THE GREAT MICROBE BOTTLENECK

THE GREAT MICROBE BOTTLENECK 

We knew it was somewhere nearby, but the area didn’t seem quite right. Driving 

through miles of seemingly endless savanna in Uganda’s Queen Elizabeth National 

Park, we saw only a dozen or so trees, and not the right kind. These were short, 

solitary, wide-crowned trees, completely engulfed by the never-ending dry grass. 

Small groups of zebra and the unique Ugandan Kob antelope dotted the landscape. 

But this did not seem like a place for a rain forest—certainly not for chimpanzees. 

It was too open, too dry, too much of a, well, savanna. Yet as we came to the peak 

of an embankment, there it was—a massive gaping vein of green in the sea of yel- 

low grass. The Kyambura Gorge. 

The gorge, while not the only one of its kind, is unusual. Cut through the center 

by a river flowing from a rain forest some hundred or so miles away, it provides a 

unique microclimate—a well-hydrated strip in an otherwise dry landscape. Along 

that strip, rain forest trees and the animals that depend upon them slowly migrated 

downstream. This occurred over tens and hundreds of thousands of years, but 

now when you sit in the middle of the usually dry savanna habitat in Queen Eliz- 

abeth Park, you see a thriving rain forest, complete with chimpanzees. Effectively, 

you see the finger of a faraway forest snaking its way into the savanna.

This gorge provides a unique interface. For contemporary researchers, it pro- 

vides a fairly easy way to follow chimpanzees. By simply driving along the 

unobstructed sides of the gorge, we can follow the calls of chimpanzees and then 

dip down into the gorge to locate them. This is a far cry from the much more chal- 

lenging work of chasing them through the forest on foot. For the chimpanzees, the 

gorge provides something much more meaningful. Instead of the few grassland 

edges in a normal chimpanzee habitat, river gorges like Kyambura provide miles 

and miles of savanna border along a relatively typical chimpanzee habitat, allowing 

chimpanzees to explore and utilize the grassland much more than their rain forest 

contemporaries. And use it they do. Some populations enjoy spending a good deal 

of time in the savanna, even hunting savanna animals. 

Sometime after the split between the lineages that would lead to chimpanzees 

and bonobos on the one hand and humans on the other, our ancestors embarked 

upon a trajectory that would include a series of changes moving them away from 

the lifestyle of the common ancestor. Sitting on the edge of Kyambura Gorge, it is 

hard not to consider one of the most dramatic of these changes: the shift from 

being a primarily forest-based animal to an animal with the capacity to live in and

utilize grassland. While the order of the events remains somewhat opaque, at some 

point our ancestors began to make forays into the savanna. This move would ulti- 

mately alter their microbial repertoire and their future. 

As contemporary humans, we normally think of chimpanzees and bonobos, if we 

think of them at all, as a side-note species. They are interesting animals, certainly, 

with much to teach us about our history, yet they hover somewhere near the brink 

of extinction and live in marginal forest habitats, certainly not species that could 

compete with humans. Shocking as it may seem, this was not always the case. If 

we could see the world millions of years ago, say around that time that human lin- 

eages and chimpanzee/bonobo lineages diverged, it would be very different. Six 

million years ago, it was an ape’s world. 

In our modern world with over six billion humans and only an estimated one to 

two hundred thousand chimpanzees and ten thousand bonobos, humans have 

reached every point on Earth, and all of the wild chimpanzees and bonobos remain 

confined to central Africa. We have to stretch our brains to imagine a world in 

which we were the minority. Yet for some periods prior to the advent of agriculture 

around ten thousand years ago, that was exactly the world in which our ancestors 

lived. 

Chimpanzees and bonobos are not fossils. Contemporary species, whether 

chimpanzees, bonobos, or humans, have all changed since this aforementioned 

ancient time. Nevertheless, around six million years ago when our own ancestors 

took their first tentative steps toward becoming human, they would have seemed 

much closer to our chimpanzee and bonobo relatives than to ourselves. Our rela- 

tives at that time were almost certainly covered with thick hair. When on the 

ground, they primarily moved around by walking on all fours but really spent the 

majority of their time in the trees. They hunted—collective and strategic hunting as 

we’ve seen. But they didn’t cook their meat, didn’t use tools that couldn’t be simply 

modified from tree branches, and kept largely to the forest.

