Hacker News Reader

Comment by mikestew | Hacker News Reader

This is one of those ideas that sounds genius on its face, until it's actually implemented. Kudzu? Grows fast, prevents erosion, let's pay farmers to till it into the top soil. MTBE? Prevents engine knock, makes for cleaner air, let's mandate its use at the federal level. Whoopsie, once it's in the water we can't get it out, and its a carcinogen.

Let's kill all of the mosquitos because we find their presence unpleasant. Well, that's done and...oh, shit. Turns out there was a value to mosquitos after all. Anyone think to save some of that DNA?

View on HN

Chairman Mao thought killing all the sparrows in China (part of his Four Pests Campaign[1]) would be a great idea because they ate laborers' grain seed. But then the population of crop-eating insects ballooned, causing the Great Chinese Famine to get a lot worse. Tens of millions starved to death.

[1]: https://en.wikipedia.org/wiki/Four_Pests_Campaign

Related: because there was a bounty on rats the farmers started growing rats secretly.

https://en.wikipedia.org/wiki/Cobra_effect

Mosquitos aren't about unpleasantness though: It is a vector for deadly disease in the tropics. We are talking hundreds of thousands of deaths a year.

Now, getting rid of all of them might not be the smartest idea: it carries plenty of risk, but to say that they are just unpleasant is a major understatement.

We should just cure these diseases. Vector control is fucking hard, we've been trying it for a century without really fixing the problem. That isn't likely to change; therapeutic interventions, however, can keep getting better and cheaper.

With the risk being that we breed immunity into the population, seeing as it's still being carried and evolving in its carriers (the mosquitos).

FWIW, we could "cure" a lot of these diseases. We have, in rich parts of the world. If the option was between "we distribute cures to forgotten diseases to everyone who needs them" versus "we take a gamble and murder all the mosquitos," I might agree the former would be the better bet.

That's not the choice we're given because not enough people in the first world want to step up in a serious enough way.

It doesn't really take that many people; these need not be expensive programs. It's a matter of political will, but a serious vector control program would be no different.

In any event, learning how to control human diseases and prevent parasites from killing us seems like a very fundamental goal of medicine which should be accomplished in any eventuality. Eradicating mosquitoes need not be.

Also, regarding resistance, it is much easier to track and defeat resistance in human patients than to track and defeat resistance in wild populations of mosquitoes, which is what we'll be doing if we try eradication campaigns.

" Vector control is fucking hard, we've been trying it for a century without really fixing the problem. That isn't likely to change"

With CRISPR and gene drive, it changes from "fucking hard" to "trivial", so it is not only likely to change, it's already changed.

Can you show me an actual program using CRISPR and gene drive that's produced some effective changes? As far as I know this is still in the realm of theory. And given that we're talking about a single resistance mechanism, I'm going to guess that it will very quickly be defeated by evolution.

DDT was the "perfect" mosquito killer, as well. It didn't last - resistance inevitably follows these attempts in wild populations.

Can you show me an actual program using CRISPR and gene drive that's produced some effective changes?

So your argument is that it can't be done because no one has done it?

No one has released gene drive systems to the wild, but there is every reason to believe that it will work fine, and no credible reason to believe that it won't.

Err, I can give you a credible reason to believe that it won't: the parasite will inevitably develop resistance.

Here is the paper where they talk about the mechanism of resistance being conferred on the mosquito: http://www.pnas.org/content/109/28/E1922.abstract

The mechanism is this: "These scFvs are derived from antibodies specific to a parasite chitinase, the 25 kDa protein and the circumsporozoite protein, respectively."

So you'll spend a ton of money building a fancy CRISPR system in your mosquitoes, release them into the wild, and in a matter of months you will have parasites with on-target mutations in these proteins that will allow them to evade your resistance mechanism. I'd lay $1000 on this without blinking.

You're basically talking about curing malaria in mosquitoes. Why not, instead, just cure malaria in humans?

