Cpn Theory: Mechanism of Resistance

Cpn Theory: Mechanism of Resistance

Chlamydia Pneumonia is an obligate intracellular pathogen associated with a variety of chronic illnesses. Like many other pathogens, it has evolved mechanisms that allow it to survive in a hostile host environment and incidentally may be directly or indirectly responsible for many chronic illnesses.

Its primary resistance mechanism probably evolved in its attempt to to thwart the host cell's immune response. When challenged with a host's cell mediated response to an intracellular pathogen, production of NO (nitric oxide), Cpn attempts to pump the damaging molecules out. It does this via efflux pumps, small proteins located in the cell wall that transport various substances across the cell membrane . Efflux pumps are metabolically expensive and this quickly depletes its stores of ATP and outruns its ability to scavenge enough host cell ATP or produce its own through anaerobic metabolism.

As ATP is depleted, Cpn convert to the cryptic (persistent) state and drastically reduce its production of proteins. It also begins producing highly inflammatory proteins, chlamydial heat shock protein 60 (HSP-60). HSP-60 induces an inflammatory response in the host which consequently down regulates production of interferon which stimulate the cell mediated response. As cell mediated response and production of HSP-60 is down regulated, Cpn is able to resume replication in its reticulate body (RB) phase. In many cases this process probably continues for an extended period until the host cell dies from apoptosis or necrosis.

The outcome of this interaction is that at any given time a significant number (probably the majority) of C. Pneumonia are in the persistent state. In this state they are not susceptible to antimicrobial agents. The exception is tetracycline class drugs which bind to the 30S subunit of microbial ribosomes. C. Pneumonia is susceptible to these in the persistent state. However tetracyclines are bacteriostatic so Cpn should have an opportunity to develop resistance.

While in the persistent state, Cpn have an opportunity to create efflux pumps to antimicrobial agents that might otherwise be bactericidal. [Anecdotal evidence suggests this process occurs in about three days.] These pumps probably allow it to reduce the concentration of the antimicrobial agents to a sub bactericidal level and in the event that energy is not available to sustain these pumps, Cpn would convert to the persistent state in which is would not be susceptible. So the combination of efflux pumps and persistent state regulated by lack of ATP should allow it to survive and arguably even thrive in in environment containing any number environmental toxins including antimicrobial agents.

- Paul

Hi Lee,

I plan on writing a few more of these including some thoughts on overcoming resistance. I am not sure it will be possible to directly block these ATP driven efflux pumps though. A lot of effort has gone into this in cancer research with little to show for it. I think the problem is these pumps are similar to other mechanisms that are essential to host cell survival.

- Paul


Maybe black pepper would work. Sounds cheap enough and I like black pepper. http://www.ncbi.nlm.nih.gov/pubmed/21051548 An abx has already been developed to include black pepper rifampin and inh. http://www.dancewithshadows.com/pillscribe/indias-new-tb-drug-risorine-…

200mg doxy daily, 500 zithromax mwf,flagyl 1000 m-fri.rifampin 2x daily,chloestryramine 2x daily

Piperine looks promising. I don't know about the calcium channel blockers. It is too vital a process in the body and may have adverse effects on health. Have any of the Cpn researchers looked at natural compounds like piperine to interfere with efflux pumps?


Feeling 98% well-going for 100. Very low test for Cpn. CAP since 8-05 for Cpn/Mycoplasma P.,Lyme, Bartonella, Mold exposure,NAC,BHRT, MethyB12 FIR Sauna. 1-18-11 begin new treatment plan with naturopath

Paul, not to stir the pot, but  are you saying this discovery negates the following? 

Excerpt from here: 

"In conclusion, we have plenty of evidence that the development of resistance in chronic C. pneumoniae infections is likely to be remote provided complementary antibiotic combinations are taken."



