Search This Blog


Monday, May 16, 2011

The Snake Oil Salesman’s Guide to Vitamin A

by Mark Dunning

You should take vitamin A. You should give it to your children. You should take it, of course, in a dose that is safe and prescribed by a physician, but you should take it. My daughter takes vitamin A and her vision has not changed perceptibly in the last four years. I have a friend who has two daughters that take vitamin A. Their vision has not changed in nearly a decade. You should take vitamin A.


Don’t act on that!

The above is true about my daughter and my friend’s children. They all take vitamin A and their vision has not changed for a number of years. That does not mean, however, that vitamin A is the reason. It might be, it might not be. We don’t know.

In spite of that first paragraph, I’m really not going to get in to the problems with the vitamin A study or argue that it should be prescribed to everyone with Usher. No, today I’m going to vent my frustration at the scientific community. Sorry, but you guys have dropped the ball.

There is little more controversial in the Usher community than vitamin A and I have to say it drives me bananas (which are not high in vitamin A, by the way). I am not a scientist or a physician. I read all I can on Usher syndrome and have learned a lot. I regularly blog about Usher syndrome, but the truth is I’m still just a dad trying to do the right thing for his child. But I don’t know what the right thing is when it comes to vitamin A.

Many doctors and researchers I trust offer a tepid endorsement of vitamin A. Many others, including a certain person that shares this space, don’t believe the current research justifies vitamin A as a treatment for Usher syndrome. And there are a couple of doctors on both sides who are fervent in either their support or their opposition to vitamin A. In the middle are families, like mine, left to make a decision with a paucity of information and, quite frankly, a ton of conjecture.

Let me take a moment to praise my friend Jennifer. Her posting on vitamin A does a good job of reviewing the studies done to date and laying out the facts. I’m paraphrasing here, but in her opinion, based on the currently available studies, she would not give her child vitamin A as a treatment for RP. I have read the same studies and have a different opinion. Hence, my daughter does take vitamin A.

But here’s where I agree with Jennifer. We need more studies on vitamin A! The study done by Dr. Berson and his colleagues is an excellent study that has followed hundreds of families for decades. The study suggests that vitamin A has the potential to be a successful treatment for retinitis pigmentosa. But it leaves more questions unanswered than it answers.

So why have there been no other studies on vitamin A? Why haven’t other scientists and researchers and physicians sought to answer the questions that have arisen from the original study? Would a potential treatment for cancer sit unconfirmed for this long? For as good as it is, there are problems with the original study. Dr. Berson and his colleagues would be the first to admit it. Heck, they spend a lot of time trying to answer the questions that remain open. But that’s part of the problem. Dr. Berson and his colleagues should not be the only ones answering the questions.

The best science is that which is proven over and over again by different studies performed by different researchers. Imagine if we had three different studies that drew the same conclusions as Dr. Berson’s original study. What would be the advice given to families then? Or imagine the flip side, where three other independent studies found that vitamin A has no effect, or worse, an adverse effect on RP. What would we do then?

Instead families are left to guess and to draw conclusions from an impossibly small sample size. We have essentially one study and the word of mouth of other families, both pro and con. Families should not be listening to me on this subject! (Seriously) Jennifer has written about sham science and I am the embodiment of that. Bella is one kid out of thousands with Usher. Maybe her experience is just the normal progression of the disease. Maybe (and I pray this isn’t true) her vision would be better if she didn’t take vitamin A.

The point is we don’t know. If I were truly a Snake oil salesman, I would present the experience of my family and of my friend’s family as proof positive that Vitamin A is THE CURE for Retinitis Pigmentosa. But I can’t sell that in good faith. Just because I write a blog on Usher and sound like I know what I’m talking about doesn’t mean my opinion on Vitamin A should matter.

The good news is that there is a way through this uncertainty. It’s called science! Do more studies on vitamin A. Answer all the questions. Take the guess work out of it and find the truth. I’ll even help write the grant proposals (writing is about all I’m qualified to contribute). Families will do the right thing once the proof is there. We just need to know what it is.

