Ask Eveleyn Publishers for a new biography of another distinguished scientist : 'A Wreath for Doctor Ramaiyya'


13.Exploring Diverse Roles of a Rice Transcription Factor bZIP48 in Light-Regulated Plant Development. By Dr Jitendra P. Khurana, Director, Department of Plant Molecular Biology, University of Delhi South Campus: 31 January 2018. 31 January 2018

Dr. Jitendra Paul Khurana, Professor of Plant Molecular Biology at University of Delhi South Campus, heads research in light and hormone signalling in the model plant Arabidopsis and crop plants like Brassica and rice. His earlier work on the isolation of Arabidopsis mutants defective in phototropism, lead to the identification of a novel blue light receptor, phototropin1, which had otherwise remained elusive for nearly a century. His group is currently carrying out work on the functional analysis of genes involved in regulating plant height and flowering time in crop species like Brassica and rice using state-of-art technologies. Professor Khurana actively participated in accomplishing the task of rice and tomato genome sequencing, as part of an International Consortia.

Sri S. P. K. Gupta, Prof. N. Raghuram, Prof. P. C. Sharma, colleagues of the School of Biotechnology and young friends.

A very good morning to all of you. It has been immense pleasure to be here before you to speak on this occasion when you are celebrating the contributions of Dr. SubbaRow.

I learnt about this venture of the School of Biotechnology way back when the series were started in the old campus. And, if I recall correctly, I also attended one of this seminars in the very first of this series.

And as Guptaji said, I think it has been a very good gesture on the part of the School of Biotechnology. This is going to keep alive the memory and contributions of Dr. SubbaRow.

Although I dont work in the area of pharmacology or medicine to which he contributed, but you will be surprised to learn that every morning I take one tablet of folic acid. This is because my throat has been itchy. Even today it is a little bit problem. So my physician advised, You take one tablet of folic acid.

Had Dr. SubbaRow been alive, he would have really appreciated where the science today stands and common components which are there in plant system, animal models and human beings -- particularly in biological clocks -- how the sensitive photoreceptors regulate that. You all know that noble prize went to the people who discovered the components of biological clock. In that sense I think there is a lot of commonality between how plants perceive light, regulate the development, also integrate the biological clock and also how other animal system also works.

I will give you one example. There is an overlap between the pathways which are common to plant systems and human systems. I have been coming to this school for the last several years and I have given talks also. One way to have a session today was may be to give a general talk, summarize the field of photobiology but then I am still at a researcher.

I thought it would be better for young students to learn what kind of work is being done in the south campus of Delhi University and in my lab particularly and also show you a few first slides. I will also able to update you where does the field of photobiology and signal transduction stand in plant systems.

Let me get started. I am going to share with you some roles this particular transcription factor bZIP48 plays in regulating plant development and most of the earlier work on light signalling was done using Arabidopsis as a system.

We are trying to see also whether in rice, which is a monocot system, whether this has standard system of like the way Arabidopsis does in terms of light signalling. The work I am going to share today with you is largely done by one of my students Naini and another one Aakanksha who are continuing this work further.

Just to give you some background on what kind of sensory receptors perceive light in regulating plant development: The first photoreceptor which was ever discovered in plants to sense the light was this pigment system called phytochrome. This is largely responsible for sensing red and far-red light and scans the solar spectrum between 600 nM to about 750 nM.

The blue part of the spectrum is sensed by cryptochromes and phototropins. The chromophore, which is responsible for sensing light and eventually regulating the activity of this particular protein, is a kinase and turns on the signal transduction. Light is sensed by open chain tetrapyrrole called phytochromobilin. Everybody knows chlorophyll which is a cyclic molecule. But this one is the same kind of tetrapyrrole bridge but is linear. This is phytochromobilin. This senses red and far-red light. As opposed to that, cryptochromes and phototropins, although they all scan UV-A light blue light but the photoreceptors, which are responsible for that, they are quite different.

In case of cryptochromes, the sensory pigment is basically MTHF, which is methylene tetrahydrofolate. This is another one called flavin adenine dinucleotide, these together are able to sense blue light to regulate activity and eventually plant development. This receptor also senses blue light but then there are two sites for particular chromophore here. This is not FAD but FMN. That is something very surprising. But this still is able to regulate different kind of responses during plant development.

