Introduction

Crimean-Congo haemorrhagic fever virus (CCHFV) is a zoonotic tick-borne human pathogen belonging to the Nairoviridae family (order Hareavirales) [1]. The virus is endemic in Europe, Africa, and Asia, and closely follows the geographic distribution of its main vector and reservoir, ticks from the genus Hyalomma. Experimental studies have shown that ticks from the genus Rhipicephalus are also competent CCHFV vectors [2] and the main drivers of CCHF transmission in regions where Hyalomma ticks are uncommon [3,4]. CCHFV is usually transmitted to humans through the bite of infected ticks, although transmission through direct exposure to viraemic bodily fluids from patients or animals has also been reported [3]. The virus contains a tripartite, negative-sense RNA genome consisting of the small (S), medium (M), and large (L) segments encoding for the nucleocapsid protein (NP), the precursor to the viral glycoproteins (GPC), and the RNA-dependent RNA polymerase (RDRP), respectively [5]. Traditionally, based on phylogenetic analysis of the S segment, there are seven different CCHFV genotypes, however due to genomic segment reassortment, which has been widely described in the literature [5–8], this number differs for other genome segments [9].

While most CCHF infections are asymptomatic or cause mild non-specific symptoms [10], a proportion progress to severe and often fatal haemorrhagic fever. Due to the lack of licensed vaccines or therapeutics for CCHF, the World Health Organisation (WHO) has previously listed CCHFV as a priority pathogen for accelerated research [11]. The recent reports of CCHF outbreaks in Europe, Africa, and Asia [12–18], and risk alerts issued in Spain and Greece [19], highlight the necessity for continued CCHFV surveillance in endemic regions. A revision of the WHO 2019 CCHF R&D roadmap published in 2024 has laid out key research priorities, including genetic characterisation and molecular surveillance of the virus in humans, animals, and vectors [20].

Most CCHFV cases in Asia are reported from central Asian countries including Kazakhstan, Kyrgyzstan, Tajikistan, Turkmenistan, and Uzbekistan [21] where CCHFV is endemic. The detection of both Asia-1 and Asia-2 (genotype IVa and IVb) CCHF viruses has been previously reported in Kazakhstan [22,23], Uzbekistan (Asia-1) [24], and Turkmenistan (Asia-2) [9].

In Tajikistan, the national system for epidemiological surveillance of infectious diseases is coordinated by the Ministry of Health and Social Protection of the Population through the country’s sanitary-epidemiological surveillance network. CCHF is among the notifiable diseases subject to mandatory epidemiological monitoring. Since 1961, CCHFV surveillance and formal diagnosis in Tajikistan have been undertaken mainly by the Ministry of Health’s Tajik Research Institute of Preventative Medicine (TRIPM), in Dushanbe [25]. The majority of CCHFV infections in Tajikistan occur in the southern region, likely due to favourable environmental factors for vector establishment [26] often following bites from H. anatolicum ticks [25]. The country currently records between 5–38 CCHF cases per year [21]. Previous studies aimed at genetically characterising CCHFV in Tajikistan have found evidence of both Asia-1 and Asia-2 viruses circulating in the region based on S segment sequence data analysis [26,27].

Recently, we have developed a tiling, multiplex PCR-based approach for recovery of Europe 1 (genotype V) CCHFV genomes which has shown better genome recovery from serum samples with low virus titre (< 10⁶ copies/mL) over non-targeted enrichment strategies [28]. The method was successfully implemented in a public health laboratory to genetically characterise viruses associated with recent clinical cases of CCHF [29].

In this study we have developed and applied a novel amplicon-based tiling scheme specific for Asian CCHFV genotypes, and genetically characterised viruses from clinical samples and H. anatolicum ticks collected in Tajikistan between 2017 and 2023. While most of the sequences generated in this study clustered within the Asia-2 genotype, we report the detection of an S segment Asia-1 reassortant virus from a clinical sample which has not been previously reported from the region. Prior to this study, only 28 published CCHFV genomic segment sequences from the region were available on the Bacterial and Viral Bioinformatics Resource Center [30] (accessed in September 2025), with no records having been published in the last decade. Our efforts, which aimed at genetically characterising CCHF viruses in Tajikistan from both clinical as well as tick samples extend our understanding of CCHFV epidemiology and transmission dynamics within the region. Furthermore, our data contributes to the broader global understanding of CCHFV, aiding in the development of effective public health strategies to predict and prevent future outbreaks which continue to pose threats to human health.

