Alkhurma hemorrhagic fever virus in Ornithodoros savignyi ticks.

Evidence for the tickborne nature of Alkhurma hemorrhagic fever virus (AHFV) is indirect because AHFV has not been detected in arthropods. One Omithodoros savignyi tick from Saudi Arabia contained AHFV RNA. This is the first direct evidence that AHFV is a tickborne flavivirus and confirms the association between human AHFV cases and tickbite history.

**********

Alkhurma hemorrhagic fever virus (AHFV) is a recently described virus within the genus Flavivirus. AHFV was discovered in 1995 in a patient with hemorrhagic manifestations and fever in Saudi Arabia (1). Subsequently, [approximately equal to] 20 symptomatic patients infected with this virus have been documented by virus isolation in this country. The clinical picture is extremely severe and the case-fatality rate is >30%, which makes AHFV one of the most deadly flaviviruses (2). Previous studies have determined that AHFV is a variant genotype of Kyasanur Forest disease virus, another biosafety level (BSL) 4 virus that causes viral hemorrhagic fever in certain regions of India (3). Accordingly, AHFV is classified as a BSL-3 or BSL-4 agent, depending on country regulations.

To date, AHFV has been isolated only from human samples. Genetic and serologic characterization has grouped AHFV with tickborne flaviviruses (1,3). A previous study associated AHFV transmission to humans with butchering of sheep and camels (1). However, no direct evidence for its association with ticks, such as viral detection in or isolation from ticks, has been documented. To investigate the tickborne nature of AHFV, ticks were collected in western Saudi Arabia and tested by reverse transcription-PCR (RT-PCR) for AHFV.

The Study

A total of 124 ticks were collected from camels and camel resting places in 3 different locations in western Saudi Arabia. The epidemiologic characteristics of the 124 ticks are shown in Table 1. Ticks were stored in individual containers at room temperature in Saudi Arabia and killed by overnight freezing at -80[degrees]C the day before shipment to France, according to the French regulations for importation. Samples were treated as previously described (4), and 200 [micro]L of clarified, crushed material was used for purification of total nucleic acid with the MagNA Pure LC system (Roche Diagnostics, Meylan, France). Pools of 10 RNA samples (5 [micro]L each) were prepared and tested by 1step RT-PCR assay with the Access RT-PCR system (Promega. Madison, WI, USA) by using primers ALKES 1 and ALK-ER (2). We used a cycling profile of 48[degrees]C for 45 min, 95[degrees]C for 5 min, followed by 40 cycles of 94[degrees]C for 15 s, 55[degrees]C for 30 s, and 68[degrees]C for 30 sec, and a final elongation step at 68[degrees]C for 7 min. The 10 specimens in the RT-PCR-positive pool were tested individually for confirmation and sample identification. PCR products were then sequenced.

The tick JE7, which was collected southeast of Jeddah, Saudi Arabia, was positive for AHFV, and sequence analysis showed 99.7% homology at the nucleotide level with AHFV strain 1176 (GenBank accession no. AF331718) in the homologous region of the envelope gene. For safety reasons, virus isolation was not attempted in our BSL-3 laboratory. To identify the tick species, morphologic studies and molecular identification were conducted. All 124 ticks were identified by using morphologic keys (5).

The complete coding sequence of AHFV-JE7 was determined by amplification and sequencing of overlapping PCR products by using the long PCR product sequencing strategy (6) (primers and detailed protocol available on request). The open reading frame sequence was 10,248 nt (GenBank accession no. DQ154114). Pairwise distance comparison with AHFV prototype strain 1176 sequence showed 0.80% and 0.70% divergence at the nucleotide and amino acid level, respectively. A total of 79 nt substitutions were observed, of which 54 mutations were synonymous and 25 mutations were nonsynonymous (Ka:Ks ratio 0.46) (Table 2). The AHFV-JE7 complete sequence has been used with other full-length amino acid sequences of other mammalian tickborne flaviviruses to reconstruct phylogenetic relationships.

No nonsynonymous mutations were observed in the envelope gene of the 11 sequences determined for human AHFV cases or in tick and human AHFV sequences. Ka:Ks ratios within human sequences (n = 11) and between tick and human sequences were 0.33 and 0.50 in the nonstructural protein 3 (NS3) gene and 0.46 and 0.33 in the NS5 gene, respectively. However, these values were not significantly different when sequences were analyzed by [chi square] test. Because of the low genetic heterogeneity observed between AHFV sequences, envelope, NS3, and NS5 sequences of the 12 AHFV isolates were colinearized and aligned with homologous sequences of AHFV and related tickbome flaviviruses to infer phylogenetic relationships, as previously reported (Figure) (2).

[FIGURE OMITTED]

As shown in the Figure, the AHFV-JE7 sequence was closely related to but unambiguously distinct from all other AHFV sequences from human isolates. AHFV-JE7 constituted a phylogenetic group distinct from human isolates and was supported by a 100% bootstrap value. The genetic distance of AHFV-JE7 from the common ancestor of AHFV isolates is less than that of any of the human isolates. This topology strongly suggests that human isolates are derived from the group including AHFV-JE7. This constitutes a strong argument for the tickbome nature of AHFV.

