Remember Dr. Sin Hang Lee?
If you don’t remember Lee, maybe you remember a while back, when the antivaccine group SaneVax touted findings that it claimed were devastating to Gardasil. Specifically, they claimed that there was vaccine-derived human papilloma virus DNA in Gardasi. Ring any bells yet? And it turns out that the guy who made this apparently horrific “discovery” was—you guessed it!—Dr. Sin Hang Lee. Back in 2011, Lee, apparently either funded by or working with SaneVax, “discovered” that there was DNA in his Gardasil. As I explained at the time, there was a lot less to this claim that meets the eye. Basically, Lee used a very sensitive nested PCR assay to amplify and detect what he claimed to be vaccine-derived sequences from the human HPV strains used to make the vaccine. Specifically, his nested system, as I explained at the time, can radically ramp up the sensitivity of the PCR assay. Of course, it can also radically increase the chances of amplifying a nonspecific strand of DNA.
Given that, it wasn’t too surprising that there might have been minuscule quantities of the recombinant plasmid used to make the protein antigens that go into the vaccine. At least, it wasn’t too surprising to me or anyone with a modicum of knowledge about how sensitive PCR is and how easy it is to get false positives. Even if Lee did everything right and actually did detect a bit of recombinant DNA from the HPV DNA used to make the vaccine. Does this matter? My answer, of course, was no, as was the answer of anyone who knew anything about vaccines and biologics. One factor to consider is how much DNA was present, which was almost certainly very, very little, given that it took nested PCR to detect it.
In fact, even if Dr. Lee’s analysis was completely correct correct and his new, allegedly more sensitive methodology had actually picked up previously undetected traces of HPV DNA from the plasmids used to make the HPV vaccine, it is, as I described before on multiple occasions, incredibly unlikely that such tiny amounts of DNA could cause problems because, as I explained, it’s incredibly difficult to get naked DNA into cells and making the proteins it normally makes, and, even if Dr. Lee were 100% correct about there being undetected HPV DNA in Gardasil, the quantities involved are many orders of magnitude less than what would be needed to have even a whiff of a wisp of a hope of the DNA getting into cells and making its protein. That’s even assuming it could pass the blood-brain barrier or that the DNA fragments were large enough to contain whole coding regions of genes with a proper promoter in front of them to drive their expression. I mean, it’s not as though whole plasmids are likely to have survived the vaccine manufacturing process, and that’s not counting the fact that Gardasil uses a yeast expression system while the other HPV vaccine Cervarix uses a baculovirus (insect) expression system, neither of which would be likely to drive significant expression in human cells.
So what we had was a fear mongering campaign derived from a nonexistent understanding of molecular biology. This campaign reached utterly ridiculous levels when the not-so-dynamic duo of HPV vaccine quackery, Sin Hang Lee and Christopher Shaw (who seems dedicated to the idea that somehow HPV DNA bound to the aluminum adjuvant in the vaccine is deadly) descended like ghouls on the corpses of two young women whose untimely deaths antivaccinationists have been trying to blame on Gardasil ever since they happened. This campaign reached a ridiculous extreme when they actually testified in an inquest in New Zealand into one of these two deaths, a young woman named Jasmine Renata, and tried, in essence, to claim that the poor woman’s body was riddled with HPV and that that HPV caused her death. It was an utterly despicable performance that Shaw, at least, followed up with an article in which he attempted to demonstrate that, in effect, Gardasil killed a young woman. It was not the least bit convincing.
And now Dr. Lee has taken his crack by publishing his paper related to the death of this same young woman in a paper published in an open-access journal I’ve never heard of entitled Detection of human papillomavirus L1 gene DNA fragments in postmortem blood and spleen after Gardasil® vaccination—A case report. Like Shaw’s paper, it’s a massive load of fetid dingo’s kidneys, and it won’t take that long to explain why.
First off, as I’ve always said before, you can tell what you’re in for in a paper by its introduction, and Lee’s introduction is a doozy. It’s full of anti-HPV tropes that would be more at home on the antivaccine propaganda blog Age of Autism (or SaneVax, for that matter). He does, however, inadvertently reveal what I’ve always suspected to be true about this case:
The parents of a formerly healthy New Zealand young woman who suffered a sudden unexpected death in sleep 6 months after Gardasil® vaccination requested testing for the presence of HPV L1 gene DNA in the post-mortem samples of their deceased daughter collected at the time of autopsy. Some of the consultants to the parents suggested that if residual HPV L1 gene DNA which is known to be present in the Gardasil® vaccine [8,9] were present in the postmortem samples, there might be a potential link between the residual HPV DNA and the un- explained death of their daughter. This paper reports the experience in developing a method for the detection and validation of minute quantities of HPV-16 L1 gene DNA in the postmortem blood and spleen obtained at autopsy. The data reported in this paper were extracted from a full report which was submitted to the Wellington coronial court at a public inquest held on August 8-9, 2012.