Now this is time to tell you that cryptochrome is a very unique molecule which was identified way back in 1993 by Ahmed and Cashmore and that it was the first ever blue light receptor discovered in any living organism. When they compared the structure of this particular protein, they realized that it is actually very close to DNA photolyase. Largely because DNA photolyases also are photo activated by blue light, but they are involved in actually repairing the damage caused in DNA because of UV light.

Some of you are familiar with the literature: If you expose any organism to UV light -- UV-B in particular or UV-C -- it causes thymidine dimer formation. And after exposure, if you put the organism in dark there is more damage. But if after exposure to UV-B light, you put it blue light, it is able to cause repair. Thats largely because the enzyme which is responsible for repairing this damaged DNA is actually activated by blue light. Thats where the similarity came between this cryptochrome, which sense blue light to regulate plant development, and DNA photolyases, which is involved in DNA repair.

The difference is the tryptophan residue, which is necessary for further activity in photolyase, is missing in cryptochromes. So largely the similarity comes in the portion of the protein to which the chromophore attaches because the site has to be conserved to which FAD and MTHF are actually attached.

So, one can collect snails. Their shells are all marvellous. If you go across the south-eastern coast of India, south of Chennai, go all the way to Rameshwaram, you will find lots of shops selling these shells. In fact it is from one of those areas, where there are lots of shops devoted completely to put in shells out onto the market that Dr. A P J Abdul Kalam, the Rocket Man and former President of India really comes.

There is one more now.

Most of you must be familiar that UV-B causes damage to living organism and sort of to the plant also. But there is also a UV-B receptor in plant has been identified now, that is able to sense UV-B light and elicit some responses which defend the plants against the damaging effects of UV-B light. UV-B is called uvr8. This is a protein which has tryptophan residue which senses the UV-B light and elicits a few responses to accumulate for anthocyanin, which can then screen and filter UV-B light and prevent damage from further exposure to UV-B light.

Likewise, there are many more receptors that are involved but I am going to focus only on phytochromes, cryptochromes, and phototropins in this particular slide just to tell you what exactly their function is during different phases of life cycle.

These are present in practically every facet of plant life cycle which is controlled by light -- right from seed germination, which is light controlled. This is largely regulated by phytochromes in red and far-red reversible fashion. If you give red light, it promotes germination immediately. After that you give far-red light and it will reverse the effect of red light and then the germination is blocked.

This is one of the first pigments on the slide. It shows red and far-red reversibility in living organism. That was the time, when it was discovered and shown that there is some such system which operates in reversible fashion, lot of people, you can say doubting toms didnt believe that such a pigment actually exists. The term used by these people was that it is the pigment of imagination. USDA scientists said such a system (does not?) exist in the plant system. Later on, we know now that such system is certainly a reality.

Like the way Mr. Gupta has been working on the life and the contributions of Dr. SubbaRow. Likewise, another journalist in the US, Linda C. Sage, was looking for some such kind of topic on which she should write a monograph. She talked to many people and somebody then told her why dont you write of the discovery of phytochrome and its role in plant development. In a very fat book of about 500 pages published by Academic Press about 20 years ago, she gave the title Pigment of the Imagination: A History of Phytochrome Research rather than 'figment' of imagination. This book is a very wonderful book. Try to read that if you have time and patience to read 500 pages.

Coming back to that light regulates germination. If you grow these germinating seeds in dark, you will find it has an etiolated phenotype which is elongated hypocotyl and typical hypocotyl hook. In contrast to that if germinating seeds are put in light, there is no elongation of hypocotyls. You will have rather find hook is open and cause the expansion of cotyledons. And then you will find further differentiation of leaves which give rise to regular rosette and plant system. This kind of transition from etiolated phenotype or etiolated life cycle to a regular photomorphogentic life cycle is also controlled again by phytochromes as well as cryptochromes.

Another feature, which is also under control of phytochrome is a process called shade avoidance. In a very densely cropping system, you actually see there is no synchrony in germination -- some seeds lag behind in germination. That will lead to such a situation where there are seeds which germinate faster and grow. Their expansion of the cotyledons of the leaves will cause some kind of shading effect on late germinating seeds. What happens is because of some kind of shading effects. This shading effect causes immediately an elongation of such a seedling and this largely because of another form of phytochromes. Actually there is not one phytochrome. There are five phytochromes in Arabidopsis, pyh A, B, C, D, E and there are three counterparts of that in rice, wheat and other monocot systems. One of the phytochromes controls these responses like that elongation response which is called shade avoidance actually is a wasteful process. Because all the energy in the seedling is taken away towards elongation of plant as such, you get a very, very poor yield. If somehow you are able to manipulate or knockout this particular phytochrome, which controls elongation, you will have very much better harvest.

