Mother-Infant Bonding, Maternal Care & Breastfeeding, Neural Basis of Hunger & Social Behavior | Marcelo Dietrich | #180

About the guest: Marcelo Dietrich, MD, PhD is an associate professor in Comparative Medicine & Neuroscience at Yale University, where his lab studies the neural basis of behavior in mammals.

Episode summary: Nick and Dr. Dietrich discuss: maternal care behavior, mother-infant bonding, and feeding in infant mammals; neural circuits regulating hunger & satiety; brain development & early social learning; behavioral variability among individuals; and more.

Related episodes:

* M&M #159: Neuroscience of Social Behavior, Pain, Empathy, Emotion, Brain Mechanisms of MDMA | Monique Smith

* M&M #108: Monogamy, Sexual & Parental Behavior, Social Attachment, Oxytocin, Sex Hormones, Childhood Psychosis, Menopause & Brain Development | Devanand Manoli

*This content is never meant to serve as medical advice.

* Full audio version: [Apple Podcasts] [Spotify] [Elsewhere]

* Full video version: [YouTube] [Odysee]

* Support M&M if you find value in this content.

* Episode transcript below.

Full AI-generated transcript below. Beware of typos & mistranslations!

Marcelo Dietrich 1:33

so my name is Marcel Dietrich. I'm an associate professor here at the a school of medicine, and my lab, it's particularly interested in the dynamic change in your circuit that happens throughout development. So from infancy to adolescence to adulthood, how neuro circuits are changing to regulate the distinct behaviors and physiological needs. So that's what we are studying.

Nick Jikomes 2:06

And on your lab website, you have a nice Lab website, you have this video and a quote from a man named Perry Harlow. Can you talk a little bit about what that quote means and who that was?

Marcelo Dietrich 2:19

Yeah. So yes, Harry Harlow, the quote is that if we have learned anything about the studies of monkeys, is that we we have to learn how to love before we have we learn how to live. And it's a long story. I will try to make it short. So Harry Harlow was a primatologist, a behavior scientist who made really pioneer studies in the 50s, 60s. And we can talk more in details if you want, but basically what he was trying to test is, what are the factors in a mother that are relevant for a baby. So he took orphan monkeys. He will actually fly these monkeys from India. He was a pioneering studies rhesus monkey. And then he would raise these monkeys with two type of Wired mothers, one wire mother that has a bottle, had a bottle of milk, and one wire mother that had a cloth. And so the cloth will provide some kind of tactile experience. And the idea was to test the hypothesis that whether the monkeys will like better a milk or tactile experience. Those studies were really mind blowing back in the day, because the monkeys would spend, like 23 hours and a half of the day wrapped around the cloth mother. So they would overwhelmingly choose the cloth mother.

Nick Jikomes 3:48

So so the idea was, well, obviously, the baby needs to eat, and the mother is the source of food. Obviously, babies love to be embraced by their mother and hear the mother's voice and all these things. But when you sort of, when you put, when you put the food, the milk, against the the physical touch sensations of the Mother, you're saying that the monkeys would spend almost all of their time getting that physical touch. They really seem to prefer that. Yeah,

Marcelo Dietrich 4:12

correct. And I think the those findings need to be put in the context of the time of the 50s and 60s, and we can also try to go through that. But I think the most fundamental experiment of Harlow is the following. We have all been a playground with kids, and we see that the kids usually look for the caretaker, for the figure that's around to see if the the caretaking figures around, and then they play. So if the caretaker disappears, if kids flip out, they so Harlow had that intuition, so he did the same thing. He put, he made a playground for baby monkeys, and he put the baby monkeys to play in the playground. But then he put the. Wire mother with a bottle one playground, or he put the wire mother with the cloth. So when Harlow put the babies in this playground and there was the wire mother with the bottle of milk, the babies will not play. They will they will be. They will freeze. They will be in tremendous distress. So that mother provide no comfort for the baby, despite the fact it provide milk when he then switched the mother, but the cloth mother, this baby's monkeys will go hug the clot mother, and they will be playing around. And then often they will look back, see that the mother figure was there, the attachment figure was there, and they will continue playing. So that was really, I think, the key experiment showing is not only a preference, but this, this artificial, inanimate mothers have actually the capacity to buffer the distress of the infant and allow the infants to explore and play and learn about the world. And so, yeah, that's a experiment that I like very much.

