An orphaned Rothschild giraffe from Dublin Zoological Gardens was presented to the University Veterinary Hospital at 14 days of age with a history failure of passive transfer of immunity, anorexia, dehydration, hypoglycemia, acidosis and persistent profuse watery diarrhea. The animal was treated on admission with intravenous fluid therapy, antimicrobials and non-steroidal anti-inflammatory drugs. The giraffe calf died 4 days post-treatment. A post-mortem examination primarily revealed a severe abomasitis and enteritis. No significant pathogenic organisms were isolated on bacteriological culture of the spleen, rumen or abomasum presumably as a consequence of intensive antimicrobial therapy. A faecal specimen was sent to our laboratory for further investigation.

Preliminary testing of the faecal specimen obtained (by Transmission Electron Microscopy) revealed the specimen to be positive for RV infection. Husbandry issues were considered to assess the potential route of RV infection. Interestingly, the calf had not been in contact with any other ruminants' prior to admission to the University Veterinary Hospital. However indirect contact with other animals could not be out ruled. Other issues for consideration were the movements of the keepers between different animal enclosures, the feeding equipment used and the apparatus used to clean the housing areas.

Transmission electron micrographs captured the appearance of negatively stained RV observed in the giraffe faecal sample. Faecal samples were prepared by mixing with 2% (w/v) methylamine tungstate negative stain on a 3.05 mm grid and allowed to stand at room temperature for 4 min. The specimen was examined using electron-optical magnification up to × 40, 000 range.

A faecal specimen from the 14-day old giraffe was sent to our laboratory for analysis. Initially this sample was diluted in phosphate buffered saline solution (1× PBS). Rotavirus dsRNA was purified from the faecal specimen using a standard phenol/chloroform protocol with ethanol precipitation 18. Briefly, 320 μl of the faecal suspension above was combined with 40 μl 10% (w/v) SDS and 40 μl 10% (w/v) proteinase K and incubated for 90 min at 37°C. The suspension was extracted once with phenol/chloroform (5:1) followed by one extraction in chloroform alone. RNA in the aqueous phase was removed to a clean eppendorf tube and precipitated with two volumes of ethanol and then stored at -20°C overnight. Purified dsRNA was recovered by centrifugation (1, 000 × g for 30 min) and the pellet containing the viral nucleic acid was suspended in 100 μl diethyl pyrocarbonate (DEPC) treated water. Purified dsRNA was stored at -80°C until required.

dsRNA segments were separated by electrophoresis in 8% (w/v) polyacrylamide slab gels, 1.5 mm thick with a 7.3 cm path length. Electrophoresis was performed using a discontinuous buffer system based on a modification previously described by Laemmli for 80 min at a constant voltage of 110 V 19.

The presence of viral dsRNA in the faecal sample was confirmed following staining with a DNA Silver Staining Kit according to the manufacturer's instructions (Amersham Biosciences, UK).

A one-step reverse transcriptase-polymerase chain reaction (RT-PCR) was performed. Purified viral genomic dsRNA (3 μl, containing approx. 50 ng) was added to 3.5 μl DMSO (Sigma-Aldrich, Steinheim, Germany), and denatured by heating for 5 min at 94°C. The tubes were placed immediately on ice to prevent re-annealing of dsRNA. RT-PCR was performed according to the modified methods described by Gentsch et al., 20 and Gouvea et al., 21 (Table 1) using 30 thermal cycles at 94°C for 1 min, 55°C; for 2 min, 72°C for 2 min followed by 10 min incubation at 72°C.

Briefly, denatured dsRNA was added to a reaction mixture consisting of 10 μl 5 × reaction buffer (Promega, Madison, WI, USA), 8 μl of a deoxyonucleoside triphosphate mixture (consisting of 1.25 mM of each dNTP) (Promega), 3 μl 25 mM MgCl₂ (Promega), 0.1 μl reverse transcriptase from avian myeloblastosis virus (5 U/μl) (Promega), 0.5 μl Taq DNA polymerase (5 U/μl) (Promega,) and 1 μl of each selected primer pair (20 pM) as detailed in Table 1. VP4 and VP7 amplification reactions produced DNA amplicons of ~880 and 1,062 bp respectively.

A semi-nested, multiplex PCR reaction was performed using primers targeting the genotype-specific regions of the VP4 and VP7 genes 2021222324.

Amplified DNA fragments were purified using a QIAGEN PCR purification Kit (QIAGEN, Hilden, Germany) and sequenced in both directions. Nucleotide sequences were assembled and analysed by DNAStar software (Lasergene, Madison, WI). All sequences were compared against those available in the current GenBank database http://www.ncbi.nlm.nih.gov/Genbank/index.html. Sequence data from the giraffe samples were entered in the GenBank database under the following Accession numbers [GenBank: EU548032 and GenBank: EU548033].

Separate alignments for the Giraffe VP7 and partial length VP4 amino acid sequences were performed. The alignments were imported into Mega4 and neighbour-joining trees were calculated with Poisson correction, deleting all sites where a gap appeared in any sequence. Statistical confidence was established by carrying out 1000 bootstrap replicates and branches were collapsed unless supported by 80% of the bootstraps 25262728.

The presence of RV in the faecal specimen obtained from the 14-day old giraffe was assessed initially by T-EM. Clusters of negatively stained RV particles with an average diameter of 70 nm were observed (Fig. 1) in the electron micrographs. These particles were similar in shape when compared to those RV of human and bovine origin (data not shown).

RV in the faecal specimen was confirmed after purification and subsequent analysis of the dsRNA banding profile in a SDS-PAGE gel. The electropherotype obtained was compared against several unrelated RV strains obtained from other animal and human sources (Fig. 2). The profile for Giraffe rotavirus (GirRV) (Fig. 2, lane 7), displayed the typical group A rotavirus dsRNA genome banding pattern (consisting of a 4, 2, 3, 2 arrangement) characteristic of the 'long' RNA electropherotype pattern.

The identification of the corresponding G- and P-types were established by RT-PCR on the purified dsRNA from the giraffe sample (using primers listed in Table 1). The complete VP7 gene was amplified using a pair of broadly specific consensus primers, located at the beginning (1–21 nt) and end (1,062–1,036 nt) of this gene. In a semi-nested PCR, using a set of internal specific primers, the G-type was determined as G10 (Fig. 3, lane 9).

Similarly, partial gene 4 sequence was amplified and subsequently genotyped using primers for the detection of widely distributed bovine P-types. These data identified a P[11] genotype (Fig. 4, lane 6).

The genotypes for giraffe rotavirus (GirRV) detected in the faecal specimen were classified as G10P[11].

Phylogenetic analysis of the amino acid sequences derived from VP4 and VP7 genes was undertaken, using a set of similar genes selected from the current databases, for comparison purposes. Comparisons with the VP4 and VP7 sequence from the giraffe sample revealed closest similarities to bovine RV sequences. Alignment studies (data not shown) showed a 98% identity at the amino acid sequence level between giraffe and bovine species.

The full length VP7 amino acid sequence was analysed and compared to 15 well-established G-genotypes previously published. GirRV was most similar to a bovine G10 gene [Accession number GenBank: X57852] (Fig. 5 and Table 2).

Comparison of the partial VP4 amino acid sequence of GirRV with 27 P-genotypes previously reported for group A rotaviruses revealed the closest branching relationship to the bovine P[11] type, [Accession number GenBank: D13394] (Fig. 6 and Table 3).