First, you're confusing wiping out the mosquito species with giving the mosquitos resistance to the malaria parasite. Those are two different things.

Second, there is no reason that a CRISPR-based system is limited to a single target. Will you also lay $1,000 against a system that targets ten species-unique sequences at once?

You're basically talking about curing malaria in mosquitoes.

Neither I nor the article is talking about that. It's a discussion of making the mosquito species itself extinct.

> First, you're confusing wiping out the mosquito species with giving the mosquitos resistance to the malaria parasite. Those are two different things.

You're the one who brought up CRISPR and gene drive. Perhaps you should read the actual paper: http://www.pnas.org/content/112/49/E6736.full; you'll see that the proposal is entirely about giving mosquitoes resistance to the malaria parasite, not about eradication. That is, the goal is to eliminate the parasite in mosquito populations (i.e., cure malaria in mosquitoes), not to kill mosquitoes.

>Second, there is no reason that a CRISPR-based system is limited to a single target. Will you also lay $1,000 against a system that targets ten species-unique sequences at once?

Yes. On-target mutations are trivial to produce, and alleles segregate independently.

"You're the one who brought up CRISPR and gene drive. "

Which can be targeted to eliminate the mosquitos themselves.

"Perhaps you should read the actual paper"

The actual paper? Like there's only one? Hint: there's more than one way to use this technology, and more than one group working with it.

"On-target mutations are trivial to produce, and alleles segregate independently."

I think you're misunderstanding what "independently" means in this context.

If the probability of a mutation that will get around one targeted sequence is (say) 1 in a million, that's almost certainly going to happen, just because there are billions of mosquitos.

However, if you target (say) ten independent sequences, the probability of any one organism having resistance to all of them is going to be 1 in (1 million)^10 = 1 in 10^60 and that is basically not going to happen. It does no good in this case for one organism to be resistant with respect to one target, while another organism is resistant with respect to another target, because all of the targets will have fatal outcomes. The only way for the organism to survive would be for it to be resistant to all of them at the same time, from the beginning.

And there's no reason why you'd have to stop at 10, either.

You should go read about MRSA, which shouldn't exist according to your logic. You also are misunderstanding how independent assortment of alleles work.

Because plasmodium is a eukaryotic species which reproduces sexually, on-target resistance mutations to any number of mechanisms can arise independently in a bunch of different organisms and accumulate through selection + allele segregation. It's also very easy to produce these sorts of mutations, since it is trivial to change an amino acid to disrupt antibody binding/recognition without altering the function of the protein.

>Which can be targeted to eliminate the mosquitos themselves.

This is incorrect; the whole point of a gene drive is that it causes increased propagation of a trait in the mosquito population. What you're describing is a very different strategy, since a trait that kills the mosquito obviously cannot propagate. It can also be achieved much more simply by using sterile males to outcompete fertile males and reduce the population ('sterile insect' technique); however, this technique only works on small populations and almost certainly wouldn't work in Anopheles or some such.

> With CRISPR and gene drive, it changes from "fucking hard" to "trivial", so it is not only likely to change, it's already changed.

The kind of hubris displayed here almost certainly belies deep ignorance. Please, doctor, tell us how trivial editing DNA is.

>We should just cure these diseases.

Much, much easier said than done.

It is just the application of money, time and effort. Total global funding for malaria research is ~ half a billion dollars. I imagine if we, say, multiplied this by ten, it would have a susbstantial effect on improving outcomes.

This could easily be framed as a misattribution of cause though.

Mosquitos occur world-wide, and even flourish in many climates where rates of infection from mosquito-borne diseases are relatively low. As an example, 90% of deaths from Malaria occur in sub-Saharan Africa, and even there, the likelihood of death from infection is radically lower for western visitors than for locals.

These diseases are largely a result of poverty and poor access to healthcare, not of mosquitos.

So we get as many cases of malaria in Europe, just nobody dies from it? Or maybe it does indeed has something to do with mosquitos after all?

