Post, Klara, Olson, Erik D. Nabuan, Gorelick, Robert J. United States: N. Copy to clipboard. United States. Nabuan and Gorelick, Robert J. Free Publicly Available Full Text. Accepted Manuscript DOE. Copyright Statement. Other availability. Search WorldCat to find libraries that may hold this journal.
Cited by: 5 works. Citation information provided by Web of Science. The numbering is based on the sequence of the mature NC protein in each case. Coloring and underlining are the same as in a. The boxes indicate the sequence comprising each ZF « less. All figures and tables 10 total. LinkedIn Pinterest Tumblr. Works referenced in this record: Nucleic acid binding and chaperone properties of HIV-1 Gag and nucleocapsid proteins journal , January Cruceanu, M.
Nucleic Acids Research, Vol. Science, Vol. Journal of Virology, Vol. Biopolymers, Vol. Although a reciprocally monophyletic clade of SIVcpz has been found in the eastern chimpanzee subspecies Pan troglodytes schweinfurthii , virus from this group does not appear to have jumped successfully into humans [9].
Surprisingly, the SIV lineage that falls immediately basal to HIV-1 group O was found in gorillas, suggesting that they might have been an intermediate host between chimpanzees and humans [10]. Answering this question would help determine the length of time SIV was in sooty mangabeys and chimpanzees before giving rise to the transmissible HIV lineages.
Determining the age of SIV would provide perspective on the spread of the virus among African primate species and the subsequent zoonoses. Knowing the age may also have implications for the evolution of pathogenicity and virulence in HIV. Historically, this lack of disease was attributed to the codivergence and coevolution of SIV and their primate hosts over millions of years [14] we use the term codivergence instead of cospeciation, because codivergence considers phylogenetic congruence irrespective of species classification, whereas cospeciation implies that SIVs infecting different primates can be classified as species complexes.
Although there is significant correspondence between the SIV and host phylogenies, detailed analysis of this relationship suggested that a preferential host switching model, in which cross-species transmissions of SIV are more likely to occur between closely related primates, could account for this correspondence [15]. Furthermore, subsequent analysis of SIV infecting various African green monkey species, thought to be exemplary of codivergence, demonstrated a lack of evidence for host-virus codivergence [16].
In addition, the codivergence hypothesis does not account for the observation that SIV is geographically confined and naturally infects only African primates. Finally, even with biologically unrealistic assumptions about a molecular clock, Sharp et al. If it were demonstrated that SIV has evolved in a clock-like manner, then we might be able to accurately determine the age of SIV.
We also provide, to our knowledge, the first estimate of the age of HIV-1 group N. Taken together, these dates suggest that SIV may indeed be a relatively young viral clade and that its transmission into humans is a natural process. The position of the root, determined by the BMCMC analysis, was also highly supported in each of the three trees. Phylogenetic inference using the three loci produces different topologies, which was expected given the observation of recombination by Apetrei et al.
Although gag was older than env , this difference was not significant. There was very high posterior support for monophyly in each of the three HIV-1 lineages as well as for the position of the root.
The three loci produced different topologies, which is not surprising given the recombinant history of HIV-1 group N [1] , [19]. In contrast, gag had the youngest date at — , and env produced an intermediate date at — We also performed additional phylogenetic inference to ensure that we captured the deepest available HIV-1 group N lineages in our analyses Figure S1. We initially thought that this discordance was due to recombination among the loci.
If recombination were responsible, the different tMRCA estimates would actually represent different times of coalescence. When examining the phylogenies, however, we found very little evidence for this scenario. An explanation of recombination would necessitate selective sweeps in gag and env , which would then go on to recreate the ancestral diversity seen in the pol phylogeny.
For example, HIV-1 group M would have a mean tMRCA around , and, approximately years later, part of the genome would have experienced a selective sweep that gave rise to the same pattern of sub-type diversity Table 1. We then explored the possibility that some of these analyses were biased.
That this discrepancy among tMRCA estimates was most pronounced in the older nodes indicated a loss of signal due to this bias deeper in the phylogeny.
We examined the demographic parameters e. Even though these genes evolved along different topologies, their demographic history, and therefore the demographic parameters inferred from them, should be the same. We hypothesized that some genes lack sufficient demographic signal to draw accurate inference about tMRCAs and that allowing the three loci to combine their demographic signal might homogenize their tMRCA estimates.