Nick Jikomes 5:59

Yeah, and it starts to tell us that, I guess, you know, there's multiple things in an infant's environment that are tied to its mother, that are of interest, that the baby needs. It needs food, it needs safety and to have protector around. It needs that physical touch. But all of those things are not weighted equally to the infant, that some of the some of them are having sort of a stronger effect on the baby,

Marcelo Dietrich 6:23

Yeah, certainly. And yeah, Harlow test that. And I think it's there's a lot to talk about Harry Harlow, but it's probably one of the saddest experiments he did a few as he progressed in his career, he got a little out of his mind and was heavily criticized for the ethics of his experiments. And one experiment that he did was to cool down the mother, so you would have he cooled down two degrees cooler than the room temperature, and ask whether the baby would attach and would get milk from the mother. And this

Nick Jikomes 7:04

is the artificial mother. Artificial, yeah, artificial. And he did. I

Marcelo Dietrich 7:08

think, I believe I might be incorrect on this one, so we have to double check and but I think he also anesthetized the mother and cool down. So if he cools down, this the mother two degrees under room temperature. The infant monkeys will never get close to them, will not get even milk, and they will die. So actually, the title of the paper is psychological death in an infant monkey. Wow, so and, yeah, very sad. So basically, in harrows experiment, he puts temperature as number one of the sensory importance of the mother. And then comes touch, tactile experiences. And then rocking, the movement is important. Milk is actually the last of the series. It's not that it's unimportant, but nursing is the lowest in importance in his experience.

Nick Jikomes 8:09

Yeah, and it's a, you know, it's potentially counterintuitive, right? You might think, well, the baby needs to eat to survive. The mother's the sole source of food. It's a little bit surprising. It's not surprising that it's the tactile sensation and the warmth are all important, but it is perhaps surprising that they're even more important than the food.

Marcelo Dietrich 8:27

Yes, certainly, and that's why I like that quote that's done on my website. Because also, at least, I think, I think if we talk to child psychologists or people that are in that field, they will not find surprising, at least now, but for us, when we start studying, this was also surprising. And I think the concept that's always on my mind, it's the ecological context in which these features evolved. So in here we are talking about mammalian evolution. We have about 6500 mammals on Earth. And all mammals necessitate a mother, a caretaker, to obtain this early, early nutrition and early stimulation. So the way I I think, and I can explain that those harlows findings in an evolutionary context is that the primary importance in the beginning might be the milk, but after millions of years evolution, the mother always predicts milk, hence milk becomes irrelevant. From the perspective of evolution, what is more relevant is the other stimulus. So the the proximal effect, as we say, it's actually touch and other stimulation, but they automate. You definitely need milk, but the correlation is always 100% in nature. Yes, you never get a mother the wild, and if you do. Have a mother, usually that provides no nutrition, the infant will die. So there is no evolutionary adapted selection. It's

Nick Jikomes 10:09

just like applying reinforcement learning within one animal's experience across, you know, millions of years of evolution. You know the the most efficient thing to do is to learn the first thing that predicts the outcome you care about. And in this case, right? The mother, the mother is always around to provide the milk. So it does make sense, when you think about it in that way, from a reinforcement learning perspective, that that evolution would sort of bake that kind of tendency in

Marcelo Dietrich 10:32

No, I agree. There's another animal that I like to talk about, because I think it's just fascinating. So as I say, we we have about 6500 mammals on earth, and the one that we know that has the shortest lactation period is one seal that lives in the Arctic. It's the so this seal the the infant, only the neonate only stays with the mother for four days. So within these four days, we'll gain, like, 200 kilos or so, and so we'll take immense amount of milk, and then the the baby will leave. But the baby cannot hunt on on their own for another month. So they survive, they fast for another month, only eating ice as a form of hydration. So actually, the physiological systems to eat and already. But ecologically, the baby cannot stay surrounding, surround by the mother, because there's a risk of predation. But even we in that extreme evolutionary context, ecological context, the attachment, the relationship of the baby with the with the mother, didn't disappear. It did shrink to four days, but it did not disappear. Maybe we need a few more millions of years to find mammals that in which that bone will be broken. But, you know, that's one of the extremes. Yeah,

Nick Jikomes 12:03

so that just gives a sense for the strength of this bond, and the importance

Marcelo Dietrich 12:08

is very it's basically a constitutive part of being a mammal.

Nick Jikomes 12:13

And, you know, one of the things we're going to get into here is how hunger and feeding work, and how some of the the circuits in the brain, develop that, regulate that, you know, but obviously, just on an intuitive sort of experiential level, obviously an infant has feeding behavior that's very different from an adolescent or an adult. You know, as adults, we eat so many different types of foods. We have to learn what we like. We get to test out all different sorts of tastes and see how we react to it. But for a newborn infant, right? It's sort of all baked in. There's gonna be one food source for some period of time. It's just the milk. The baby doesn't need to learn about the milk really. The suckling reflex is gonna be there right from the start. And as you said, you know the mother, the mother is obviously there when it comes to mammals. That's just part of being a mammal. Can you talk a little bit about how sort of hunger and food and nutritional needs change from infancy to adulthood, and, you know, sort of general ways that cut across all mammals. You know, obviously the milk has a certain nutritional composition. It's presumably tuned to the baby's needs early in life. Then those are going to shift over time. But what are some of the ways, the major ways, and most are all mammals that sort of the food needs and the hunger needs of an individual change from infancy, infancy onward.

