkers

Medical imaging and clinical examination are crucial in disease diagnosis. However, clinicians often rely on biomarkers to confirm the presence of infection. These biomarkers are generally classified into two main groups. The first group includes pathogen biomarkers, which directly detect microorganisms or the substances they produce. Classical methods include microbial culture, whereas more targeted techniques detect specific compounds secreted by the pathogen [38].

The second group includes host biomarkers that reflect how the body responds to infection. These include simple laboratory measures such as total and differential white blood cell counts as well as proteins like C-reactive protein (CRP) [39]. Another key biomarker is procalcitonin (PCT), the precursor of calcitonin normally produced by the thyroid gland, but during infection, is synthesised by multiple tissues and regulated by pro-inflammatory mediators such as TNF-α, IL-1, and IL-6. PCT helps differentiate bacterial infections from viral infections [40].

Despite their clinical utility, host-response biomarkers are still not specific enough, particularly in large populations and in immunocompromised patients. This limitation highlights the need for more accurate biomarkers. Recent studies have suggested several promising molecules, including human neutrophil elastase, matrix metalloproteinases (MMPs), serum myeloperoxidase, and xanthine oxidase, detected in wound exudates [41].

3. Omics Approaches in Host–Pathogen Interaction

Despite the continued relevance of standard microbiological, immunological, and molecular methods, these approaches often provide an incomplete and fragmented view of infection dynamics and disease processes. Elucidating the mechanisms underlying infectious diseases and developing targeted diagnostic and therapeutic strategies requires a comprehensive understanding of host–pathogen interactions. Omics approaches, including genomics, transcriptomics, proteomics, metabolomics, epigenomics, single-cell omics, spatial transcriptomics, radiomics, and microbiomics, have profoundly advanced biomedical sciences by enabling high-resolution and system-wide analyses of both pathogens and their hosts [42,43]. These technologies allow the identification of molecular signatures, characterisation of immune responses, determination of immune repertoires, discrimination between infected and uninfected cells, and identification of pathogen entry receptors and their expression patterns [44].

Fungal multidrug resistance determinants were identified by using a pan-genomic and pan-transcriptomic analysis, enabling the prediction of highly multidrug-adapted fungal strains that may pose a significant clinical concern [45]. Novel virulence mechanisms associated with key requirements for pathogen survival during infection were identified by using spatial transcriptomics, facilitating the further identification of potentially druggable targets [46]. An integrated analysis combining virus–host interaction network and quantitative proteomic approaches revealed that SARS-CoV-2 evolved multiple strategies to control fundamental host cellular processes and evade the immune responses, providing a powerful resource for the development of therapeutic strategies [47]. Recently, Hildebrandt et al. [48] provided a comprehensive examination of the host’s response to Plasmodium infection using a combined approach involving spatial transcriptomics and concluded that omics approaches could be crucial to determine the mechanisms involved in the coordination of immunity in both partial and full immunisation events. A study combined metabolomics and transcriptomics analysis provided insight into metabolic regulation of immune cell function, revealing the required key metabolites for macrophage M1/M2 polarisation states [49,50].

While single-omics approaches are primordial in understanding the host–pathogen interactions, they are generally insufficient to describe the complex interactions and the exact molecular targets in physiological and biological systems [42].

In contrast, integrated multi-omics strategies provide a more holistic understanding of disease mechanisms, functional interactions, and therapeutic targets [42,51].

Single-omics methods often reveal what is present but fail to explain the complex mechanisms of disease or the interconnected biological pathways involved, whereas multi-omic integration offers a systems-level perspective on host–pathogen interactions and molecular communication networks [52].

A comparative multi-omics study integrating transcriptomic, proteomic, and metabolomic analyses of Salmonella Typhi and S. Typhimurium infections revealed distinct pathogen-associated molecular and metabolic signatures. The analysis identified 141 differentially expressed genes and 263 proteins, mainly involved in metabolic pathway modulation and cell cycle regulation. Alterations in central carbon metabolism, including inhibition of the Krebs cycle and enhanced conversion of glucose to pyruvate and lactate, were associated with increased production of reactive oxygen species. These findings indicate that S. Typhi employs metabolic reprogramming strategies to manipulate the intracellular environment and limit competing cellular processes, whereas S. Typhimurium is characterized by molecular profiles associated with strong immune activation and inflammation-related metabolic responses [53].