There was malaria in Sweden a couple of hundred years ago, and we still have a few mosquitos. It was never a big problem but malaria went away when it couldn't spread efficiently, largely because of better houses and hygiene such as not keeping the cattle in the same house (room) as the people. Also a lot of breeding ground was dried out and converted to farm land.

I'm all for releasing CRISPR on the worst species of mosquitos though. And ticks. They can't serve any function other than killing a few persons each year due to TBE.

It's sad when a chain of logical reasoning points out (again) how weird it is that we cannot solve the real problem, which is "Why is there a third world in the first place?"

Why does there have to be a region in permanent squalor? Why can't every place on earth where civilization exists be a decent place for humans to live, with peace, opportunity and sanitation?

But ... back down to earth. Since we can't fix that larger problem, at least maybe we can stop malaria by killing all the mosquitoes.

Malaria and other mosquito-borne diseases are actually a big reason why there is a third world.

Really? How so? (Not attacking ... sincere question).

Seems to me that good second-world sanitation might alleviate it. But third world countries lack the sanitation and hence suffer Malaria. Seems the problem is political/social.

anti-mosquito sanitation is relatively expensive, compared to other forms of sanitation.

I'd guess that the climate in Sweden helps too, so I still think we need CRISPR.

Coincidentally, there are quite many areas in Sweden were it is impossible to go outside in the summer without mosquito nets, and it is worsening.

Actually there are some in Crimea and some in Bulgaria. Literally less than 100 cases in 10 years, mostly imported, not endemic. Treated in hospitals.

The WHO recently declared Europe "malaria free". http://uk.reuters.com/article/us-health-europe-malaria-idUKK...

It seems that at least some of the reason why malaria is so prevalent in sub-Sahara Africa has to do with poor housing, poor management of drainage / sewers, and poor roads.

http://www.hindawi.com/journals/jtm/2012/819563/

I'm not ruling out eliminating mosquitoes per se, but the United States and Europe managed to largely eliminate malaria in the 1930s-1940s without eliminating mosquitoes. So before I'd start advocating the elimination of entire subsets of species, I'd ask how the US and Europe were successful in their efforts, and wonder why we can't apply those lessons to the rest of the world. Maybe the money spent in targeting the elimination of mosquitoes (which cause a lot of problems) would be better spent in general health and sanitation (the lack of which also causes a lot of problems), for instance.

I'm not sure where I suggested there are as many cases in Europe?

Mosquitoes do exist in Europe; they don't spread malaria because it isn't there to spread, and it isn't there to spread because of effective healthcare.

(Europe isn't a great example as the mosquito population is not exactly high, which I guess is climate-related, but the above is also true of parts of the developed world where their populations are higher)

Exactly, we are talking about millions of people if you go back a few thousand years.

Millions of people if you go back a few years

You know, mathematically speaking, each of us had about 20 million ancestors, if we go back in time. All of them dead now. So what?

People living in malaria zones find mosquitos a little more than "unpleasant".

Same with Dengue Fever. I spent a month in Bali last summer with a group of 40 people, 4 of which got Dengue. 10%.

What were you doing there?

Hopefully vacationing. Bali is beautiful.

Mongolia is beautiful too. No dengue, no malaria.

Being a "digital nomad"? :)

MBTE is unconfirmed as a carcinogen except at very high doses and can be efficiently and economically removed from water with simple bioreactors. It is better than either lead additive or the health effects of increased pollution from inferior combustion from increased knocking. It is in fact used medically to dissolve gallstones by way of injecting it into the gallbladder. It is not required to be added to gasoline; while there are laws requiring oxygenates to be added, MTBE is not required and there are alternatives. Do your research better.

MTBE...can be efficiently and economically removed from water with simple bioreactors.

That wasn't the case fifteen years ago. Mandating something, then hoping we come up with an inexpensive way to clean up the mess later probably isn't a good long-term plan.