Marcelo Dietrich 13:33

Yeah, it's a fantastic question. There's so much to unpack. There so many things come to my mind, maybe, maybe we break down in peace, and you can reflect on what I say. Since we are talking about different animals, different mammals and examples of that, I will continue giving these examples, because I think this one is mind blowing too. So marsupials, some marsupials, so mammals that will have a pouch and the embryo will actually attach to the nipple within the pouch and then grow within the pouch. Some are Supers, will have multiple layers of nipples. So for example, we will have the vagina will have three chambers. So the first embryo climbs from the chamber and attached to the first layer of nipples, and then once it grows, it climbs to the second layer of nipples, and then the embryo from the other chamber will come to the the first layer, so they are synchronized in the climbing the layers, and each layer will have a different milk, which will be important for the growth and nutrition and the immune system, immune protection of the growing infant as it grows in the. Journal pouch of the marsupial man.

Nick Jikomes 15:04

So in marsupials, the embryos developing, and it's always in the pouch, but there's literally different sections of the pouch, and the embryo is kind of climbing through each section as it grows and as it grows while it's still, you know, in the in the pouch in the womb, I guess it's attaching to different nipples and the composition of the milk in each nipple is actually different

Marcelo Dietrich 15:24

Exactly. So I give that example to illustrate this remarkable features is the same other different nipples, each nipple with the different milk composition. So what we do know is that new composition chains throughout lactation in animals and primates and humans, and that match the nutritional and also the other needs. For example, milk will have many antibodies that will be important for immune protection of the infant early on, for example, IgA, that's mucosal antibody will be very high early on, and then we will start paid as lactation goes on, as also the infant start produce their own levels of antibody. So there are all these features that are matching the physiology of the infant. So if so that's talking about that. So

maybe you Yeah, maybe you

Nick Jikomes 16:26

Yeah. So, so basically, I mean obviously, obviously, the nutritional needs of a baby are different than an adult. But what you're saying is, like even within, even within the preterm period, even within early infancy, the infancy, the nutritional needs are changing. So the composition of milk, you know, from when a baby is born on day one will be different, you know, just a few weeks or months later. And that's all tuned to the baby's needs. There's the nutritional needs, there's the sort of immunological effect here. Before the baby can prove its own antibodies, it's literally ingesting them from the mother. And so all of these things sound like they're sort of beautifully orchestrated to perfectly match what the baby needs each step of the way.

Marcelo Dietrich 17:06

Yes, certainly. And so you know one.

And then the question you ask also the question about regulation, right? There's a regulation of feeding in adults. Well, the that we know quite a bit about that, and how is the ingestion regulated in an infant or in a new need, we don't know fully that's the real the, I think the most accurate answer. And there are many theories of whether breastfeeding and intake of solid food, for example, utilize same neuroscircuit same brain processes, or parallel brain processes and things like that. So we still don't know the answer to that. However, we know a few things, mostly from rodent models. I would say rats, mostly, some work in mice that very earned in development right after birth. There, the neonates don't seem to have a mechanism to stop ingestion when they are ingesting like breast milk. In other words, if you take a infant rat and the infant attached to the nip of the mother. But instead of getting milk from the nipple, you pump milk in the mouth, this baby will ingest constantly, until almost literally exploding, Oh, wow. We only stop when the stomach is in huge, huge, huge age. So they

Nick Jikomes 18:55

it sounds like that in the infant rodent, at least the satiety mechanisms that are developed in an adult are not there

Marcelo Dietrich 19:03

yet. Yeah. So when, again, when you put that in perspective of the relationship of what is happening in breastfeeding, it makes sense from the following perspective, the regulation of the intake is actually done by the milk ejection reflex. And the milk ejection reflex is not constant. In fact, when the baby is attached to the nipple, is stimulating the sensory neurons in the nipple in the rat. People have done that experiment, the ejection of milk from that stimulation is going to be a random event, really, within, within a certain distribution of time, yeah, within 30 seconds to three minutes or four minutes, I don't remember exactly the data, when the milk is going to be ejected. Is a random probability. Yeah. And I think that's really fascinating, because you remove the need for the infant to have a regulatory process, yeah, again, the same evolutionary process. There is no need for regulating because there will not never be enough. And whatever comes comes, it comes with the timing that's definitely sufficient for absorption, opening the stomach, going to the lower gut, and getting more milk, more milk. So the mother regulates the process. Again, you give to a third party the regulation of the self,

Nick Jikomes 20:36

yeah. So, so, so when you see an infant mammal suckling milk from its mother. It's not actually getting a con. It's not like if we, if you and I, were to have a cup of milk with a straw in it, and we can constantly suck it out at a constant rate. It's that's not what it's like. The milk is sort of coming out haphazardly. And this is actually serving, uh, an adaptive function. It's preventing the the baby from ingesting too much, essentially, exactly,

Marcelo Dietrich 21:02

and that's one part. And the other adaptive function is when the baby's there, in the breast, in the nipple, many other things are happening beyond the ingestion of milk, much like the experiments of Harlow. So in that in that moment, for example, when you record the autonomic nervous system of an infant, the autonomic nervous system, the sympathetic tone of a baby, is much more regulated, is much calmer when it's attached with the mother, so the baby

Nick Jikomes 21:39

feels safe, the stress levels go down. The baby's presumably not anxious or anything.