Multi-omics analyses have also revealed strain-specific host–microbiome interactions. Seong et al. [54] combined metagenomic and metabolomic analyses to investigate fecal microbial diversity in infants with atopic dermatitis, identifying two distinct subclades of Bifidobacterium longum. Subclade I, predominant in healthy infants, harbors genes encoding proteins essential for immune maturation and intestinal barrier integrity, whereas Subclade II, enriched in infants with atopic dermatitis, contains restriction system elements and a toxin gene (K06218), highlighting the importance of strain-level determinants in host–microbiome interactions and disease susceptibility [54].

In the context of infection-associated cancers, multi-omics approaches have demonstrated significant diagnostic and prognostic potential. Mohamed et al. [55] applied transcriptomic and genomic analyses combined with advanced bioinformatics to investigate Helicobacter pylori-associated gastric cancer and identified 43 differentially expressed genes correlated with disease progression. Overexpression of TPX2, CTHRC1, and SLC12A2 was significantly associated with poor overall survival, underscoring their potential value as prognostic biomarkers [55]. Similarly, integrated genomic, proteomic, and bioinformatic analyses in brucellosis have identified stage-specific biomarkers and revealed persistent inflammatory signatures distinguishing chronic from acute disease phases. Five proteins (SAA2, ASS1, MARCO, sICAM-1/CD54, and HSP27/HSPB1) were validated as effective markers for disease diagnosis and stage differentiation [56].

4. Technologies for Detecting Emerging Infections

The detection of emerging infectious agents relies on a range of complementary diagnostic technologies that differ in analytical performance and operational complexity (

Table 1. Summary of pathogen detection methods: principles, advantages, and disadvantages.

Method	Principles	Advantages	Disadvantages
Culture-based methods	Growth of microorganisms under favorable conditions for isolate colonies.	- Efficient in proving the viability of the species. antimicrobial susceptibility testing; strain typing - Low costs	- Detect only viable microorganisms. - Inappropriate for fastidious species. - Time-consuming and labor-intensive. - Delayed pathogen detection. - Low sensitivity to specific pathogens
Protein-based methods (e.g: MALDI-ToF MS)	Matrix-assisted laser desorption/ionisation time of flight mass spectrometry.	- Rapid identification. - Convenience and accuracy.	- Limited database. - Depends on prior culture. - Absence of antimicrobial profile of the pathogen.
Chemometric methods	Rely on variations in chemical bonds, such as C-H, O-H and N-H, and analyzing spectra obtained from techniques such as FTIR, SERS, NIRS or GC-MS.	- high sensitivity. - Efficient and rapid analyzes of big data. - Detection hidden relationships between variables	- Complexity of statistical analyzes. - Depends on the size of the data. - necessity of sophisticated data analyze software.
Polymerase chain reaction methods	In vitro enzymatic amplification of a targeted DNA sequence by repeated thermal cycles (denaturation, hybridization, and renaturation).	- Directly use on clinical samples. - High sensitivity. - Differentiation to strain level.	- Depends on primer design. - Highly sensitive to contamination.
Loop-Mediated Isothermal Amplification	Amplifying DNA or RNA at a constant temperature, usually 60 to 65 °C.	- Hight capacity for detecting low quantities of DNA.	- Highly sensitive to contamination. - Primers design highly complexed.
Fluorescence In Situ Hybridization	Use fluorescent probes binding to targeted DNA or RNA sequences.	- Direct detection of DNA/RNA sequence. - Localisation of genes on chromosomes, label ribosomal RNA.	- Target a limited number of microbial groups. - Complex sample preparation.
High-throughput sequencing (IonTorrent platform)	Based on the release of hydrogen ions during DNA synthesis.	- Direct detection of pathogens from different samples. - No need for prior culture. - Able to detect microbial variant.	- high cost. - sophisticated equipment.

). These include culture-based, protein- and chemometric-based, molecular, and high-throughput sequencing approaches, as well as emerging artificial intelligence–assisted digital health tools.

