Seven or eight years ago, I was a SICU attending at the Hospital of the University of Pennsylvania by the faculty there. But eight years ago, I took a position as a full-time bench, slash, animal, slash, translational researcher at the Feinstein Institute for Medical Research in New York. I've been asked to talk today to address the question, what do steroids do to the critically ill? And the answer is, of course, it's complicated. Now, I'm making the assumption that you will, that the other lecturers will provide all the information you could possibly want about the use of steroids in lung injury and in sepsis. I'm gonna function, I'm gonna focus on something slightly different by asking a slightly different question, which is, why do we give steroids to critically ill patients? And an even better question, in my humble opinion, which is, nevermind what do steroids do to the critically ill, what do they do, period? The caveat here is that when we say steroids, we mean glucocorticoids. We're not talking about aminoralocorticoids or androgens or anything else. The focus here is on the synthetic glucocorticoids that we so frequently give to patients and what we'll, from here on in, call BCs to some. So, to go to the questions I said I was gonna address, why do we give glucocorticoids to critically ill patients? Well, there are a couple of reasons. In particular, we give them from immunosuppression, we give them to enhance cardiovascular activity, and we give them to limit pulmonary fibrosis, which I'm not really gonna touch on too much here. In terms of the immune effects of glucocorticoids, they are primarily anti-inflammatory, and certainly that's the reason that we give them. So, for example, they block activity and function in natural killer cells and neutrophils. They inhibit cytokine release in macrophages and in T cells as well. In T cells, in the adaptive immune system, they inhibit IL-10 release, they promote the Th2 phenotype and block the Th1, or pro-inflammatory phenotype in T cells. But they also have, glucocorticoids also have some pro-inflammatory effects. So, for example, they increase adhesion and phagocytosis, and they also cause mobilization of white blood cells from the bone marrow in the spleen. And as we have been told more and more in the previous couple of decades, immunosuppression in the setting of a critically ill patient may not be a very good thing. In terms of the cardiovascular effects, glucocorticoids, they enhance contractility in both cardiomyocytes and endothelial cells by enhancing functions of other mediators, particularly catecholamines. So the result is an increase in cardiac output and blood pressure, they decrease capillary leak, and by effects on the kidney, they cause salt and water retention. Now, that's good news in some aspects of critical illness because it improves cardiac performance and will increase blood pressure in a hypotensive patient, but they reduce vasogenic edema. The bad news is that it increases cardiac work, it'll reduce substrate delivery, increase hydrostatic edema, and the net result can be negative. There are also some good reasons why we shouldn't give glucocorticoids to critically ill patients, in particular, their effects on metabolism. So glucocorticoids enhance hepatic gluconeogenesis, they enhance protein degradation and decrease glucose uptake, glycogen synthesis, and protein synthesis in skeletal muscle. They increase glucagon activity and secretion by the pancreas and reduce insulin activity. They enhance lipolysis, the net result being an insulin-resistant hyperglycemic state that's not unlike the metabolic syndrome that plagues so many older American adults. There are also effects in skeletal muscle. Clearly, loss of skeletal muscle is not a great thing, critically ill, and there are bone effects, leading to a loss of bone matrix, also not a particularly great thing in a critically ill patient. Now, to get a better handle on this, let's talk about how glucocorticoids work. Glucocorticoids are lipids, they can traverse the cell membrane unimpeded. Once they enter the cytosol, they combine with a protein complex that includes the glucocorticoid receptor and a bunch of other stuff. The glucocorticoid receptor is then phosphorylated. The other stuff is jettisoned, leaving just the glucocorticoid receptor with glucocorticoid bound. Once the combination of the glucocorticoid receptor and the bound glucocorticoid traverse the nuclear membrane, and we'll talk about how that happens a bit later, they can do a number of different things that will affect transcription. They can directly bind to DNA via a glucocorticoid regulatory element, a GRD, and this can lead to both enhancement and repression of transcription. They can bind, they can tether themselves to another transcription factor, again, both enhancing or repressing transcription, and they can bind to a glucocorticoid element and to another transcription factor, again, either enhancing or repressing transcription. Now, the non-classical pathway is one that occurs in the cytoplasm, and in this case, the glucocorticoid complex either alone or in complex with a series of proteins can stimulate intracellular pathways that activate cellular activities, in particular, for example, the pro-inflammatory effects of the MAP kinases. Finally, this glucocorticoid complex can bind to mitochondria. So this complex can both enhance and reduce the activity of a bunch of different pro-inflammatory pathways, for example, NF-kappa-B, P38, junk, and ERK. One more point to make about these glucocorticoid complexes is that most of them are dimers. So as you can see in both the direct and composite binding for both enhancement and repression of transcription, a dimer is the functional complex. However, in the cytoplasm and for repression and tethering, the