Marcelo Dietrich 21:44

And in that context, the baby learns. In that context, the baby's so it counts, stress response is low. The baby learns about the environment very early days, we learn all the associations. We start learning sound associations. And then when visions start to get better in the human baby, we start learning facial expression.

Nick Jikomes 22:07

Yeah, and this would imply that, I mean, there's a lot you could unpack here, and there's a lot of things you could think about in terms of, you know, modern human caregiving for infants and stuff. So, you know, just ask you one thing here, you know, you could imagine two human babies. One of them is getting breast milk directly from the Mother, the natural way and doing things the natural way. Another one, you could imagine giving an infant formula from a bottle, same nutrition value, same calories and everything. The implication here is that baby would actually have some learning deficits, maybe because they're not going to have this change in the sympathetic nervous system that's going to affect their stress levels and that's going to have knock on effects for their ability to learn faces and to become familiar with their environment.

Marcelo Dietrich 22:53

Yeah, it's absolutely a generalization, but I think is fairly correct, and I think, sadly, we learned quite a bit about that process from, I think your audience is familiar with that, probably the Romanian, Romanian orphanage. So in the late 80s, with the fall of the Berlin Wall in Romania, Romania in the 80s, was in an economical crisis, so need more babies. So there was a governmental communist incentive to have more babies, so families will have a lot of babies. And there was a selection process, so babies that were kind of misfit were put like in this warehouse, so they receive nutrition, but really no stimulation, no caretaking. And so when scientists got there after the the fall of the dictatorship, they found this, hundreds of 1000s of babies. So there was nothing that could be done. There were no families to adopt all these babies. So they run the the only clinical trials that we probably will ever be able to run, which was selecting some kids and give to foster care. And what we learned from that, that if you do take

infants

during a sensitive period, roughly the first two years of life, and you provide that care, that comfort, that tactile stimulation, they will develop quite well. But if deprive them so, if you only give the bottle you suggest, then this infants will develop with very severe, uh, neuropsychiatric symptoms and social deficits, impairment, capacity to socialize. Yeah? So, yeah,

Nick Jikomes 24:53

wow, yeah, yeah. So that that early period development is is super important, and it's not just the sort of raw. Physical needs of the body in the form of, you know, getting food and hydration and stuff. It's all of this sort of these, these sensory and social effects that put the nervous system in a state that actually just enables learning. Generally, if you don't have that, you're going to have all sorts of problems. Yeah.

Marcelo Dietrich 25:19

Yeah, absolutely, that's

Nick Jikomes 25:24

And so you've, you've done studies in in young rodents that have to do with disentangling how the feeding system develops and things like that. Before we get into some of that work that you've done, I want to give people a general sense for how the sort of hung we'll just call the hunger circuitry in the brain works in an adult first so I'll let you sort of take it where you will. But the hypothalamus is going to be a very important brain region here, and there are specific neurons in the hypothalamus that that are really, really cool and really important here, that drive hunger and satiety. Can you give us a basic understanding for what some of the key circuitry is that controls hunger in mammals, in adult mammals,

Marcelo Dietrich 26:06

yeah, absolutely. And I will certainly focus on on the hypothalamus, because I think we make sense for our discussions later on. So in the actually the most basal part of the brain. So above our palates, our soft palate, they will be our pituitary, the gland they regulate many of the hormones, like HPA, stress response, and right above that, there's a layer of thin skull, and then the hypothalamus. So they have in the very base of the brain and the nucleus, the part of the brain that's the most base, that was called the archway nucleus in the in the humans, in prime is the infundibular nucleus, and in this region there are some neurons that are important in the regulation of feeding and metabolism. We start to suspect this more than 100 years ago, because that region, the pituitary gland, is commonly, commonly affected by tumors. So the tumors go and because the hypothalamus is right there, it pressures the hypothalamus, so you start causing some people have symptoms in energy metabolism, obesity or things like that. So in the 19th century, people start to suspect about the role of the hypothalamus in feeding. So fast forward 100 years, we know there are two populations of neurons there. One is called AgRP. HRP comes from a good related peptide, a good related protein, which is a gene, the AGP gene that produce this peptide by these neurons. And when these neurons are activated, or when the animal is deprived of food, these neurons will be activated, and they will promote feeding and many of the metabolic responses to fasting or to food deprivation. Next to these neurons, there's a second population that's called palmci neurons. They produce propio melanocortin, and the propial melanocortin is a large molecule, large protein that can be chopped in several small peptides, and some of these peptides have strong satellite effects like Msh alpha, SHM and also Beth endorphin, which is an opioid. So the palm syndromes are taught to be the counterarm of the AGP neurons having a little more delayed effect, but when they are active, they promote satiety, and AgRP neurons, when they are active, they promote hunger. That's an overly simplification, but I think it's a bare one. So,