It could be argued that several mosquitoborne flaviviruses have been isolated from ticks, e.g., Saint Louis encephalitis virus and West Nile virus (7,8). Phylogenetic analysis showed that these viruses are closely related to other Culex-associated viruses and that their evolution is determined by mosquitoes, not by ticks. Thus, the evolution of AHFV is clearly determined by ticks because all the most closely related viruses are also known to be associated with ticks, although in at least 1 case (Powassan virus) there is some evidence that the virus is also found in mosquitoes (9). Because ticks blood-feed on vertebrates, detection of AHFV RNA could be due to AHFV in the blood of the camel; however, this virus may not replicate in the tick. Because no obvious sign of blood was noted in the JE7 tick, detection of a full-length open reading frame sequence is convincing evidence that the virus had replicated in the tick, thus increasing the likelihood that the tick is a vector for the virus.

Conclusions

In the arid ecosystems of Saudi Arabia and other parts of the Persian Gulf, Ornithodoros savignvi, the sand tampan, has been associated with camels, their resting places, and to a lesser extent, other domestic and wild animals (5) found in camel resting places. O. savignyi is a multiple-host-seeking, nocturnally active, cryptic tick that commonly attacks humans and other animals resting under trees (10), which supports its role as a vector and transmitter of AHFV. The closely related Kyasanur Forest disease virus, which is endemic in certain regions of India, is the only tickborne hemorrhagic fever virus that has been isolated from Ornithodoros spp. (11). AHFV is the first human pathogenic RNA virus to be detected in O. savignvi. Recent reports of mosquito transmission of AHFV (12) remain unverified (13) but merit investigation by virologic analyses of field-collected mosquito pools. Our study provides the first unequivocal evidence that AHFV is a tickborne flavivirus, confirms previous phylogenetic analysis linking human AHFV isolates to a tick source, and shows the value of molecular techniques in rapidly and safely detecting arboviral activity in local arthropod fauna.

Wide-ranging implications arise from the identification of Ornithodoros ticks, a well-established genus in the Persian Gulf region in the transmission cycle of AHFV. First, the ability of the unfed tick to remain dormant in harsh conditions for long periods may give it an extended reservoir role for AHFV. Second, the wide distribution of Ornithodoros ticks in the region suggests that the geographic limits of AHFV may be larger than presently assumed. A recent clinical case from Najran, which is >600 km from the Jeddah-Makkah area, supports this view (14). This finding reinforces the need for physicians in the region and elsewhere to consider AHFV in the etiology for undifferentiated fever.

The relationship between Ornithodoros ticks and military activity-related tickborne disease incidence has been recently documented in the Middle East (15). A large number of foreign military and civilian personnel are presently based in the region, thus expanding the opportunity for exporting clinical disease or infected ticks. The association of JE7 with camels further supports the role of camels in AHFV transmission cycle as well as the zoonotic nature of the disease. Larger studies involving more tick species are needed to better understand AHFV ecology and transmission dynamics. Investigations to obtain better knowledge of the geographic distribution of AHFV are necessary in countries near Saudi Arabia.

Acknowledgments

We thank Shelley Cook for help in improving the manuscript and Stephanie Ostrowski for field assistance and tick transportation.

This work was supported in part by the Comparative Structural Genomics of Viral Enzymes Involved in Replication (VIZIER) integrated project of the European commission in the Sixth Framework Programme for Research and Technological Development (contract no. LSHG-CT-2004-511960), the RiVigene PCRDT6 project, and the Covibio-Diag project.

References

(1.) Zaki AM. Isolation of a flavivirus related to the tick-borne encephalitis complex from human cases in Saudi Arabia. Trans R Soc Trop Med Hyg. 1997;91:17-81.

(2.) Charrel RN, Zaki AM, Fakeeh M, Yousef AI, de Chesse R, Attoui H, et al. Low diversity of Alkhurma hemorrhagic fever virus, Saudi Arabia, 1994-1999. Emerg Infect Dis. 2005;11:683-8.

(3.) Charrel RN, Zaki AM, Attoui H, Fakeeh M, Billoir E Yousef AI, et al. Complete coding sequence of the Alkhurma virus, a tick-borne flavivirus causing severe hemorrhagic fever in humans in Saudi Arabia. Biochem Biophys Res Commun. 2001;287:455-61.

(4.) Charrel RN, Izri A, Temmam S, de Lamballerie X, Parola E Toscana virus RNA in Sergentomyia minuta flies. Emerg Infect Dis. 2006; 12:1299-300.

(5.) Hoogstraal H, Wassef HY, Buttiker W. Ticks (Acarina) of Saudi Arabia Family Argasidae, Ixodidae. In: Wittmer W, Buttiker W, editors. Fauna of Saudi Arabia. Vol. 3. Basel: C iba Geigy Ltd; 1981. p. 25-110.