Methods

Sample collection, RNA extraction and quantification

All samples were collected by TRIPM as part of Tajikistan’s national system for epidemiological surveillance of CCHFV. Clinical serum samples were collected between 23 June 2017 and 12 July 2023 and inactivated in AVL buffer (QIAGEN) as previously described [28]. Tick samples were collected by flagging between 7 May 2023 and 26 May 2023, washed in 300 µL of PBS and homogenised using a mortar and pestle. Tick collections were conducted primarily in Bulyoni Poyon and nearby areas where recent human cases had been reported and where H. anatolicum populations are abundant. All samples were processed by TRIPM Virology Laboratory following national CCHFV instructions and legislation. Sampling locations, which represent the principal endemic focus rather than a uniform national sampling distribution, have been indicated in Fig 1.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Sampling locations for the clinical (blue pins) and tick-derived (green pins) samples.

Map image was generated using ArcGIS software [31].

https://doi.org/10.1371/journal.pntd.0014204.g001

RNA was extracted with the QIAamp Viral RNA kit (QIAGEN) following the manufacturer’s spin protocol and eluted in 60 µL of AVE buffer. RNA was shipped to the UK Health Security Agency (UKHSA) and stored at -80 °C. RT-qPCR was performed on duplicate 10-fold dilutions of the RNA extracts as previously described [32]. Copies of CCHFV genomic RNA were determined based on a quantified synthetic S segment RNA standard curve and normalised taking into account extraction and elution volumes. 11 tick homogenate samples and 11 clinical serum samples with a cycle threshold (CT) value between 23.7 and 34.3 were included in the study (Table 1). These sequenced samples represent a subset of RT-PCR–positive detections identified through the national surveillance programme during the study period; comprehensive surveillance statistics will be reported separately.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. CT values and RNA copy numbers of CCHFV clinical and tick samples. *Samples collected from patients with reported contact to C18.

https://doi.org/10.1371/journal.pntd.0014204.t001

Bioinformatic analysis

Raw Illumina sequencing reads were processed as previously described [28]. Illumina reads were mapped against reference sequences (accession number KJ676542.1 for the S segment, ON500501.1 or KC344856.1 for the M segment, and KX013456.1 for the L segment) utilising BWA MEM [34]. Nanopore sequencing data was filtered based on quality (minimum 8). Nanopore sequencing reads were mapped against the reference utilising minimap2 [35].

Consensus sequences were derived from mapped reads using BCFtools [36] masking regions with no coverage. The dual primer pool barcoding used for sequencing each sample allowed for consensus sequences to be called only from regions which displayed the expected tiling amplification pattern, excluding regions affected by background contamination. For targeted enrichment data, primer sequences were first removed from sorted BAM files with BAMClipper [37] prior to consensus generation. Reported consensus reference coverages (Table 2) were calculated excluding gaps.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Reference coverage and genotype of the analysed CCHFV samples. *Sample RNA enriched with SISPA followed by Illumina sequencing.

https://doi.org/10.1371/journal.pntd.0014204.t002

All generated consensus sequences have been deposited in GenBank with the following accession numbers: PV390603- PV390627; PV738566- PV738575; PV953843-PV953859.

Results

Genomic segment recovery from both Asia-1 and Asia-2 lineages was achieved for clinical as well as tick homogenate samples with the newly developed tiling multiplex PCR methodology.

Phylogenetic analysis of the S, M and L segments confirms detection of an S segment Asia-1 reassortant virus in a clinical sample

Except for a single clinical S segment sequence (PV390605) derived from sample C14 which clustered with an Asia-1 (genotype IVa) 2017 strain from China (MG659724.1), all other generated S segment sequences clustered within the Asia-2 (genotype IVb) clade, supported by high bootstrap values (Fig 2A).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2. Maximum likelihood phylogeny for the S (A), M (B) and L (C) genomic segments.

Sequences have been colour-coded based on genotype affiliation of the S segment as follows: III (Africa-3, red), IV (Asia-1/IVa, blue; Asia-2/IVb, purple). Sequences generated in this study are indicated by black/white circles (clinical samples) or black rhombi (tick homogenate samples). White circles indicate sequences recovered from patients with reported contact. Sequences belonging to the identified Asia-1 S segment reassortant virus are indicated by red triangles. Genotype numbers are indicated next to the relevant clades. Except for single nodes which have been colour coded as described above, clades not relevant to this study have been compressed (black triangles).

https://doi.org/10.1371/journal.pntd.0014204.g002

Notably, the M (PV390616.1) and L (PV390623.1) segment sequences derived from sample C14 both clustered within the Asia-2 genotype (Fig 1B and 1C), indicating reassortment of the Asia-1 S segment.