It is in fact used medically to dissolve gallstones by way of injecting it into the gallbladder.

Arsenic has medicinal uses, too. I'll swing by the house later to drop some in the water inlet of your house.

MTBE is not required and there are alternatives

When the alternatives are more expensive, MTBE has been effectively "required". Perhaps it is not the current case, but it most certainly was in years past.

Do your research better.

You mean do my research such that my conclusions match yours? Or did you just have a nice, cold glass of Uncalled-for-Snark(tm) with your breakfast this morning?

I did some more looking around. Oral MTBE is a carcinogen at levels above 200mg/kg/day in rats. When inhaled it has acute toxic effects before it has carcinogenic effects. MTBE tastes bad enough to render drinking water unpalatable at levels around 10ug/l. So you basically can't drink enough to give yourself cancer.

Calling it a persistent pollutant is perfectly fine. It's bad enough to make entire aquifers taste bad until you put in the hardware to clean it out. What it isn't is a problematic carcinogen.

"Because of the intense odor (and taste) of MTBE, humans will not tolerate either air or water concentrations sufficient to produce the cytotoxic precursors required to promote cellular proliferation." http://onlinelibrary.wiley.com/doi/10.1111/j.1539-6924.1997....

> Arsenic has medicinal uses, too.

Not for its macroscopic physical properties it doesn't.

> I'll swing by the house later to drop some in the water inlet of your house.

Just be sure to stick to things like the EPA exposure limits and so on.

> That wasn't the case fifteen years ago. Mandating something, then hoping we come up with an inexpensive way to clean up the mess later probably isn't a good long-term plan.

That's fair.

> You mean do my research such that my conclusions match yours? Or did you just have a nice, cold glass of Uncalled-for-Snark(tm) with your breakfast this morning?

I'll admit that a little bit of it is snark. I find that the kind of person that's most vulnerable to misleading science is also pretty vulnerable to snark. But, no, I really did mean that you needed to do more research.

I am all for environment and sustainability, but in case of mosquitoes, even I would draw an exception! You have to be in a tropical country to know how many nights they spoil and how much effort it is, everyday, to keep yourself insulated -- all those Aerosols, sprays, nets... So much effort and resources. They are under the desks, in corners, in gardens, everywhere.. they seriously affect the lifestyle in a bad way.

There will be side effects, but could it really be worse than the alternative? That is, one million deaths per year.

They very well could be. That's just it -- we don't know. And when we kill them all off, are we going to be able to reverse it when we find out that the result is causing two million deaths per year?

Which is why people are studying the issue, writing papers and popular articles, instead of just doing it.

Despite the link-bait headline, no one is talking about killing off all mosquitos.

And yes, gene drives can be reversed.

Under what hypothetical scenario would killing off mosquitos cause two million (human?) deaths per year?

What if the diseases that depend on mosquitoes to spread mutate/evolve themselves to spread via air...

You know, "life finds a way" and all that..

Easy. Untold side effect of MTBe caused infertility in 50% of cases in second generation of black women in treated regions. Entire country was depopulated in 25 years. Like that research article from 2056?

The worst man-made ecological disasters in history caused tens of thousands of deaths. We can say with certainty that killing off a few species of mosquitoes will not be worse.

You're suggesting that because it's never happened before (that we know of), we can say with _certainty_ it won't happen in the future? Can we go back 100 years and say that about climate change please?

1 million is a really big number, which is what makes the certainty easy. That's even bigger than projected deaths from climate change.

You certaintyometer needs to be tuned. Around 129.000 people died in Hiroshima and Nagasaki and at least 200.000 people killed in Chernobyl until data. And this is a child's play compared with the >85 millions of humans killed by a worst ecological disaster. We can say with certainty that nobody thought in ancient times that those small and clean rodents, living peacefully in the grasslands and steppes of Central Asia were something to worry. Rats in Europe were not unlike the good species of mosquito if we think about it, with their own diseases and new random relationships...

