Hacker News new | ask | show | jobs
by randomstudent 3216 days ago
This is a very important question, which I'll try to answer as clearly as possible.

First, a word of warning: This doesn't look like the appropriate kind of article to be posted on HN. As a doctor, it's hard to understand all that's going on, and even harder to appreciate what the real impact of this research is. Laymen should completely ignore this kind of articles, and read instead clinical trials or meta-analysis that report practical outcomes people are actually interested in, like life expectancy, cure rates, exercise capacity, weight loss, etc. This paper is about some molecular mumbo-jumbo that is not very interesting in itself, unless you're a researcher in the area.

Now, to answer the question. Classically, inflammation was defined by Celsus and Galen as 5 cardinal signs (in latin):

- Tumor (swelling) - Calor (warmth) - Rubor (redness - more noticeable in light skin; don't forget that Celsus and Galen worked primarily with patients from Europe and the Middle East) - Dolor (pain) - Functio laesa (loss of function)

These are clinical signs, that is, something the doctor can observe at the bedside. Think of when you fall and hurt your knee: the area where the impact happens swells. It might become warm to the touch. The skin might become red (different from bleeding). You'll feel pain, and for a while, your knee won't work very well because it hurts (loss of function). You don't need any fancy equipment to analyse these signs.

Now, these are adaptations of the body to injury: redness and warmth appear because the blood vessels in the skin dilate in response to injury. Blood brings red blood cells, which caus redness, and heat from the inside of the body, which causes the skin to be warm to the touch. The blood vessels also "open holes", which causes the liquid part of the blood to escape the bloo vessels and get between the cells, thus causing swelling. The injured cells release signalling substances, which are detected by neurons in the skin, thus causing the individual to feel pain. Finally, some or all of these changes, might cause the limb to lose it's usefullness (because it hurts and it's swollen). The goal of inflammation is to bring white blood cells to the place of the injury in order to fight a possible infection and assist in the repair of the injured tissue. Pain doesn't help protect or repair the tissue directly, but it notifies the organism to the fact that something wrong is happening, which causes the brain to act appropriately: remove the limb from a hot surface in the case of a burn, protect a wound, etc. People that are not sensitive to pain have serious problems from childhood because they lack these indirect protective responses (they won't remove their hand from burning coals, even as they smell their flesh burning, for example)

This is the classical picture of inflammation, caused by a wound or trauma to the body. It's simple, and it's functions are (nowadays) obvious.

But we can go deeper... Let's talk about pneumonia. Pneumonia happens when a bacteria or virus overwhelms the defenses of the lungs, and starts reproducing in the normally microorganism-free lung. This causes blood vessesls to upen up (thus causing swelling), to dilate (thus causing redness), but the lung probably doesn't get any warmer - it's already pretty warm. The patient will not feel pain. After all, the inside of the lung doesn't have nerves capable of feeling pain! (the outside does, and if the inflammation gets near the peripheral part of the lung, it might be VERY painful). Maybe the lung won't get any warmer (it's already pretty warm already). You will run a fever, but that's not the same kind of warmth we've discussed. You will feel breathlessnes, that is, loss of function of the lung. So, while some things are common the picture gets murky. Also, the body is working very hard to defend itself, and is now producing lots of molecules that circulate in the blood and which we associate with "inflammation". We now need some fancy equipement to detect these molecules: our 5 senses are not good enough. Even though the swelling if the lungs can be detected by auscultation (with a stethoscope), the doctor probably use somethin like an X-Ray machine to visualize the swelling.

Let's go deeper. Some bacteria enter the bloodstreem and latch to a valve of the heart. This kind of infection is called an endocarditis. The valve has no blood vessels, so you won't get swelling. You won't get redness or warmth either (again, you ill run a fever, but that's different). The valve also has no nerve endings, so you will feel no pain! The bacteria will happily eat your valve away (and when it's eaten away you'll die), but it may take time, and meanwhile, the valve can work relatively well. So, no loss of function. Yet, your body is fighting in overdrive to clear the infection and kill the bacteria. In fact, lots of mollecules associated with inflammation are running wild all through the body, sometimes causing as much damage as the infection itself! These are the same mollecules your body produces in pneumonia, and also in other kinds of "deep" infections. So, in a sense, even though we have amost none of the 5 classical signs of inflammation, we have a response of the body that aims to defend it against infection. So we might decide to call it inflammation. Again, we can't see the inflammation here. We need some fancy machines to see the damaged valve (even though we can listen to the valve failing with a stethoscope), and fancy lab equipment to detect these molecules, as well as the consequences of the inflammation inside the body.