As our own lineage changed and began to display some of the features that we 

equate with humans, the world was a different place. The use of grasslands was 

perhaps not completely foreign, and even today some small groups of chim- 

panzees utilize mosaic environments with forest as well as grasslands, like the 

Kyambura chimpanzees, yet they did not likely make long journeys into these habi- 

tats. At that early time, the individuals that spent time in grasslands were the odd 

ones. 

Often when groups of individuals veer into new areas, they do so to escape in- 

tense competition, and as our own ancestors moved to savanna habitats, they 

probably did so less to break new ground than simply to find somewhere they 

could exist with fewer rivals. Such habitat moves often led to marked inefficiencies, 

and when our early ancestors relocated, they probably experienced profound disad- 

vantages. Being unsuited, at least at first, to function in grasslands, our early 

ancestors suffered a number of consequences that likely included smaller popu- 

lation sizes. Or near extinction. 

Determining historic population sizes, particularly prior to periods in which we 

have written records, is fraught with difficulty. But studies suggest that our ances- 

tors’ population densities were sometimes very low, with lower numbers than 

those of current gorilla and chimpanzee populations, and at least once teetered on 

the brink of extinction. Our ancestors would have been an endangered species. We 

believe this to be the case because our genes maintain some of these records, and 

by comparing the genetic information among contemporary human populations 

with that of our close ape relatives, we can tease out inklings of relevant infor- 

mation. 

The information revealed is striking. Analyses of the human mitochondrial 

genome, a region of genetic information that passes only from mother to daughter, 

as well as studies of mobile genetic elements that accumulate in regions of the 

genome in a clocklike way provide clues to our historic population size and

suggest it was much smaller than we might expect. 

Our preagricultural ancestors likely lived in small groups, which is not neces- 

sarily surprising. Most of our evolutionary history as primates was spent in forest- 

ed environments. And while exact timelines for the main events remain unknown, 

moving from forested environments to savanna habitats, shifting from largely fixed 

territories to a more nomadic lifestyle, and adapting to the various new conditions 

imposed by these changes must have been traumatic. An apt comparison might be 

the idea of contemporary humans living on Mars. The generations of our ancestral 

populations that confronted the savanna frontier probably did so at some cost. But 

our interest in small population sizes here is less about the consequences for hu- 

mans and more about the consequences for microbes. 

Low population densities, such as those exhibited by our ancestors, have a marked 

impact on the transmission of infectious agents. Infections need to spread. If 

population sizes are low, it is much harder for this to happen. The scientific term 

for substantially reduced population sizes is population bottlenecks, and when 

population bottlenecks occur, species should be expected to lose their microbial 

diversity. 

Microbes can be largely divided into two different groups, acute and chronic, 

and each group is impaired in small host populations. In the case of acute agents 

(like measles, poliovirus, and smallpox) infections are brief and lead either to 

death or immunity from future infections: they kill you or make you stronger. Acute 

microbes require relatively large populations; otherwise, they will simply burn 

through the susceptible individuals, leaving only the immune or the dead. In either 

case, they go extinct. If there’s no one left to infect, that’s the end of the line for a 

microbe. 

Chronic agents (like HIV and hepatitis C virus), unlike acute agents, do not lead 

to long-lasting immunity in their hosts. They hang on to their hosts, at times hold- 

ing on for a host’s entire lifetime. These agents have a better capacity than acute

suggest it was much smaller than we might expect. 

Our preagricultural ancestors likely lived in small groups, which is not neces- 

sarily surprising. Most of our evolutionary history as primates was spent in forest- 

ed environments. And while exact timelines for the main events remain unknown, 

moving from forested environments to savanna habitats, shifting from largely fixed 

territories to a more nomadic lifestyle, and adapting to the various new conditions 

imposed by these changes must have been traumatic. An apt comparison might be 

the idea of contemporary humans living on Mars. The generations of our ancestral 

populations that confronted the savanna frontier probably did so at some cost. But 

our interest in small population sizes here is less about the consequences for hu- 

mans and more about the consequences for microbes. 