JeanneRoz ~ DX'd w/ CPN 4/2007; 6/07 -"officially" dx'd w/CFIDS/FM; also: HHV6, EBV, IBS-C, 100 Doxy:BID; 500 mg Biaxin BID; Tindamax Pulses, B12 shots, ERFA Dessicated Thyroid,Cortef, Iodoral 25 mg, Vit D-6,000 uni

Hi Jeanne,

This is really more about semantics but I essentially agree with that statement. It is indeed very unlikely that Cpn would develop any of the types of resistance associated with genetic mutation and subsequent selection if multiple antimicrobial agents are used. The efflux pump mechanism is different as Cpn probably does not develop this trait. It already has the capability to make efflux pumps, just not the need to until potentially damaging agents are encountered.

- Paul

This efflux pump theory seems to me like castles in the air: guesses on top of conjectures on top of guesses. Okay, it'd be a pretty rare organism that didn't have pumps to pump stuff in and out of it; but that it has pumps that operate on the particular antibiotics we use hasn't been established.

In biology, there are always alternative explanations. In describing some results on HSP60 production from Stratton's lab here, recently, you used efflux pumps as the explanation; but a simpler explanation is that the antibiotics used shut down protein synthesis in Cpn, that HSP60 is a protein, and that thus the antibiotics reduced the quantity of it. The three-day delay might just be due to the time taken for the HSP60 to leak out of the Cpn into the cell, and then out of the cell into the culture medium where it could be measured. Indeed, I think that result argues against efflux pumps: if HSP60 production had first gone down, then increased again, you'd have a basis for saying "okay, the cell at first was inhibited from producing HSP60 by the antibiotics, and then it recovered, probably due to efflux pumps pumping the antibiotics out and letting the cell resume production of HSP60". But that isn't what happened: instead the production went down and stayed down.

As for that Hammerschlag book, the "search within this book" feature of both Google and Amazon yields only two results for the word "efflux", one being the page you pointed to, and the other being a reference to cholesterol efflux, which isn't relevant here. The page you pointed to says nothing relating to clinical or in vitro experience with efflux pumps; the only things it says about efflux pumps relate to the results of genome studies, looking for efflux pump genes in the Cpn DNA. And what it says, to me, is that they didn't find any, despite hard looking. Let's go through the "hits" the paper lists, which I suppose were the best hits their computer found:

ErmA (BAB42746, 243 amino acids (aa)), a 23S rRNA ribosomal methylase, has 24% indentity[sp] and 42% similarity over its first 215aa to the C. pneumoniae dimethyadenosine transferase gene, known as the kasugamycin resistance gene (ksgA, 565 aa).
Those numbers are completely lousy. The one protein is twice the size of the other; the two proteins have only 24% of their amino acid sequence the same, and only 42% even "similar" -- and that is only over a restricted range (albeit, in this case, not very restricted). Now, even a single amino acid change can entirely destroy the functionality of a protein; a protein which is twice the size and of which less than half of the half that overlaps is even "similar" is vanishingly unlikely to have the same function. The DNA sequence may share an evolutionary origin, from back in the deep past, but function? Forget it. And, indeed, in this case the function of both proteins is described, and the two functions are different from each other (and neither is an efflux pump).
Second, the macrolide efflux gene product (MefA, AAL58635, 405 aa) has homology to a C. pneumoniae putative efflux protein (YgeD, NP_224932.1, 565aa), within its N-terminal one-third.
In this "hit", only a third of the one protein has any resemblance to the other -- and other than that the resemblance numbers are, it seems, not worth printing.
Remarkably, the S. aureus NorA multi-drug efflux protein (BAB56857, 388 aa), though not homologous to MefA, also has 23% identity and 43% similarity to YgeD, in a central region of the protein spanning 195aa.
This is comparing the same Cpn protein to a different efflux protein, and finding another "hit", this time over a different restricted range. The numbers for the "hit" are again utterly lousy, and are especially lousy in context: only half of the gene "hits" in the first place. These are the sort of "hits" that one can get just from random chance, when one runs a computer program that does massive data mining, testing for billions or trillions of possible hits.

