From Cdk5 to Alzheimer’s disease: future challenges

▪ What first inspired you to study neuroscience & was this always the career path you intended to follow?

Whilst growing up, I always liked science and I actually went to a vetinary school in my native country, Taiwan. Now, when people ask me if I always knew I wanted to be scientist, I realize that when I was a very young child in Taiwan, I never considered becoming a scientist. But one thing I do remember is that my parents would buy us a lot of books to read, some of them were science books like basic biology and astronomy. Some were nonscience, more like novels and fiction, but I have always been drawn to those scientific books. I loved reading about the insides of plants and animals and I actually thought that everyone was the same, but they are not – having my own child has made me realize that not all children are drawn to science, people have very different interests. I now realize that I always liked science and am passionate about it.

My PhD study was in the cancer department; I studied an autocrine growth factor that stimulated cancer cell growth in petri dishes. Even after my PhD study, I continued cancer research by joining a laboratory that was concerned with the identification of antioncogenes (later the more common name, tumor suppressors, was coined). I identified a large group of protein kinases that showed homology with the kinases in yeast that promote cell division. I became quite intrigued by one of the kinases, known as Cdk5, simply because it was difficult for me to pin down where the kinase acted. I had trouble detecting the active form of the kinase in all the cancer cells that we used in the laboratory. So, eventually, I used tissues from the entire mouse and basically went cell type by cell type, organ by organ, tissue by tissue, determined to really figure out whether this kinase is actually chemically active at all. At the time, I found that there is only one organ in which I was able to see the activity of this kinase, and it was the brain.

Then I started to read papers and books about the brain and about neuroscience and I was fascinated by it. At the time, I also collaborated with a child neurologist, Verne Caviness, who studied cerebro-cortical development. He really inspired me to get into neuroscience; he taught me a lot of things, both on the research side and also things that are not necessarily science-related, so he was like a mentor to me. I was drawn to studying the brain ever since.

▪ Can you give an overview of your very first studies & how your research interests have developed since then?

After I established my own laboratory, the first thing I did was to create a mouse model for Cdk5 loss-of-function and it turned out that this mouse displayed very interesting brain development abnormalities. I was very proud of myself! I remember looking at the brain sections of this mouse and being able to tell that there was something wrong with the distribution of cells within the cerebral cortex and in the hippocampus. When I first showed those sections to a pathologist, who is very experienced in looking at mouse tissues, he told me that my mouse had no abnormalities and no phenotype. But I was not discouraged; I bought myself a microscope and trained myself to be a pathologist. I looked at section after section as I was sure that there was something very wrong. I went back and showed them to him and in the end he had to agree with me. Those were the fun days! I’ve studied brain development ever since.

I think my venture into Alzheimer’s disease (AD) was unexpected. By studying Cdk5, I accidentally found that this kinase can become hyperactivated and dysregulated under neurotoxic conditions. This started a long journey for my laboratory, involving investigations into AD-related mechanisms and studies into the underlying pathology, as well as ways to ameliorate the pathologies and symptoms in mouse models. So it has been a fantastic journey. My laboratory has been working on AD since the late 1990s so it has been almost 15 years now.

▪ How would you describe your motivations & attitude to research?

I really feel that it is a privilege to be able to have my own laboratory and have funding to do the things I really enjoy doing. Every day I come to work knowing that I am going to learn something new and that is such a motivation, especially as I know that the work I carry out will make an impact and contribute to better understanding of such devastating diseases and will perhaps help realize novel ideas for therapeutic intervention. That gives me a reason to be very motivated and very happy with what we do.

▪ What are currently the main areas of research in your laboratory?

In terms of AD, we try to look at the changes that happen in the very early stages of the disease and the mechanisms behind the progression of the disease. In particular, we are really interested in the epigenetic arena because we have found that there are many alternations in the nucleus, such as in gene expression, and I think that this is not just changes in the expression of one or two genes, but a pattern. There is a pathway that alters the expression of our genes involved in the whole pathway.

At the Picower Institute for Learning and Memory (MA, USA), where I am heavily involved, many scientists are studying brain circuits that underlie a particular behavior in a particular part of the brain and I am also interested in trying to figure out how different parts of the brain and different parts of the circuit may contribute to different aspects of Alzheimer’s pathology. So we’re using multidisciplinary approaches to study AD, from molecular, cellular and genetic approaches, to symptoms and behavioral neuroscience. So those are the areas that we are particularly interested in.

▪ What is the greatest advance that you have witnessed during your time in the field?

There have been quite a few. I would say the increasingly advanced and refined genetics that allow us to manipulate rodents and to specifically ablate or express a particular gene in very particular cell types in a defined temporal time point; I think that really revolutionized how we can approach the brain. I think that is very powerful. Additionally, the RNAi technology that permits researchers to study loss-of-function effects, not necessarily through mice knockouts, but you can just use hairpin RNAi to knockout a particular gene product very easily. I think that has really made a huge impact. Also, recently, this new technology using light-activated or -inactivated channels to manipulate brain circuits really helps a lot. I should also mention GFP and the identification of all those fluorescent proteins that allow one to easily label or mark particular cell types. I think that a combination of these advances has really made studying the brain possible.