Inspection of mapped reads revealed no mixed nucleotide positions above the 5% minor-allele frequency in the reassortant sample (C14). Read-depth profiles were even across all three segments and no dual mapping signatures were observed. These findings indicate the presence of single, homogeneous viral populations, supporting reassortment rather than dual infection as the most plausible explanation.

The S segment sequences derived from two clinical samples and three tick samples clustered with a sequence derived from a clinical sample isolated in Tajikistan in 1990 (AY297692.2). Five clinical samples all isolated in Jayhun formed a separate cluster, with the closest sequence also being AY297692.2. A clinical sample (PV390603.1) was more closely related to a 1973 tick S segment sequence from Turkmenistan. All other S segment sequences clustered with previously published sequences derived from clinical samples isolated in Central and South Asia (Fig 2A). A single partial Asia-2 S segment sequence (PV738575.1) was excluded from phylogenetic analysis due to low reference coverage.

All M segment sequences generated in this study cluster within the Asia-2 genotype with high bootstrap support (Fig 2B). M segment sequences derived from the tick samples as well as four clinical samples cluster with a 2021 tick sequence from China (ON500501.1), whilst M segment sequences from the other four clinical samples cluster with a 2012 clinical sample sequence from Afghanistan (KC344856.1). Three partial Asia-2 M segment sequences (PV953844, PV953853, PV390620) were excluded from phylogenetic analysis due to low reference coverage.

All L segment sequences generated in this study also cluster with other Asia-2 genotype. Most L segment sequences (including one sequence derived from a tick sample) are most closely related to an L segment sequence derived from a historical 1969 clinical sample isolated in Tajikistan (KX013444). Two sequences, one from a clinical sample and one from a tick sample are most closely related to a 1990 clinical isolate sequence from Tajikistan (AY720893.2). Four partial Asia-2 L segment sequences (PV738567, PV738569, PV738571, PV738574) were excluded from phylogenetic analysis due to low reference coverage.

The M segment displayed the lowest amino acid conservation within clinical and tick samples

Minimum and maximum pairwise identities at both nucleotide and amino acid level are presented in Fig 3 for each sub-group. Since except for a single Asia-1 S segment sequence all other sequences were Asia-2, only the S segment Asia-2 within group pairwise identities are reported.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 3. Within group minimum and maximum pairwise identity (%) for both nucleotide (A) and amino acid (B) sequences for each CCHFV genomic segment.

https://doi.org/10.1371/journal.pntd.0014204.g003

The average nucleotide sequence pairwise identity for the S segment was 94.9%, namely 94.7% for the human samples, 96.5% for the tick samples and 95.5% for the Asia-2 samples. For the M segment nucleotide sequences, the average pairwise identity was 91.2%, namely 90.7% for the human samples and 99.1% for the tick samples. For the L segment amino acid sequences, the average pairwise identity was 97.5%, namely 97.9% for the human samples and 97.4% for the tick samples.

The average amino acid sequence pairwise identity for the S segment was 99.3%, namely 99.4% for the human samples, 99.1% for the tick samples and 99.4% for the Asia-2 samples. For the M segment amino acid sequences, the average pairwise identity was 92.5%, namely 92% for the human samples and 96.1% for the tick samples. For the L segment amino acid sequences, the average nucleotide pairwise identity was 98.9%, namely 99% for the human samples and 98.8% for the tick samples.

Discussion

The presence of CCHFV in Tajikistan has been known for decades, with the first records of clinical CCHF cases dating back to the 1940s, when CCHF was first recognised as a disease [25]. It was not until 2004 (for the S and M segments) and 2005 (for the L segment) that the first studies on genetic characterisation of circulating viruses in the region were published in the Russian literature [41–43], thus, providing evidence of the presence of Asia-2, and later in 2013 Asia-1 viruses, based on S segment analysis [27]. Only two historical strains isolated in 1969 [9] and 1990 [42–44] have been fully genetically characterised from the region, the former belonging to the Asia-2 genotype and the latter being an M segment Asia-1 reassortant virus.