We need a consensus on vitamin A. My daughter’s vision, and my sanity, depend on it.

Friday, May 6, 2011

Dispatches from ARVO: Day 5

By Jennifer Phillips, Ph.D.
The last day of the ARVO meeting was short and sweet, and the very last presentation I saw before heading to the Ft. Lauderdale airport was the one I’m choosing to recap here. The talk was by Hari Jayaram of the University College London’s Institute of Ophthalmology , who described a collaborative research project in which cultured human retinal cells were implanted into a rat model of retinal degeneration. At first glance it might sound like the premise of a Mad Scientist thriller, but it was actually quite a well-designed and relevant study. Here’s an overview of the experimental rationale, set-up, and outcome:

Main goals of the research: The primary task was to optimize methods of culturing a type o f human retinal cells that had the ability to develop into photoreceptor cells. The subsequent goal was to actually implant these cells into a rat model of retinal degeneration and observe the cellular activity and the effects, if any, on the visual function of the rat.

Methods of achieving the first experiment: Müller Glial Cells (MGCs) were isolated from donor human retinas and cultured with the addition of various growth factors and other molecules. The researchers were seeking to find the best ‘cocktail’ of ingredients that would stimulate the MGCs to do what they are known to do in intact, living retinas under certain conditions, specifically to de-differentiate and divide to create a population of cells with the potential to become other retinal cells. In this case the goal was to influence these new “cells with potential” to begin developing into rod photoreceptor cells.

The challenge in such an experiment is not only to find the right ingredients to add, but also the right stage of cell specification to achieve. Highly differentiated cells, e.g. fully formed photoreceptors, tend not to be very happy in cell culture, and usually die off pretty quickly. Minimally differentiated cells can live quite happily in cell culture for an amazingly long time, but for the purposes of transplantation, these cells may not have sufficient guidance (by way of internal and external molecular cues) to become the cell type of the researcher’s choosing.

Results of the first experiment: In addition to being able to culture MGCs in a relatively undifferentiated (read: having the potential to become a number of different cell types) state, the investigators also hit upon a successful combination of growth factors and molecules known to stimulate rod photoreceptor development. They confirmed their success by analyzing the molecules being produced within the differentiated cultured cells and found that they were churning out molecular markers of rod precursor cells (RPCs)—specialized enough to know to develop into rods, but not so specialized that they would become unhealthy before transplantation time.

Methods of achieving the second experiment: The investigators selected a rat model of retinal degeneration, known as the P23H rat, bearing a mutation in rhodopsin, the light-sensitive protein found in rod outer segments. In both rats and humans this mutation causes RP due to rod photoreceptors death. Unlike the Mertk mouse model described in yesterday’s post , in which photoreceptors die because of a protein dysfunction in the neighboring RPE cells, the retinal degeneration in the P23H rat is due to a defect in photoreceptors themselves. This is important because the researchers needed to be sure they were targeting the right population of cells, replacing the particular cell types that were defective, namely rod photoreceptors.

P23H rat retinas received transplants of either the MGC cultures or the RPC cultures. Each rat in the experiment received cells in only one eye, with the other eye remaining untreated for use as an experimental control. The researchers waited 3 weeks, during which time the rats received drugs to suppress their immune response so that they wouldn’t reject the donor tissue. After this waiting period, visual function was analyzed by ERG, and retinal tissue was examined to evaluate the growth and development of the implanted cells.

Results of the second experiment: When the researchers observed the retinal tissues of the rats that received the “undifferentiated” MGCs, The researchers observed that these cells had incorporated into the retinal layer usually inhabited by MGCs and had developed an MGC-like morphology. These cells send processes throughout all the other cells layers of the retina, and have a very distinctive shape.

In the retinas of rats receiving “differentiated” RPCs, the investigators reported that these cells incorporated into the photoreceptor cell layer of the retina. No outer segments appeared to form in these cells, suggesting that they did not complete the process of becoming mature rod cells. However, they did form connections with other retinal neurons, which demonstrates the extent to which they were incorporated into the existing retinal structure. The researchers also determined that these cells were producing proteins exclusive to rod cells. So, although the transformation from RPC to fully specialized rod cell was incomplete, the cells did display a “rod-like” molecular profile and cellular behavior.