Low population densities, such as those exhibited by our ancestors, have a marked 

impact on the transmission of infectious agents. Infections need to spread. If 

population sizes are low, it is much harder for this to happen. The scientific term 

for substantially reduced population sizes is population bottlenecks, and when 

population bottlenecks occur, species should be expected to lose their microbial 

diversity. 

Microbes can be largely divided into two different groups, acute and chronic, 

and each group is impaired in small host populations. In the case of acute agents 

(like measles, poliovirus, and smallpox) infections are brief and lead either to 

death or immunity from future infections: they kill you or make you stronger. Acute 

microbes require relatively large populations; otherwise, they will simply burn 

through the susceptible individuals, leaving only the immune or the dead. In either 

case, they go extinct. If there’s no one left to infect, that’s the end of the line for a 

microbe. 

Chronic agents (like HIV and hepatitis C virus), unlike acute agents, do not lead 

to long-lasting immunity in their hosts. They hang on to their hosts, at times hold- 

ing on for a host’s entire lifetime. These agents have a better capacity than acute  population bottleneck: a diverse population (top) is greatly diminished by a near- 

extinction event (middle), resulting in a more homogenous population (bottom). 

(Dusty Deyo) 

Sometime following the split between the chimpanzee/bonobo lineage and our 

own, another important change occurred in our ancestors that would have dra- 

matic consequences for our microbial repertoires: they learned to cook. Not 

Michelin three-star cuisine, of course, but cooking nonetheless: using heat to pre- 

pare food. Exactly when our ancestors harnessed the power of fire remains a mys- 

tery. Presumably, fire first provided warmth and security from predators and com- 

petitors. Yet it appears to have quickly become a profound way of altering food. 

Richard Wrangham, my mentor from Harvard, discusses cooking and its conse- 

quences in depth in his well-researched book Catching Fire: How Cooking Made Us 

Human. Among other things, he analyzes in detail cooking’s origins. 

When our ancestors began to cook extensively, in addition to the advantages 

that cooking offered them by making food more manageable and palatable, they also benefited from its remarkable ability to kill microbes. While some microbes 

can survive at incredible temperatures (such as the hot spring microbial hyperther- 

mophiles that grow and reproduce at temperatures above the boiling point of 

water), the vast majority of microbes that make their living off of animals cannot 

survive the temperatures associated with cooking. As microbes are heated during 

cooking, their normally solid, densely packed proteins are made to unfold and 

open, allowing digestive enzymes quick and easy access to destroy any capacity to 

function. As with the population bottlenecks that our ancestors swung through, the 

cooking that became their standard way of life served to again diminish their up- 

take of new microbes, helping limit their microbial diversity. 

The earliest solid evidence that humans controlled fire comes from archaeo- 

logical finds in northern Israel where burned stone flakes dating back almost eight 

hundred thousand years were found near fire pits. This is almost certainly an un- 

derestimate. African sites dating to over a million years ago contain burned bones 

that could be the remains of cooking, yet the lack of archaeological evidence makes 

these finds more ambiguous. In Wrangham’s analysis, the evidence of cooking 

goes back much further. By examining the remains of our ancient ancestors, pa- 

leontologists have found physiological clues indicating that they consumed cooked 

food. For example Homo erectus, a human ancestor from 1.8 million years ago, had 

exactly the larger bodies and smaller digestive tracks and jaws to imply that they 

consumed higher-energy diets that were easy to chew and easy to digest—in other 

words, foods that had been cooked. 

Whatever the exact date of our ancestors’ culinary dawn, it has certainly ex- 

ploded since then. Cooked foods make up the vast majority of contemporary diets. 

In my work with hunters around the world, I’ve had a chance to sample from a vast 

range of these foods—from roasted porcupine and python in Cameroon to fried 

wood grubs in rural DRC. On one occasion, my “friendly” Kadazan collaborator in 

Borneo even gave me dog stew as a practical joke (I didn’t really see the humor). 

I’ve had a chance to sample food far beyond the beef, lamb, and chicken staples 

that I grew up eating in America. Yet no matter what I’ve eaten, or where I’ve eaten it, one thing can be certain: if the food has been cooked sufficiently, the likelihood 

that it will make me sick is small. 

The dual factors of diminished population sizes and cooking were not the only 

things that served to decrease the microbial repertoires of our early ancestors. The 

transition from rain forest habitat to a savanna habitat meant different vegetation 

and climate but also an entirely different set of animals to interact with and hunt. 