To test this hypothesis, we compared a partition analysis where the concatenated genes shared a single demographic scenario to analyses where that scenario was inferred for each gene independently. In all cases, allowing the three loci to share demographic information homogenized the tMRCA estimates such that there were no longer significant differences in the age of the root among the phylogenies.
Since tMRCA estimates from g ag are the most robust to combining demographic parameters, these dates should be taken as the best estimates.
Given the different selective regimes that these loci experienced, it is unlikely that the differences in the tMRCA estimates among the three loci were due entirely to variable demographic signal.
Nevertheless, accounting for this variation in demographic signal appears to have resolved the majority of the discrepancy among the tMRCA estimates. In addition, even if the differences among the tMRCA estimates were real, and due to recombination, all three loci suggest root ages that are of the same order of magnitude. We next sought to determine if the dates we obtained were the result of clock-like signal within SIVsm or whether SIVsm had no clock-signal and we were inadvertently extrapolating HIV-2 rates across the entire tree.
We compared the date estimates from gag , pol , and env to analyses where all non-SIVsm sequences were excluded. Furthermore, there was no significant difference between the SIVsm gag substitution rate we estimated of 1. This similarity indicates that SIVsm does indeed have sufficient clock-like signal to date tMRCAs, and it does not appear to evolve at a different rate than HIV-2 group A, despite differences in host biology and pathogenicity.
In a population of constant size, the most basal lineages are consistently lost due to normal coalescent processes; the age of the root is expected to be approximately two times the effective population size [20].
However, if the population is expanding exponentially, the basal lineages will be maintained until a carrying capacity is reached. To determine if exponential growth explains the SIVsm population dynamics better than a constant population size, we looked at the exponential growth rate in alignments containing only SIVsm sequences.
Thus, it seems probable that pol failed to reject a constant population size because it simply lacked sufficient demographic signal. Therefore, it is likely that SIVsm has not been evolving at a constant population size for the past years.
We note, however, that a tMRCA estimate will tend to post-date the actual introduction of viral lineages into a new host if genetic diversity has since been lost or is not fully sampled. We believe such comparisons still provide useful information as long as this caveat is recognized.
Ninety-five percent HPD intervals are larger for these estimates than for other single clades because the age estimates for any two clades are not perfectly correlated. The findings presented in this study indicate that the tMRCA of SIV in sooty mangabeys and chimpanzees is — and — , respectively, assuming the relaxed molecular clock is unbiased. This date suggests that HIV-1 group N is the youngest transmissible HIV lineage and the only lineage to have originated in the second half of the twentieth century though the possibility of a deeper history cannot be excluded given the sparse sampling.
The dispersed timing of these transmissions to humans implies that no single external factor is needed to explain the cross-species transmission of HIV. This observation is consistent with both of the two prevailing views of the origin of the HIV epidemics. The first is the bushmeat hypothesis [22] , whereby SIV is transmitted to humans during the slaughter or butchering of infected primates.
The second is that the growth of sub-Saharan African cities allowed for these nascent lineages to gain a foothold [5] , [7]. According to the second hypothesis, SIV may have been jumping into humans since it first infected chimpanzees and sooty mangabeys. A change in human ecology then may have altered the evolutionary dynamics, whereby a virus that historically may have only infected a few individuals and then died out now has the potential to become an epidemic lineage.
It does not seem farfetched to venture that SIV will continue to be transmitted to humans well into the twenty-first century. There are several arguments suggesting that SIV has been present in sooty mangabeys and chimpanzees longer than our results indicate. First, coalescent processes or selective sweeps might have removed the deeper lineages from the phylogeny. While we cannot discount the latter, our finding that the SIVsm population has not evolved under a constant size suggests that deep SIVsm lineages may still be present.
It remains unclear whether coalescent processes may have removed deep SIVcpz branches. A second argument is that our sampling was not thorough enough, and deep SIV branches were not included. While possible, other studies that included additional non-dated SIVsm and SIVcpz sequences did not uncover additional deeper branches [8] , [18] , [23].
As a part of this study, we demonstrated that one major SIV lineage evolves in a clock-like fashion and at a rate indistinguishable from HIV. This difference is likely because we had far fewer SIVcpz sequences that were sampled over a relatively small window of time.
Lastly, it is possible that our relaxed-clock models are biased and therefore unable to accurately date SIV coalescent events.
We cannot dismiss this possibility, but the accuracy of these methods has been previously confirmed by other studies predicting the year of sampling of older HIV isolates from and [4] , [5]. All other SIV lineages are blue. Tree is midpoint rooted. This difference is expected to be minor, since mutation of F16 to W has little effect on HIV-1 NC NA chaperone activity and intravirion reverse transcription and no effect on infectivity [ 58 ].