Who were these “consultants”? One wonders. Were they perhaps Sin Hang Lee and Christopher Shaw, aided and abetted by SaneVax? One wonders, one does. In any case, I wrote about the ridiculousness of letting Shaw and Lee testify at this inquest. One hopes that the committee listened politely and then completely ignored the pseudoscience as embodied in this paper.
So let’s take a look at what Lee did (or claims to have done). He took DNA isolated from these tissue samples and subjected him to his own new super-duper, super special, super sensitive nested PCR, looking for a 190 base pair sequence from the HPV L1 gene. As you might expect, given Lee’s background, there were…problems with his methodology. Rather than critiquing the exact primers he chose and how he went about doing his PCR, let’s step back and look at the experimental design from a bird’s eye view. Basically, he tested DNA from tissue samples from one young woman. There are no controls. How many people in the general population would test positive using Sin Hang Lee’s methodology? We don’t know because he hasn’t made an attempt to find negative controls, and without negative controls we don’t know that every tissue sample would test positive when subjected to his methodology. In fact, the only negative controls I saw mentioned anywhere were negative water controls. That’s perfectly fine to rule out nonspecific amplification that doesn’t depend on DNA (such as artifacts like primer-dimer, but it doesn’t tell you anything other than that.
Another issue is that troubles me is the way that Lee uses degenerate oligonucleotides. It’s worth going back to images I’ve sued before to illustrate nested PCR:
And here’s nested PCR:
Nested PCR can be very, very sensitive, even more sensitive than “simple” PCR, depending upon the number of amplification cycles used in each PCR step. It’s that sensitivity that allows nested PCR to amplify very tiny amounts of target sequence. Now, Lee used a combination of degenerate primers and non-degenerate primers. A degenerate primer is a primer in which some of the positions in the sequence contain more than one base; i.e., there is a mixture of primers with different nucleotides at that place. The reason this is done, generally, is to amplify sequences for which we know there are variations in the sequence. Alternatively, it is done when trying to amplify sequences based on the protein sequence. Because of the degeneracy of the genetic code, most proteins can be coded for by more than one codon of three nucleotides. Usually, the variable base is the third base in the codon, but not always. To account for those nucleotides, degenerate primers are sometimes used as a way of putting a “wildcard” in the positions that can have more than one base in nature. However, such primers have a downside. The more “wild cards” a researcher puts in his primer sequences, the larger the number of potential sites to which those primers can bind. What one gains in sensitivity for potential coding sequences to be detected, one loses in specificity. It’s a tricky balancing act that is not as straightforward as those without much experience in PCR realize.
That’s why in general we avoid using degenerate primers except for this sort of purpose: Trying to isolate a sequence where we know that certain positions can have different bases. If the target sequence that a researcher is trying to amplify by PCR is known (and, make no mistake, the HPV-16 L1 gene sequence used to make HPV vaccine is known), it’s usually a bad idea to use degenerate oligonucleotides, because doing so will decrease specificity and greatly increase the chance of amplifying what I like to call crap. Basically, I can’t figure out why Lee would use the method he’s using. Arguing that there is variability in the natural HPV-16 L1 sequence and he wants to pick that up won’t wash. In fact, he is aiming to detect the vaccine strain sequence; so detecting any natural HPV that might have come from warts or other HPV infections would actually be counterproductive to what he’s trying to accomplish: To detect the vaccine HPV-16 L1 sequence postmortem in Jasmine Renata. It’s almost as though he’s intentionally trying to muddy his findings. Maybe more than almost.
In any case, the same-nested procedure used by Lee, as far as I can tell, involved using degenerate primers for the first round of PCR, and then selecting one set of specific primers that make up the degenerate primer mixture and then using them to repeat the amplification. The idea is to start out less specific and then get more specific by removing the degenerate primers in the second round of PCR. The problem is, of course, garbage in, garbage out. The other problem is, as I said before, we don’t have any negative controls from tissue samples from people who were not vaccinated with Gardasil. The reason this is important is inadvertently described in Lee’s discussion:
Since the human genomic samples contain numerous DNA fragments which are substan- tially complementary to the base sequences of the HPV PCR primers, co-amplification of non-target DNAs of the human genome invariably occurs in the same-nested PCR settings when PCR amplicons are re-amplified with the same primer(s).