And different animals meant different microbes. 

While we still understand very little about the ecological factors that lead to 

microbial diversity, there are some key factors that certainly play a role. We know, 

for example, that the biodiversity of animals, plants, and fungi supported by trop- 

ical rain forest systems is higher than any other ecosystem on land. When our 

ancestors left the rain forest, they entered into regions with diminished biodi- 

versity. The diversity of microbes would almost certainly have been reduced, as 

would the diversity of the host animals that they infected. So the savanna grassland 

habitats likely housed fewer animals and a lower diversity of microbes capable of 

infecting them, which in turn contributed to lower microbial repertoires for our 

ancestors. 

The kinds of animals living in the savanna also differed in critical ways from 

those in the forests, including a marked contrast in the diversity of apes and other 

primates. Simply put—primates love forests. The king of the jungle is a primate, 

not a lion. While some primates, like baboons and vervet monkeys, live very suc- 

cessfully in savanna habitats, forest regions trump savanna regions in terms of pri- 

mate diversity. When we consider the microbes that could most easily infect our 

ancestors, the diversity of primates in any given habitat plays an important role. 

They are certainly not the only species that contribute to our microbial reper- 

toires—in my own studies, I focus not only on primates but also on bats and ro- 

dents—but they do play an important role. 

Some years ago, I began considering what factors might improve or decrease the 

chances that a microbe would jump from one host to the next successfully enough 

to catch on and spread in the new host. It may seem that bats and snakes, for 

example, would provide similar sources for novel microbes. Yet there is a strong 

argument against this idea. Long evident to those doing work on microbes in lab- 

oratories is the fact that closely related animals have similar susceptibility to cer- 

tain infectious agents. So a mammal, like a bat, would have many more microbes 

that could be successfully shared with a human than a snake. If not for the logis- 

tics and ethics, chimpanzees would make the ideal models for studying just about 

every human infectious disease. As our closest living relatives, they have nearly 

identical susceptibility to the microbes that infect us. Over time, less and less labo- 

ratory research on human microbes is conducted in chimpanzees, but this is large- 

ly because of the valid ethical concerns associated with conducting research on 

them and the difficulty of controlling these large and aggressive animals in cap- 

tivity. 

Closely related animal species will share similar immune systems, physiologies, 

cell types, and behaviors, making them vulnerable to the same groups of infectious 

agents. In fact, the taxonomic barriers that we place on species are constructs of 

our own scientific systems, not nature. Viruses don’t read field guides. If two dif- 

ferent hosts share sufficiently similar bodies and immune systems, the bug will 

move between them irrespective of how a museum curator would separate them. I 

named this concept the academically accurate but unwieldy taxonomic transmission 

rule, and it holds up for chimpanzees and humans as it would for dogs and 

wolves.¹ The idea is that the more closely related any two species are, the higher 

the probability that a microbe can successfully jump between them. 

Most of the major diseases of humans originated at some point in animals, 

something I analyzed in a paper for Nature, written with colleagues in 2007. We 

found that among those for which we can easily trace an animal origin, virtually all

came from warm-blooded vertebrates, primarily from our own group, the mam- 

mals, which includes the primary subjects of my own research, the primates, bats, 

and rodents. In the case of primates, while they constitute only 0.5 percent of all 

vertebrate species, they seeded nearly 20 percent of major infectious diseases in 

humans. When we divided the number of animal species in each of the following 

groups by the number of major human diseases they contributed, we obtained a 

ratio that expresses the importance of each group for seeding human disease. The 

numbers are striking: 0.2 for apes, 0.017 for the other nonhuman primates, 0.003 

for mammals other than primates, and a number approaching 0 for animals other 

than vertebrates. So as our early ancestors left the primate-packed rain forests and 

spent more time with lower overall primate biodiversity in savanna habitats, they 

moved into regions that likely had a lower diversity of relevant microbes. 

