Microbial Diversity and Evolution
Transcript of Part 2: Microbial Respiration of Arsenate [As(V)]
00:00:00.00 Hello, and welcome to iBioSeminars. My name is Professor Dianne Newman, 00:00:04.05 and I am a professor at the California Institute of Technology in the divisions of Biology and Geological and Planetary Sciences. 00:00:11.28 And I am also an investigator of the Howard Hughes Medical Institute. 00:00:14.28 So in this second part of my lecture I am going to give you an example of microbial diversity at work in 00:00:22.14 environments all over the world today, where we are going to talk about 00:00:27.12 how bacteria are able to respire a toxic compound called arsenate, arsenic V, 00:00:34.01 and how the ability of these organisms to do this can affect the geochemistry of their environment in a very significant fashion. 00:00:41.03 Now the reason that we are motivated to study microbial respiration of arsenic is because arsenic is toxic, 00:00:49.21 and the activities of microorganisms are very relevant to liberating arsenic 00:00:54.27 into water supplies that can lead to skin cancers such as that shown here 00:01:00.04 on the hands of a woman from Bangladesh. 00:01:03.00 Now these lesions that you see are coming from keratosis of the skin, 00:01:08.11 but at a more molecular level we know that arsenic is extremely harmful to cells 00:01:14.13 regardless of what oxidation state it exists in. 00:01:18.03 Arsenic can exist either as arsenate, arsenic V, and in this form it is very similar to phosphate. 00:01:25.22 And because of this the cells get confused, and they take up arsenate 00:01:30.15 when they mean to be taking up phosphate, and instead of adding inorganic phosphate onto ADP to generate energy, 00:01:38.19 they will sometimes add arsenate on. 00:01:41.22 And this is unstable and rapidly hydrolyzes, and leads to an interruption in energy generation. 00:01:49.01 Now arsenic can also exist in another form commonly within the environment and within living cells, 00:01:56.12 and that is as arsenite, arsenic III. And this form is even more deleterious to cells 00:02:02.15 than arsenate because as the more reduced form, arsenite, 00:02:07.05 it is capable of reacting with reduced sulfur groups found on proteins very readily. 00:02:14.10 And when it does this it binds and distorts these proteins 00:02:17.27 and denatures them and causes them no longer to have their normal function. 00:02:22.26 Now in the United States we have strict rules on the drinking water standards 00:02:27.10 in terms of the maximum concentration of arsenic in drinking water supplies. 00:02:31.20 Last I checked it was around 10 micrograms per liter, although this number may indeed even be smaller now. 00:02:39.29 But in other parts of the world, for example in Bangladesh, 00:02:44.04 where there is really a tragically high level of arsenic in the drinking water supplies in that country, 00:02:51.09 the concentrations can go up to 1000 micrograms per liter. 00:02:55.29 And what I want to show you here is both in the United States and in Bangladesh how significantly elevated these concentrations are 00:03:05.26 in various parts of these countries. 00:03:09.11 So in the United States... what you are looking at now is a graph that was produced by the USGS several years ago, 00:03:16.01 and it is plotting regions where arsenic concentrations get quite high. 00:03:21.06 And so in areas that are red and orange we have concentrations that are exceeding the levels 00:03:28.26 that are approved for the drinking water supply. 00:03:32.13 And I am going to focus later in the talk on a part of this map in Northern California 00:03:38.05 where arsenic concentrations are very large, 00:03:41.11 and I will describe the relevance microorganisms have in terms of affecting the geochemistry of arsenic in these habitats. 00:03:49.05 All right, so now you are looking at a plot of Bangladesh, 00:03:53.26 where the arsenic concentrations in drinking water supplies are 00:03:57.02 extremely high, very elevated, and indeed in many locations 00:04:01.15 well above what would be considered the safe level by EPA guidelines within the United States. 00:04:07.04 And a dramatic illustration of this is simply when you are looking at this plot here. 00:04:11.15 In the regions in purple the probability where the concentration of arsenic in the drinking water 00:04:16.05 is greater than 50 micrograms per liter is over 70%. 00:04:21.02 It is at least 70% in these regions, and this is a non-insignificant area of the country. 00:04:27.26 So how does arsenic get into the drinking water in these environments, and what is the connection to microbial activity? 00:04:34.29 The connection is driven by microbial metabolism and specifically anaerobic metabolisms 00:04:40.29 that occur in the sediments once oxygen is depleted from these environments. 00:04:46.20 What I am showing you here now is a very simplified view of the geochemical cycle of arsenic in a freshwater environment. 00:04:55.27 What I want to focus on is this part of the slide here 00:04:59.25 where you see this box that is depicting the transformation between arsenate, arsenic V, and arsenite. 