Of course, Lee did sequence several of his products. The first sequence was clearly not a pure sequence, but was contaminated with additional sequences, which prevented identification and validation of the PCR product. A couple of the PCR products that Lee sequenced (Figures 6 and 7) were in fact genomic DNA, and it took a lot of fiddling with different primers and conditions for Lee to get fragments that sequenced as HPV 190 base fragments. This suggests to me findings that aren’t robust, which suggests to me that it’s more likely that he’s amplifying contamination than anything else, which, given the multiple rounds of PCR would be very easy to have happen if even a single prep of the plasmid containing HPV L1 were done in the same lab as the PCR of the DNA from tissue samples. That’s yet another reason why controls from tissue samples derived from people who were not vaccinated would be important.
Here’s another thing that would be important. Lee claims that vaccine-derived HPV L1 DNA is somehow hanging around in the body and that it, in essence, might have killed Jasmine Renata. It’s not as if the sequence of the plasmid that is used to make the protein antigen used to make the vaccine isn’t known. Lee only looks at HPV L1 sequence that is inserted in the plasmid. If fragments of the original plasmid used to make the vaccine were in fact still in the vaccine and somehow magically did continue to hang around in the body after vaccination at concentrations detectable by PCR, then it should more than just L1 there. There should be random fragments derived from the whole plasmid, not just the L1 insert. There should be readily identifiable plasmid and promoter sequences. In fact, the sequence I’d look for is a sequence containing overlapping the promoter used in the plasmid. That way, you’d detect HPV L1 sequence attached to the specific yeast promoter used in the manufacturing process connected to known plasmid sequence. It’s the obvious thing to do, because, if that sequence were found, it would be very hard to explain any other way than coming from the plasmid used to make the vaccine. It would even be hard to explain by plasmid contamination because the plasmid containing the HPV-16 virus that Lee bought from ATCC to use as his positive control for PCR reactions because that virus is in Bluescript, which is an E. coli plasmid.
Lee didn’t choose to do that. One wonders why. He even acknowledges that he knows about this issue:
The presence of HPV-16 L1 gene DNA fragments of a vaccine origin indicates possible co-existence of other companion microbial DNA, such as DNA fragments of the plasmid pGAL110 and yeast cells which are used in the vaccine production by the manufacturer . A poten- tial consequence of these viral and microbial DNA frag-ments with their unmethylated CpG motifs in macro-phages [41-46] is to cause release of various cytokines, including tumor necrosis factor (TNF), a recognized myocardial depressant [47-51]. TNF-induced hypoten- sive shock is a documented observation among animals [52,53] and humans [54,55]. To answer the question whether the quantity of these persistent viral or microbial DNA fragments can stimulate the macrophages to release enough TNF to generate a significant pathophysiological impact following Gardasil® vaccination needs expanded research.
Uh, no. Lee should have done the research right in the first place if he wanted to convince anyone by actually looking for the sequences I described above, specifically an amplicon (the DNA target sequence to be amplified by PCR) that encompasses part of the L1 gene, the promoter region used, and pGAL110 sequence that is directly attached to the insert. He could also look for yeast genomic DNA sequences, as he himself suggested. The combination of finding HPV-16 L1 sequences from the plasmid used to make the HPV vaccine plus yeast genomic DNA would be very supportive of his hypothesis. Instead, Lee claims to have amplified L1 DNA fragments “of vaccine origin.” Yet he hasn’t proven that they are of vaccine origin. To do that, he would actually have to amplify some pGAL110 plasmid sequence as well as L1. Then at the end he concocts a handwaving explanation to justify why he had so much trouble amplifying L1 from Jasmine Renata’s tissues when he can amplify HPV L1 DNA from the blood of women infected with HPV-16 in other studies, in which the HPV DNA isn’t in the normal B conformation or is somehow stabilized by binding to aluminum adjuvants. It’s utter nonsense.
Much like everything I’ve seen published by Lee on this topic. No wonder SaneVax and Age of Autism love it so. It’s kind of sad, given that Lee clearly has some skill at PCR, that he chooses to use it for such a silly application in the service of antivaccine fear monger. As a physician with these mad PCR skillz, he really should know better. He should even know just how tiny the amount of HPV-16 L1 DNA there could possibly be in the vaccine, given that he could only detect it with a very sensitive nested PCR test and that that amount is so tiny that it is incredibly unlikely to hang around in the body for more than six months and be detectable in the tissues postmortem, much less cause harm.
Yet Lee chooses not to know.