Nick Jikomes 28:56

so a bit of a simplification, but in essence, at the bottom of the brain, in what's often referred to as the arcuate nucleus of the hypothalamus, there's two neurons. And basically, if you turn on one type, type of neuron, the HRP neuron, animals look like they're very hungry, they will eat a lot, etc. Then if you turn on the other palm C neurons, you basically get a counteracting effect, exactly. And so, so naturally, you know, you would think, you know, what do we know about these neurons in terms of their activity? If you've got sort of a hunger neuron type and a satiety neuron type, you'd imagine that turning on, say, the age of appearance, drives feeding behavior, that when animals are hungry and they're looking for food, those neurons might be very active, etc, etc. Is that essentially how they work?

Marcelo Dietrich 29:38

Yeah, it's, yes. That's a

so in the early 2000s

was shown that the AGP neurons were response in that predicted manner. You feed the private animal, you have increased activity of AGP neurons. You feed the animal back and the activity to decrease the Tech. Nix used back then were markers. They have a very slow dynamics, which means matter of hours to find those activity chains and in 2015 using new technologies, basically three labs, almost simultaneously. Scott stern so Janelia farms, Mark Anderman at Harvard, and Zachary Knight at UCSF. Using three distinct techniques, they were able to record the activity of these neurons in the live mouse, and they actually show that once the animal actually sees the food or perceives that the food is available in the environment, the activity of these neurons will suppress, will decrease in an infecting anticipation of eating after eating, there's a further suppression that comes from gut signals. But that that anticipatory drop was something that was somewhat unexpected. If you think, like, only we call omniostat, like, if you're changing your temperature in the house and but like, 72 and so the thermostat we want to switch on and off at that point. But here you're saying that, as is reaching, is predicting Yes, yes, future. So it's actually switching anticipation to Yeah.

Nick Jikomes 31:25

So if Yeah, you're saying, if these neurons were merely regulating hunger in a purely, like homeostatic fashion, they would be pretty simple. They wouldn't start to change their firing until an animal eats the food and the calories get into the body, and they're actually, like, ingested but, and that does happen to some extent, but what you're saying is, when an animal, even hungry animal, even sees food, these neurons are immediately responding, and so they're doing something more sophisticated and complex that has to do with, you know, understanding that food is about to come even though you haven't actually ingested it yet.

Marcelo Dietrich 31:59

Yeah, absolutely. I think that comes back again to the complexity of the brain and of mammalian systems in general that we are not dealing with engineer like machine that turns on and off, and it's a much more complex system in which one of the main advantage of this complexity is really the capability to predict what's going to happen at different timescales, seconds, minutes, hours, days and so And that really gives a lot of adaptive value to the system, because, yeah,

Nick Jikomes 32:45

and, and, you know, oftentimes the caricature that people have of the hypothalamus is it's deep in the brain. It's part of our so called reptilian brain. It's sort of a stupid area of the brain compared to maybe the cerebral cortex or something like that. What does this start to say about, you know, the sophistication of the hypothalamus and how much that's been maybe underestimated over time, and to what extent do these circuits actually tie into learning and memory systems that we think of as being higher up in the brain? Normally?

Marcelo Dietrich 33:18

Wow, that's a very good question. I'm trying to think how to answer this question without diverging too much, because I think there's a lot to unpack in this HRP,

I think one way that I can answer, I think some of we still don't know fully, that's I think it's a fair answer of how these hypothalamic circuits, how they interact with higher order learning circuits. Now let's say an animal sees a sound that sound has no predictive value. It's the first time the animal sees the sound. Yeah, the age opinions will not do anything you record the activity, they're not going to

Nick Jikomes 34:08

do anything. They have no relevance to food. And so then

Marcelo Dietrich 34:11

you pair that sound with food. You give the food. So every time the sound rings, there will be food that will come with the right after so then when you give the sound, the EGP neurons are going to start to respond. So there's definitely a learned component. So the end was learning that the sound now is a predictive cue of food, and that cue is actually coming from higher brain orders sensory systems all the way to the AGP we know a little bit about the circuits that engage that cue evoked suppression of the AGP activity. However, we don't know fully yet how the this memory. Yeah. How the memory of that context is actually going all the way? And I think that's a very important question, and I think there are many laboratories right now trying to actually figure that out. Yeah,

Nick Jikomes 35:17

yeah. But I guess you know, at a high level, the circuit is not purely responding to hard coded, instinctive food related things. It is actually receiving information that's learned through experience.

