Published online before print May 18, 2007
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Laboratory of Allergic Diseases, National Institute of Allergy and Infectious Disease, National Institutes of Health, Bethesda, Maryland, USA
1 Correspondence: hrosenberg@niaid.nih.gov
The manuscript, "Endogenous pathway to SIRS, sepsis, and related conditions" was selected as a Pivotal Advance because it features an original hypothesis regarding the role of TLRs, notably TLR4, in maintaining homeostasis and in modulating responses to endogenous stimuli. Specifically, the authors present data and a novel mechanism to explain the way in which TLR4 interacts with the ubiquitous endogenous ligand, heparan sulfate.
Dr. Platt, much of the clinical focus of your work is on transplantation. How did this interest lead to your work in Toll-like receptor biology?
JLP: Our work on TLRs originated with a question about the pathogenesis of transplant rejections. More than 15 years ago, I undertook to find out how vascular pathology, which characterizes how all forms of grafts are rejected, might arise. Wisely or not, I questioned initially whether changes in the metabolism of heparan sulfate proteoglycan (HSPG) might be an underlying factor in the pathophysiology of vascular disease [1 ]. HSPG was then, and still is, thought to be important for controlling coagulation and complement activation, to be a crucial component of the endothelial barrier, and to be an essential anti-oxidant, all of which might undergo dysregulation during the genesis and manifestation of vascular disease. Interestingly, I discovered that nearly every immune and inflammatory process results in shedding of HS. This observation led my graduate student, Lucy Wrenshall, and I to question whether the HS shed in this way might be biologically active, particularly in the immune response to transplantation. My student found that soluble HS activates macrophages in much the same way as does LPS [2 ] and thus we speculated that such an action of HS might explain how immunity, rather than tolerance, occurs after transplantation. This observation also led to us wonder exactly how HS activates antigen presentation. Ultimately, this question was answered by another student, Geoff Johnson, and a fellow, Yuzo Kodaira, who found that HS activates TLR4 [3 ]. At the time that we made this observation (around 2000) no one wanted to consider the possibility that an endogenous substance might activate TLRs, as these receptors were thought to recognize microbial polymers pretty much exclusively. Of course, our finding that an endogenous substance could activate TLR4 led us to modify the simple, canonical model of an agonist–receptor interaction.
The last point that you bring up leads to one of your more intriguing findings, the demonstration that proteolytic activity, generated by the addition of the protease, elastase, is required for TLR4 activation, and its ability to respond to ligands. Is this consistent or inconsistent with many previous studies, in which TLR4 responses are measured in the absence of protease?
JLP: I don’t think it’s inconsistent, probably just something that has been missed. The prototypical TLR4 agonist, LPS, activates complement, which includes proteolytic enzymes that may contribute to TLR4 activation. Consistent with this concept, purified lipid A, the component of LPS that activates TLR4, which should have greater agonist activity per unit weight than LPS, actually has much less activity—indeed, by orders of magnitude. Besides explaining how the activity of TLR4 agonists is amplified, the role of activation by proteases also explains how TLR4 can respond to an endogenous agonist some of the time, but not all the time, or why, as I have noted in my lectures, that I am still standing at the podium, and not lying on the floor owing to the constant presence in the same place and time of TLR4 and endogenous agonists. The canonical model of TLR activation requires an element of constitutive control, and TLRs must have some regulatory mechanisms that prevent ongoing signaling in response to ever-present endogenous ligands. One such regulatory mechanism is the requirement for proteolytic activation, as we have shown.
One feature that has not been discussed directly so far—HSPGs are well-known as having great structural diversity. Has the question of specificity been addressed? In other words, is it clear that TLR4 responds to all forms of HSPG, or perhaps only to specific structural variants?
JLP: We have thought rather a lot about this question, and our manuscript describes the extent to which we have explored this to date. HSPGs have been characterized as having greater structural diversity than nucleic acids. We know that TLR4 is responding to the carbohydrate elements of heparan sulfate proteoglycan and not to the protein core, as shown by our finding that heparanase, which destroys the carbohydrate, completely abrogates signaling. However, there are two important questions that need to be answered, specifically— 1) how does this carbohydrate become available to TLR4; and 2) what is specific about the interaction? These are issues to be addressed in the near future.
A chicken and egg question—which came first? Did TLRs begin as molecules designed to recognize endogenous products of inflammation, and then generate pathogen recognition capabilities, or vice versa?
JLP: I wish we had the means to answer this question but of course we do not. I can offer speculation, which we have in abundance. We suspect that TLRs originated as one receptor that functioned to recognize breakdown products of tissues. In our recent manuscript [4 ], we discussed recognition of tissue breakdown products by receptors as one of the central themes in the biology of plants lower animals and mammals. This mechanism might have arisen to regulate development of multi-cellular organisms and may be related to a concept that we have found (as did others) that degradation of proteoglycans is essential for organogenesis [5 ]. Consistent with this concept is the use of Toll and its receptors in the development of Drosophila. On the other hand, recognition of breakdown products might have been a means for the recognition of infection and indeed Toll is now known to sense infection in Drosophila as well. We cannot (yet) go back in time to determine which of these (or some other) functions of Toll and TLRs arose first. We can, however, learn from the broad representation of these mechanisms in biology. Overall, our basic and clinical observations suggest to me that TLRs evolved to respond to endogenous more than exogenous signals. This concept may be usefully applied to our understanding of how TLRs maintain the health of tissues and how abnormal TLR function leads to a range of diseases, many non-infectious [6 , 7 ].
As a final note, can you tell the readers of JLB a bit about what led to you to a career in research from a more clinical medicine and surgical background? What were some of your specific influences and directions?
JLP: I was always interested in research but, having attended a clinically-focused medical school and having undergone training in the more clinical field of pediatrics, I did not really know what research was until I was introduced to it by my mentor, Dr. Alfred Michael, during my postdoctoral fellowship in pediatric nephrology at the University of Minnesota. Al, as did other researchers in the 1970s and 1980s, thought many, if not most, diseases could be explained in one way or another by the actions of the immune system. However, it became clear to us that in order to understand the role of the immune system in disease pathology, one might not find it particularly advantageous to characterize the complex manifestations, as they exist in already established diseases. Instead, he encouraged me to study the rejection of transplants since, in this context, one could know when, how, and exactly by what the immune system was stimulated. Al taught me that science should not be limited by conventional fields, and he encouraged me to pursue many avenues of investigation, to make many errors, and to correct at least a few of them. This freedom made my entry into what might be considered surgical research a natural, if unconventional, pathway.
 View larger version (130K): [in a new window] |
Figure 1. Dr. Platt received his M.D. degree from the University of Southern California in 1977, and trained in pediatrics at the Children’s Hospital of Los Angeles. He was a Fellow in Pediatric Nephrology and was promoted to Assistant, then Associate Professor of Pediatrics and Associate Professor of Cell Biology and Neuroanatomy at the University of Minnesota. He joined the faculty at Duke University as the J. W. and D. W. Beard Professor of Surgery in 1992. In 1998 he moved to the Mayo Clinic, Rochester, where he is currently Professor of Surgery, Immunology, and Pediatrics, Head of the Transplantation Biology Program, and Vice Chair for Research in the Department of Surgery. Dr. Platt is member of numerous organizations including the Association of American Physicians and the Institute of Medicine.
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||