Another controlled process is phototropism. This is controlled largely by phototropin-1. I am the first to isolate a mutant of that which is not phototropic. That mutant was utilized by others to clone the gene that led to the discovery of this particular phototropin called phototropin-1. When the gene was available, at the same time, the genome of Arabidopsis also was getting sequenced.

Looking at the genome sequence, they found one more sequence similar to phototropin-1. They gave it the name phototropin-2. Later, they created a mutation of that and showed that particular phototropin only partly controls phototropism but largely controls chloroplast relocation in the cell. What happens is photo-oxidase damage to the chloroplast because of the presence of phototropin-2 when you give low light. Every cell has about hundred and hundred fifty chloroplasts. They keep rotating in a kind of streaming movement and get uniformly distributed.

When you have low energy of light, they accumulate in a certain orientation so that they can gather more light for photosynthesis of the system. But on exposure to high energy light, they will have a different orientation so that they are not damaged by the high-energy intensity and light energy diffuses through the system and the photo system is not damaged. This chloroplastic movement or chloroplastic accumulation response or chloroplast avoidance response is controlled by phototropin-2.

Another response regulated by blue light in addition to phototropism and chloroplast relocation is stomatal opening and closing. Everybody knows about the role of protein reflexes and overall stomatal turgidity and gas issue opening of stomata with gaseous exchange. But people didn’t really know what exactly is the light system which is responsible for the stomatal opening. Once I isolated that mutant, and a group in Japan was able to isolate a mutant of this. It was surprising that in mutant called phototropin-1, this response was normal. So was the situation also in mutant of phot1 where this response was normal. It came as a surprise that how come you have one photoreceptor regulating the phototropism and another photoreceptor sensing the same kind of light regulating chloroplast relocation and if these mutation don’t influence closing and opening of the stomats.

Does that mean there is a third photoreceptor? It turned out that you have a double mutant of phot1 and phot2 that was defective in this response, indicating some degree of overlap in the function of phot1 and phot2 in regulating stomatal movement, which is so very vital for overall plants productivity by gaseous exchange of oxygen and carbon dioxide. As I mentioned in the beginning, there is a strong biological clock in plants. There are many examples. Many flowers open in the morning and close in the evening

We have many blooming trees. We also have those pinnies closing evening and opening in the morning. This kind of clock is very, very versatile in the plant system. But to be able to train a biological clock, we need to have light as signal. Temperature has also a role to play in that. That light is sensed by phytochromes and cryptochromes. In the sense, both red and far-red light as well as blue light and regulate and entrain the biological clock. This is the last part of the light cycle of the plant system which is very, very necessary to complete the light cycle.

We have phytochromes and cryptochromes responsible for sensing light signalling. In photoperiod sensing, the leaves sense the photoperiod and the signal is transmitted to the apical meristem and that causes transition to flowering at the apical meristem. Its conversion of apical meristem to floral meristem. Photoperiod sensing in the leaves is largely done by phytochromes and cryptochromes.

Now you have a fair amount of idea that every major developmental stage of plant life cycle is under the control of light and that these sensory photoreceptors have overlapping function to regulate plant development. Some cases, they have specific roles. In other cases, they have an overlapping role. In a way that may be the system that has been evolved is that in case a plant is growing under a shaded canopy, you wont find red light penetrating down to the seedling because red light is being absorbed by the leaf for photosynthesis and the same light is also activating these phytochromes. So essentially you require some other kind of light that might be necessary to regulate plant development where the seedlings and plants are under the shade conditions. May be blue light takes a part of the job which is otherwise done by phytochromes.

Our understanding of the signal transduction has largely come through work on Arabidopsis as a model system. It is a very, very complex signalling cascade. Many of the pathways get affected by photoreceptors that are photoactivated. I am showing you one of the major pathways in a very, very simple way. We talked about different photoreceptors. There are five phytochromes, three cryptochromes, two phototropins and one uvr8 -- and of course light effectors.