monomer can be functional. So to recap, we have a classical pathway, transcription, a direct or tethered or composite binding to enhance gene expression, and direct tethered, composite, and monomer binding to repress gene expression. There's a non-classical pathway in the cytoplasm which can both activate and suppress pro-inflammatory protein pathways. And enhancement activation versus repression and suppression is a nice explanation for both the variability in glucocorticoid effects and for changes in glucocorticoid effects under both baseline conditions following normal perturbations and in the critically ill. But there's still a lot of questions. For example, under normal circumstances, how come glucocorticoids decrease functions in white cells but increase function in cardiomyocytes and thalassites? Same thing, same compound, why does it do something different? And how does an individual cell toggle back and forth between enhancement activation, repression, and suppression? And how about in critical illness, and particularly in sepsis, how does this play out in the clinical trial? And particularly in sepsis, how does this process become dysregulated given that a dysregulated host response is one of the defining characteristics of sepsis? And finally, how come in one of the early slides, I wrote GCs bind to a protein complex that includes the glucocorticoid receptor and a bunch of other stuff that gets jettisoned? Hint, hint, it's time to talk about glucocorticoid receptors. Now glucocorticoid receptors are really, well, I think they're really interesting, but I'm a basic science nerd, and I think they're really interesting, and it's my job to convince you that they might not only be interesting, but maybe even clinically important. So let's start by talking about the structure. The glucocorticoid receptor consists of an N-terminal transactivating domain, a DNA binding domain, a hinge region, and a ligand binding domain. Importantly, activation functional domains exist in both the NTD and the LBD. These are the sites where glucocorticoid binds to glucocorticoid receptor. And the nuclear localization region, that is the region that's responsible for allowing this protein slash glucocorticoid complex to get into the nucleus through a membrane via receptors and pores, exists in combination in both the DBD and the LBD. This is a depiction of the glucocorticoid receptor with a little more three-dimensional structure. You can see that the glucocorticoid can bind to the two activation function domains. And you can also see the ligand binding domain, which is where, as previous slides showed, Hsp90 holding the rest of this molecular chaperone complex in place binds as well. Looking at this three-dimensionally, you can see again that the ATF1, ATF2 bound together by the glucocorticoid itself can alter the confirmation of this complex and allow a couple of things to happen. Once the glucocorticoid-glucocorticoid receptor complex has entered the nucleus, a couple of things can happen. The first is that there can be binding to the glucocorticoid regulatory element via the DNA binding domain. There's dimerization in the ligand binding domains and the ligand binding domains is also a place where co-regulators can bind leading to gene transcription. Well, it turns out that most of the effects of glucocorticoid binding don't actually occur on transcription at the site where this complex gets bound to DNA. Indeed, they tend to be fairly far downstream most of the times. Now, they can be as close as 100 base pairs away or as far away as 100 kilobases, 100,000 base pairs. In general, what is believed is that these co-regulators, once bound, cause the DNA strand to form a hairpin, to form a loop. And it is by looping and bringing the GRE into close contact with the glucocorticoid regulated gene that transcription is actually affected. Importantly, depending on the nature of the co-activator, the co-regulators, you can get activation or you can get repression. Okay, now let's talk about the molecular biology of glucocorticoid receptor. So glucocorticoid receptors are encoded by a single gene locus. It's on chromosome five in humans, chromosome 18 in mice. The initial transcription of this gene locus creates a primary RNA transcript, which consists of introns, which are cleaved and eliminated, and exons, which are the numbered areas, which are spliced back together again to form the RNAs that are actually used to encode proteins. Now, it turns out that there are a number of different splice alternatives, different ways to put segments back together again when the primary RNA transcript for the glucocorticoid receptors is spliced. And this leads to what we call splicing alternatives or splicing isoforms, GR-alpha, GR-beta, GR-gamma, GR-A, and GR-P. Of particular importance to us in this discussion is GR-alpha and GR-beta. The differences between the two line primarily in this region with the differential cleaving sites indicated here, and we'll come back to that in a minute. Well, it turns out also that in the GR-alpha mRNA, you can start protein synthesis, you can start translation at different sites. And as a result, there are a whole series of different GR-alpha isoforms that can function as glucocorticoid receptors, GR-alpha A, B, C1-3, D1-3. The GR-alpha isoforms are present in different concentrations in different cells. I'll show an example of that in a minute. And it may well be that these differential concentrations, both within cells and between different cell types, may account for specific responses in different cells, why one thing is a repressor and one thing is an activator, and may also account for the reason that one cell is activated by glucocorticoids and another cell is turned off by glucocorticoids. Coming back to the GR-alpha and GR-beta transcripts and