▪ What has been your greatest achievement to date?

We have identified a signaling pathway in which hyperactivation of Cdk5 contributes to neurodegeneration and AD-like pathology and have created a very powerful mouse model that really shows profound neuronal loss and neurodegeneration. We can utilize this mouse model to identify early mechanisms underlying disease progression and also to explore new ideas for therapeutic intervention and I think that now leads to the whole epigenetic idea of regulation and dysregulation in AD. So I think that has been my main contribution.

▪ As director of The Picower Center for Learning & Memory, what are the main aims of the center & what do you hope to achieve?

The center really aspires to understand the mechanisms underlying memory formation, learning and cognition. We have investigators using hugely diverse but complimentary experimental systems to achieve this goal and to start to think about how to deal with devastating brain disorders, such as AD, Parkinson’s disease, schizophrenia and autism. First, we have to understand how the brain works and how the neurons communicate and coordinate with each other to ensure that a particular behavior and sensory response occurs. So this is our goal – to understand the fundamental aspects of the brain to gain better insight into brain disorders.

▪ What are the underlying links between your far-reaching research areas of psychiatric disorders, Alzheimer’s disease & learning?

AD is well known for learning and memory impairments and eventually dementia. However, it is probably not as well known that psychiatric disorders, such as schizophrenia and autism, also severely impact cognitive function. For instance, cognitive impairment and learning deficits are common features of schizophrenia, and also of autism, although there are exceptions. Thus, they share some very similar fundamental symptoms and underlying mechanisms: there may be a theme underlying the cognitive aspects of these different diseases.

▪ How useful are mice as model organisms & when do you foresee your research being translated into human trials?

Mice are the most highly used organisms because of the ease of genetics; one can freely manipulate gene expression in mice and a lot of behavior paradigms and electrophysiology techniques are well established for mice, so in a way, they are very accessible to use. The observations made in mice, especially disease-related observations, are going to have to be tested in humans one day to assess how meaningful the results are when translated into humans. That is a big question because not all the chemicals that tested beneficial in mice also show meaningful effects in people.

▪ Can you provide some background on your recent Nature paper on the effects of HDAC2 on genes & the result on Alzheimer’s pathology?

This is a continuation of our first paper published in 2007 in Nature, where we made a very surprising finding that in our powerful neurodegeneration mouse model, we can see very profound neuronal loss in the hippocampus and in the cortex and the mice suffer very severe memory impairments [1]. If we treat them with HDAC chemical inhibitors, we can very significantly recover their learning and memory capacity, despite the profound neuronal loss. In that report we showed that even though there is profound neuronal loss, the remaining neurons seem to function at a higher capacity: they develop more synapses and have more abundant dendrites.

So we basically wanted to further validate those observations to try to understand how HDAC inhibition can lead to beneficial effects on cognition. We decided that we really needed to identify the precise enzyme that regulates learning and memory. We identified HDAC2 to be this enzyme, so we used a combination of genetic and pharmacological approaches to demonstrate that HDAC2 plays a major role in regulating cognition [2]. It binds to the promoter and regulatory elements of many genes involved in learning and memory. Also, HDAC2-deficient animals appeared to be refractory to the effects of HDAC chemical inhibitors.

In this current Nature paper, we found that HDAC2 itself is actually involved in the pathogenesis of AD; we found that HDAC2 is increased in two different mouse models of AD and in post-mortem human AD brains [3]. This increase in HDAC2 results in a greater association with its target genes as well as downregulation of the expression of many genes essential for learning, memory and synaptic plasticity. HDAC2 increase can be readily upregulated by a very well-characterized AD culprit, β-amyloid.

This result demonstrates that in AD, there is a prominent blockade of gene expression at the epigenetic level and it shows that β-amyloid, previously shown to have acute effects in synaptic plasticity by inhibiting NMDA receptor function, can also exert a chronic effect on cognitive function by inducing epigenetic blockade of synaptic plasticity and gene expression. I think that while this study reveals a very interesting new mechanism to explain cognitive deficits in AD, it also suggests that if cognitive deficits are the result of gene expression blockade, then one could argue that if you reverse this blockade by inhibiting HDAC2, it may be possible to recover some of the cognitive function lost in AD patients.

▪ How do you foresee the field progressing in the next 5 years? What major challenges need to be faced?

We need to discover novel ideas and novel targets for therapeutic intervention of AD now more than ever. I think that more individuals now suffer from AD than many other major diseases, and as the population ages, the number of AD patients is going to significantly increase: the prediction is that the number is going to double or triple within the next 20–30 years. I think that we must come up with some meaningful way to either delay or at least attenuate the symptoms of the disease. There has certainly not been a lack of effort in this area as individuals have been trying to do this for many years already; but we have not been able to come up with an intervention that shows significant promise. This is clearly not going to be easy, this is going to be the next major challenge.