Marcelo Dietrich 35:29

100% Yeah, 100%

Nick Jikomes 35:34

and so what happens? Just to give people a little bit more sense of how important these neurons are, let's say you take an adult animal, and I know that there's ways to do this, and we don't need to go into the technical methodological details, but scientists have the ability to say, destroy HRP neurons in an adult and people have done this, what happens if you destroy these neurons in An adult animal?

Unknown Speaker 35:56

So

Marcelo Dietrich 35:58

is another good question. It's, the findings have been a little controversial in the last couple years, but I would tell you the the originally in 2005 the laboratory ritual parameter, they create a transgenic mouse line that allow eliminating the AGP neurons in the brain acutely. So expressing the receptor for differ toxin, diphtheria toxin, so when you give the toxin, will bind to the receptors in the AGP neurons only, and kill these neurons. When he did that the animals will starve to death with within a week. And so that was thought to be the case until recently. Now, when he inject, did the same procedure in a 2345, days old animal, so I knew Nate, nothing happened. So originally in his paper was a paper published in Science. I remember vividly the day I the paper came out. And so what started entrepreneurs were essential for feeding in the adults, but not in the units. Now that also has been put a little bit in check recently with another study that used several of these techniques and said that maybe the effect of the ablation had some indirect effects. Was not exactly by killing the AGP nearest.

Nick Jikomes 37:41

So it wasn't fully, maybe not a fully clean experiment. It was doing something other than ablating The AgRP, maybe.

Marcelo Dietrich 37:46

And it's a, I think it's a very, it's still very controversial, because many people have replicated those findings. And, yeah, I think these techniques all have their caveat. Now, in general, if you don't have AGP neurons or inhibit the AGP neurons, the animals will at least have a deficit. Yeah, they eat less. They will have a death in eating. If you food, deprive them and give food, they will have less of that effect. So in general, that's the

Nick Jikomes 38:19

yeah and, and, I guess in any case, you know technical issues aside, one of the key results there was that you see a big difference in the type of effect you get in babies versus adults,

Marcelo Dietrich 38:32

absolutely.

And they're always stuck to me in my brain, and it was one of was a key finding to stimulate us many years later. So I start my lab in 2014 and I was interested in that question, though, what regulates the intake of breast milk, of calories early on, if you can, if you can, kill the Egypt and nothing happened, yeah,

Nick Jikomes 39:06

and I mean, immediately I'm thinking of what you told us before, where the the baby, there's this mechanism in place in the mother to make sure that not too much milk comes out, because the baby actually doesn't have the ability to self regulate its intake. And so that would imply that the maturation of certain parts of the brain that have to do with feeding and satiety takes time and has not developed in the early infant.

Marcelo Dietrich 39:31

Yeah, absolutely. Yeah. This concept, it's much clearer to me, and I think to us in general, today than it was 10 years ago, but that was somewhat what we tried to test. Basically, we asked the question whether this AGP, neurons that promote hunger would regulate the intake of breast milk or will have any importance for the injection. Positive Behavior of infant that's still in the full dependency on the mother. So the mother is the regular regulator. And so we did those experiments right when I started the lab, trying to play with the neurons, activating the Egyptian neurons, inhibiting the Egyptian neurons. And it was quite striking that we would find no change in ingestion of breast milk, for example, if we doesn't matter what we did with the neurons. But when we would play with the neurons in the babies, for example, in an infant mouse, there was a long we could make these babies cry more or cry less. So the infant mice vocalize in the ultrasonic range, and so we could have really big effects on the amount of this cry behavior, or the amount of ultrasounds that the babies will emit when they were alone. And we thought that that was quite interesting. And those ultrasounds are known to function as a cry, which means, if the mother listens, the mother comes, yeah, yeah to the mother.

Nick Jikomes 41:18

So, so just to unpack a couple things here before you continue. If you take an adult mouse, yeah, and you turn on the HRP neurons, but there's no food around, they will start looking for food. They'll sniff in the air. They'll dig through their bedding. They'll they'll search around for food, and they it looks like they're trying to find food. If those neurons are active and there's food around, they'll basically stop what they're doing and eat it. But what you're saying is, when you do this in an infant, what it starts doing is basically calling for its mother,