In all cases, numbering refers to the nt position in gRNA. In view of the critical role of NC chaperone activity in reverse transcription [ 18 , 19 , 37 ], we investigated the activity of SIV and HIV-1 NCs in model systems that recapitulate the reactions required for minus-strand transfer Fig. Note that in our systems, the minus-strand transfer assay includes the annealing step as well as DNA elongation.
Annealing of the complementary R regions is indicated by vertical lines. The diagram is not drawn to scale. Representative gels can be found in Additional file 1 : Fig. Error bars represent the standard deviation SD from three or more independent experiments. To compare the annealing activities of the two NCs Fig. As the NC concentration was increased, the extent of annealing was also increased.
For example, when 1. In contrast, comparison of the rates of annealing showed small, but more significant differences between the activities of the two NCs Table 1 a.
For example, at 1. Again, there was a small, but more significant difference in the reaction rates with the two NCs. With 1. Table 1 b. Not surprisingly, minus-strand transfer with the HIV-1 substrates was slightly more efficient than with the SIV substrates, but in this case too, the end point values with 1.
The rate of the reaction was 2. Note that self-priming products [ 18 , 19 , , ] were not formed under the conditions used for these assays. Error bars represent the SD from three or more independent experiments. As shown in Fig. To quantify these characteristics, we calculated the transition slope, which reflects the degree of intercalative binding by NC to DNA, and the hysteresis area ratio, which reflects the amount of strand separation, as a function of concentration [ 58 , 65 , 66 ] Additional file 3 : Fig.
S3, Additional file 4. The lines in b are fits to a simple binding isotherm Additional file 4 : Eq. Error bars are standard errors for three or more measurements. The slope of the force-extension curve is measured near the midpoint of the transition by averaging over the extensions between 0. This makes it possible to obtain a value for the hysteresis area ratio Additional file 4 : Eq. The results of these quantitative analyses as a function of concentration are shown in Fig. Analysis of the hysteresis as a function of concentration also shows that for the lowest concentrations tested, HIV-1 NC binding resulted in a larger hysteresis area ratio compared to SIV NC.
At all other concentrations tested, the proteins behaved in a very similar manner Fig. This additional force at low extensions is referred to as the DNA compaction force F c. The magnitude of the F c reflects the ability of the protein to attract dsDNA, which normally results in DNA aggregation in the absence of applied force [ 67 , 68 ]. To quantify this compaction force, we used the method described in the legend to Fig. This corresponds to a difference in compaction energy of 0.
F c is calculated in the low force-extension regime denoted within the gray box in the inset and magnified in the main figure. In addition to the reactions with NC, a no NC control was included. The average size of the NA aggregate formed in the absence of NC was 0. These values agree with previous reports using different NA substrates [ 45 , 69 , 70 ].
Each curve represents the average of at least three independent experiments. Retroviral NC proteins interact with NAs using both specific and non-specific modes of binding [ 27 ]. Fluorescence anisotropy FA salt-titration assays have previously been shown to be capable of distinguishing the relative contribution of specific vs. Thus, a protein-RNA complex that is more dependent on electrostatic interactions than specific contacts, dissociates at lower salt concentration relative to a complex that is characterized by specific non-electrostatic binding interactions.
K d 1M represents the K d of the protein-RNA interaction when all electrostatic contacts have been screened out and only specific ones remain e. Similarly, Z eff values of 5.
S6, Additional file 8 : Fig. S7, Additional file 9 : Table S1, and Additional file This issue is of great importance in view of the essential role of the NC protein in retrovirus replication [ 18 , 19 , 20 , 76 ] and the widespread use of simian model systems for studies on HIV-1 pathogenesis, vaccine development, and drug resistance [ 77 , 78 ], and more recently, the development of HIV-1 eradication and CURE strategies [ 79 ]. Here, we focus on two important events in the virus life cycle: the minus-strand transfer step in reverse transcription that is required for synthesis of a full-length copy of the viral RNA genome; and selective gRNA packaging directed by the Psi structural element.
However, we also demonstrated that despite an overall functional similarity, quantitative differences in NA aggregation and compaction capability distinguish the two proteins, which may be relevant to the infection process.