They are all activated by different monochromatic light. As a result of that, they are able to activate different signalling components. Some of them are called phytochrome interacting factors, phytochrome signalling substrate, phytochrome-A. Most of these factors eventually activate or regulate activity of the factor called COP1, which is part of E3 ligase involved in protein turnover. Eventually, this COP1 is the one which is able to regulate the activity of HY5 and HYH. In dark, it is able to degrade that, on photoactivation. This is not able to bind them and they are photo activated. HY5 is one of the major factors involved in light regulation of the transcription of various genes, which are light regulated.

Most of this work on COP1 and HY5 interaction has been done by Xing-Wang Deng at Yale. Many components of E3 ligases system and 26S proteosomes were discovered first in the plants and later on they were discovered in animal system also. Deng has shown how HY5 and COP1 interact to regulate plant development. Under dark conditions, COP1 interacts with SPA1 and this particular complex is then able to recruit HY5. HY5 is ubiquitinated and is finally degraded. As a result of degradation, HY5, basically a bZIP transcription factor, is not able to bind to the promoter of the genes, which are light-regulated and terminate photomorphogenesis.

Thats the reason you have an etiolated phenotype. It is photomorphogenesis in dark because HY5 is not free and is degraded. As a result of that, there is skotomorphogenesis in the dark. Under light -- whether it be red light, far-red light or blue light -- COP1 has signal for cytosol localization. Under these conditions, COP1 is shunted outside of the nucleus. There is some kind of titration in the nucleus and in the cytosol. Once COP1 is shunted out of the nucleus, COP1 is not able to make a complex with SPA1. As a result, HY5 is free. It is not bound to COP1/SPA1 for degradation and it is accumulated in a large amount in the nucleus and binds the promoters of many genes which are light regulated, leading to photomorphogenesis.

Deng has also shown by CHIP-CHIP assay that HY5 binds to the promoter of roughly about 3000 genes which is very, very remarkable. It means either up regulation or down regulation to regulate plant development. With this background, I would take you to the kind of work we are doing. Most of this work has been done on a dicot system like Arabidopsis, which is a plant from the family Brassicaceae to which mustard also belongs. We wanted to see whether the overall phenotype and the morphology of monocot, which is quite different from that of dicot, is also regulated by similar components or whether there are some unique features of this component or different novel components which might regulate plant development. We moved on rice because we were involved in the sequencing of the rice genome way back in 2000-2001.

Until 2005 when our paper, International Rice Genome Sequencing Program 2005. The map-based sequence of the rice genome was published in Nature (436: 793-800) and the picture I am showing you now was taken in the University of Tsukuba, when Dr. Meenu Kapoor and Dr. Sanjay Kapoor were participating in the deliberations of a consortium for genome sequencing. We were there almost once in a year and they were very precious hosts for us and we had a good time. In this particular programme my colleague Akhilesh Tyagi was coordinator and Dr. N. K Singh, DG ICAR and Dr T. Mohapatra from IARI also participated. Incidentally, I happened to be present in that particular meeting where no other member went and this picture was clicked. So it was incidental that I had to make the presentation. In its 20th August issue, Science published a small story that the genome of rice has been sequenced.

Once we completed the sequencing of rice genome, the next question arose, what all these genes are doing? You will be surprised that we claimed in this paper that there are about 37,000 genes, which are coding for proteins. The original estimate was that there were 55,000 genes and the remaining 18000 could be of the non-coding type. When we got better annotation tools, it turned out that there are about 31,000 genes, which actually are coding for the proteins in rice.

Likewise when the genome of Arabidopsis was published, they claimed about 25,000 thousand genes and as annotation improved, the gene number in Arabidopsis actually came up 19,000,000. Actually, rice and Arabidopsis are not much too different. And this is also true for the human system. The number of genes must not to be different between these two plants and human systems although modifications may take place in terms of altered splicing and lead to many different kind of products in the human system as compared to the plant system.

With Dr. Sanjay Kapoor, Prof. Tyagi, Dr. Usha Vijayraghavan, Veluthambi at different places in the country, We had a major project sanctioned by DBT where we tried to focus on understanding the genes and their regulation during various stages of panicle development and seed set in rice. My lab focused on part of these three stages and lastly on, some of the stages involved in early transition to flowering.

Dr. Sanjay Kapoor focused on panicle stages and Dr. Tyagis lab focused on seeds stages. Students in-house carried out microarray analyses of all these different 19 stages of reproductive development including panicle transition to seed development. The data taken from all our students was largely computed by Dr. Sanjay Kapoor to look at the different genes expressing at transitional stages of the development of panicle or seed set in rice.