The only interesting thing here is that the Cpn gene is, in the first place, described as a "putative efflux protein". That might mean that they have some other reason, besides the gene matching, to think it's an efflux protein. Or it might not.

Finally, the virginiamycin resistance gene product (VgaB,U82085, 552 aa), an ATP-binding efflux pump, shares 32% identity and 53% homology over its C-terminal two-thirds with the chlamydial yjjK gene product, a putative ATP-binding transporter.
These are the strongest numbers so far, but they're still not at all strong. Virginiamycin isn't a drug we use, anyway. (Nor do we use the other drugs in its class).
Of note, the macrolide-streptogramin resistance gene products MsrC of E. fecalis (NP_815134) and MsrA of S. epidermidis (P23212) are homologous to both VgaB and YjjK in a similar region.
No numbers here, but by now we know how low the book's standards for "homology" are.

These gene-matching results just don't prove anything; and Hammerschlag doesn't claim they do. Even if she did, I'd be suspicious: the numbers are just too lousy. If biologists had a long tradition of decoding genomes and guessing the function of genes, that'd be something to respect; but as it is, these sorts of computerized gene-matching results, which mine massive databases of gene information, are very new. Biologists know that when testing several hypotheses, one should downgrade the statistical significance of one's results, via the Bonferroni correction (or similar corrections). But these gene searches are a new world: they test millions of hypotheses per second, and it's likely that nobody has figured out how exactly to determine when a gene bank "hit" is truly statistically significant, and when it's just random chance. And as I said, even it's not just random chance, it doesn't mean the two genes have the same function.

Hi Norman,

Margaret identified 3 potential efflux pumps. I agree the homology is inconclusive but the fact remains that what is observed in vitro, a 4-fold increase in MICs requires an explanation. Since the known resistance mechanisms were investigated and did not exist one is left with either a completely novel mechanism of resistance (possible) or efflux pump resistance (probable). The more recent data and observations over many years also support this idea. Anyway the forum topic is "Cpn theory" which I think is more than fair considering the high probability of this in fact being the mechanism of resistance.

- Paul

Looking into this a bit further, it turns out that to get the full story, one has to go to the paper cited by that book, which has considerably more details, and is available for free online:
They didn't merely do a computer search; they also took the genes the search found, and sequenced them out of the isolated samples -- the ones which had shown the 4-fold increase in MIC. (Each of those samples, the paper says, maintained its MIC when cultured in the laboratory.) They found that none of their hypotheses worked: the genes they sequenced just weren't any different between the low-MIC sample and the high-MIC sample. That includes the putative efflux pump genes, the ribosome genes, and the ErmA gene; none of them changed, although the MIC changed. As the authors put it, "Thus, no mechanism was found to account for the shift in the MIC for these isolates." But that doesn't mean that some "completely novel" mechanism is required; they didn't really rule out any of the mechanisms they postulated -- neither the efflux pumps, nor the others. Their report is of the "we checked the tree for low-hanging fruit, and didn't find any" variety, not the "we went over the tree from top to bottom, and can guarantee there is not a single fruit" variety.

I don't have anything against the idea of Cpn having efflux pumps; their existence seems plausible even in the absence of any particular evidence for them. But I have yet to encounter such evidence. And it could easily be that, living a sheltered life inside animal cells, Cpn hasn't been exposed to the sort of microbiological competition that results in the development of efflux pumps for antibiotics.

Hi Norman,

As I understand it there would be no genetic differences between a bacteria that had genetic sequences that were capable of creating these pumps and one that had created them. The genome would be identical. The only differences would be in the cell wall where the efflux proteins that were created in response to a stimulus were located. This process probably takes a day or two which while proving absolutely nothing does track well with observations.