00:05:08.11 Now these are the two primary forms of inorganic arsenic. 00:05:11.22 Other parts of this diagram that I am not going to describe 00:05:14.29 now further depict pathways that are involved in the methylation of arsenic species, 00:05:21.13 and this is an interesting subject in itself, but let's restrict our attention here now 00:05:27.04 to this inorganic cycle because this is likely to be the most important part in controlling arsenic geochemistry 00:05:35.03 in potable water supplies in many environments, such as in Bangladesh. 00:05:39.00 What we know is that in these environments 00:05:41.05 arsenic V is able to be trapped often times in association with iron minerals 00:05:47.29 where it comes down into the sediments. 00:05:50.18 And it stays there bound up to these ferric oxyhydroxide species, where it is either co-precipitating 00:05:59.11 or adsorbed onto these iron minerals, 00:06:01.04 and this arsenate because of its anionic charge likes to stick to these positively charged minerals 00:06:07.26 very well. So it remains associated with them. 00:06:10.25 And down in the sediments it will stay quite happily 00:06:14.13 if it weren't for the deposition of an electron donor, an organic carbon source, oftentimes fed by latrines 00:06:21.23 for example in Bangladesh, that provides a fertile substrate to promote microbial growth. 00:06:26.08 Now in these sediments, of course, there are other electron acceptors that can be used for respiration 00:06:31.20 and these include oxygen as a primary example, sometimes nitrate, sometimes other substrates, 00:06:39.01 but many occasions arise where these other electron acceptors that have a higher redox potential are depleted. 00:06:46.15 And once this happens, there still is an abundance of organic carbon that needs to be oxidized 00:06:51.23 and that is burned by utilizing arsenic and iron in these sediments as electron acceptors for microbial respiration. 00:07:00.04 Now when bacteria begin to respire arsenic, they reduce it from arsenic V to arsenic III, 00:07:06.09 so again that is arsenate going to arsenite. 00:07:08.26 And arsenite is uncharged at the pH of natural waters, 00:07:12.28 and as such it doesn't adsorb as well under most conditions, although it still can adsorb. 00:07:18.28 It would be too simplistic to say that it just becomes purely soluble, 00:07:23.04 but by and large I think it is fair to say that when bacteria reduce arsenate to arsenite, 00:07:28.24 this does affect its mobility, allowing arsenite to enter into the water supply, leave the sediments, 00:07:36.20 and be carried there in groundwater systems into drinking water supplies. 00:07:41.24 So knowing that arsenic respiring microorganisms are catalyzing this crucial step, 00:07:47.28 it became of interest to be able to predict their activity. 00:07:51.23 And yet, many years ago, well, not that many years ago, about 10 years ago 00:07:56.02 this was very difficult to do because there was nothing known about 00:08:00.01 the molecular mechanism underpinning microbial respiration of arsenic. 00:08:03.03 All that was known was that there were diverse organisms from the tree of life. 00:08:07.14 So if you remember from my first lecture that big phylogenetic tree I showed you, 00:08:11.12 here now is a reduced version of that where most of these organisms 00:08:15.08 you see belong to the bacterial family, although there are a few here in the archaea, 00:08:20.05 but all of these organisms are ones that interact in some fashion with arsenic. 00:08:23.29 Yet, going back ten years ago, we didn't know, as I said, how they did this. 00:08:30.15 What was the molecular biology underlying how they were able to utilize arsenic in respiration? 00:08:36.21 So my group set out to answer this question. And what we needed at the time 00:08:41.23 was a strain that we could utilize as a model system 00:08:45.20 in order to gain insight into this process. 00:08:47.27 And so we went to an environment where we thought we could isolate an arsenate respiring organism 00:08:53.17 that would be easy to work with in the lab. 00:08:55.21 And by easy to work with I mean one that we could grow on the bench, 00:09:00.05 under regular atmospheres, where we didn't have to worry about keeping it strictly anaerobic all of the time, 00:09:05.22 which up until the isolation of this organism, had been the case for the previously described arsenate respiring bacteria, 00:09:13.03 these organisms that were only able to grow under anaerobic conditions. 00:09:17.26 Now this bacterium came from a pier in Woods Hole. 00:09:23.14 And this is Woods Hole, Massachusetts. It is where the Marine Biological Laboratory is. 00:09:27.26 And a student in the microbial diversity summer course for her independent project helped me isolate an organism 00:09:35.10 from wooden chips from this pier. And the reason that we went to this wooden pier is that arsenic historically was used as a wood preservative. 