Visual function was tested by measuring the ERG response to various light intensities. In the eyes treated with RPCs, the investigators detected a greater photoreceptor response to bright light compared to controls. Even without evidence of forming outer segments—the part of the rod cells where rhodopsin functions, the inclusion of RPC-derived cells resulted in a boost in visual function. Interestingly, researchers saw a less dramatic, but still significant increase in photoreceptor response to bright light in the eyes containing undifferentiated MGC transplants. No evidence of additional rod or rod-like cells originating from the transplanted human cells was detected in these eyes by histology, yet the presence of the new Müller-like cells seemed to improve function nonetheless.

Conclusions: These experiments have clear implications for future work toward cell-replacement therapy for photoreceptor degeneration. The experimental design might seem unusual, particularly the idea of putting human cells into a rat retina, but it was important to ascertain how human retinal cells might behave in a complex, living system. Clearly, there were some limitations to this experiment—the cell differentiation was incomplete, in that neither the MGC nor the RPC transplants appeared to mature to 100% fully functional cell types. This may indicate that the cell culture conditions need to be optimized further, but it may also be due to the limitations of cross-species transplants. As you might imagine, human retinal cells are somewhat larger than their rat counterparts, so there may have been some physical limitations to cell growth and differentiation.

Although further optimization and additional testing in animal models will be required before this treatment is ready for clinical trials, this work will likely make a significant contribution to the field of retinal cell replacement.

So, as I wrap up my final report from ARVO, 20,000 feet above Middle America, I can say unreservedly that this is one of the best, most diverse, career enriching meetings around. I’m full of ideas for my own experiments, and I’m already looking forward to the 2012 Annual Meeting.

Jen’s final ARVO stats, by the numbers:
  • 5 science-filled days
  • 38 talks
  • 75 posters
  • 4 walks on the beach
  • 2 swims
  • 7 airplanes
  • 1 missed flight
  • 3 delays
  • 18 hours in flight
  • 100 new ideas

Thursday, May 5, 2011

Dispatches from ARVO: Day 4—Valproic Acid revisited

by Jennifer Phillips, Ph.D.

Last month, I blogged on some peer reviewed research describing a clinical observation of the effects of Valproic Acid in patients with a particular form of retinitis pigmentosa (not Usher). Today at the ARVO meeting I saw a nice poster presentation that may add a bit more data to that story. The authors, from the same research group at U Mass that produced the original patient study published in the British Journal of Ophthalmology, now report the results of dosing mice with a severe form of RP with Valproic Acid.

This particular mouse model has a mutation that affects the Retinal Pigmented Epithelium (RPE), the cell layer that lies adjacent to the photoreceptor outer segments. The RPE is important for keeping the outer segments—and, by extension, the entire photoreceptor—healthy, and defects in the molecular pathway responsible for the RPE-Photoreceptor relationship often lead to retinal degeneration (the RPE65 gene, which causes Leber’s Congenital Amaurosis when mutated, is another example of an RPE gene crucial for retinal cell health).

This mouse strain, with a mutation in a gene called Mertk, has an RPE defect that causes RD very early on in life, with the standard flattening of the ERG and thinning of the neural retina as cells die in large numbers. Unlike the human patients reported in the study by Clemson et al., this form of RP is autosomal recessive in both mice and humans. These researchers gave daily doses of Valproic Acid to Mertk mice over a period of four weeks, beginning when the mice were about 1 month old. ERG amplitudes and retinal thickness were recorded from a group of about 20 control and experimental Mertk mice at the beginning of the period, and again after the four week treatment ceased.

At the end of the trial, the retinal thickness in the Valproic Acid treated mice was significantly greater than that of the control group, indicating that fewer cells had died during the four week treatment period. There also appeared to be some benefit to the ERG response, although not as dramatic as the retinal thickness differences.