The first spur in activity of gene changes was when there was a transition from shoot apical meristem, which is vegetative meristem, to initial transition, which is called P1 stage. There is another spur between P3 and P2. Another one at P5, P4 and so on. There were some changes also during seed development after fertilization.

What exactly these stages represent?

This particular transition is largely because about 2500 genes are differentially expressing when the organogenesis take place in panicle development followed by meiosis, followed by mitosis, differentiation, fertilization, and embryo development. This is the stage where seed desiccation takes place. In seed maturation, there are very few genes are involved in that.

We were largely involved in also taking care of abiotic stress genes during seed maturation. What exactly are the genes which are showing this kind of differential expression?

This is focussed only on the transcription factors which are highly expressing in panicle development. Focussing on this particular part of the heat map, we realized that there are some transcription factors whose expression is high. Red colour shows high expression, dark one shows median expression and green shows very low expression.

Look at this particular gene here. These variable genes are transcription factors expressing very high in P1 and P2 but the expression goes down in later stages. There are other genes which turn on P3 and go down later on. Thats the way different transcription factors activity is coming up, on and off during different stages of panicle development. When we see the nature of this transcription factor, we find that they include the Dof domain containing transcription factor, PSD, SRL, bZIP, polycomb and bHLH and homeodomain genes also which are showing differential expression.

Our groups in the South Campus decided to focus on different families of these transcription factors. My lab is trying to analyze the entire bZIP gene family in rice and how different genes of family are expressing differently at different stages of panicle development in vegetative tissues.

This is one representative example, we published in a paper way back in 2008. It has right now more than 300 citations and drawn the attention of many workers all over the world. These are the expression stages for vegetative tissue. The green one shows hardly any expression and you see that in particular bZIP47, 62. They are high in SAM, still higher when the transition takes place and gradually go down during later stage of panicle development and seed settings. The other basic bZIP transcription factors are not expressing at any stage either in vegetative tissues or organogenesis stages. They come up only in later stage of seed development and the role might be lastly in seed maturation.

We were surprised that the first ever bZIP protein, identified about 30 years ago by Ralph Quatrano, that was involved again in seed maturation and regulating expression of LEA protein genes. It has been shown for the first time that they may be involved in regulating seed desiccation and seed maturation. I have talked about broadly about bZIP proteins.

In the very first slide I showed you, I just focused on bZIP48 which does not have much of a change in expression at some stage -- specifically in the floral organs but also expression in the vegetative tissues.

These kinds of data give an idea where this gene is expressing and is likely to have function also in the same tissue. Thats why have selected 25 different genes for functional validation.

I am going to focus on only on those involved in regulating light-induced plant development and on those which are orthologous to HY5 in case of Arabidopsis. For, HY5 is also a basic leucine zipper class of transcription factor. When we did this phylogenetic analysis, we realized that there is only one AtHY5 in Arabidopsis but that this is also homolog for AtHYH, which is also there. When we put in the sequences of rice bZIP proteins, it came to our knowledge that there are actually three bZIP proteins, which show homology with AtHY5. They are bZIP1, bZIP18, bZIP48. There is no ortholog like AtHYH in rice and other monocots. It came as a surprise that there are three orthologs of AtHY5.

And what exactly might be their role? Do they have similar function? Do they have overlapping function? Have they acquired a novel function during the course of evolution into neofunctionalization? Work is going on two of them -- bZIP1 and bZIP48. I am going to only talk about the work largely done on bZIP48 because that one was able to clone. WE had a little difficulty with bZIP1. We still have a problem cloning cDNA of bZIP18 although the genomic clone is working. That is how it gets processed and how it expresses and functions differently from other two bZIPs.

I said bZIP48 is orthologous or homologous to HY5. It is actually orthologous to HY5. There are two different ways to assign a function to bZIP48. One is to obtain the mutant of HY5 and see whether it is possible to complement that. The other way is to raise the transgenic which may knock out the function of bZIP48 in rice or to overexpress that and see how it alters the phenotype. We took care of all three approaches. This particular slide shows this mutant of Arabidopsis. This is a young light grown wild type seedling just about four or five days old. The short hypocotyl, cotyledons are expanded. Thats normal phenotype.

When you mutate this particular gene HY5, it causes elongation. It grows in light. I showed you the dark-grown phenotype, which is etiolated elongated hypocotyl although it has expanded cotyledons not typical hypocotyl hook. It was very, very useful to have been able to identify the function of this particular gene. If knocked out, HY5, it will not be able to give photomorphogensis responses like the way wild type does.

