|
|
 |
| TRANSCRIPT OF DR MILES DAVENPORT’S VISIT ON WEDNESDAY THE 28TH MAY |
| Dr. Kathy Takayama |
|
Download PDF |
| Dr. Miles Davenport, Head of Complex Systems in Biology Laboratory (UNSW), visited VSG for a live chat on Wednesday the 28th May from 09:00 - 10:00 (Omnium Time). Course participants can read the transcript below or download it as a PDF; however visitors to VSG do not have access to documents in this area. |
|
kathy: Welcome to Omnichat, Miles, and thanks for your time. We have some participants who are eager to talk to you... can we start off with some questions?
miles: yes. I should perhaps start by saying that there seems to have been a lot of interest in vaccines in the discussions I have seen on-line so far. This is the current area of my work, so I am happy to field Q on this also
kathy: Can we have a question from a team member?
mary: Perhaps you could comment on what approach you believe from your modeling experience is most favorable at the moment in terms of HIV control---
miles: I'm not sure if you mean at a population level (control of the epidemic), or at an individual level - control of virus within a host. I work on both. The anti-retrovirals are doing well at controlling virus within individuals (at least until they get drug resistance). Within a population it is less clear what the impart of HAART will be. In some senses, the use of these drugs in the west is not a good test, since the epidemic is not growing so fast we can really tell.
miles: In terms of immune control - the evidence from monkey vacination trials suggests that CTL can control the level of virus in chronic infection - although they seem quite unable to prevent actual infection.
mary: I was thinking primarily of the individual level--but of course this affects the population.
miles: Well, at the individual level, combinations of drugs work very well to control virus to low levels and allow CD4 t cell 'recovery', and even when drug resistance occurs it seems that the virus is less 'pathogenic' (ie: CD4 cells are not as depleted as you would expect with normal virus)
mary: Is there long term control in the simian studies, or do they also develop resistance?
miles: I don't think anyone has treated monkeys for that long to do a good study of this - largely because it is so difficult to treat them (- they won't swallow tablets). there are some interesting studies where treating during primary infection can lead to long term viral control (presumably immune control)
amy: Could you explain more about how T cell senescence functions in HIV infection? Is it connected with the reason the infection progresses from R5 strains to X4?
miles: Hmm. I haven't drawn a direct relationships between the two. Senescence is discussed in the context of the CD8+ CTL response to the virus, and that high avidity 'dominant' clones may senesce. By contrast, the CXCR4 argument relates to division rates of CD4+ t cells - particularly the difference between naive and memory division rates.
miles: the senescene of these high avidity cells may mean that the overall avidity of the response declines with time. Since viral load is set by the equilibrium between T cell responsiveness and viral proliferation, you could imagine that the gradual increase in viral load is due to a gradual decrease in average avidity.
miles: However, no one thast i know of has good enough data on the CTL response to HIV to really be certain that senescence is occurring, although the data that exist are generally supportive.
kathy: Anne, from Team Rana, had posted a question on the message board for you, Miles as she isn’t here this morning, I will forward her question to you:
"I understand that R5 strains are seen more in the early stages of infection and then there is a changeover to X4 strains. Are both these strains present during initial introduction of the virus or is this a changeover that occurs during the intial stages of infection? This is a repeated pattern seen in HIV-1 infection, would there be a realistic way to break this pattern as the R5 stage of infection seems less virulent?"
miles: In response to Q on CCR5 on web:
1) there is no reason that CXCR4 tropic virus isn‚t transmitted early on ˆ However it seems likely that this virus would only grow very poorly, because the turnover (division rate) of naïve cells is so low.
2) CXCR4 tropic virus doesn‚t really emerge until late in infection, and even then, it‚s not as if the CCR5 tropic virus totally disappears (the two tend to coexist)
3) There is some debate what proportion of individuals develop CXCR4 tropic, but probably the majority.
4) Breaking the pattern - There is conflicting evidence that HAART may reverse the dominance of CXCR4 tropic strains (by increasing CD4 T cell numbers and reducing naïve cell division rates)
I have a mathematical model for this Œcell turnover and cell tropism‚ hypothesis that supports the above conclusions (if anyone likes models)
kathy: Would you be able to post your model for us on the site? We could perhaps put it up in a Symposium
miles: Hmm, a bit difficult - actually the figure on my web-site is one schematic incarnation of the model, but like most models it is pretty indecipherable without text. There is supposed to be a talk on the model on-line at the dimacs site (http://dimacs.rutgers.edu/Workshops/Infectious/program.html)
but I have never found where they put the actual talks (just the abstract)
carlos: So, in every steps of the disease, the virus atacks the CCR5 or CXCR4 in function of the cell numbers that have this receptors, and it is variable and a fuction of the disease, it sounds complicated.
kathy: Ok, sounds great. Does anyone else have a question following up on the CCR5?
amy: What about the CCR5 tropic strains makes them more manageable than the CXCR4 strains? Is the difference the degree of discrimination between receptors?
miles: some people argue that CXCR4 virus is more 'pathogenic' because it kills more cells faster in culture. other studies suggest that it is equally pathogenic on a per cell basis (ie: once a cell is infected with either strain it dies just as fast).