00:09:44.01 And so we reasoned that if we could find bacteria attached to this wood here 00:09:47.09 they would be likely organisms that would be able to resist arsenic, because this wood was full of arsenic, 00:09:53.09 and also able to tolerate being exposed to the atmosphere, because we were going to collect 00:09:58.22 them right at the air/water interface where they would be seeing a lot of oxygen 00:10:03.20 or at least so we thought or so we hoped. 00:10:05.09 And our first enrichment was one where we put them under anaerobic conditions 00:10:11.08 in a bottle where we gave them a food source and arsenate as their sole electron acceptor for respiration. 00:10:17.08 And we asked if they could grow and reduce arsenic. 00:10:21.01 And the way we could tell if they were reducing arsenic was by whether or not the bottle was turning yellow. 00:10:26.08 And this is a very simple test. It's a test for the production of arsenic trisulfide, As2S3, 00:10:34.08 that is the molecular formula for arsenic trisulfide, 00:10:37.27 which is a yellow mineral that is produced only when arsenate is reduced to arsenite, 00:10:42.26 and then arsenite combines with sulfide to generate this yellow mineral 00:10:48.12 under the pH conditions relevant for the bottle. 00:10:51.13 So seeing this we were able then to tease out the organism catalyzing this reaction, 00:10:55.16 and we did a very targeted isolation where we required that 00:10:59.10 this organism was also capable of growing in a rich medium on a plate on the bench. 00:11:04.16 And by going back and forth between the anaerobic bottle and the rich plate, we were able to 00:11:09.29 ultimately isolate a bacterium, shown here, 00:11:13.00 that we called Shewanella sp. strain ANA-3. And the reason we named it Shewanella 00:11:19.00 is because when we sequenced its ribosomal DNA, in particular the 16S subunit, 00:11:25.00 which I mentioned before is used oftentimes, is used many, many times, in microbial phylogeny 00:11:32.11 to understand the evolutionary relatedness of an organism to other organisms. 00:11:36.13 It became quite clear that this strain that we isolated grouped within a larger genus called Shewanella. 00:11:42.06 Now, ANA-3 was a fantastic find for us because, as I told you, we demanded that it was able to grow easily on plates, 00:11:51.08 and this opened up the possibility of performing experiments 00:11:55.01 where we could do genetic manipulations and select for single colonies 00:12:00.04 on these plates like the ones you see here that had a specific mutation. 00:12:06.10 Before we entered into the whole prospect of doing mutagenesis 00:12:11.25 and genetically manipulating these organisms, 00:12:13.27 however, we wanted to verify that this organism actually could respire arsenate. 00:12:18.12 And the more rigorous way of doing this beyond just seeing whether the bottle turned yellow 00:12:23.08 was to measure the coupling of the reduction of arsenate to arsenite 00:12:29.25 to the oxidation of lactate to acetate. 00:12:33.13 Now I am only going to show you the data of arsenate reduction, 00:12:36.14 but you can take my word for it that the data also indicates 00:12:39.26 a very nice stoichiometric relationship of the molar amounts of lactate oxidizing to acetate 00:12:47.19 as arsenate being reduced to arsenite 00:12:49.18 in a one to two ratio as predicted from this metabolic formula. 00:12:54.21 So here what you see is that beginning at time zero we gave these organisms ten millimolar concentrations 00:13:02.29 of arsenate, which they then reduced and concomitantly converted to arsenite 00:13:10.10 and at the same time, they grew by many generations because this is now a logarithmic plot. 00:13:18.29 And this was what we would expect from an organism that could respire on arsenate. 00:13:24.03 So we were excited because in this bottle the only electron acceptor for growth was arsenate. 00:13:29.19 Now going forward what we wanted to do was identify what the enzymatic system was that catalyzed 00:13:36.26 the conversion of arsenate to arsenite. 00:13:38.22 And so I told you in my introductory lecture that one way that organisms have of generating ATP 00:13:44.01 is by coupling the transfer of electrons from an electron donor 00:13:49.13 in this case lactate, which is being oxidized to acetate, 00:13:53.00 through the membrane that contains a respiratory chain 00:13:57.06 made up of proteins and small molecules that can not only pass electrons through the membrane, but can also concomitantly 00:14:04.19 translocate protons out of the membrane generating this gradient that can be harnessed to make ATP. 00:14:10.12 Now at the end of this chain lies what our target was, 00:14:15.15 and that is the enzyme responsible for ultimately receiving these electrons from the chain 00:14:20.06 and passing them onto arsenate, reducing it to arsenite. 00:14:25.08 This was unknown when we began. 