To validate the STLV-1 prevalence in JMs, we first examined for positivity of the anti-STLV-1 antibody in plasma of 280 JMs from five independent troops originating from inhabitants of different areas in Japan. The plasma samples were obtained from JMs at the routine health checkups during the period from 2015 to 2016. We found that 180 of 280 macaques (64%) were seropositive for STLV-1 (Table 1), which was generally consistent with previous reports [41, 43, 44, 52]. We then determined the variation in the seroprevalence among the troops. The numbers of seropositive individuals were 59, 17, 36, 34, and 34, with a frequency of 68%, 55%, 63%, 56%, and 77%, respectively, for each troop (Table 1). In addition, the rearing population density in the free-range facility differed in each troop but was not correlated with the seroprevalence, suggesting that the density did not influence the high STLV-1 prevalence (Table 1).
We hypothesized that a substantial proportion of STLV-1-infected JMs might play a critical role as viral super-spreaders for frequent horizontal transmission and eventual high prevalence of STLV-1, possibly because of their abnormally high viral loads and poor humoral immune response against STLV-1. In fact, our recent incidence of an outbreak of infectious malignant thrombocytopenia in JMs by simian retrovirus type 4 (SRV-4) demonstrated that some of the monkeys developed asymptomatic SRV-4 infection with persistent viremia in the absence of SRV-4-specific antibody response and became viral super-spreaders [54], [55]. Taking this unexpected result into account, we evaluated ABTs and PVLs in the JM cohort. We found that the ABTs among 180 seropositive macaques were normally distributed with a geometric mean of 4076 and an ABT of 8192 at the maximum number of individuals (Fig. 1A, Additional file 1: Figure S1). We observed no significant difference in the titers among the five troops (Fig. 1B). We also examined the STLV-1 PVLs in PBMC samples and found that the PVLs of 171 proviral DNA-positive macaques were normally distributed and ranged from 0.01%–20% with a geometric mean of 0.62% and PVLs from 0.64% to 1.28% at the maximum number of individuals (Fig. 2A, Additional file 2: Figure S2). Again, we observed no significant difference in the PVLs among the troops (Fig. 2B). The data regarding ABTs and PVLs from the 183 macaques positive for either value (herein tentatively regarded as ‘STLV-1-infected’) were plotted as shown in Figure 3. Among the JMs, 168 were positive for both values, whereas three were negative for ABTs but positive for PVLs and 12 were positive for ABTs but negative for PVLs. Among the STLV-1-infected macaques, we did not observe any individuals with abnormally high PVLs and poor ABTs (Fig. 3). It is notable that the three ABT-PVL+ monkeys belonged to either of two troops (two macaques in troop C and one in troop D), and their PVLs were comparable or less than the mean PVLs. It is therefore unlikely that only three monkeys caused the high prevalence in all the independent troops. Rather, we observed positive correlation between ABTs and PVLs (R = 0.50, p < 0.0001) (Fig. 3), suggesting that humoral immunity was properly induced in response to the increasing proviral loads in these macaques.