primary transcripts, as I said, the differences lie primarily in the ligand-binding domain. The GR-alpha is formed from these elements, the GR-beta from these elements. And the main difference, the really important difference, is that little green area at the end. Why is this important? Well, it's important because in this case, the GR-alpha contains both the AF1 and the AF2 site so that this is the areas where glucocorticoid can bind. In contrast, the glucocorticoid-beta does not have AF2. As a consequence, GR-beta can't bind glucocorticoids, and therefore it can't activate transcription or intracytoplasmic pathways. So it can bind to the DNA, it can bind to proteins, it can bind to all kinds of things. What it can't do is activate anything because it lacks the glucocorticoid itself. And this absence of GC means that it functions as a dominant negative repressor of glucocorticoid function. Now, what does this mean functionally? Well, functionally means that a state where GR-alpha goes down, there'll be less glucocorticoid activity. And a state where GR-beta goes up, there'll be less glucocorticoid activity. Let's look at the functional consequences of this in, well, not really in sepsis, but in cecal ligation and puncture, which is the best animal model of sepsis we have. I'll add, for completeness sake, that it's a bad model. This is some work done by Mabel Abraham in our lab a couple of years ago, where basically we looked at the abundance of GR-alpha and GR-beta as a function of time following cecal ligation and puncture. These are the data for alpha. And basically you can see that in heart, lung, skeletal muscle, and liver, it's decreased over time. I'd also point out that the liver expresses multiple GR-alpha isotypes, isoforms, explaining the myriad of functions that glucocorticoids can induce in hepatocytes. In contrast, looking at GR-beta, at least in heart and the lung, the abundance goes up. Again, as a function of time past CLP. Functional consequences, well, yeah, as it turns out. Cardiac output. Under normal circumstances, the cardiac output goes up in response to glucocorticoids, in this case, dexamethasone. Following CLP, not only is the function, unstimulated function, lower than in the baseline, but the response to dexamethasone is almost nil. Similarly, looking at hepatic gene transcription, in this case, the gene encoding glucose 6-phosphatase. Again, activity goes up in response to, transcription goes up in response to dexamethasone at baseline. Transcription levels to start are far lower than at baseline following CLP, and there's almost no response at all to dexamethasone. Well, how about sepsis for real? So this is tissue provided to us by Richard Hodgkiss at Wash U. His group basically did warm-up autopsies on a series of patients who died in the ICU. Almost immediately after death, they sampled tissue on non-septic patients and septic patients, and here's what they found. Looking at cardiac tissue, GR alpha is lower in the septic patients than in the non-septic patients. GR beta is higher in the septic patients than in the non-septic patients. Similarly, GR alpha is higher in the liver. I'm sorry, GR alpha is lower in the liver in the septic patients than in the non-septic patients, and GR beta is higher. So alpha is lower in patients who died with sepsis than it is in critically ill patients who died from other causes, and GR beta is higher in patients who died with sepsis than in critically ill patients who died from other causes. And finally, let's talk about the clinical relevance. Different responses in different tissues and cells and maybe within the same tissues and cells reflect differential abundance and expression of the glucocorticoid receptors. In critical illness, critical illness-induced changes in glucocorticoid activity and function may well reflect changes in glucocorticoid abundance and expression, as we showed you for those septic patients in cardiac antibiotic tissue. What makes this clinically relevant is that there are drugs that are known to alter the relative abundance of glucocorticoid receptor expressions in tissues. The best example are psychoactive medications, particularly those that are used to treat depression. It turns out that a number of these antidepressants work in part by altering glucocorticoid-mediated processes in the brain, and some recent data suggests, or some not so recent data suggests that these alterations reflect changes in the glucocorticoid receptor abundance. So endocrine signaling-related responses to antidepressants in cells reflect that, reveal that imipramine and decimipramine reduce the levels of GR-beta mRNA and C-gene activation in U937 blood cells. Similarly, looking at SY5Y and GERKAT cells, a Japanese group found that paroxetine and decipramine increased GR-alpha. So there's a potential that there are drugs currently approved for use that may well affect glucocorticoid receptor abundance, changing the balance of alpha to beta and potentially reversing some of the changes that are seen and that directly affect glucocorticoid responses in critically ill patients. And the final piece of clinical relevance, we started with a question, so let's end with a question. Does it make sense to conduct trials with glucocorticoids when the issue lies with something else, that is the glucocorticoid receptors? That note, I thank you very much. I guess we're not gonna be able to ask questions, but my email address is on this slide. Feel free to contact me if you have any questions at all. Thank you very much.

glucocorticoids

anti-inflammatory

immune response

molecular biology

glucocorticoid receptors

efficacy

Pharmacology

Pulmonary

Sepsis

Pharmacology

Sepsis

Presentation

Select

2022

Acute Respiratory Distress Syndrome ARDS

Foundational

Intermediate

Advanced

Sepsis Resources