00:14:27.15 However, before us investigators looking at different organisms 00:14:32.09 had established very beautifully that there was a separate 00:14:35.19 detoxification system in many organisms that allowed them to resist arsenic, 00:14:40.14 and this worked as summarized as shown here where in these 00:14:46.02 detoxification schemes...when these bacteria would see 00:14:49.19 arsenic V, and it would enter into the cell via some type of phosphate transport system, 00:14:54.15 they would reduce it to arsenite via a protein called ArsC, 00:15:00.04 and then this arsenite would somehow be recognized by an efflux pump 00:15:06.09 comprised of various proteins ArsA and B that would expel arsenite from the interior of the cell. 00:15:16.05 Now this was well known, and because of this we could make use of 00:15:22.10 systems where people had previously taken strains that used to be resistant to arsenic, 00:15:28.03 genetically knocked out the genes that were required to confer that resistance, 00:15:33.09 and we could then provide genes from our organism, 00:15:37.22 Shewanella ANA-3, and ask whether or not we could restore the ability of these organisms to 00:15:44.27 resist arsenic or even to respire arsenic, whether we could confer that ability upon these organisms. 00:15:50.14 And so this was a gain of function approach, where we took DNA from Shewanella ANA-3, 00:15:55.09 and we cut it up into pieces that we could clone into a vector. 00:15:59.20 And then each of these vectors we put into different 00:16:02.23 cells of our host strain that on its own was incapable, in this case when we started, of being able simply to resist 00:16:10.25 arsenic at high concentrations. And so in screening a variety of these strains 00:16:15.18 we found a clone that conferred arsenic resistance to E. coli 00:16:19.24 as you can see here. So what you are looking at now is the optical density of overnight cultures of E. coli 00:16:28.01 that contain in this case either no plasmid at all, and that is shown here with the triangles. 00:16:36.19 And as the arsenite concentration is raised greater than 0.1 millimolar, 00:16:43.09 you see they rapidly are unable to grow at all. 00:16:48.22 Whereas with a special plasmid we found with genes from Shewanella ANA-3, 00:16:52.08 they were able to resist and grow quite nicely, even at concentrations all the way up to ten millimolar. 00:16:59.14 And here we just have a control of the vector without any DNA from our ANA-3. 00:17:05.29 Now were these genes also required for the ability to respire arsenate? 00:17:13.19 So we had identified through this gain of function approach 00:17:17.19 the Ars system that Shewanella ANA-3 had because it could complement the mutant that had been deleted in its own 00:17:26.13 native system of these Ars genes, as I told you, ArsA and ArsB and ArsC 00:17:32.03 are necessary for the detoxification pathway. 00:17:34.16 So in order to answer this question whether or not this detoxification pathway was required for respiration in ANA-3, 00:17:41.27 we wanted to answer this question because we wanted to see if there was yet another pathway out there that might 00:17:49.09 be important for arsenate respiration that was completely independent of this pathway. 00:17:53.08 We generated a transposon insertion into ArsB that encodes the protein that is involved in the efflux of arsenite from the cell, 00:18:02.20 And this had the effect of simultaneously knocking out the expression of the downstream gene, ArsC, 00:18:10.14 that encodes that arsenate reductase that I told you is inside the cell 00:18:15.22 that had been previously shown to be necessary for arsenate reduction linked to detoxification. 00:18:21.08 So when we did this now- all of this genetic manipulation was done in Shewanella ANA-3- 00:18:26.14 we found that the answer was no, that this mutant that didn't have a functional ArsB or ArsC 00:18:32.16 was still capable of respiring arsenic. And here you can see its growth 00:18:36.21 in red in optical density over the course of several days, 00:18:41.20 and yet as arsenate is being reduced in the wildtype the culture is growing quite well, 00:18:51.17 but in this mutant that we made that was able to still grow and here in red 00:18:59.06 while it could reduce arsenate, it stopped, it plateaud out and wasn't ever able to grow to the same 00:19:09.09 density as the wildtype cells. So what this indicated to us was that even though there was clearly 00:19:15.15 another enzymatic system necessary for the respiration, 00:19:18.11 of arsenate, these genes involved in detoxification were still important. 00:19:22.11 And that is intuitive. As the organism is churning through more and more arsenate, 00:19:27.15 as arsenite is accumulating within the cell, 00:19:29.16 the organism needs a mechanism to get rid of it, and this arsenic detoxification 00:19:34.20 pathway then becomes relevant whether or not 00:19:37.27 the reduction is necessary is another story but the machinery to pump out the product, arsenite, 00:19:44.06 is indeed of some relevance at high concentrations. 00:19:47.15 So now the hunt was on for the genes that encoded the respiratory arsenate reductase, 00:19:52.21 having established that there was an enzymatic activity independent of the detoxification 00:19:59.13 arsenate reductase. And here we got very lucky. 