Our study contributes significantly to the understanding of the circulation and genetic diversity of CCHFV in Tajikistan and more broadly in Central Asia by providing genomic insights into circulating CCHF viruses in clinics and ticks. Most of our genomic sequences cluster within the Asia-2 genotype for all genomic segments (Fig 2). A single S segment sequence from a clinical sample collected in Vahdat (C14) clustered within the Asia-1 genotype (Table 2, Fig 2A). The Asia-1 S segment had an average pairwise identity (%) of 89.66% and 97.96% at the nucleotide and amino acid level, respectively, with the other recovered Asia-2 S segment sequences. The incongruent phylogenetic pattern observed for the M and L segment sequences recovered from the same sample, which clustered with other published Asia-2 sequences, is consistent with S segment reassortment. The absence of other Asia-1 S segment sequences in our dataset together with the low proportion of minor variants sites (~ 0.5%) supports this as authentic viral reassortment rather than sample contamination or mixed infection. To our knowledge, this represents the first documented Asia-1 S segment reassortant virus identified in Tajikistan using full multi-segment genomic analysis, likely reflecting the historical scarcity of complete CCHFV genomic data from the region as well as the lower prevalence of Asia-1 viruses in Central Asia [45,46].

The functional implications of S-segment reassortment are potentially important. The S segment encodes the nucleoprotein (NP) which plays a central role in viral RNA encapsidation and replication, and is also the principal antigenic and diagnostic target for most serological and molecular detection assays [8]. Substitution of an Asia-1–derived S segment into an Asia-2 genomic background could therefore alter epitope composition and primer-binding efficiency, potentially influencing assay sensitivity or cross-reactivity. Although the Asia-1 and Asia-2 NP sequences share high amino-acid identity, the observed divergence at the nucleotide level (~10%) may affect the performance of lineage-specific RT-PCR or LAMP assays designed for Asia-2 viruses. These findings underscore the need for ongoing molecular surveillance to ensure that diagnostic reagents remain inclusive of circulating reassortant viruses.

Phylogenetic analysis of sequences from samples C19, C20, and C21 which were collected from patients who reported direct contact to C18 confirmed their genetic relatedness for all genomic segments (Fig 2). Notably, sequences from sample C16 which was collected from the same area three days earlier (Table 1) also cluster with these sequences (Fig 2), providing insights into a possible preceding transmission event between C16 and C18.

The lower average within group pairwise identity observed for the M segment at both nucleotide (Fig 3A) and amino acid level (Fig 3B), with averages of 91.2% and 92.5% respectively, is not surprising and is likely driven by immunological pressures [47]. Previous studies have also found the M segment to be the least conserved between different CCHFV genotypes at both nucleotide and amino acid levels [47], in accordance with our data. The relatively lower nucleotide pairwise identity of the S segment compared to the amino acid pairwise identity is indicative of a higher number of synonymous mutations in the S segment which translate into highly conserved nucleoproteins (NP).

One of the limitations of the study lies within our targeted amplicon-based enrichment design, which although highly effective for recovering near-complete CCHFV genomes from samples with low viraemia, was optimised for genotype IV (Asian lineages). As such, it may have limited sensitivity for more divergent genotypes, potentially biasing detection against strains from other genotypes. Due to the utilisation of a sensitive RT-PCR in our study, able to detect all CCHFV genotypes, the risks of false negatives have been mitigated. Nonetheless, to address the challenges associated with genetic divergence across CCHFV genotypes, we are developing a broader targeted sequencing panel incorporating conserved primer sets designed to capture all recognised CCHFV genotypes, enabling future studies to detect and characterise any emergent or translocated lineages in the region. Furthermore, while only H. anatolicum ticks have been collected in this study, further studies should aim to investigate other Hyalomma tick species such as H. asiaticum or genera such as Rhipicephalus ticks, which are present in the region and have been known to harbour different CCHFV genotypes [21,48].

This study provides critical insights into the genetic landscape of CCHFV in Tajikistan and highlights the value of integrating vector and human surveillance within a unified One Health framework to strengthen early detection of emerging CCHFV variants. By combining genomic data from both clinical and tick-derived samples, we demonstrate an approach that can reveal new transmission patterns, reassortment and movement of viral genotypes across ecological and administrative boundaries. Such integrated genomic monitoring provides an adaptable model for cross-border surveillance in Central Asia, where animal trade, pastoral mobility and shared vector habitats facilitate regional virus circulation. These efforts align directly with the WHO R&D Blueprint for Crimean-Congo haemorrhagic fever (2024–2030), which identifies integrated molecular surveillance, vector ecology, and data sharing across sectors as core priorities for global research coordination and outbreak preparedness. Strengthening this coordinated genomic surveillance capacity will be critical for anticipating viral evolution, refining diagnostics and guiding evidence-based control strategies at the human-animal-environment interface.

Supporting information