So, although there are one or two more controls I would like to see in this particular data set, I can’t find too much fault with it, and it does seem to lend more credence to the clinical trial of Valproic Acid treatmemt for Autosomal Dominant RP that is currently recruiting participants, in that it provides another situation in which Valproic Acid appeared to slow the rate of photoreceptor death in a progressive retinal degenerative disease. I’ll pass on further developments in this story as they are reported.

Tomorrow is the final day of the ARVO meeing—a shorter day to accommodate everyone’s travel plans, but a number of interesting presentations on the program nonetheless. Tune in Friday for one last dispatch from this wonderful meeting.

Tuesday, May 3, 2011

Dispatches from ARVO: Day 3

By Jennifer Phillips, Ph.D.

Today’s cool Usher science story comes from Kate McCaffrey and colleagues at Rosalind Franklin University, who are making some interesting discoveries about a new potential therapy for Usher type 1C.

In the past when we’ve talked about gene replacement therapy, the focus has been to find ways of replacing a gene product that is absent or faulty in some way. The exact nature of the mutation causing the absence or faultiness doesn’t usually matter—as long as you know which gene is affected, substituting a good copy should solve the problem.

Be that as it may, Usher gene replacement has barely gotten going in animal models yet, let alone human patients. And, as we’ve discussed, some Usher molecules will likely be very difficult to work with, due to their large size or other constraints.

Another approach to this that has met with some preliminary success in the hands of McCaffrey et al. is to target a particular mutation for correction, rather to replace the whole gene. The mutation in question is the USH1C G216A mutation, which is prevalent in Acadian populations as well as in people with Ashkenazi Jewish ancestry[this is only common in Acardian populations as far as we can tell]. This is a mutation that alters the normal splicing of the USH1C gene.

So what is ‘normal splicing’ exactly? Briefly, splicing occurs in the transitional molecule between the DNA code housed in the nucleus and the protein product that goes out into the cell and does stuff. The transitional molecule is called messenger RNA, and its job is to faithfully copy the genetic code from a particular address on the chromosome and then serve as the blueprint for assembling the protein. But before it can provide a coherent recipe, the content has to be condensed.

The information on the chromosome that codes for the protein, called exon sequences, is interspersed with other DNA sequence that isn’t part of the instructions, called intron sequences. When the messenger RNA is first synthesized, it contains both exonic and intronic bits, which need to be excised before the message will make any sense as protein building instructions. Think about an article on how to build a gazebo in a Home Improvement magazine. If the instructions were spread out on consecutive pages, it would be straightforward to build. But if after every page or two there were instructions for a different project, or page after page of advertising, you would probably either include extra, incorrect steps in your building project or perhaps abandon it altogether, leaving it unfinished and non-functional.

Fortunately, the cellular machinery that processes messenger RNA is pretty clever about locating the parts that need to be deleted (or, if you will, the pages that need to be torn out) and then splicing together the ends of the exons so that the instructions are arranged in one continuous string of code. How does this machinery know how to tell exons from introns, you might ask? It’s all in the code! DNA base sequences indicate where the junctions are between the parts to keep and the parts to delete.

So, sorry for the protracted explanation (I’m not doing to well at ‘short and pithy’, am I? ) but if you followed it, it now shouldn’t come as much of a surprise that while mutations in the protein coding, exonic sequence of Usher genes can and do lead to Usher syndrome, so can mutations affecting the splice sites of these genes. The G216A mutation is an example of this. It creates an extra splice site in the middle of the code, which leads to some of the exon sequence being lost and the protein being incomplete, which in turn leads to a severe form of Usher syndrome.