00:20:03.14 So on this initial plasmid that we cloned what we found was, as I told you, all of these genes that encoded 00:20:10.14 the system necessary for detoxification, 00:20:12.19 the Ars system. But right next door were a couple of genes 00:20:18.21 and very happily for us we picked up just the very beginning end of one of these genes in our initial clone. 00:20:27.26 And we were motivated to keep sequencing because when we sequenced this portion of our clone here 00:20:32.28 we noted that the amino terminus had some homology to enzymes in the dimethylsulfoxide reductase protein family. 00:20:42.04 And this is a protein family that is responsible for 00:20:46.12 different types of anaerobic microbial respirations, enzymes that are serving as terminal electron acceptors 00:20:53.22 for substrates other than oxygen, including dimethylsulfoxide, DMSO. 00:20:58.27 So in sequencing further we found there were these two genes downstream 00:21:04.05 shown here in red, which we named arrA and arrB. 00:21:09.03 And we surmised that because these genes were right next door to the arsA, B, C genes 00:21:16.24 that they might encode the genes for the respiratory arsenate reductase. 00:21:20.21 So this is one of these moments in science 00:21:23.01 where you just get quite lucky, and indeed it turned out to be the case. 00:21:26.04 And here is the proof. So the way to demonstrate this is to knock those genes out 00:21:31.05 and then do the same experiment I mentioned before where we gave the cells only arsenic as the electron acceptor 00:21:37.25 and asked could they reduce it and could they grow. 00:21:41.02 And here the data I am showing you is the arsenate reduction data 00:21:44.22 where the wildtype in red is capable of reducing arsenate 00:21:49.05 very rapidly whereas if we knock out either the arrA or the arrB 00:21:54.23 gene there is no reduction. And what I am not showing you is that there was no growth here in these cultures either. 00:22:01.22 And then when we complemented, when we gave back the wildtype version of these genes, 00:22:06.03 we were able to restore growth and restore arsenate reduction activity. 00:22:09.27 So having found these what this suggested to us is that 00:22:15.12 the pathway could be diagrammed as shown here in terms of electron transport. 00:22:21.16 So as I have mentioned before, the oxidative phosphorylation 00:22:25.28 process depends on electron transfer from an electron donor to an electron acceptor. 00:22:31.11 In this case, that is arsenate, which comes into the cell and gets reduced to arsenite 00:22:38.05 and the coupling of electron transfer from donors such as lactate being oxidized to acetate 00:22:45.26 through enzyme complexes and small molecules such as quinones 00:22:52.10 or menaquinones ultimately to these enzymes at the tail end that are the acceptors of these electrons. 00:23:00.14 And as these electrons are passed protons are translocated, 00:23:04.03 and this generates that battery as I explained in my first talk around this membrane 00:23:08.24 that can then be harnessed through the ATP synthase to drive ATP formation and oxidative phosphorylation. 00:23:13.18 Now what I told you at the beginning was that the ArsC protein, which is the 00:23:21.29 enzyme involved in arsenic reduction 00:23:24.19 for detoxification is inside the cell. 00:23:27.06 And what we surmised from looking at the sequence data 00:23:29.26 of our arrA and arrB proteins was that they had motifs that suggested that they were likely to be 00:23:38.02 localized here in this compartment, which is called the periplasm. 00:23:41.21 And that is the space in between the outer membrane and the cytoplasmic membrane 00:23:46.13 in gram negative bacteria. Now knowing that arsenate could be pumped into the cell 00:23:51.12 through outer membrane... or porins, transporters, so not necessarily pumped, 00:23:59.16 but just taken in through these big porins. The transporters that pump it are located here in this inner membrane. 00:24:06.06 We surmised that this enzyme, if it were located here in the periplasm, 00:24:11.16 would be identifiable if we were able to put a fluorescent tag on it as a ring around the cell. 00:24:18.19 So if the enzyme were within the cell, if you did a microscopic view of the cell 00:24:24.04 everything really, effectively, would look solidly green, 00:24:28.20 whereas if that protein were just within this outer compartment, it would visualize as a ring. 00:24:34.26 And that is what we did. And here you see sure enough, as we predicted, rings being formed. 00:24:41.01 And we verified this through more classical biochemical techniques and fractionation 00:24:45.00 to demonstrate that this activity indeed was periplasmic. 00:24:48.00 Now why does any of this matter? 00:24:50.03 Okay, at this point we have identified a novel enzyme. 00:24:53.07 We know something about where it resides, but really the motivation for this project began with 00:24:57.15 trying to understand an environmental problem, and how to develop a marker for 00:25:02.03 the activity of these organisms in the environment. 