They aren't a random niche animal. Mosquitoes are pollinators and a major food source for many, many other animals. We know that "pollinator" and "major food source" can be critical roles in an ecosystem.

+1. But they are not the only ones. However, I 'd rather observe tropical ecosystems for 10 more years before making such a radical effort. Like I said earlier, no one is eradicating cars just because more than a million humans die in car accidents every year.

Of course. Do you eat pesticides for breakfast?

How do you supposedly kill mosquitoes without chemical dangerous products? Products that will kill people with cancer, asthma or birth defects.

How do you kill mosquitoes without killing bees and other useful insects that pollinize most of the agrarian production?

Most people ignore that on their own(using just wind to pollinize) the food that we cultivate the food production would be 5 percent or 10% of the current production. 50% of the world population will starve.

Bees already died en masse after the introduction of new pesticides, that had to be delayed in lots of places to study how they were accepting bees.

The healthier bees today live on cities like Paris which had forgotten pesticides in parks. The irony.

Ignorant people are very dangerous. Most people, specially experts on a specific field, are ignorant in most of the other areas.

Near my house there is a river 30 years ago some smart ass though it was a good idea to introduce an alien species of fish into it. No problem, the new fish was small and won't eat the local fish, they said.

Genius! The new fish eat all the eggs of local native population, that basically become extinct as a result.

Read the article before commenting.

https://en.wikipedia.org/wiki/Sterile_insect_technique

I'll tell you how. Species-specific virus, like the one we developed in 1974 for controlling populations of gypsy moth.

At a certain point you're saving people so they can starve to death. That's the worst case. Best case: saving them speeds up how soon we run out of non-renewable resources. What's more merciful to mankind as a whole in the long run? I'd say it's keeping the planet in good shape, and I don't see how eradicating malaria can do anything but hurt that goal.

When mortality rates fall, birth rates typically fall more.

I'm not a scientist, but I know from movies that the amber is filled with mosquitos.

Now: to ensure we're only extracting the mosquito DNA and not the dino DNA...

Perhaps do a pilot test in a limited area and see the impact?

If I recall correctly, that was done in Panama by the canal builders; one of the reasons they were successful was because they wiped out mosquitoes in the region, and protected the workers from malaria.

Their method was to put a bit of gasoline in all of the swamps. The thin layer of gas on the water surface killed the larvae. It probably caused lots of other environmental damage too though, so a different approach would be needed today.

Malaria used to be endemic in much of the US.

http://www.cdc.gov/malaria/about/history/elimination_us.html

I guess they were pretty close to eradicating mosquitoes in transmission areas once they started the control effort.

A similar tactic was used in Cuba and New Orleans to combat yellow fever. [1] They used a layer of oil rather than gasoline though.

[1] http://exhibits.hsl.virginia.edu/yellowfever/new_strategies/

I used to think Kudzu was really bad until I read this:

http://www.smithsonianmag.com/science-nature/true-story-kudz...

Not saying it was a good idea to plant Kudzu but there are far worse invasive plants to deal with.

Yeah I'm surprised that most sources still cite those junk stats from the 1970's and 90's.

Kudzu is great for erosion and its invasiveness in the Southeastern US is GREATLY exaggerated. Apparently the 1996 survey that is most cited greatly overestimated the number of hectares covered.

Stuff in the south in general grows really, really fast.

What if they keep a bunch of unaltered mosquitoes in captivity while eradicating them in the wild? Then if it does have some unanticipated effect the change may be reversible. (Of course it's possible the re-released mosquitoes will be out-competed by whatever's moved in in their absence, but you've got a better chance than if you didn't keep a backup.)

> we find their presence unpleasant

It's a bit more then "unpleasant" I would say

Or it could be like air bags and seat belts: certainly responsible for its share of issues, but far fewer than what it prevents.

Reply on news.ycombinator.com