Marcelo Dietrich 41:48

exactly, they call for their mother and and then we did the

sequence was more or less I'm telling the sequence that we did the experiment. So then we we look for if these neurons were active when the infant is alone, right for the mother. And actually they get very activated. They are very active as much as one would predict after depriving adult from food for like a day. So we were very surprised to separate the baby from the mother for half an hour, an hour, an hour and a half, and the activity is very high. Again, we've first we use these techniques that have no temporal resolution. And so the neurons were active, and the infant is alone, the AGP neurons, and when you inhibit these neurons, they cry less. When you over activate them, they cry more, and that changes the mother behavior. And then we took some of these techniques that I mentioned, that these labs used in 2015 and then we applied them to record the infant mouths, which is quite tricky, because infant mice are very small, and they grow from one day to the next. But yeah, I also remember those experiments, and basically it was quite striking when we are recording the neurons and the infants are with the mother, with the siblings in the nest, the neurons are quiet as soon as we take everybody out and they are alone, the activity of Egyptians start firing immediately within seconds. So it's a very rapid effect. And as soon as we put any social partner back with the siblings or the mother, the activity suppressed to what it was before separation, within 30 seconds. It's a very rapid alarm system, like turn on, turn off, that's very different from what we understood from how the activity of these neurons is modulated in the adult, because in the adult, the activity is slowly increases as the animal is deprived of food.

Nick Jikomes 43:59

Yeah, yeah. It's a slow signal that's tied to basically how hungry the animal is, yeah,

Marcelo Dietrich 44:03

and here is within seconds. So, and what was interesting, the experiment that we did, very, very much inspired by Harlow, was separate baby mouse activity of the AGP neurons is high, and then put it back different kinds of foster mothers and see what is important for suppressing the activity of the

Nick Jikomes 44:32

and I guess naturally to anticipate some of this, naturally knowing what we knew about AgRP neurons and the relationship to hunger, these are, you know people call them the hunger neurons. Basically, you might predict, I might have predicted, I think, that if you took the baby away from its mother, the mother predicts food for the baby very reliably. So the adrenals become more active. And if you then give it milk without the mother present, the neurons might get silent again, because the AgRP neurons are. Supposed to be all about hunger, exactly.

Marcelo Dietrich 45:01

So we did that the exact experiment that you said. So we pump milk back. Basically was an experiment that we have to take mother's milk. So we pump milk from the mother and give the milk back for the pub. So the pump will just their own

Nick Jikomes 45:18

milk. Yep, they're getting their natural mother,

Marcelo Dietrich 45:21

natural mouse milk. So we did that, and we also did a cow, whole milk from cow to have a different source of calories to control for that, and milk didn't suppress the activity of the neurons. So the babies, we filled the we filled their stomach while they were alone, and he did not into the activity of the neurons. And then we were puzzled, because he broke the broke our instinct. So

Nick Jikomes 45:46

these neurons remain active, the so called quote, unquote, hunger neurons remain active even though the animal's not hungry anymore, because you just gave them all the milk that their stomachs

Marcelo Dietrich 45:55

could hold. Exactly. It was around that point that the student that was with a grad student that was working my lab. His name is also Marcelo Marcel Zimmer, so we call him m2 and we were puzzled by that data. And then reading Harlow and all those studies, we say, how can we test that he was then, that we did the experiment that I was going to tell you before we isolate the infant, and then we gave different foster mothers. So instead of milk, we provide different forms of caretaking behavior. And we thought first we would provide a foster mother that we the mice, like females, when they are not lactating, the nipples are not protruded, so the mice cannot attach to it. They don't have nipples that the baby mouse can attach. But if you take a grandmother type of thing, they will have no nipples. But they are caretakers, so they really

Nick Jikomes 47:02

still feel the warmth of the Mother, the feelings of her fur, the tactile sensations, but there's no nipple to attach to.

Marcelo Dietrich 47:07

They will leak, they will groom. They will do everything that a mother will do, but there is no attachment to the nipple. So one type of foster caretaker was the grandma. We call the grandmother, the other was female. We know from our own experience with influence, but also from the animal research and the animal world that what's called the pacifier effect, and infants that's alone is in a distress. You give a pacifier, you stimulate the mouth and that calms down the baby. I often give that example for bat orphanage. In bat orphanage, one great way to raise bats that are orphans is actually to give a pacifier. They grow badly, so we thought that we need to give a pacifier to see if a pacifier will pacify the Egyptian. So one foster caretaker was a female mouse that had nipples, so now the babies could attach, but the nipple produced no milk. We experimentally generate that kind of female mouse. And the third type of foster caretaker was another mother, a different caretaker that had nipples and had milk. And then we asked which of the mothers will cause the largest inhibition against the prediction being that, if these neurons are hunger neurons, the presence of milk should suppress when it comes from the natural environment, the breast, when we pump, we had no effect, because was artificial, and the results were striking. If we give a caretaker a grandmother, no oral stimulations, just caring, we suppress the activity of the Egypt it was, let's say, 30% if that foster caretaker has nipples, therefore oral stimulation we suppress, let's say, 80% now, if the foster caretaker, if has nipples and milk, is still suppressed, 80% the presence of milk has no effect in further suppressing the AGP neurons. And that was really remarkable at that time, because you say, Wow, it's really it's that concept that we discussed before. It seems that the milk is insignificant to modulate the activity of this hunger neurons during the sensitive period of life in which caretaking nutrition is coming from the mother. It's being related by the mother. Yeah,