(6.) Emonet S, Grard G, Brisbarre N, Moureau G, Temmam S, Charrel R, et al. LoPPS: A long PCR product sequencing method for rapid characterisation of long amplicons. Biochem Biophys Res Commun. 2006;344:1080-5.

(7.) McLean RG, Francy DB, Monath TP, Calisher CH, Trent DW. isolation of St. Louis encephalitis virus from adult Dermacentor variabilis (Acari: Ixodidae). J Med Entomol. 1985;22:232-3.

(8.) Lawrie CH, Uzcategui NY, Gould EA, Nuttall PA. Ixodid and argasid tick species and West Nile virus. Emerg Infect Dis. 2004;10:653-7.

(9.) Gould EA, Moss SR, Turner SL. Evolution and dispersal of encephalitic flaviviruses. Arch Virol Suppl. 2004;18:65-84. (10.) Hoogstraal H. Argasid and nuttallielid ticks as parasites and vectors. Adv Parasitol. 1985;24:135-238.

(11.) Bhat UK, Goverdhan MK. Transmission of Kyasanur Forest disease virus by the soft tick, Ornithodoros crossi. Acta Virol. 1973;17:337-42.

(12.) Madani TA. Alkhumra virus infection, a new viral hemorrhagic fever in Saudi Arabia. J Infect. 2005;51:91-7.

(13.) Charrel RN, Zaki AM, Fagbo S, de Lamballerie X. Alkhurma hemorrhagic fever virus is an emerging tick-borne flavivirus. J Infect. 2006;52:463-4.

(14.) Memish ZA, Balkhy HH, Francis C, Cunningham G, Hajeer AH, Almuneef MA. Alkhumra haemorrhagic fever: case report and infection control details. Br J Biomed Sci. 2005;62:37-9.

(15.) Sidi G, Davidovitch N, Balicer RD, Anis E, Grotto I, Schwartz E. Tickborne relapsing fever in Israel. Emerg Infect Dis. 2005;11:1784-6.

* Universite de la Mediterranee, Marseille, France; and ([dagger]) King Abdul Aziz University, Jeddah, Saudi Arabia

Remi N. Charrel, * Shamsudeen Fagbo, ([dagger]) Gregory Moureau, * Mohammad Hussain Alqahtani, ([dagger]) Sarah Temmam, * and Xavier de Lamballerie *

Dr Charrel is a virologist at the Universite de la Mediterranee in Marseille, France. His research interests include arthropodbome and rodentbome viruses that cause disease in humans, with a special interest in emerging and reemerging viruses such as flaviviruses, arenaviruses, and phleboviruses.

Address for correspondence: Remi N. Charrel, Unite des Virus Emergents, Universite de la Mediterranee, 27 Blvd Jean Moulin, 13005 Marseille, France; email: [e-mail]

Table 1. Epidemiology of ticks collected in Saudi Arabia

Location           Type of location   Ticks isolated

Northeast Jeddah   Camels and camel   Ornithodoros savignyi nymphs and
                   resting place      adults and Hyalomma spp.
Southeast Jeddah   Camels and camel   O. savignyi nymphs and adults and
                   resting place      Hyalomma spp.
Kilaakh (50 km     Camels             H. dromedarii
from Taif)

Location           No. ticks   Date of collection

Northeast Jeddah       64           Jun 2004
Southeast Jeddah       32           Jun 2004
Kilaakh (50 km         28           Jan 2005
from Taif)

Table 2. Genetic differences between Alkhurma hemorrhagic
fever virus coding sequences of human (strain 1176
AF331718) and tick (strain JE7, DQ154114) origin *

                 No. mutated sites/
               no. sites in gene (%)

           Synonymous     Nonsynonymous
Gene        mutations       mutations     Ka:Ks

VirC      2/291 (0.69)          --         --
CTHD           --           1/60 (1.67)    --
prM       2/267 (0.75)          --         --
M         4/225 (1.78)     1/225 (0.45)   0.20
E       6/1,488 (0.41)    2/1488 (0.13)   0.25
NS1     4/1,062 (0.32)   3/1,062 (0.28)   0.43
NS2a      8/687 (1.16)     1/687 (0.15)   0.11
NS2b      2/393 (0.51)     3/393 (0.76)   0.60
NS3     9/1,863 (0.48)   5/1,863 (0.27)   0.36
NS4a      2/378 (0.53)     2/378 (0.53)   0.50
2K         1/69 (1.45)          --         --
NS4b      4/756 (0.53)     4/756 (0.53)   0.50
NS5    10/2,709 (0.37)   3/2,709 (0.15)   0.11

* Ka:Ks, ratio of nonsynonymous-to-synonymous nucleotides; VirC;
mature virion C protein; CTHD, C-terminal hydrophobic domain; prM,
premembrane; M, membrane; E, envelope; NS, nonstructural; 2K,
transmembrane domain.