In our initial approach, we examined the minus-strand transfer reaction, since the rate-limiting step, i. Assays of annealing or annealing plus DNA elongation Figs. S1, Additional file 2 : Fig. However, a more quantitative analysis revealed a significant difference in the compaction forces, F c s, generated by these two proteins at extensions less than the DNA contour length Fig. SIV used to induce dimerization [ 83 ]. It is important to note that the minimal Psi packaging element has not yet been unambiguously identified in HIV-1 [ 25 , 30 ] or SIV [ 61 , 62 , 73 , 74 ], as it has for other retroviruses such as MLV [ 85 , 86 , 87 ] or Rous sarcoma virus RSV [ 88 , 89 , 90 ].
Nevertheless, despite the high structural and functional homology, our studies also clearly reveal subtle differences in the NA chaperone functions of HIV-1 and SIV NC proteins that can be explained by differences in their NA aggregation capabilities and DNA compaction energies.
Based on biochemical assays and quantitative biophysical analysis, we demonstrated that despite a high degree of similarity between SIV and HIV-1 NC proteins, modest differences in their nucleic acid chaperone activities were observed, which reflect differences in DNA compaction energy and ability to aggregate NAs.
Taken together, the vast similarities and only subtle differences observed in NC functional assays help to further validate SIV as a useful vehicle for development of new therapeutic strategies in the fight against the devastating consequences of AIDS. AF [ 94 ]. M [ 12 , 13 ], which was obtained from Dr. In addition, there is functional identity at position 4, with K for Mne and R for Mac SIVmac NC was used for some of the initial minus-strand annealing and strand transfer experiments; however, the results obtained with either SIV NC were the same within uncertainty data not shown.
M obtained from Dr. The dsDNA product was run on a 2. RNAs were prepared by in vitro transcription and purified as previously described [ ]. For comparison, identical conditions were used for the HIV-1 system.
HIV-1 RT 1 pmol and 0. The standard reaction was scaled up as needed. One bead was held in an optical trap, while the other was immobilized on a micropipette tip attached to a flow cell placed on a translational piezoelectric stage [ ]. By gradually moving the fixed bead while recording the extension and the force exerted on the single DNA molecule, the force-extension profile of a dsDNA in the absence of protein was obtained.
Data were analyzed using the Dispersion Technology Software provided by the manufacturer, and sizes were plotted as volume percent versus particle size. The average size of the aggregate population produced was calculated by taking the product of the aggregate volume at each particle size sampled and averaging over the total volume of the population.
The average diameter was calculated as the mean of 3—5 measurements with the standard error indicated. The salt-titration binding assays were performed essentially as previously described [ 27 , 75 ]. To correct for the effect of increasing NaCl on RNA anisotropy independent of protein binding, separate salt-titration assays of the RNA in the absence of protein were carried out with every trial.
The no protein control values were then subtracted from the data obtained for protein-containing reactions. The corrected data were then analyzed as described [ 27 , 75 ]. In Eq. Z eff represents the number of electrostatic contacts involved in the interaction. Substituting Eq. Sample concentrations ranged from 3. Scattering data were acquired and buffer subtraction was performed by the SIBYLS beamline staff as described [ , ]. Subsequent data analysis and ab initio envelope generation were performed largely as previously described [ 28 ].
Briefly, the SAXS data collected at different exposure times for a given concentration of sample were examined separately and exposures with clear evidence of radiation damage were discarded. High quality exposures for each RNA concentration dilution were merged and then analyzed using Guinier analysis [ ] to calculate the radius of gyration R g and the extrapolated scattering intensity at zero scattering angle I 0 using the program PRIMUS [ ].
Kratky analysis [ ] was also performed for each RNA concentration dilution to confirm the extent of folding. If the R g was found to increase upon increasing RNA concentration indicative of concentration-dependent effects or if the Kratky plots suggested that the RNA was not well folded, the data were not analyzed further.
If samples passed these quality control analyses, then the data sets from the three concentration dilutions were scaled and merged into a single curve. The maximum inter-electron distance D max was varied until the P r decayed smoothly to zero and the experimental data fit well. Then for each of these D max values, five ab initio envelopes were generated in fast mode with no symmetry restraints imposed using the ATSAS suite of programs as described [ 28 , ].
Isolation of a T-lymphotropic retrovirus from a patient at risk for acquired immune deficiency syndrome AIDS. PubMed Article Google Scholar. Hybrid origin of SIV in chimpanzees. Chimpanzee reservoirs of pandemic and nonpandemic HIV Cold Spring Harb Perspect Med.
0コメント