Multiple factors likely conspired to decrease the microbial repertoires of our early 

ancestors. As they spent more time in savanna habitats, our early ancestors inter- 

acted with fewer host species, and those hosts were on average more distantly re- 

lated to them. The advent of cooking increased the safety of meat consumption 

and stopped many of the microbes that would have normally crossed over during 

the course of hunting, butchering, and ingesting raw meat. And the population bot- 

tlenecks that our ancestors went through further winnowed down the diversity of 

microbes that already infected them. All in all, the conditions associated with 

becoming human served to decrease the diversity of microbes present in our an- 

cient relatives. Though many microbes undoubtedly remained in our early ances- 

tors, there were likely far fewer than those that were retained in the separate lin- 

eages of our ape relatives. 

During the time that our own ancestors went through their microbial cleansing, 

their ape cousins continued to hunt and accumulate novel microbes. They also 

maintained microbes that would have been lost in our own lineage. From a human 

perspective, the ape lineages served as a repository for the agents we’d lose—a microbial Noah’s ark of sorts, preserving the bugs that would disappear from our 

own bloodlines. These great ape² repositories would collide with expanding human 

populations many centuries later, leading to the emergence of some of our most 

important human diseases. 

Perhaps the single most devastating infectious disease that afflicts humans today 

is malaria.³ Spread by mosquitoes, it is estimated to kill a staggering two million 

people each year. Malaria has had such a profound impact on humanity that our 

own genes maintain its legacy in the form of sickle cell disease. Sickle cell, a ge- 

netic disease, exists because its carriers are protected from malaria. Protection was 

so important that natural selection maintained it despite the debilitating disease 

that appears in approximately 25 percent of the offspring of couples that each carry 

the gene. People who are afflicted with sickle cell have their origins almost exclu- 

sively in one of the world’s most intensely malaria affected areas—west central 

Africa. 

My interest in malaria is both personal and professional. During my time work- 

ing in malaria-infested areas of Southeast Asia and central Africa, I was infected by 

it on three separate occasions. On the last of those occasions, I almost died. The 

first two times I’d had malaria were both in regions where malaria was common. 

I’d exhibited all the typical symptoms—severe neck ache (similar to how you’d feel 

if you slept in a strained position) followed by intense fever and profuse sweating. 

On each of my first two bouts, I simply went to a local doctor and received a quick 

diagnosis and treatment. While the pain and illness were miserable, they both re- 

solved reasonably quickly. 

I was in complete denial at the time I had my third round with this deadly dis- 

ease. I wasn’t in the tropics; I was in Baltimore! I had returned from Cameroon to 

do research at Johns Hopkins University, and I had very different symptoms, led by 

intense abdominal pain. I must have also had fever since I remember complaining 

to friends who were putting me up in their local bed and breakfast that my room was too cold. These new symptoms and the fact that I’d left Africa many weeks ear- 

lier fed my denial that this could possibly be malaria. I finally realized I needed ur- 

gent care while sitting half delirious in a tub of scalding water and watching the 

overflow hit the floor of my friends’ bathroom. Although I recovered after a few 

days in the hospital, the illness brought home for me the huge impact that this dis- 

ease has on the millions of people who are regularly sickened by it. 

My professional interest in malaria had started much earlier. As a doctoral stu- 

dent studying the malaria of orangutans in Borneo, I’d had the good fortune to 

spend a year working with some of the world’s foremost experts on malaria evolu- 

tion at the CDC in Atlanta. There I had the luxury of spending afternoons with Bill 

Collins, perhaps the world’s greatest expert on the malaria parasites of primates, 

discussing how malaria might have originated. Among the prominent themes of 

our chats was the importance of wild apes. 

At the time, we knew that wild apes had a number of seemingly distinct malaria 

parasites. One of them was particularly intriguing. Plasmodium reichenowi was 

named after a famous German parasitologist, Eduard Reichenow, who had first 

documented the parasites in chimpanzees and gorillas in central Africa. Reichenow 

and his contemporaries saw a number of these particular parasites, collector’s 

items for the German researcher, and correctly identified them through exami- 

nation by microscope as closely related to our own Plasmodium falciparum. In the 

1990s, during my time at the CDC, molecular techniques were paving the way to 

detailed examination of these parasites, allowing us to compare them accurately to 

our own parasites and providing much greater evolutionary resolution than a 

microscope could ever offer. Sadly, all of the parasites of Reichenow’s time had 

been lost, and all that remained was a single lone specimen. 