4.1. Culture-Based Methods

Culture-based methods were among the earliest techniques used for detecting, isolating, and identifying pathogens. They became standardized after Louis Pasteur’s invention of the first artificial liquid culture medium in 1860, followed by the development of the Petri dish in 1887 for solid cultures [57]. These methods involve cultivating microorganisms under controlled conditions to isolate colonies with distinct morphological aspects confirmed by a range of biochemical tests [58,59].

Although widely used, culture-based methods have several limitations that reduce their effectiveness for infectious disease surveillance. They can only detect viable microorganisms [60], whereas certain species cannot be cultured under routine laboratory conditions because they require specific growth requirements [61]. Furthermore, culture techniques are both time-consuming and labor-intensive. Preliminary results usually appear after a few days, while final confirmation may take more than a week. For fungal infections, culture may require several weeks before visible growth occurs [62]. The sensitivity of these conventional methods is also limited for some pathogens; for example, Campylobacter may remain undetected in nearly one-third of positive stool samples [63]. Such delays and inaccuracies hinder the implementation of appropriate treatments and increase healthcare costs.

4.2. Protein-Based Methods

Protein-based methods detect microbial proteins via specific interactions. Among the most widely used techniques are Matrix-assisted laser desorption/ionisation time of flight mass spectrometry (MALDI-ToF MS), Enzyme-Linked Immunosorbent Assay (ELISA), Chemiluminescence ImmunoAssay (CLIA) and serum neutralisation assay (SNA) [64]. Other approaches use fluorochrome-labelled antibodies to capture antigens, as in the Immunofluorescence assay (IFA), or rely on the polarisation of light emitted by a fluorescent molecule when excited, as in Fluorescence Polarisation Assay (FPA) [65].

MALDI-ToF MS can be used for the rapid identification of a wide range of clinical pathogens based on their protein profile. Compared to traditional techniques, this method has many advantages, including its convenience, speed and accuracy, especially thanks to the enrichment of the database and the optimisation of algorithms [66].

This technique still has significant limitations, particularly when it comes to non-cultivable pathogenic microorganisms, as MALDI-TOF only analyzes well-purified bacteria. In addition, it is highly dependent on databases, making it very limited for the identification of rare pathogens that are not yet recorded [67].

4.3. Chemometric Methods

Chemometric approaches are increasingly used to detect clinical pathogens by analyzing spectra obtained from techniques such as Fourier transform infrared spectroscopy (FTIR), surface-enhanced Raman spectroscopy (SERS), near-infrared spectroscopy (NIRS), or gas chromatography–mass spectrometry (GC-MS) [68,69]. These methods rely on variations in chemical bonds, such as C-H, O-H and N-H, which reflect the molecular composition of microbial cells [70].

4.4. Molecular-Based Methods

Various molecular techniques have been developed for clinical pathogen detection and identification. These techniques detect pathogen-specific nucleic acids (DNA/RNA) through hybridisation with synthetic oligonucleotides, allowing the detection of target genes [71]. The most common methods include PCR and its variants, isothermal amplification, fluorescence in situ hybridisation (FISH) and next-generation sequencing (NGS) [72].

4.4.1. Polymerase Chain Reaction-Based Methods

PCR enables the exponential amplification of a specific DNA region of a pathogen’s genome using sequence-specific primers and a thermostable DNA polymerase. The reaction produces millions of copies of the target sequences, which can be visualised by gel electrophoresis. When the goal is the identification, the PCR product obtained from the amplification of a universal region can be sequenced using the Sanger method and compared with reference sequences in genomic databases [73].

In clinical Microbiology, this technique has become a routine tool due to its high sensitivity and specificity and its ability to detect even small amounts of pathogen DNA, including from non-viable cells. Recently, advanced variants such as multiplex PCR and Real-Time PCR were developed. Multiplex PCR allows for the simultaneous amplification of multiple gene targets. This makes pathogen detection and differentiation faster than conventional PCR [74]. The success of this method largely depends on primer design, which must have similar melting temperatures and minimise the formation of dimers to ensure efficient and accurate amplification [75]. Multiplex PCR can be used directly on clinical samples [76]. However, if the pathogen is present in very low concentration, conventional PCR or culture is needed to confirm the results [77].