We used this rice gene and complemented the mutation. The HY5 overexpressed with bZIP48 in rice will show the same phenotype as the wild type does, indicating that functionally the rice gene is able to complement mutation in the Arabidopsis. It may have analogous function to perform.

Normally such a gene, which controls plant height, if overexpressed, should cause further suppression in growth. That did not happen. We were not surprised too much for a reason. Even when Arabidopsis HY5 was overexpressed, it did not cause further reduction. That is something very, very unusual.

People working on HY5 in Arabidopsis are still trying to find the answer. Although we knock it out, we have elongated phenotype. When you overexpress it, it does not alter further height but causes compression of that. The same thing must be happening also in rice overexpression. Rice bZIP48 is almost orthologous to Arabidopsis HY5.

Nainy, who started this work and finished her Ph.D. a couple of years ago, noted two more phenotypes in Arabidopsis HY5 mutant which was not reported earlier. At least 3, 4, 5, dozens labs all over the world must be working on HY5 and I do not know how they escaped what she noticed, namely, the wild-type seedling has more of this position of cotyledons which are in that particular fashion. HY5 has a duping kind of phenotype. It does not happen when HY5 is overexpressed. Complementing the mutation partially takes care of minor changes. A large amount of complementation is taking place to make the phenotype do like the way wild type does. Same thing was also noted by Nainy. That root hair shows the gravitropism in this particular HY5 which also could be rescued by the overexpression of the rice bZIP48. This is something much more striking.

Look at the phenotype of the flowering plant of Arabidopsis. There is a certain angle of silique on the vertical stem which is called the bolting stem. You see the HY5. It has actually a wider angle of silique. Complementing that with the rice bZIP48 gives phenotype like the way wild type does.

These three phenotypes were not noticed by the researchers who initially worked for almost twenty five years in Arabidopsis. We are reporting it for the first time and it should be published within a few days time. It is already available online in Plant Physiology. After being sure that bZIP48 is able to functionally complement HY5 mutant, we also went about bZIP overexpression lines of this particular gene in rice as well as RNAi-lines. This is the phenotype of young seedlings of overexpression lines, the vector control which is more like wild type. Unlike Arabidopsis, where the overexpressed gene did not alter the phenotype in the wild type, but in case of rice, there is a drastic reduction in the height of young seedlings. If you see the adult plants also, in the three independent transgenics, there are compression in the internodes is very, very dramatic compared to the wild type which is much normal.

We were very excited looking at this phenotype. Is this gene involved in the Green Revolution gene, which is called semi-dwarfism in wheat? We were a little surprised. This is something some of you are not familiar with although you have heard of Norman Borlaug, who was responsible for discovering this semi-dwarf wheat, which led to the Green Revolution eventually.

Later, we found that this was the gene which was mutated in a component in this particular semi-dwarf wheat. That is signal transduction component of GA pathway. And. GA needs to degrade that to be able to turn on. That particular negative regulator of GA signalling was actually mutated. As a result of that, GA turns on the GA transcription. The GA effect in elongation was compromised and resulted in the semi-dwarf phenotype. We thought HY5 might be doing something like that. But it cannot be a surprise to us that all this causes the shortening of the stem in case of rice. In the wild type, there is one internode. There are two internodes in case of its overexpression line in rice. We found to our surprise that the stem will be more stout and steady if there is a reduction in the height. That didnt happen. Actually the stem was much thinner compared to the wild type. Also, there was reduction in the cell length as well as reduction in the old vascular bundles. This turned out to be something much more dramatic than what we thought. It might be working closer to reduction genes.

This picture on the screen also shows the stem picture and reduction in the vascular bundle in the wild type. All the three are independent transgenic lines. To our surprise, this secondary cell wall thickening was very, very thick in case of the wild type. There is an unequal reduction because they are all independent lines. May be the expression level is different, depending upon the insertion of the transgene. But all the three dramatically have much less secondary cell wall thickening. That is why the stem cells are actually very, very brittle and not steady although they are much reduced in terms of their height.

Looking at the phenotype, where we have raised RNAi lines, what we find is that the internodes are much longer compared to the wild type. I dont know how many of you have seen wheat or rice growing in the field. If you have seen that, you will find that there are many tillers coming from one seed and those tillers actually appear from the compressing internodes which are buried inside the soil.