Interestingly, while also proving nothing, Cpn exposed to Zithromax for prolonged periods of time develop an irregular shaped cell wall. My guess is that Cpn has a difficult time pumping out Zithromax as its structure is resistant to these types of mechanisms (which why it has a 68 hour half life). So I am guessing that it does not get pumped out and repeatedly stimulates production of additional pumps until the cell wall has the strange looking irregular shape that has been noted. If you cannot find EM's of some of these I will try and dig them up. They have small dots covering the cell walls. One could almost imagine them having thousands of tiny pumps... ;)

- Paul

The thing is, both the old and the new MICs were "stable upon subculture in antibiotic-free medium", according to the paper. I take that to mean that they repeatedly cultured the bacteria, growing them until they produced EBs, taking those EBs, and adding them to new cells to infect those cells, and that doing that didn't change the MIC. That would rule out the possibility that this is merely a "phenotypic" change, such as the production of efflux pumps. Instead it would mean that there's some genetic change somewhere; it's just not in the places that they looked for it, but rather somewhere else in the genome.

(These days, with sequencing rapidly becoming cheaper, one might sequence the entire genome, to find where it changed. But that paper only sequenced a tiny fraction of it.)

I've seen some pictures of strange-shaped Cpn, but there are lots of explanations for strange shape. The explanations I would turn to first involve structural proteins that help the germ maintain its shape, rather than membrane proteins such as efflux pumps. (It is the membrane that they're located in; most of the sources I've encountered say that chlamydiae don't even have a cell wall, although I suppose there are some structures that one could call by that name.)


That "stable upon subculture in antibiotic-free medium" does sound like they may have produced EB's and infected a fresh culture. I wish they would state that one way or the other. If they actually used EB's that would change things considerably as I do not think EB's use any of the cell membrane. But I could be mistaken about that and some of the cell membrane might condense down into the EB and preserve some proteins located there.

Anyway Margaret did this research and probably has more data available than was published. And from her writing it is clear that she thinks Cpn have efflux pumps that are capable of expelling antimicrobial agents although she was unable to prove it. I do not think we are going to accomplish a lot by continuing to argue it with no new data available.

- Paul

Surely the main point is that antibiotics become less and less effective over time. Whatever the mechanism involved, that is the bottom line (efflux pumps is as good an explanation as any).

Antibiotics didn't cure me. Only made me worse in the end.

Hunter: Don't think - experiment

Paul, that's not at all clear to me: looking at that paper, I can't tell whether she thinks efflux pumps are hugely probable or whether she just thinks they are a remote possibility which was nevertheless worth investigating on grounds of thoroughness, so as to remove a possible criticism of her paper. Of course, if you are referring to stuff she's written elsewhere, that's different; but in that case a citation would be appreciated.

By the way, you don't have to postulate something like efflux pumps remaining in the membrane into the EB stage, and somehow generating more efflux pumps when it comes time for the germ to expand and grow again. Instead, you can jump on the "epigenetics" bandwagon, and suppose that although the genetic sequence remains the same, parts of it are methylated to increase the expression of the efflux pump genes -- and that there's some mechanism to make that methylation persist through cell division events. (It'd naturally be retained in the EB stage; what is harder is to make it be retained through the process of reproduction of RBs, since that involves replication of DNA, which tends to strip off added methyl groups, or at least not add them to the new copy. Or if you don't like the idea of methylation, there are other mechanisms which can change gene expression, and which could persist.)

There's also another explanation of her results, which doesn't involve any genetic mutations or epigenetic changes in the germ. That is that those particular patients were infected with a mixture of two varieties of Cpn, which had different antibiotic sensitivities, and that the first time she sampled, she got the more sensitive one, and the second time, the less sensitive one. Adding to the plausibility of this is that even the less-sensitive strains were within the range of sensitivities she found in other patients.

Hi Norman,

I was not familiar with epigentics thing so I have some homework to do. However I like it because it allows one to explain all of the observations. As you state it might cover the case of reduction down to an EB. In the case of cell division you would not need this explanation as there would be many (perhaps hundreds or thousands) of these proteins in the cell membrane and they would be roughly evenly split among the cells when they divide.