As it happens, we molecular biologists have several tools for inhibiting splicing at a particular site on a messenger RNA molecule. We can generate small molecules that physically interfere with the splicing machinery, blocking a particular part of the code that indicates ‘splice here’ such that it will skipped over while the splicing machinery moves on to the next splice site. Sometimes we do this when we want to interfere with normal splicing and create a poorly constructed protein, but in this case, McCaffrey and colleagues used it to block that mutated extra splice site, G216A. They tested their interfering molecule in cell cultures containing the human USH1C G216A mutation and in a mouse model containing the same mutation. When they did this, they found that splicing at the wrong site caused by the mutation was effectively blocked, and both the human cells in the dish and the tissues they examined from the mouse were able to revert back to making the normal USH1C protein again.

The research presented was mostly a test of whether or not this technique would work at the molecular level, and it did, rather impressively. More work is needed to decide how and when to deliver a dose of this molecule to the affected tissues in the mouse model, and then to examine the effects on the observed hearing and vision defects, but it’s a great start.

Dispatches from ARVO: Day 2

by Jennifer Phillips, Ph.D.

Today was a 12-hour juggernaut of talks, poster presentations (mine included) and really good scientific and social conversations. While many of these situations would make great blog fodder, one series of talks really had the wow factor. This session was entitled “Optogenetics, Visual Function and Restoration”. I’ll skip over the highly technical side of this field and cut to the chase as quickly as possible here:

I know I’ve mentioned opsins before on the blog, but to recap briefly, opsins are light-sensitive molecules in photoreceptors that change shape in response to light of a particular wavelength, which causes the cell membrane to change polarity, which in turn causes in an electrochemical signal to be transmitted to nearby ‘receiver’ neurons. As it happens, light sensitive opsin-type molecules are present in a lot of different multicellular organisms, some of which completely lack eyes, and even in various species of bacteria. The principal function of the molecule is the same, all throughout the evolutionary tree: Light activation creates an electrochemical response that allows a polarity change in the cell membrane—negative to positive, or the other way around. Changing polarity is useful component of cell biology, and cells of all sorts use polarized membranes, and channels or gates within those membranes, to regulate the flow of charged particles in and out.

So here’s wow factor part A: scientists have figured out how to isolate these molecules from bacteria. When introduced into neurons, these molecules can be stimulated to either activate or silence a neuronal signal. There were numerous patents mentioned in these talks, as these molecules are impressively multipurpose for manipulating electrochemical cellular responses of all imaginable types.

Now here’s wow factor part B: The last talk discussed the application of this technology as a possible treatment for RP patients. In most forms of RP, including that seen in Usher syndrome, rod photoreceptors die off, causing night blindness and tunnel vision. Changes occur to the remaining cone cells throughout the eye, most particularly in the outer segment of the photoreceptor where these light-sensitive opsins are housed. For some time, it’s been assumed that at advanced stages of RP, severe loss of vision correlates with severe loss of photoreceptor cells of all types. Recently, though, clinicians and researchers have discovered that a respectable number of cone cells in these eyes remain viable, although not functional. So, to recap: cone photoreceptors in an RP patients eye still exist—they are not dead—but they do not function because the portion of the cell containing the opsins that kick off the whole signaling cascade throughout the retina has degenerated.

This technology could potentially enable us to replace the degenerated native opsins with a light sensitive ‘alternative’ opsin that would serve basically the same purpose: open a channel in the cell membrane, allow charged particles to flow through and signal to the neighboring neurons. One research group has already done this replacement experiment in a mouse model of retinal degeneration, and by observing the behavior of mice navigating a maze, they were able to record an increase in visual function following the treatment. Wow!

Cooler still, as some of these bacterial opsins have been found to respond to specific wavelengths, we may be able to target them to particular subtypes of cones to give RP patients some degree of color vision!

Based on what I heard today, it seems like more preclinical testing will be required before this technology approaches readiness for Phase I Clinical trials, but the power of the system is undeniable. I predict we will be hearing a lot more about this in the coming years.

In the meantime, please take a moment to appreciate the contribution of basic research to this potentially groundbreaking advance in human health. Bacteriologists are the ones who first discovered these species, living happily in extreme environments of high salt or high methane levels that would be instantly lethal for most other critters on the planet. These basic scientists were not out to find a cure for blindness, but only to better understand the diversity of life. In doing so, they identified these molecules and introduced them to the larger scientific community which is now only just beginning to understand how to harness their potential. As I’ve said before, you never know where the next great discovery might come from.