We then sought the possible route(s) of transmission by which this high prevalence occurred. It was previously reported the age-related increase of STLV-1 prevalence in JMs, suggesting the frequent horizontal transmission whereas the mode of transmission remains to be elucidated [41, 43, 44, 52]. If mother-to-child transmission (MTCT) were the main route, the infection rate should drastically increase at around one year of age, followed by a gradual increase with age. On the other hand, if horizontal transmission were the main route, the infection rate should be low in younger ages, followed by a steep increase with age. To verify these possibilities, we examined the age-dependent change of seroprevalence in the cohort. The frequencies of seropositive individuals in each age group were 17%, 33%, 58%, 79%, 93%, 100%, and 95% at age groups of 0, 1, 2, 3–5, 6–8, 9–11, and ≥12 years, respectively (Table 2). We also analyzed the age-dependent change of proviral DNA prevalence and found that the frequencies of proviral DNA-positive individuals in each age group were 11%, 31%, 58%, 75%, 89%, 98%, and 91% for the respective age groups (Table 2). The infection rate was over 30% at one year of age and was dramatically increased over 50% at the age of 2 years, over 70% at the age of 3-5 years, followed by infection among almost all of them over 9 years of age, irrespective of the either value of positivity. Considering that the rate of MTCT in the case of long-term breastfeeding is approx. 20% when the children of 3 years old and over of HTLV-1 carrier mothers were tested [56], these results appeared that STLV-1 may be frequently transmitted via both maternal and horizontal routes. Of note, large numbers of younger JMs with relatively low prevalence (i.e., among 280 JMs cohort, 102 JMs were 0–1 years of age) reduced the apparent prevalence rate of the entire cohort (64%), although almost all the adult JMs elder than 9 years of age were positive (Table 2). In addition, there was no significant difference among troops regarding ABTs and PVLs (data not shown). We also examined the dynamics of the ABTs and PVLs with age among the STLV-1-infected JMs. It was found that the ABTs and PVLs were gradually increased from 0 to 2 years of age, followed by a slight increase of ABTs and mostly stable PVLs with age (Table 2).
Results described above suggested that STLV-1 was frequently transmitted via both maternal and horizontal routes. However, it is still possible to speculate that the JMs of long-term (e.g. more than 3 years old) latent MTCT but not horizontal transmission might gradually convert to be positive, which led to the result as shown in Table 2. In order to validate this possibility, we conducted a retrospective study of the STLV-1 seroprevalence in this cohort. First, the cohort during the period from 2011 and 2012 was examined for the seroprevalence and found that 224 of 366 JMs (61%) were seropositive (PBMC samples during this period were not available), which was mostly comparable with the result during the period from 2015 to 2016 (64%) as shown in Table 1. We then selected 139 monkeys whose plasma samples were available in both periods. In 2011-2012, 111 of 139 JMs were seropositive, whereas 28 were seronegative. Four years later, 24 (86%) of the seronegative 28 monkeys in the former period were seroconverted for the antibody (Table 3). Remarkably, among ten seronegative monkeys of four years of age and older (i.e., sexually mature monkeys) during the period from 2011 to 2012, eight were seroconverted within four years interval (80%), which was comparable with the monkeys of three years of age and younger (i.e., sexually immature monkeys) in the former period (16/18, 89%). Although seroconversion occurred mostly in the younger ages, the frequency of seroconversion between the seronegative immature and mature monkeys was comparable (89% and 80%, respectively), which was consistent with the results shown in Table 2. These results strongly suggest frequent STLV-1 transmission occurring among JMs via both horizontal and maternal transmission routes.
It has been shown that a certain degree of heterogeneity of HTLV-1 genome was present among the virus-infected individuals in the same community[57, 58] while the heterogeneity of the viral genome between mother and child was minimal[59-65]. If it is the case with STLV-1-infected JMs, then it could be possible to differentiate whether the STLV-1 infected in a monkey is derived from mother through MTCT or from any other monkeys through horizontal transmission. In order to examine the possibility, we compared the nucleotide sequences of STLV-1 LTR, tax and env regions from randomized 12 JMs in a troop D as a representative. It was found that almost all the sequences of 3’LTR and env region were identical in all the monkeys except for only a unique heterogeneity in some monkeys (for 3’LTR: C333G for A1671; G346A for A2594, for env: T1218C for 8 monkeys, respectively) (Additional file 3: Fig. S3a and S3c). Furthermore, no heterogeneity was observed in the tax region of all monkeys (Additional file 3: Fig. S3b). These results indicated that STLV-1 genome was highly conserved among the JMs in the troop, with consistent results in terms of other troops (Kuramitsu et al., unpublished observation). Consequently, it was not possible to determine the transmission route of STLV-1 on the basis of the heterogeneity of the virus genome, which was unlike HTLV-1.