Real-time PCR, also called quantitative PCR (qPCR), is a method that detects and quantifies DNA as it is amplified [78]. It uses specific primers and fluorescent probes for rapid and specific detection [79]. The cycle threshold (CT) represents the number of cycles needed for the generated fluorescence to exceed the background signal, indicating the initial amount of DNA in the sample [80].

4.4.2. Loop-Mediated Isothermal Amplification

Loop-Mediated Isothermal Amplification (LAMP) is a method for amplifying DNA or RNA at a constant temperature, usually between 60 à 65 °C, without the need for thermal cycling [81]. Four specific primers are used for the amplification, recognising six different sequences. The internal primers, forward inner primers (FIPs) and backward inner primers (BIPs), help form a stem-loop structure, while the external primers (F3 and B3) ensure strand displacement [82]. The detection of amplicons from a LAMP analysis can be carried out by several methods, such as detection by turbidity, agarose gel electrophoresis, fluorescence through the incorporation of dyes (such as SYBR Green, EvaGreen, calcein AM, hydroxy naphthol blue, etc.), or by lateral flow strips [82,83]. LAMP has been successfully used to detect pathogens like Neisseria meningitidis in cerebrospinal fluid and Plasmodium falciparum in plasma [84].

4.4.3. Fluorescence In Situ Hybridisation

Fluorescence In Situ Hybridisation (FISH) is a technique that uses fluorescent probes designed to bind to specific DNA or RNA sequences, especially the small subunit ribosomal (SSU rRNA). It allows the direct detection and localisation of genes on chromosomes and can also label ribosomal RNA in various pathogens in complex matrices without culture [85,86]. It also allows multicolour labelling, making it possible to study microbial distribution and interaction within tissues [78,86,87].

Although FISH is less affected by inhibitors than PCR-based methods, it can only target a limited number of microbial groups at the same time [78]. To overcome this limitation, an improved approach known as Combinatorial Labelling and Spectral Imaging-FISH (CLASI-FISH) was developed. This method uses up to 28 color combinations derived from eight fluorochromes, allowing the simultaneous identification of multiple microorganisms within a single sample [88].

4.4.4. High-Throughput Sequencing

High-throughput sequencing (HTS) technologies have revolutionized pathogen detection, genomic surveillance, and infectious disease research by enabling rapid, cost-effective, and culture-independent identification of microbial agents [89]. These technologies are broadly classified into three generations: first-generation, second-generation (next-generation sequencing, NGS), and third-generation sequencing, based on their sequencing principles, throughput, and read length. First-generation sequencing was initially introduced in the 1970s by Maxam and Gilbert, relying on chemical cleavage of DNA at specific nucleotides. Subsequently, Sanger sequencing became the dominant method, providing highly accurate but low-throughput data, and it remains a reference technique for targeted analyses and validation [90]. Second-generation sequencing, including platforms such as Illumina and Ion Torrent, introduced massively parallel sequencing and has become the standard approach for whole-genome sequencing, antimicrobial resistance profiling, and outbreak investigations [91]. More recently, third-generation sequencing technologies, such as Oxford Nanopore Technologies and Pacific Biosciences, enable real-time long-read sequencing, facilitating genome assembly, structural variant detection, and detailed epidemiological surveillance [91,92,93]. Extensive metagenomic and transcriptomic research has demonstrated that NGS technology accurately profiles the antimicrobial resistance of multiple multidrug-resistant pathogens, thereby facilitating targeted precision treatment management [94]. Collectively, these approaches support the discovery of novel microbial strains, the study of pathogen evolution and zoonotic transmission, and the tracking of transmission routes during epidemics, thereby strengthening public health surveillance and response capabilities