I think that both of these are probably true, and the reason that CXCR4 appears more pathogenic both in vitro and in vivo is that it kills naive T cells - and these are the main source of 'replacement' t cell sin late infection. Therefore each naive cell killed has a much greater effect on CD4 t cell numbers
mary: I am wondering about syncytium-inducing subsets in general, and their overall effect on the person being infected and the division rate of the virus?
miles: well again, some people have argued that CXCR4 tropic is more 'virulent' and grows faster. (ie: produces more virus per infected cell) It is hard to differentiate this from number on f infected cells. i wouldn't get hung up on the issue of syncytia, as they are probably only an in vitro observation
miles: ... number OF infected cells...
miles: overall effect on person being infected - arrival of CXCR4 virus coincides with rapid depletion of CD4 t cells - i suspect rapid CD4 depletion causes CXCR4 to dominate (by leading to higher naive cell turnover) and also causes rapid CD4 depletion (by killing naive cells)
amy: What do you think is most promising in the realm of vaccinations?
miles: sorry, getting sloppy with my proof reading on this tiny little box (selects for short comments)
..... and CXCR4 tropic virus also causes rapid CD4 depletion (by killing naive cells)
miles: PROMISING IN THE REALM OF VACCINATION:
Hmm. I think we are quite a way from a vacine in humans, but are getting somewhere in monkeys. In particular making a 'disease attenuating' vaccine (one that can't prevent infection, but can slow disease progression). I have been working on an epidemiological model of this that suggests that such a vaccine may still play a role in slowing the epidemic. there are now several examples of CTL-induign vaccines in monkeys that can do this.
(see Barouch, D, Science 2000, Shiver, J, nature 2002)
kathy: Miles, here is another question from the Message Board, from Shelley (Team Rana)
"I read over the abstract on T cell subdivisions and how the progeny are more prone to die as the number of cell divisions it undergoes increases. I was wondering if there has been any research as to what exactly causes this weakness in the T cell response and if there is any possibility of trying to improve upon this? Is one of the methods currently in place involving increasing the avidity of those T cells that dominate in the primary and secondary response so that may compensate for the death of other T cells as division continues?"
miles: kathy - to pre-empt you asking re epidemiological model ON LINE abstract at
http://www.retroconference.org/2003/Abstract/Abstract.aspx?AbstractID=1338
miles: 1) senescence is not particular to t cells all cells undergo it because of telomere shortening etc. There is some literature that suggests T cells can express telomerase in acute infection and therefore avoid this mechanism of senescence
miles: 2)avidity. We currently don't understand much about how to manipulate avidity - except that low doses of antigen seem to lead to higher avidity, so there may be a trade-off between cell number and cell avidity) - see
Bullock, T. N., D. W. Mullins, et al. (2003). "Antigen density presented by dendritic cells in vivo differentially affects the number and avidity of primary, memory, and recall CD8(+) T "; J Immunol 170(4): 1822-9.
kathy: We have about 5 minutes remaining in the session... does anyone have a final question to propose to Miles?
mary: What types of modeling are appropriate to controlling HIV infection in populations in general, as you alluded to earlier?
miles: I am not sure what you mean by types of modelling. Simple ordinary differential equation models have been used with great success to illustrate behaviours of epidemics (ie: Anderson RM and May RM). more recently 'uncertainty analysis has been used with good effect to try and ask about outcomes (Blower, SM is the major author here).
I think modelling has the ability to make our handwaving assumptions about behaviours of epidemics more rigorous (just as they can for our assumptions about the dynamics of immune cells / virus within the host)
miles: Sorry if I am starting to speak in authors / references. I only arrived in US on Sunday and am still pretty jet lagged. However, if you are interested in modelling, those 3 authors are a good start.
kathy: I think, if there are no further comments, we may wrap up this session, and allow Miles to go have some dinner... Thank you very much, Miles, for a great discussion, and we hope you'll stay in touch with the VSG project ...
mary: Thank you for interacting with us, and also for all the good info & references
miles: Thanks
|
|