If we go even deeper, we'll "discover" that heart attacks (technically "myocardial infarctions") are the result of inflammation too. Even though what the patient feels is just pain, possibly followed by loss of function: breathlessness and death.

So, at each step, we get further removed from the intuitive description of inflammation, which you can feel with your hands and see with your eyes, into a scary realm only accessible with the help of lab equipement and thick books. At this point, inflammation is actually a buzzword that means little...

This paper's conclusion doesn't mean that the hypothalamus (a part of the brain) is red, swollen, warm, painful or has stopped functioning. That would be an encephalitis, an extremely serious disease that will land you in a hospital or a coffin (see the movie "Contagion" for some realistic depictions of encephalitis). It means that some cells, vaguely associated with inflammation, are producing some mollecules which in most parts of the body are associated with inflammation.

In most situations, inflammation is either bad or signals something bad. In this situation it certainly seems bad: the concrete changes that inflammation brings to the brain seems to be similar to the ones caused by diabetes (according to my cursory reading of the article). But I agree that the word has been used for so many things, that you can't actually tell whether inflammation is good or bad.

On the other hand, inflammation is a defense mechanism of the body. So what is a person with no inflammation at all? It's a person really close to death, because Nature is scary and full of bugs, and someone incapable to mount an inflammatory response is what bacteria, viruses and fungi call dinner. People like these exist (either from birth or due to unfortunate circumstances), and we can only keep them alive by keeping them in very clean rooms and pumping them full of antibiotics until their immune system (and the capability to mount an inflammtory response) recovers.

Using the same word for what protects us from infection and for what turns our brains into something similar to the brains of a person with untreated diabetes is confusing. We're at a point where when a doctor talks about inflammation, it has a dompletely different thing in mind from laymen. A doctor might even have something completely ifferent in mind from another doctor or from a researcher (which is more serious).

In the end, I have nothing to tell you except that you're right to be confused, and if it serves as a consolation, I'm often similarly confused when I read the titles of these papers (and sometimes the abstracts), even though I've been studying these things for years.

1 comments

Hi, Endocrinologist and obesity specialist here.

Let me confidently assure you that inflammation in the hypothalamus is a dominant feature of obesity, and obesity is currently one of the top diagnosed medical problems in America. This stuff isn't hypothetical, but well accepted within the clinical research field for metabolic disorders.

Please don't recommend laymen avoid reading articles like this. That's rather senseless. While this stuff hasn't become standard in medical school curricula yet, it definitely will be some day, particularly when we are more confident in interventions that prevent hypothalamic gliosis.

EDIT: also, don't discount hypothalamic inflammation based on what you know about encephalitis. Some parts of the hypothalamus (the arcuate nucleus, in particular) are highly susceptible to inflammation because they are surrounded by a characteristically porous part of the blood brain barrier (behind the hypophyseal portal system). This allows bulky blood-born protein hormones (like leptin and insulin) to make it into the sensory neurons.

Indeed this part of the brain doesn't become red and painful, but something more relevant happens. Microglia surround the neurons responsible for appetite suppression and kill them off, which permanently changes the body fat setpoint that the brain strongly defends. There is definitely a profound and sometimes irreversible loss of function here, and that's the current focus in research. Hopefully some day we will find ways to drive repair of these circuits.

> Let me confidently assure you that inflammation in the hypothalamus is a dominant feature of obesity

I certainly believe you (I'm not qualified to agree or disagree, as I'm not a specialist like you). The point of my rambling comment is to make sure the readers understand that the kind of inflammation we're talking about here is very removed from what people usually think about when they think of inflammation.

> Please don't recommend laymen avoid reading articles like this. That's rather senseless

Hm... Here I'll have to disagree. I think medical research should be consumed rather judiciously, especially by laymen or by doctors that are not specialists in the research area in question.