In this particular case, because we have knocked out the function of HY5 ortholog in rice, which is bZIP48, it is not able to compress the internodes. That causes the elongation of that. What you find is the elongation of the internode. You also see roots emerging from the aerial part are elongated. Roots are also green, which is something very remarkable that happens in this particular case. It simply shows that bZIP48 causes reduction in height when overexpressed. When you knock it out, it is able to cause elongation in internodes. In the meantime, we also searched the database in Korea, which has a large collection of T-DNA tagged mutants.

We were able to fortunately get the same gene we called bZIP48. We had another surprise when we analysed that particular mutant. We got the line which was a mutant heterozygous and we had to segregate them multiply back. In segregating population, what we realized was that there are some which are as good as in terms of the plant height as for wild type and there are others which are highly reduced in terms of overall development. We actually noted that although there was not much elongation in shoot, there was extensive root proliferation which also is true in the case of HY5 mutant in Arabidopsis. This particular seedling, unlike HY5 in Arabidopsis, is able to mature and set seed but it did not grow beyond this stage and finally perished. We had therefore to maintain the segregating population heterozygous lines multiply back and keep analysing. There again was a surprise: In case of Arabidopsis, if you overexpress HY5, it does not cause any change in height and if you knock it out, it causes elongation. Here very strong phenotype of knockout T-DNA tagged. This one was not able to survive beyond a certain stage of development. And, in case of overexpression, we had height reduction. We observed major changes as a function of bZIP48 vis-Ã -vis HY5.

Since we had already in-house facility of doing microarray, we compared all these transgenic lines, overexpression, RNAi, mutant lines and everything. In case of rice, we did some analyses of different assays. We found that the expression of genes involved in GA biosynthesis, BR biosynthesis, jasmonic acid and many other pathways were affected.

I focus here on these particular genes which are involved in conversion of thiopurine to pyrenol and further on to glutaraldehyde, which is precursor for GA3 eventually through the function of GA20 oxidase. Then we searched for GA1 and GA4, depending on the species, which are functional GAs in case of plant system. What we saw very clear is that in vector control, which is like wild type, expression of these genes are very, very high and in overexpression line, expression of these particular genes are very, very low.

So in the very first or second gene involved in GA biosynthesis, its expression is low. It means you have low GA activity. Thats the reason why you have reduction in plant height.

When we sent this paper for publication in Plant Physiology, the editor said: If you claim that the expression of this gene is down in the transgenics, why dont you show the binding of this particular transcription factor to the promoter of that. We had to do this experiment for six months before we got a response. Many other things also had to be done in meantime. We were able to finally analyse the promoter of this gene called KO2, pyrene oxidase, and realized there are three elements called light-responsive elements (LREs) and these are G-box elements and Nainy was able to show that G-box2 is bound to the bZIP48 protein which we were able to express and purify in bacterial expression. We were also able to knockout this particular site where G-box is located. If you have to have functional G-box2 to be able to show retardation in EMSA and if you mutate that, it is not able to bind that.

This was the clinching issue to show that how exactly HY5 is able to regulate gene in plant system finally contributing to the alteration in the height of this particular plant. In addition, we were able to show what is the expression of this gene in transgenics, knockout or overexpression lines. Once this was compared to KO2, we also realized we can take N15, one of the mutations in KO2 which gave semi-dwarf rice in Japan. This also correlates very well that this particular gene has a very important role in the regulation of plant height otherwise also.

In addition to this promoter interaction with particular protein, we focused on the transcripts and other things about the protein. Because protein is degraded in case of HY5-COP1 interaction in Arabidopsis, why dont you show similar things is happening in case of rice? Fortunately, I had got this antibody made about 3-4 years ago but Nainy was not able to standardize the proper western blot. I was keen to use this antibody to carry out CHIP-CHIP assay to see what all genes are bound to. As that would be a very complicated thing, she decided to finish her Ph.D. and then do those experiments. I got luckily a post-doc by that time starting for revision. Another student, Aakansha, joined. She was already standardizing and was able to very nicely standardize this protocol. Within a span of two months, she was able to generate much more data. I will show you later what happened in case of Arabidopsis.