00:25:03.25 And the environments that we are talking about are sediments such as those shown here 00:25:08.23 as I said exist in Bangladesh and in many other parts of the world 00:25:11.27 where arsenic is in high abundance adsorbed onto iron minerals 00:25:15.11 that collect in these sediments. Now in the beginning I told you that the state of knowledge about arsenic respiring organisms 00:25:25.17 when we began this project was that they were phylogenetically diverse. 00:25:29.24 And so this is just a little image that illustrates that concept, 00:25:33.15 and the names are not important. Just pay attention to the dots here 00:25:36.13 The point is that the dots are everywhere on this tree, 00:25:38.27 and these are organisms that themselves are metabolically versatile. 00:25:42.16 So one would never be able to predict simply by 00:25:45.19 looking for a signature, a genetic signature of a specific species whether or not arsenate respiration 00:25:51.20 were occurring because these species are capable of respiring other metabolites, so you wouldn't know necessarily 00:25:58.25 whether they were respiring arsenic in situ 00:26:00.28 and moreover close relatives of any individual species that can respire arsenic sometimes can't, 00:26:07.06 and so the species indication is not nearly good enough for tracking this activity. 00:26:12.04 Where we got very lucky was that in our model system of Shewanella ANA-3 00:26:16.24 the gene and the protein encoded by that gene that we found 00:26:20.16 catalyzed the key step in the conversion of arsenate to arsenite, namely the arrA protein together with arrB, 00:26:29.19 was remarkably well conserved over this very evolutionarily diverse group of organisms, 00:26:35.23 and that as we continued in our studies to look into the enzymology of this family of respiratory arsenate reductases 00:26:44.10 we found that they formed this nice tight group together. 00:26:48.06 And because of this we reasoned we might be able to go after this group using probes that were 00:26:53.24 specific to detecting the expression of the gene encoding this enzyme in the environment. 00:26:58.27 So to do this we took a close look at the arrA gene, and we divided it up into sections here, 00:27:09.05 and we looked at these sections for the ones that were the most diagnostic as being unique to 00:27:15.05 the enzyme family of the arsenate reductases, and not similar in these particular regions 00:27:21.27 to other enzymes that were more broadly within this DMSO reductase family because 00:27:26.22 we didn't want a marker that would pick up this family writ large. 00:27:30.23 We wanted a marker that would specifically be able to identify 00:27:33.21 this branch of the family, the arsenate reductase branch. 00:27:36.24 So we took a look at these two domains here that we called E and F, 00:27:42.18 and we designed PCR primers that could pull up a very short fragment 00:27:49.07 in a PCR reaction that we could sequence. And we hoped that this particular stretch would be one 00:27:58.09 that would only be amplified in cells that contained 00:28:01.23 the arrA enzyme, rather the gene, and then of course that gene encodes the enzyme that catalyzes the reaction. 00:28:08.21 So what you are looking at here now are controls of bacteria that have proteins in the DMSO reductase family. 00:28:15.03 This is this whole group here in blue, but do not contain the arrA gene. 00:28:20.14 And then numbers 5 through 19, rather 6 through 19 00:28:25.14 contain or belong to organisms that are arsenate respiring organisms. 00:28:32.08 And so we took the DNA from all of these different bacteria 00:28:37.28 and we did a control to make sure we could amplify just by looking for the product of primers to target the 16S ribosomal DNA gene 00:28:46.03 and happily in our negative controls we were unable to detect a product, 00:28:52.16 yet in almost all but these last two lanes here, 00:28:57.18 even if it is very faint, take my word for it, 00:28:59.20 we did indeed see the amplification of the product we were looking for. 00:29:04.16 And so for the vast majority of these bacteria, number 19 is actually an archaean, so we weren't that concerned 00:29:10.20 about this because we would infer potentially that the arsenate reductases 00:29:15.26 might have evolved along a slightly different pathway in the archaea 00:29:19.00 than in bacteria. We nonetheless appear to have a fairly robust 00:29:22.14 short sequence that we could use to track the presence of this gene or its expression. 00:29:28.06 So with this tool in hand we then wanted to go to an environment and ask 00:29:32.20 whether or not this gene was important in controlling the geochemistry at that site. 00:29:37.16 And so the first thing that we needed to establish was whether or not 00:29:41.20 this gene and the enzyme it encoded was required for arsenate to be converted in iron rich sediments. 