I'd rather have people read distilled accounts of research (textbooks, meta-analysis, etc.) than studies in knockout mice.

EDIT: Answering your edit. I'm not discounting anything. I think the study is as relevant as it gets for a study in mice, and the role of hydrocarbons of the diet is an important one. After all, is it just calories in VS calories out or is there something else?

I do understand what the study is talking about when it talks about microglia and inflammation. I'm merely trying to answer the question about "What is inflammation IRL" in a way that laymen may understand - most people don't know that microglia are derived from cells of the immune system (in a sense, they are the "inflammatory cells" of the brain), and can't draw the parallel between neutrophils and monocytes happily blasting away through the extra celular matrix under the skin in search for pathogens and microglia killing healthy neurons.

In no way I want to be skeptical of the article or discount the importance of its findings. I only want to draw the distinction between the "naïve" idea of inflammation and the kind of inflammation the article is talking about.

EDIT2: The poster asked why he didn't feel "inflammed". He would certainly feel "inflammed" if he had an encephalities, so there is definitely some confusion here. My comparison between this "low level" inflammation and encephalites was meant to highlight that. Do you think I was unclear?

I concede with many of your points. We are living in an era where nutritional dogma is really strong in a lot of areas, and I think research articles like this do drive lots of amateur dietary advice from people who over-interpret it.

Pardon my passionate writing, obesity is a personal matter for me. I'm looking forward to it being pushed further out of the lab and into the clinic, but I'm not sure what those steps look like just yet.

More important is the focal nature of this specific kind of inflammation. It is also not quite a typical inflammatory response if just messing with adhesion factor and AGE receptor changes it this much. (Different from even chronic inflammation.)
I'm in medical school right now, we talked about metabolic disorders and the role of the hypothalamic nuclei in obesity. Wouldn't be that hard for a medical student who remembers what microglia are to extrapolate from there. I also agree that laymen can indeed read these kinds of articles, if comfortable with the content. If not, there's plenty of secondary sources commenting on these new findings.
The one prospective test to be run in humans (interventional study) is if we have some potent AGE formation inhibitor and what that would do to prevent obesity. People working on SENS might be the ones to ask (esp. GlycoSENS).

Expensive to run of course.

> Hopefully some day we will find ways to drive repair of these circuits.

Any idea on how we may do it? AFAIK, currently we have no good mechanisms to "repair" individual brain circuits: even in strokes, we mainly exercise the affected area so that neural plasticity repurposes other neurons (usually in the cerbral cortex) to take over lost functions.

Does the hypothalamus have this degree of plasticity?

I saw a study once that raised mice on both standard and high fat chow. Predictably the high fat fed mice developed central obesity. (High fat diets are still a reliable way to make a mouse fat).

They then put the fat mice back on the normal chow, and they lost weight, but importantly they didn't ever reach the healthy weight of the mice fed on normal chow from the start.

The explanation they offered was that the GABAergic Agrp/NPY neurons (that promote hunger and lower metabolism) had formed stronger connections, and many of the Glutaminergic POMC/Cart neurons (that promote satiety and increase metabolism) had died off.

Whenever I go digging for this study I can't find it. Allison Xu's lab however has written a few studies now talking about this neural subtype ratio in various models, as well as the inflammatory mediators linked to their changes over time. http://www.pnas.org/content/110/8/E697.long

Im not sure about the plasticity questions, or neural regeneration, but that would be ideal if possible. I have however seen more crude suggestions for treatment that involve killing off neurons in the hypothalamic nuclei that promote hunger and feeding using techniques such as gamma knife radiotherapy. https://www.ncbi.nlm.nih.gov/pubmed/15695926

But it would be a LONG time before that becomes a viable surgery in humans!

Thanks. I'll read these papers he I have the time. I agree that gamma knife surgery on the hypothalamus is probably not yet ready for prime time (and it might never be if we can find something better)
Hi,

Endocrine patient here.

I've seen some research that suggests a phenomena of hypothalamic obesity due to decreased oxytocin secretion. Know much about this?

Could Inflamation of the hypothalamus decrease / increase sensitivity of the negative feedback mechanisms of the endocrine systems that are controlled via the hypothalamus / pituitary?

Thanks for your input doc.