4.5. Integration of AI and Digital Health Tools for Outbreak Detection

Digital health technologies play a key role in outbreaks by bringing health services to the patient, improving diagnostics, surveilling outbreaks, and controlling the spread of infections [95,96,97]. Artificial intelligence, telehealth services, telemedicine mobile health applications, big data, health information exchange services and the internet of things are among the emerging health technologies and digital practices in health care [96]. Various digital health tools, with specific reference to the Zika outbreak, such as computational modelling, followed by big data, then mobile health, and finally other novel technologies, were used for several purposes, including disease monitoring, diagnostics and treatment [95]. Digital surveillance, considered as a crucial complement to the existing standard and official surveillance structures, was used to early detect Ebola and Polio epidemics and to efficiently respond to the public health emergencies [98]. Salim et al. [99] found that digital health tools such as mobile and web applications, machine learning, geographic information systems, social media, and Google Trends were integrated in Dengue surveillance by several countries. Similarly, a recent study underscores that digital health technologies strengthen the global outbreak preparedness by improving early outbreak detection as well as communication and continuity of health care service. However, it emphasises ethical considerations and challenges such as confidentiality and protection of personal health information [100]. Relying on digital health innovations, artificial intelligence (AI) offers advanced capabilities in the prediction and diagnosis of outbreaks. It could be used for public health surveillance, enabling the capability to investigate diverse data sources in real time [101]. Rashid et al. [102] emphasise that AI is progressively being used in numerous healthcare fields such as disease diagnosis, imaging, patient care, surgical and procedural practices, and public health. The AI models used to prevent the tuberculosis outbreak have demonstrated their efficiency in forecasting emerging tuberculosis hotspots with high accuracy, enhancing the early detection and intervention, and optimizing resource allocation to support the zero-tuberculosis vision [103]. An integrated approach involving an AI model predicted precisely the monkeypox cases with high probability to improve early intervention and outbreak control [104].

5. Modeling Emerging Infections

5.1. Traditional Two-Dimensional (2D) Cell and Tissue Culture

Tissue or cell culture consists of growing living cells from a host tissue or organ within a controlled artificial environment, replicating in vivo conditions. This approach enables the study of various interactions between pathogens and their hosts, particularly for pathogens with obligatory intracellular multiplication, such as viruses and certain bacteria [105]. Several 2D organotypic models have been developed to study host–pathogen interactions in various organs and tissues, including pulmonary, cutaneous, renal, pancreatic, hepatic, and intestinal epithelial models [106]. The process starts by collecting tissue from the target organism and carefully dissecting it to isolate the desired cells. The cells are separated using mechanical or enzymatic methods, expanded in culture to form a monolayer, purified and then cryopreserved for later use [106].

In the 2D monolayer model developed by Roodsant et al. [107], organoids are cultured in specialised Petri dishes to study infectious interactions at the intestinal barrier, including viral infections by EV-A71 and bacterial infections by Listeria monocytogenes. These dishes contain a porous membrane, often coated with animal collagen, which facilitates cell adhesion and attachment of intestinal organoids or other relevant cell types [107].

In such a system, tissue or cell cultures are exposed to the target pathogens, enabling direct observation of cellular processes such as microbial entry into host cells, changes in metabolic activity, activation of immune-related factors, and alterations in cell morphology [108]. A study based on this model highlighted the invasion of the bacterium through the epithelial barrier, causing disruption of the cytoskeleton and an intense inflammatory immune response [107].

Despite the utility of these 2D culture models, they lack structural complexity and do not allow the study of infection over extended periods due to the rapid cell deterioration [109]. Additional limitations include the loss of physiological stimuli, the absence of diffusion-mediated molecular transport, and the inability to replicate cellular interactions that depend on a 3D microenvironment. Data obtained from these models may fail to accurately reflect the complexity of human infectious diseases [108,109].

5.2. Advanced Three-Dimensional (3D) Technologies

Three-dimensional (3D) cell or tissue cultures are laboratory-grown structures derived from human or animal pluripotent or adult stem cells that simulate the structure and function of an organ, a tissue, or specific physiological processes of a living organism [110]. The production of these tissues can be ensured and facilitated by 3D bioprinting technology [111]. The latest is generally carried out on culture plates using bioinks, mainly hydrogels, which reproduce the biochemical properties of the target tissues [110].