00:29:48.21 So for example in this site here where we have arsenic adsorbed onto ferric oxyhydroxide minerals, 00:29:56.29 would we see the reduction of arsenate to arsenite in these environments 00:30:00.04 in the absence of arrA? If the answer to that was no, then if we were able to detect 00:30:10.13 the presence of the arrA protein, or in this case what we were really doing was detecting 00:30:16.26 the expression, the messenger RNA of the gene encoding that protein, 00:30:21.24 we could have a very good reason to associate any geochemical profiles in these sediments 00:30:29.10 that showed reduction with a microbiologically catalyzed process 00:30:33.28 driven by this enzyme. So, with that as the important logical first step, 00:30:41.00 we then could go and look to see whether or not those genes were indeed present 00:30:45.16 and expressed in the sediments of interest. 00:30:47.21 So, the way we got into this is we first were able to synthesize in the laboratory 00:30:53.25 sediments that mimicked those in the natural environment that I just showed you. 00:30:57.18 And I am not going to show you all of our controls, but just one representative data file here. 00:31:02.09 But what we found was very exciting, and that was, 00:31:05.19 as we had hoped, the transformation of arsenic, where now we are looking at 00:31:09.18 arsenates again going away and being transformed to arsenite, only occurred when we had functioning arrA activity, 00:31:20.12 and that was revealed to us in these experiments by looking at the ratio of RNA and DNA for the arrA sequence, 00:31:29.03 and seeing a peak of expression maximally just at the time period where arsenate reduction to arsenite was at its maximum. 00:31:38.26 And in controls where we added organisms that we genetically deleted for their ability to express this protein, 00:31:47.17 we didn't see any arsenate reduction under this condition. 00:31:51.27 And so in the absence of some chemical reductant a biologically catalyzed process was absolutely required 00:32:00.18 for arsenate transformations under these very environmentally relevant conditions. 00:32:05.05 So let's now go to a real environment 00:32:07.06 and see how this plays out. 00:32:09.10 OK. So we chose to go to the Haiwee Reservoir, which is north of the city of Los Angeles, 00:32:14.18 and it is in the southeastern part of the state of California. 00:32:21.01 Los Angeles being right around here and San Francisco up here. 00:32:25.23 And Mono Lake over near the Nevada border, 00:32:28.28 that I told you in my introductory lecture has very high concentrations of arsenic, 00:32:32.13 feeds into this ultimately through the California aqueduct. And this aqueduct is carrying very high loads of arsenic. 00:32:44.18 So these concentrations are well in excess of what is necessary to meet EPA guidelines 00:32:50.08 for the concentrations in potable water supplies that I'll remind you are 10 micrograms per liter. 00:32:56.03 Something needs to happen, and one of the ways that arsenic is reduced 00:33:02.09 in concentration in the water is by reactions with ferric chloride 00:33:08.05 that is added to the water as it is coming down the aqueduct 00:33:12.14 at the Cottonwood Treatment Facility that is several miles upstream of the Haiwee Reservoir. 00:33:18.17 Now at the Cottonwood Treatment Facility iron chloride is injected into the water. 00:33:24.26 And when it hits the water because the pH is circa neutral 00:33:29.08 iron III in this iron chloride rapidly converts into rust, ferric oxyhydroxide amorphous iron minerals, and as I've been telling you 00:33:38.21 arsenate likes to adsorb and stick onto these minerals very well. 00:33:42.24 And so arsenic gets bound up and co-precipitated with these iron oxides 00:33:47.16 and they together come down through the LA aqueduct and 00:33:51.11 wind up sedimenting out in this channel here, leading into the Haiwee Reservoir, and this is a view of the channel. 00:34:00.15 Now it was into this particular environment that students in my lab and that of 00:34:05.12 a former colleague of mine at Caltech, Professor Janet Hering went. 00:34:09.00 And here are two students, although they have both now moved on. 00:34:14.06 They are postdocs at different locations. My student Davin Malasarn and Janet's student, Kate Campbell, 00:34:21.01 both of whom are obscured in this photo, but they went into the Haiwee sediment channel 00:34:26.11 and they took these cores. And Kate was a geochemist and she and others in Janet Hering's laboratory had been studying 00:34:33.12 these sediments for years and were very familiar with the mineralogy and the aqueous geochemistry 00:34:38.23 within these sediments. They have done extensive studies. 00:34:41.17 So Kate and Davin teamed up together, and what Davin had done was to develop, in collaboration with Chad Saltikov in my lab, 00:34:49.25 this probe to look for respiratory arsenate reductase within real sediments. 00:34:55.11 So the first thing we wanted to understand was whether the geochemical 00:34:58.14 profile within this core might indicate that microbial activities were occurring. 00:35:04.00 And as I told you, in these types of sedimentary environments, 00:35:07.10 if we see arsenate being transformed to arsenite 00:35:11.11 that is a very, very powerful indication that microbial activity must be present 00:35:19.06 because in the absence of any other type of chemical reductant which in these particular sediments is not present 00:35:26.09 arsenate would otherwise be stable. 