Re: Re: Re: A modest proposal 5 June 2003
Previous Rapid Response Next Rapid Response Top
Brian T Foley,
HIV Researcher
Los Alamos National Lab, Los Alamos NM 87545

Send response to journal:
Re: Re: Re: Re: A modest proposal

Jamie Mills asks:
"Did either Bess et al. or Gluschankof et al. approach 80% or even 98% purified virus, or what looks like purified virus?"

And the answer is YES. In the Bess paper, the HIV-1 isolate MN, clone 4 virus grown of CEM-SS cells produced a preparation that was more than 99% pure. They discuss why there are differences in purity depending on which cell line and which virus is used.

Gluschankof P, Mondor I, Gelderblom HR, Sattentau QJ.
Cell membrane vesicles are a major contaminant of gradient-enriched human immunodeficiency virus type-1 preparations.
Virology. 1997 Mar 31;230(1):125-33.
PMID: 9126268

Bess JW Jr, Gorelick RJ, Bosche WJ, Henderson LE, Arthur LO.
Microvesicles are a source of contaminating cellular proteins found in purified HIV-1 preparations.
Virology. 1997 Mar 31;230(1):134-44.
PMID: 9126269

"...how is that proof of an exogenous and infective retrovirus?"

No single experiment is "proof" of anything, really. There are whole textbooks written about virology and retroviruses. I can't spell it all out here. Endogenous retorviruses do exist, they are found in the genomes of nearly all eukaryotes. If the virus in not found in the germ-line DNA of an organism, it is not defined as endogenous, it is exogenous. No complex retrovirus (such as the lentiviruses and the T-cell leukemia viruses) which carry regulatory genes (tat, rev, nef, vif, vpu, vpx etc) in addition to the gag pol and env retroviral genes, have ever been found in an endogenous state in any host. A few endogenous viruses have one or two potenital open reading frames in addition to gag, pol and env, but none are truly complex. The closest relative to lentiviruses found in the human genome are endogenous retroviruses such as HERV-K, and these retroviral elements are very clearly not HIV-1 or any other lentivirus.
As for infectious, the global pandemic of infections with many different subtypes and circulating recombinant forms derived from the HIV-1 M group of viruses pretty much answers that question. Whether one studies infectious molecular clones of HIV-1, or transmission chains such as the case of Nushawn Williams who infected 13 young women, or local or country-wide epidemics, the answer is always consistent.

"... our local laboratories use proteins isolated from sucrose gradients,(as mentioned in accompanying leaflets), ..."

How were those proteins grown before the sucrose gradient prpearation? Was it clone whole virus grown on human cells? Or was it just viral genes grown in E. coli? Do you observe a large false-positive rate with the tests you are using? Just because sucrose gradients of HIV-1 whole virus particles are contaminated with varrying levels of cellular vessicles when the virus is grown on human cell lines, does not mean that all sucrose gradients are useless for all purposes.

"...If we then move on to western blots as a means of confirming a diagnosis, how many bands and which do you need to confirm a diagnosis? This varies considerably around the world and means a person may be HIV positive in Africa but hop on a plane and have the same test in Australia and be HIV negative! What does that mean for our patients? ..."

No test of tests for any infection or any other medical condition (such as pregnancy for an example) is always 100% perfect. What this means for any patient, is that some human intelligence is required in interpretting results. There are dozens of possible logical reasons for either a false negative (very recent infection is the major one) or false positive (the patient has elevated levels of some antibody or antibodies that bind to HIV proteins) HIV ELISA test. This is why a single ELISA is not the recommended standard for diagnosis. A person truly cannot be HIV positive in one country and HIV negative in another. They can be "officially" diagnosed as such, but they are either infected or not regardless of that diagnosis. If a gets a positive pregnancy test result with one test and negative with another, it doesn't mean she is "partially pregnant", and the same principles hold true for diagnosing HIV infection or any other infection. More than 99% of people who are infected with HIV-1 develop antibodies to many different HIV-1 proteins over time, such that some bands on the western blot typically appear first, followed by others. Also, very late in AIDS progression antibodies to some ands on the western blot are prone to decline. The end result is that we should expect a temporal variance in the number of reactive bands within any one infected individual. It should also be noted that "reactivity" to a protein band on a western blot is not a binary condition (wither "yes" or "no") but an analog condition varying from weak to strong.

The end result is that both patients and doctors need to use other data besides the ELISA and WB tests and human intelligence to interpret that data. If a person has never had sex, never had a transfusion, never injected drugs and has a weakly positive ELISA and two weakly positive western blot bands, the sensible thing to do is to wait a few months and re-test to see if this was just a spurious problem. If the results are identical months later, then it might be worth finding out what is causing them (perhaps infection with HTLV, which is one of the closest relatives of the lentiviruses). On the other hand if reactivity has increased to a strong ELISA positive plus 3 or 4 strongly reactive western blot bands, there is little (but still some) doubt that the person is truly infected, and further tests should be done to find out for sure, and if so, to find out how this person could have become infected (perhaps they were not fully honest about sex, or IV drug use, or perhaps they will be the clue that uncovers a new route of transmission as Kimberly Bergalis was in the famous "Florida Dentist" case).

Competing interests:   None declared