These 3D models may include one or multiple types of cells integrated into a matrix, as not all cells in an organism can grow except on a basement membrane that serves as a matrix, with the exception of blood cells [112]. This matrix must be biocompatible and possess properties similar to those of the extracellular matrix in vivo to ensure cell viability and prevent a reduction in compound concentration in the medium due to adsorption [112]. These characteristics enable the reproduction of biologically relevant phenomena similar to those observed in a living organism [113]. These in vitro 3D tools provide a promising alternative to reduce, or even completely replace, the use of animal models in preclinical studies (Figure 2) [114]. Among these technologies, one can distinguish.

5.2.1. Organ-on-Chip

Organs-on-chip (OOC) are microfluidic models integrating living cells within bioengineered microchambers that are continuously perfused, allowing for the control of various parameters such as concentration gradients, dynamic conditions, cellular and tissue boundaries, as well as interactions between tissues and organs [115]. The design of an OOC relies on the reverse engineering of human tissues and physiological systems. It varies depending on the type of organ, tissue, or cells being modeled, as well as the biological phenomenon under study. This results in major differences in chip geometry, including size, diameter, the number and shape of channels, and the materials used [116]. An OOC is characterized by three fundamental properties: (i) the nature and three-dimensional organization of the cultured cells and tissues, (ii) the integration of various cell types (notably epithelial, stromal, vascular, and immune cells) to reflect the in vivo tissue composition, and (iii) the presence of biomechanical forces relevant to the modeled tissue (for example, stretching forces for lung tissues, contractions, and hemodynamic forces for vascular tissues) [117]. An organ-on-chip (OOC) typically consists of a small fragment of simulated tissue or organ placed within a network of microfluid channels that function like a vascular network, for nutrient delivery and metabolic waste removal, electromagnetic pumps to generate organ dynamics and porous membranes lined with endothelial cells to mimic biological barriers [118,119]. Sensors can be integrated to monitor the environment in real time, while the culture medium ensures cell survival. This technology offers precise control of cellular and tissue architecture, reproducing chemical gradients and biomechanical forces. Moreover, multiple OOC models can be interconnected to stimulate complex physiological systems, enabling the study of organ interactions, drug pathways and functional responses [120,121].

Several organ models have been developed to mimic different tissues such as the lymph nodes, bone marrow, kidney, lungs, thymus, spleen, skin, liver, intestine, myocardium, renal tubules and tumor microenvironment [122]. These systems have improved our understanding of how immune responses vary across tissues during infection [123]. Organ-on-a-chip technology has enabled researchers to gain a deep understanding of the protective role of pulmonary surfactant against infection by Mycobacterium tuberculosis [124].

In another study based on a rat liver-on-chip model of HBV infection, hepatocytes were able to maintain their differentiation, morphology and key functional markers such as ALB, TFN, HNF-4α, β-actin [125]. Later, a human liver-on-chip model replicated HBV infection and accurately mirrored patient host responses, including albumin secretion and Cyp450 enzyme activity, and elevated levels of IL-8, MIP-3α, SerpinE1 and MCP-A [126]. Similarly, a human intestine-on-chip model allowed the study of polarised Coxsackie B1 virus infection, highlighting villous destruction and cytokine secretion (IP-10 and IL-8) that classical models could not reproduce [127]. OOC systems have also provided insights into SARS-CoV-2 pathophysiology, demonstrating rapid viral detection in endothelial cells normally resistant in 2D monocultures [128].

Other studies have used a kidney-on-chip model to mimic pseudorabies virus-induced kidney injury, further showing how these systems can faithfully reproduce complex, tissue-specific infection processes. This platform revealed, for the first time, the pathogenesis of this virus, demonstrating that it compromises the reabsorption barrier, leading to a reduction in Na⁺ reabsorption through altered expression and localisation of Na⁺ transporters. These findings could explain the electrolyte imbalances observed in patients affected by this infection [129].

5.2.2. Spheroids

Spheroids are self-assembled three-dimensional (3D) cell cultures formed when cell–cell interactions dominate over cell-substrate interactions [130]. They have a hollow center surrounded by actively dividing cells, which closely mimics cellular signalling and interactions with the extracellular matrix [131]. Unlike organoids, spheroids generally do not contain stem cells, which limits their ability to self-renewal. They are useful for studying microbial interactions, tumor infections, and immune response because their growth pattern resembles what happens in vivo tissues [132]. Several techniques can be applied to generate spheroids, including polymer encapsulation, microfabricated plates, ultra-low attachment culture, and the hanging drop method [133,134,135,136].