Initial work with this lone P. reichenowi parasite showed that in fact it was the 

closest of the many primate malarias to our own deadly human malaria, P. 

falciparum. Yet with only a single specimen, it remained impossible to say much 

about the origins of these parasites. Perhaps, long ago, the common ancestor had 

a parasite that over millions of years had gradually evolved into distinct lineages of P. reichenowi and P. falciparum, a hypothesis favored by some at the time. Or per- 

haps the ape parasite simply resulted from the transmission of the common 

human parasite to wild apes at some point in fairly recent evolutionary history. A 

third possibility, neglected by most considering the huge number of humans and 

the incredible proliferation of P. falciparum among them compared to the existence 

of only a few dozen known parasites in apes, was that perhaps P. falciparum was in 

fact an ape parasite that had moved over to human populations. 

Bill and I understood that to truly address the evolutionary history of these para- 

sites we’d need to get more samples from wild apes, ideally many. As a young doc- 

toral student, I was ambitious yet still naïve about the difficulties associated with 

getting these kinds of samples. But I promised Bill I’d do it and set about planning 

ways to sample apes in the wild. 

Unbeknownst to me at the time, I was about to be called away by my soon-to-be 

postdoctoral mentor Don Burke to conduct research in Cameroon. I was unaware 

at the time that I’d spend nearly five years establishing a long-term 

infectious-disease-monitoring site there in Cameroon. Eventually, though, I did fol- 

low through on my promise to Bill and got those ape samples. Ultimately, in col- 

laboration with sanctuaries in Cameroon that helped to provide homes to orphan 

chimpanzees, we discovered that ape malaria parasites were not as uncommon as 

people had suspected. By teaming up with Fabian Leendertz, a veterinary virologist 

who had done similar work in the Ivory Coast, molecular parasitologist Steve Rich, 

and the legendary evolutionary biologist Francisco Ayala, we took an important 

step toward cracking the origin of this disease. 

Together we were able to compare the genes in hundreds of human P. falci- 

parum samples that already existed with around eight new P. reichenowi specimens 

from chimpanzees in locations throughout west Africa. The genetic comparison 

surprised us all. Amazingly, we found that the entire diversity of P. falciparum (the 

human malaria) was dwarfed by the diversity of the handful of P. reichenowi chim- 

panzee parasites we’d managed to uncover. This discovery told us that the most 

compelling explanation for P. falciparum was that it had been an ape parasite and only jumped over to humans through a bite by some confused mosquito, some- 

time after our split with the chimpanzee lineage. Human malaria had, in fact, origi- 

nated in wild apes. In the years that followed our work, a number of researchers 

documented more and more of the parasites in wild apes. 

Subsequent work by my collaborators Beatrice Hahn and Martine Peeters (the 

same scientists who have done work on SIV evolution) has shown that the malaria 

parasites infecting wild apes are even more diverse than our study indicated. They 

have shown that the ape parasites most closely related to human P. falciparum exist 

in wild gorillas, rather than chimpanzees. How these parasites have been main- 

tained among wild apes and whether or not they’ve moved back and forth between 

chimpanzees and gorillas remain questions for future studies. Either way, there is 

no longer any doubt that human P. falciparum moved from wild apes into humans 

and not in the opposite direction. 

That malaria crossed from a wild ape into humans makes great sense when viewed 

from the perspective of the evolution of our lineage. The microbial cleansing that 

resulted from habitat change, cooking, and population bottlenecks among our own 

ancestors had cleared our microbial slate, decreasing the diversity of microbes that 

were present before. Perhaps the many years with leaner microbial repertories had 

also decreased selective pressure on the many innate mechanisms that we have to 

fight against infectious diseases, effectively robbing us of some of our protective 

disease-fighting tactics. 

In more recent times, as our population sizes began to increase, wild ape dis- 

eases, some of which we’d lost millions of years earlier, had the potential to infect 

us again. When these diseases reentered humans, they acted on us like uniquely 

suited novel agents. Malaria was not the sole microbe to make the leap from apes 

to modern humans, and the stories of others, like HIV, tell a strikingly similar tale. 

The loss of microbial diversity in our early ancestors and the resulting decrease in 

their genetic defenses would make us susceptible to the microbial repositories that our ape cousins maintained during our own microbial cleansing. While we con- 

tinued to change as a species, yet another part of the stage would be set for the 

brewing viral storm.