The selection idea is not too far out either. While I would presume they would use EB's in their culture that were from a specific strain they could have been burned by having an existing chlamydia infection in the cell cultures. Years ago Dr. M and Dr. S. found this to be the case in all of the cell lines they ordered and had to pretreat them before infecting them. So it is possible that Margaret was inadvertently selecting for the more resistant strain. This could also explain the treatment observations that have been noted as people may carry multiple strains. I still think efflux pumps are the issue though. And even if this is the case a 4-fold increase in Mics and a tendency to keep Cpn in a persistent state is annoying but hardly insurmountable.

- Paul

In this case they were culturing germs from patients, so they wouldn't have been using EBs from a specific strain; instead they would have been sampling whatever mixture the patients carried. Also, out of forty-some patients, only two of them showed this change in MIC after antibiotic treatment. So even if infections with a mixture of Cpn strains aren't usual, it's not unreasonable to guess that the infection was a mixed infection in those two cases.

With cell division, even with the population of efflux pumps being split between cells, the germ still would need some continuing instructions to make more; otherwise they would be repeatedly diluted until there weren't enough of them. One could suppose a situation where without any efflux pumps, an antibiotic comes in and shuts the cell down so hard that it can't make any more, but that a few efflux pumps slow down the onslaught enough for the germ to have time to make more efflux pumps. That depends on some tricky timing, though; it seems easier to suppose that there's some internal signal the germ has, to make efflux pumps -- which is where epigenetics can come in. (This weblog post seems like a decent introduction to the topic.)

I feel a bit guilty encouraging you in this, since it seems a bit like you have efflux pumps on the brain, and are seeing efflux pumps behind every bush, but hey, if you're going to do it, you might as well do a good job of it.

Hi Norman,

The continuing instructions are a molecule that stimulates production of a new protein. Cpn like all bacteria make lots of these pumps all the time or they could not survive. Metabolic and environmental toxins would otherwise build up and kill the organism. The only question is whether they make ones capable of pumping out antimicrobial agents and we will simply have to agree to disagree on that question.

- Paul


Oh, well. Many people with progressive MS who take long-term antibiotic treatment seem to stop their progression. This against the natural history of the disease at this phase. I speak from some clinical experience. Cellular processes are still almost unknown; cellular interactions are almost completely obscure. It's a Byzantine puzzle: the more you promote your view the more you find yourself in uncharted territory. For myself I enjoy playing Voltaire's innocent Candide. I also like having a living wife.
D W - [Myalgia and hypertension (typically 155/95.) Began (2003) taking doxycycline and macrolide and later adding metronidazole. No medication now. Morning BP typically 110/75]

David, how wonderful to see your post - hope you and Sarah are doing splendidly!  You are missed here.

Your post is coincidentally humorous (no "u" here), in that we just a few days ago named our beautiful young Sorcerer kid, A.I. from our famous buck of 25 years ago, Candide.  He was to go to Canada, but no A.I.s "took" this year so he will live here. 

I read Candide fifty years ago, but not since.  I loved it, and should read it again.


3/9 Symptoms returning. Began 5 abx protocol 5/9 Rifampin 600, Amox 1000, Doxy 200, MWF Azith 250, flagyl 1000 daily. Began Sept 04 PPMS EDSS 6.7 Now good days EDSS 1 Mind, like parachute, work only when open. Charlie Chan  In for the duration.&am

Even better, see Leonard Bernstein's musical. Fabulous music.

PPMS-misdiagnosed 2001-diagnosed 2006. Probably caught cpn in birth canal but it didn't pass BBB until my 40s. Minocycline 7 mos.- resulting bronchitis 5 months.Go to private m.d. out-of-plan. Wheldon CAP 3/2/07 Stopped 12/12; resumed 12/13

I am glad to see you, David. I always appreciate your clearly written posts.

Stratton/Wheldon protocol 02/2006 - 10/11 for CFS and many problems 30 years