**3. Results**

#### *3.1. Direct Sequencing and Phylogenetic Analysis*

Both generic primers for F and G genes were successful in amplification of the strains, showing relatively high viral load for 13 of the 59 samples (≥1 × 103.5 copies/mL). The F gene domains of the hRSV-A isolates possess about 10% nucleotide variability. The signal peptide, however, showed higher nucleotide variability. Partial F gene sequencing revealed that the Saudi strains in the current study belong to the NA1 genotype (6/13, 46.15%), ON1 (5/13; 38.46%), and GA5 genotype (2/13; 15.38%) (Figure 1a). Partial G gene sequencing using strain-specific primers for these strains confirmed the correct strain classification obtained by using F gene sequencing, and similar clustering of the hRSV strains was obtained using both genes (Figure 1a,b). The F gene domains of the hRSV-A isolates possess about 10% nucleotide variability. The signal peptide and transmembrane domains, however, showed nucleotide variability to be higher than detected in the HVRs of the G gene (data not shown).

#### *3.2. Deduced Amino Acid Sequence and Sequence Analysis*

Five N-glycosylation sites, N27, N70, N116, N120, and N126, were found to be conserved among all the thirteen Saudi strains, except for strain Taif-103, that has lost one of the N-glycosylations at N120 (Figure 2). The fusion protein of Taif-20, Taif-23, Taif-110, and Taif-112 showed S-to-N substitution at position 105 just before the first cleavage (Figure 2), with no potential e ffect on the function as demonstrated by the polymorphism phenotyping prediction tool (data not shown). Seven amino acid substitutions—L3F, T11I, I19T, T13A, L15F, A16T, and I19T—were detected in the signal peptide, and three—T120A, T125N, and V127I—amino acid substitutions were detected in the p27 domain (Figure 2).

A duplication of amino acids (QEETLHSTTSEGYLSPSQVYTTS) was found in two regions within the G protein which reflects an 72 nt insert characteristic to the ON1 genotype (Figure 3). L298P, V303A, and Y304H amino acid substitutions were detected in all ON1 Saudi strains, Y273N was detected in Taif 23, and L274P was detected in Taif-24 and Taif-103 strains (Figure 3). The P310L substitution, equivalent to P286L in viruses that have no duplication, was recorded in 10 out of the 13 Saudi strains. There is an existence of accumulated signature coding changes in the epitope regions of the G protein (Figure 3). Variations among the N-glycosylation site of HVR2 of the Saudi strains' attachment proteins were detected. Interestingly, all Saudi NA1 and ON1 strains except Taif-24 and Taif-103 showed N237D amino acid substitution that resulted in loss of the N237 N-glycosylation site. N250 was found only in Saudi GA5 strains, while N251 was detected in 2/5 ON1 and 6/6 of the NA1 Saudi strains. N273 was detected only in Saudi GA5 strains, as well as a single ON1 Saudi strain. N294 (N318 in ON1 genotypes) was detected in the majority of Saudi ON1 and NA1 strains, but none of the Saudi GA5 strains or the A2 prototype strain (Figure 3).


**Figure 2.** Deduced amino acid sequences of partial fusion proteins from different Saudi hRSV-A strains. Highlighted are the signal peptide (1–22) (blue box), heptad repeat domain C (75–97), cleavage site-1 (CS-1) and cleavage site-2 (CS-2) at Arg109 (NSRARR↓E) and Arg136 (KKRKRR↓F) (red color), p27 (110–136) (green box), fusion peptide (FP) (137–155), and heptad repeat domain A (153–end of the current sequence). The N-glycosylation sites, NXT/S, where X is not a proline, are underlined.


**Figure 3.** Deduced amino acid sequences of the partial attachment protein (G) from different Saudi hRSV-A strains. The N-glycosylation sites, NXT/S, where X is not a proline, are underlined. Duplicated regions in the ON1 strains are shown in the boxes.