Nick Jikomes 49:42

yeah. So, I mean, it is sort of gets this idea that these neurons, they're obviously about hunger, but it's less about it's not completely about the presence or absence of calories. It's about the presence or absence of sensory stimuli that will predict whether you can acquire those. Things. And although this is counterintuitive from one perspective, based on what you said before, if you think about it in evolutionary terms, from an infant, you know, there's a 100% correspondence between the presence of the mother and the presence of the milk, so it sort of makes more sense that it would be sensitive to these predictive sensory cues.

Marcelo Dietrich 50:15

Yeah, absolutely. I think that's that's correct. Of course, there's some experiments that we haven't done and they should be done and they are very difficult to do. Yes, yes, hence why they haven't been done. And one alternative, so what you just said is the evolutionary explanation that we discussed before. So breast milk becomes irrelevant. The mother is a predictor. Hence, the proximal signal for the infant is the presence of the mother the milk. Milk has no regulatory function. Another explanation will be what we will call developmental explanation, or onto genetic explanation. It will be somewhat like this, the infant, the EGP neurons actually are responding to deprivation of nutrients, or, let's say, a metabolic challenge, yeah, that's somewhat similar to food deprivation, and very early, the mother provides the milk so fulfills that need and provides all of the sensory experience the animal learns, learns so rapidly, so strongly that for now only, only Use the proximal learning cues, and doesn't use the nutritional cues anymore because they are irrelevant. So we'll be, it will be that same evolutionary concept, but now placed in a developmental concept. So from birth, from birth to that point. So then we learn that contingency, learn that pairing, and now is not using the breast milk as a as a cube, so we don't know that. We don't think that's the case because of all these studies that I told you, but it's very hard to prove that something doesn't happen or something doesn't exist, and I think they're really the experiment that will be important to do is to be able to raise these infants in the absence of a mother, for example, with an artificial care to us, bottle feeding them from birth and then uncoupling maternal caretaking from ingestion, and see what happens. The problems with that, those are very challenging experiments to doing animal models. Second, when you do that, you cause a lot of collateral damage, because you, you are, you are breaking a contingency. That's, as we said, from the high roads work and others, is very important for development.

Nick Jikomes 52:57

Yeah, so, but I think another thing that's a little tricky here too, is we're talking about rodents. So they're smaller creatures. And obviously baby rodents are even smaller than adult rodents, and their metabolism is very fast, and they're very sensitive to, you know, things like temperature. So I would imagine that at least in principle, the brain of an infant's probably gonna learn early on that when you're detached from the mother, obviously you're you're away from the source of milk, but you also immediately have considerations, you know, for like hypothermia here, and so the brain could learn very quickly, when the temperature drops, I'm about to be nutrient deprived because mom's not around, or things like that,

Marcelo Dietrich 53:41

yeah, the way we think is very similar to that, but it's broader in terms is not only nutrient deprivation, but it's as is a metabolic challenge in general, the infant, if needs to produce more Heat, right? Will be a net. That's it. And originally, when we published that study in 2019 now, five years ago, we did test that idea that temperature was a an important component of that it is so. And how did we do that? Basically, remember, if we socially isolate the baby mouse, AJP neurons, the hunger neurons, quote, unquote, are highly active. So what we did was to provide a warm environment. If you provide a warm environment, that's the temperature of the nest, then you prevent that activation, or you block substantially that activation I see, we also did an experiment to test whether there was learn. In other words, did they learn that when they are alone, it's cold and that generates a temperature shock that needs to amplify metabolism. So it's a learn. So. What we did was to raise but mothers at thermonuclear temperatures. Thermal neutral is when you don't need to produce heat from your body to warm yourself up. So we raised mothers at the at 35 Celsius, so the temperature of the nest. They delivered the babies at their temperature. So the babies never saw anything lower than 35 Celsius. So they were always warm. They never had to produce heat. And 10 days later, around the time that we always that we did the other experiments, we separate them from the mother,

Nick Jikomes 55:32

yep, just like you did before, but now temperature is not a variable, yeah.

Marcelo Dietrich 55:35

So we separate at 35 Celsius, so they are just alone, or we separate them at the colder temperature, room temperature. So if we put these babies alone at 35 Celsius, we have a very minimum activation of the neurons, really, really, really minimal. So almost no activation. If we are if we put the babies at room temperature, we have an incredible activation, so they don't need to learn. The contingency is something that the system is capable of responding without experience in their gut. And

Nick Jikome