Monday, May 2, 2011

Dispatches from the ARVO Annual Meeting: Day 1

By Jennifer Phillips, Ph.D.

Hello readers! I’m here in sunny Ft. Lauderdale attending the Association for Vision Research and Ophthalmology meeting—that’s right the teeming nerd hordes are at it again. But rather than waiting until the end for a long, post-hoc recap as we’ve done with past scientific meetings, this time I thought I’d try to mix it up with some short, pithy daily notes from the event. I realize I’m not exactly known for ‘short and pithy’ around here, but I’ll do my best.

Topic of the day: Ethics of Genetic Applications in Ophthalmology

This was a talk session with presentations by four speakers, a mix of genetic counselors, clinicians, medical ethics specialists, and researchers. One common theme was the psychological and emotional impact of a diagnosis via genetic testing. There was a bit of friendly debate about the benefits vs. the harm of testing and informing children of later-onset genetic diseases and some general recommendations of having qualified genetic counselors—or clinicians with sufficient training in such counseling, be involved in delivering this news to families and following up with them to make sure they had adequate support. We’ve talked quite a bit on this blog about ways to make this clinician-patient-family interface more successful, so this isn’t anything new, but hopefully it will reassure some of you to hear that the problem is being recognized and discussed in this kind of professional setting.

The most interesting point of discussion, though, by which I mean, the one that made me perk up and say “hey, I should totally blog about that!” was put forth in a very nice talk by Bernard Lo, a professor of medicine and director of the medical ethics program at UCSF. He talked generally about the ethics of clinical trials, and specifically used the RPE65 gene therapy story as an example. I’ve described a few clinical trials here, so the basics—random assignment of participants into a control group or an experimental group, reasonable efforts to prevent the patients from knowing which group they’re in until the end of the study, etc.--should be familiar to most readers. However, the ethical conundrum this speaker put forth that attracted my attention was something neither Mark nor I have discussed in our various posts about clinical trials: any participant in a clinical trial, particularly a Phase I trial, is taking a personal risk so that researchers might benefit from new knowledge and so that future patients might benefit from the studied treatment. There is absolutely no guarantee of any personal gain—it’s altruism, pure and simple. Moreover, conducting such studies runs quite counter to the primary goals of most medical professionals, in that treatment is not being personalized and is not designed to improve the condition of the participants.

The RPE65 story was used as an interesting illustration of this ethical morass. As you may recall, the ‘control group’ did not consist of a separate group of people, but instead patients received treatment in one eye, and had ‘sham surgery’ on the other. Is sham surgery, in this case sticking a needle deep into someone’s eye without delivering any potential treatment, ethically justifiable? Thinking about it from a pure research point of view, it is an important control. The placebo effect is a powerful factor in any clinical trial and must be accounted for in order for any results to be meaningful. Even seemingly objective things like eye chart tests can be influenced by patient effort or the belief that he or she should be seeing better after treatment. And interestingly, the more invasive the intervention, the more powerful the placebo effect. Injecting a completely inactive substance elicits a stronger placebo response from patients than orally consuming the same substance. And the placebo response to surgery has been shown to be more powerful still. Thus, it’s pretty important to control for this with any new treatment being tested. That said, sham surgery is a heckuva lot riskier than taking a sugar pill. Is it ethically permissible? And, if so, under what circumstances?
This is an intriguing question, and one that should prompt a bit of reflection about the risks and benefits of participating in a clinical trial. The potential benefits are great, but not necessarily for the participants. The risks are not inconsiderable, and are very personal. Yet people routinely consent to participate in these trials. Just as the clinicians conducting the trials must subvert some basic tenets of their training regarding individualized care for any given condition, the participants are also willing to forego personal gain so that others might benefit from whatever the study shows. That fills me with awe.

So, for those of you who have already participated in trials, and those who are planning or considering participation in upcoming trials, THANK YOU. You are all truly my heroes.