We grew these 4 days old seedlings in light and also in dark and light grown seedlings were exposed to dark for different hours. Like-wise, the dark grown seedlings were exposed for four days to light for different hours. We did not find any change what so ever. There was transition from light to dark and dark to light. And there was also recognition of two bands parallel. But this is known in case of HY5 which is phosphorylated. The upper form which is phosphorylated moved smaller or lesser as compared to the other one. It is very, very clear that the upper band could be because of phosphorylation of this particular transcription factor. The lower one is not phosphorylated. This was the major surprise.

Why this particular protein is not degraded like the way HY5 does because of COP1 interaction. This is the data taken from the published literature. This is what happened in case of Arabidopsis and what you see here is again light to dark transition. In light, you have a very high accumulation of HY5. As you put it in dark, this protein is degraded within 15 to 20 hours. The data we are finding in case of rice is quite dramatic compared to what is happening to this. And the same thing in the dark grown seedling, there was very low accumulation exposed to light in order to recover accumulations. Because of COP1, we have stability of HY5 accumulation but we did not find anything like that.

We sent this data back to reviewers and comment came that we have such a wonderful set of antibodies and everything else is now fine but have you tested in your Arabidopsis transgenics which you complemented what happens to the protein? The handling editor wrote: If you prove this and show what has happened, that will clinch the issue and the paper will be accepted for publication.

Aakansha had to do two more experiments to show that. This is a positive control containing protein, which is recognized by the antibody. You then have HY5 mutant as well as wild type Arabidopsis where bZIP48 antibody does not recognize anything because they dont have anything in terms of rice. So, Arabidopsis HY5 is not getting recognized by antibodies of bZIP48 of rice. We had put in the transgenics and overexpress rice gene in case of Arabidopsis. There also we did not find any change in these two bands when they exposed them to different hours of darkness. So, it looks that the rice bZIP48 protein does not somehow interact with COP1 and does not get degraded. We found the alteration in phenotype in the case of rice and Arabidopsis.

This was the clinching evidence we provided and the paper was accepted. It should be published any time in this week. The paper was accepted in August and was due for publication in September but the editor of the journal wrote to us that they are bringing out a special issue on energy, focusing on light and oxygen. Would you mind waiting or would you want to put our article in the focus issue which would be published in February 2018. We agreed for that because we will have more visibility.

(See the article OsbZIP48, orthologous to AtHY5, exerts pleiotropic effects in light-regulated plant development. Burman, N., Bhatnagar, A., and Khurana. Plant Physiology Focus Issue Energy: Light and Oxygen 176 : 1262-1285)

I am going to show a few more slides since its a transcription factor that is nuclear localized. It also makes homodimer. This was confirmed by BiFC and shows up by FRET assay. We were again able to show dimer formation interaction like that. But for whatever reasons, the data is not very clear. By yeast two hybrid assay, we were not able to show any transactivation. May be it involves other transcription factors to be able to regulate transcription. This is a good evidence to show why bZIP48 protein is not getting degraded. There is some evidence that we can show that this COP1 is expressed in nucleus and also somewhere in cytosol and this bZIP1. We showed you that bZIP48 goes to the nucleus. Now, bZIP1 and the other gene Aakansha is working on, accumulates in the nucleus. But if you have interaction by basically BiFC, we show that COP1 and bZIP show interaction. It means COP1 is able to interact with bZIP1. If you do this, there will not be any interaction at all. This again is clinching evidence that bZIP48 gives a different kind of phenotypes.

We are characterizing bZIP1 more in detail to have a closer look at the HY5 function in Arabidopsis. That is the reason why bZIP48 is not able to interact. We also believe we will be able to deduce, looking at the sequence, that the domain which is known to be interacting in COP1 and HY5 in case of Arabidopsis is mutated in bZIP48. Thats the reason why we do not find any interaction.

The final slide shows why this particular protein is degraded, why there is etiolated phenotype and why this protein has some function. These are the major features now that bZIP48 is able to complement HY5 looks like it is analogous function to perform but, unlike HY5, it is able to cause reduction in plant height when overexpressed and is able to cause lethality and root proliferation when you knock it out. Of course this is a dark and light stable protein, which is different from what is known in case of Arabidopsis.

I would like to thank you for your attention. And, this is a bouquet of flowers from Tsukuba. They are cherry blossoms. (Omitted is the portion of the presentation that Professor Khurana asked not to be published on the website

Lecture video recorded by Vishal Tuli and Soham Bharadwaj and transcribed and edited by Dr Dinesh Jaiswal with Dr. Irum Rizvi.