Torque Teno Virus (TTV) was first discovered in the blood of a post-transfusion non A-G hepatitis patient which led to TTV being repeatedly linked as the causal factor of cryptogenic hepatitis. There have been numerous reports about the TTV viremia in patients of Hepatitis B and C from all over the world but there is severe lack of data on epidemiology of TTV and its sequence analysis from Pakistan. Therefore the present study was designed to clarify the rate of prevalence of TTV in hepatitis infected population vis-à-vis healthy subjects, outline a relationship of TTV prevalence with the incidence of hepatitis and to finally analyze the full length genome sequence of TTV from Pakistani isolate.
The findings of this study report a TTV incidence rate of 89.7%, 90.0%, 92.5% in the Hepatitis B group, Hepatitis C group and control population respectively. Our statistical analysis refutes the hypothesis that the frequency of TTV incidence is influenced by the factors such as age, gender and ALT level in hepatitis patients.
The high prevalence of TTV alludes to the likelihood of a latent, possibly lifelong infection. Some of the patients in our study were less than 8 years of age. Our findings are complemented by other reports of TTV infection in infants and children
[28, 30–32] which provide evidence that TTV is a mother-to-child transmitted virus, as suggested by Yokozaki et al. (1999) rather than a blood-borne virus.
In any epidemiological study of TTV prevalence with a PCR based method, the design and target position of primers is crucial in determining the frequency of positive detection of virus. Due to the high sequence diversity among TTV genotypes it is improbable that a single set of primers will detect all types of TTV. For this reason the primers used in this study were chosen on the basis of extensive review of data pertaining to surveys of prevalence of TTV. It has been emphasized by the findings of multiple reports that primers from UTRs will have greater potential to detect multiple genotypes of TTV rather than those designed against the ORFs
[33, 34]. The reason being that there is higher sequence conservation in the non-coding regions than in the coding regions, where in some stretches of the UTRs more than 90% sequence identity has been observed
Based on the above mentioned facts, for the present study primer pair NG133/NG132
 was selected. A thorough analysis of TTV detection strategies by Hu et al. suggests that a group of four primer sets have the ability to screen all genotypes of TTV and TTV-like viruses except genotype 21. Of this recommended array of primers, NG133/NG132 is a part
The results of the present study are in parallel with data published from other parts of the world. For example, in the study mentioned above, the positive rate of TTV detection was found to be as high as 93% with 5’-UTR primers but only 10% with primers designed from ORF region in the healthy population of Japan. Another study which surveyed the TTV infection across different geographic locations (Japan, USA, Mayanmar, Korea, Nepal, Egypt, Bolivia, Ghana) revealed the frequency of positive result between 70-98% from UTR primers, both in healthy and disease infected population
. The only previous report of TTV infection rate from Pakistan was made in 1999 in blood donors from Aga Khan University, Karachi. The reported prevalence was only 16% with primers from the N22 region which belongs to the ORF1 of TTV
. In separate studies from Russia, Iran, and Hong Kong, infection rate in healthy population was 94%, 80% and 90% respectively
[7, 37, 38]. In contrast to the above reports of high prevalence, TTV was observed in only 2.9% of healthy Iranian population emphasizing the fact that ORF primers have less conserved sequence which makes them incapable of matching multiple genotypes
The only studies where the rate of infection in healthy population is relatively lower than our findings, when similar primers have been used, are from India and UAE
[39, 40]. In these studies nested PCR approach was followed where NG133/NG147 were used in the first round while NG134/NG132 amplified in the second round. Out of 50 healthy volunteers, only 26 tested positive which is a 52% infection rate. The same study shows that 80.6% of patients with fulminant hepatic failure and 76% of patients with Acute Viral Hepatitis tested positive for TTV infection. A different research that tested the prevalence in patients on hemodialysis and healthy subjects; the positive infection rate was 83% and 43% respectively
. The study from UAE
 reported that 97.9, 95.7% of patients with HBV and HCV tested positive for TTV infection which aligns closely with our results of 89.7% and 90.0% of HBV and HCV patients testing positive for TTV infection. In contrast a mere 34.9% of healthy population was found to be carrier of TTV DNA which differs from our finding of 92.5% of control population that was observed as TTV positive.
The possible reasons for this marked variation in the frequency of infection in control population in our study and the above cited reviews could be the difference in the number of sample population, the fact that the prevalent genogroup in these areas is different from Pakistan or simply that TTV is more ubiquitous in Pakistan than in other countries.
The TTV isolate cloned in our study named TPK01 has sequence similarity spanning over most of the genome with the previously reported genotype 22 and 23 isolates Kt-08f and Kt-10f from Indonesia
. It is worth mentioning here that the previous isolates from Pakistan which were short clones of N22 region (200 bp approx.) belonged to genotype 1a in majority
. TPK01 also shows significant similarity to the PMV isolate from United Kingdom of genotype 17
. Stretches of 5’-UTR of the PMV sequence are up to 92% identical to TPK01. Other sequences that show identity with TPK01 over some stretches of genome are ViPi04, isolate from Italy
, tth6 and tth20 isolates of a Hodgkin’s disease patient from Germany
. The entries of genome that related most closely to TPK01 were all isolates belonging to genogroup 2.
The arrangement of ORFs, promoter and poly-A sites is similar to genotype 22 and other members of group 2. In TPK01 nt1362-1610 show no similarity with any of the TTV genomes assembled in NCBI Nucleotide database, this region corresponds to the HVR of ORF1 and has therefore hypermutated. In a study where TTV isolates were cloned from Hodgkin’s patients, the HVR extended over nt1425-1693 and showed 80 to 99% variation among the isolates that were cloned, whereas the overall variation in the sequence of the cloned TTV genomes was only 2%
. This high rate of variation that translates to divergence at the amino acid level could suggest immune-evasion by hypermutation
[27, 42]. Premature stop codons have also been observed in ORF1 of our isolate which splits the coding sequence into three shorter translation products. The occurrence of interrupted ORF1 has been observed in other TTV isolates as well. ORF1 splits in two potential reading frames in isolates tth7, tt32b8, tth8 while the isolate tth22g4 is interrupted twice which translates to three putative proteins
. TCHN-a, TCHN-c1, TCHN-d2, US32 also have premature stop codons which can give rise to two translation products instead of one
In one study, subviral molecules could be isolated from a Hodgkin’s lymphoma cell line when it was transfected with full-length TTV genomes. It was found that additional splicing events and rearrangements within the genome raised ORFs that could not be explained by the parent genomic sequence. These subviral particles varied in sequence length from 172 bp to full length genome. Similar kinds of molecules were also isolated from sera of pregnant mothers
. These findings present the question that whether the multiple genotypes and types of TTV (Torque Teno Mini Virus, Torque Teno Midi Virus) are due to wide intra-genomic rearrangements or are actually separately evolved genotypes.