Spheroid models better mimic physiological conditions compared to traditional 2D culture. Initially, cells adhere to one another through integrins, then compact via E-cadherin, forming a structure that ensures the supply of nutrients and oxygen to the peripheral cells while the central cells may exhibit some deficiency [137,138]. In co-cultures of macrophages and melanoma cells, Seo et al. [139] reported that spheroids reproduced direct cell interactions and cytokine secretion patterns (TNFα, TGF-β1) compared to in vivo responses, unlike 2D systems. Similarly, spheroids developed by Petrucciani et al. [140] to model tuberculosis infections showed enhanced macrophage activation and stronger CD8+ T-cell responses. Together, these findings highlight the critical role of 3D architecture in capturing immune mechanisms and complex biological processes near physiological conditions. In a study published by Thomas A. Sebrell et al. [141], researchers were able to model the gastric sphere as a spheroid in vitro to mimic H. pylori infection, which induces in-creased activation of spheroids, increases the migration of dendritic cells to the epithelium, and causes overexpression of chemokines, explaining the significant attraction of immune cells, This model provides a physiologically relevant and more accurate inter-action between the H. pylori and the gastric environment host [141].

5.2.3. Organoids

Organoids are miniature three-dimensional structures derived from stem or progenitor cells that self-organise in the laboratory to replicate the architecture and certain functions of a target organ [106]. These micro-organs are far more stable than spheroids and can faithfully mimic both the morphology and functional activity of the tissue being studied [142,143].

Organoids are cultured within a three-dimensional matrix composed of collagen, extracellular matrix extracts, or synthetic materials such as hyaluronan hydrogels or polyethene glycol [144]. Their formation is influenced by multiple factors, including intracellular signaling, intracellular interactions and the biochemical and physical properties of the surrounding matrix [145]. Hydrogels and other biomaterials play a key role in regulating the physical conditions of organoid culture, helping to maintain proper structure and function [145,146].

Different types of organoids have been employed to model a variety of infections. For instance, human organoids have been used to investigate immune responses to commensal bacteria and Shiga toxin-producing Escherichia coli [147]. In another example, HIV-infected cerebral organoids, composed mainly of neurons, astrocytes and microglia, have provided a platform to study the virus’s neuropathogenesis and the molecular mechanisms underlying neuroinflammation [148]. A recent study by Millen et al. [149] concluded that organoids provide a more precise and personalised insight into patient responses to treatment, compared with conventional two-dimensional models. Furthermore, they enable genome editing using CRISPR-Cas9, enhancing the accuracy of predicting disease-associated biomarkers.

6. Conclusions

An effective diagnostic framework for emerging infections must reflect the complexity of pathogen emergence. Although traditional culture remains essential for confirmation and reference, it is often slow and incomplete. Molecular approaches, including PCR, qPCR, LAMP, FISH and NGS, offer rapid and specific detection, but still face limitations in multiplexing, sensitivity at very low pathogen loads and lack of standardisation. Protein-based assays and MALDI-ToF mass spectrometry provide fast and cost-effective tools for routine identification, yet their accuracy depends on sample quality and a comprehensive reference database. Meanwhile, chemometric analysis of spectroscopic data shows potential for pathogen characterisation and biomarker discovery, though its clinical implementation requires improved preprocessing, interpretability and safeguards against overfitting. Physiologically relevant 3D models, organs-on-chips, spheroids, and organoids, address gaps left by 2D cultures by restoring tissue architecture, gradients, and multicellular immune responses, enhancing studies of invasion, immune evasion, replication, and clearance. Future progress will hinge on integrative pipelines that combine rapid front-end detection with confirmatory sequencing, database-aware proteotyping, and hypothesis-driven infection modeling, all under harmonized quality frameworks and One-Health surveillance. Such convergence may enable faster, more accurate diagnostics and a deeper mechanistic footing for therapeutics and public-health response.