00:35:28.24 And so what Janet and former students had been able to show using 00:35:31.25 very sophisticated X-ray spectroscopic techniques that I am now going to explain very simply to you 00:35:37.19 is that as you go down through these columns that I just showed you in the previous image the arsenic speciation changes. 00:35:47.22 And so you can see that here. These are just spectral profiles of a curve you get using a fancy form of X-ray spectroscopy 00:35:54.17 called XANES for arsenite, arsenic III, and arsenate. 00:36:00.22 And now what you are looking at are the profiles at different depths in centimeters going down into that core. 00:36:07.19 And while in the very upper layers, in the first over two centimeters 00:36:13.01 worth of the sediment, most of the arsenic is lining up with the arsenic V hump. 00:36:18.20 Very rapidly as you go below two and a half centimeters you see that the peak shifts and now it is aligned with the arsenite, arsenic III. 00:36:26.22 So this was compelling geochemical evidence that some type of microbial process was likely occurring, 00:36:33.04 and therefore the icing on the cake to help put it really together 00:36:38.02 would be to demonstrate that within these sediments 00:36:40.12 this gene, arrA, was not only present, but also being expressed into messenger RNA, 00:36:47.03 which then later of course would be converted into the protein that would catalyze the transformation of arsenate to arsenite. 00:36:53.09 And so what Davin's main contribution was, together with Kate and colleagues, 00:36:58.20 was to go into these cores, extract the DNA and extract the RNA, 00:37:03.13 sequence it, and what he found that was very impressive 00:37:08.16 was that 62 to 97% of the sequences that he found 00:37:13.00 were identical in their sequence to those from known arsenate respiring organisms. 00:37:19.16 And so here we showed indeed that messenger RNA for this arrA gene, which was our robust indicator 00:37:27.04 of the ability of the organisms to respire arsenate, was not only present, 00:37:30.23 but expressed in Haiwee sediments, and because we knew that in these sediments 00:37:34.14 the expression of this gene was required for the transformation of arsenate to arsenite 00:37:39.18 we could very firmly be able to conclude that 00:37:44.08 microbial activity and microbial respiratory activity was catalyzing this process. 00:37:48.10 So to summarize, what I hope I have taught you in this lecture 00:37:53.02 is that many microorganisms respire arsenate, 00:37:55.22 and now many more are known beyond those that I showed earlier in the talk. 00:38:00.14 The list grows larger and larger every year. 00:38:03.20 But remarkably, evolution has conserved the mechanism 00:38:07.14 whereby the vast majority of these organisms are able to catalyze that activity. 00:38:11.01 And because of this, we can look to a specific gene, 00:38:17.08 the arrA gene, as a marker for this process. 00:38:20.06 And we can generate probes to detect this gene in a variety of environments, 00:38:25.16 and these probes have been optimized more recently by Chad Saltikov who was a former postdoc in my laboratory, 00:38:32.04 but now is a professor at UC Santa Cruz who has been able to take our early work 00:38:37.09 with these primers and extend it much farther into a variety of environments. 00:38:41.10 And now this is really an established way of tracing the activity of arsenate respiring organisms in the environment, 00:38:48.26 and we hope that this will be useful in a variety of countries around the world 00:38:53.28 where arsenic is a problem, where it would be of relevance to understand 00:38:57.27 if microbial activities indeed were occurring, because that might affect remediation strategies 00:39:04.12 to limit the conversion of arsenate to arsenite in those environments. 00:39:09.23 So to acknowledge as I have been doing throughout the talk, and now here by showing 00:39:15.20 their images, of course, all of these types of projects 00:39:18.04 are performed by very talented students and postdocs 00:39:21.02 in the laboratory, and I was very lucky to work with two great guys: 00:39:25.04 Chad Saltikov who was one of the first postdocs in my lab, 00:39:27.26 and he is now a professor at UC Santa Cruz in the environmental toxicology department, 00:39:32.25 and Davin Malasarn, a graduate student in biology at Caltech who is now a postdoc at UCLA. 00:39:40.15 And together with our colleagues and collaborators at Caltech, Janet Hering and her student Kate Campbell, 00:39:46.07 and a colleague in Britain, Joanne Santini, we were able to make the discoveries that I summarized in this short talk. 00:39:54.15 All of our work has been supported by a variety of agencies, and I would like to just briefly acknowledge them here. 00:40:01.08 For this particular project, the Luce Foundation and the Packard Foundation were 00:40:04.23 very important at the beginning, and HHMI later in the work, and Chad was supported through his independent 00:40:10.11 fellowship from the National Science Foundation. 00:40:14.10 Thank you, and this is the end of the second series in the iBioSeminars lecture on microbial diversity.