Stopping FGFR3 production to treat achondroplasia (part 1)

Introduction

This blog currently has 24 followers (that's what Google tells me...). These resilient readers likely think I repeat the same information too much, as the articles often come with some basic concepts about achondroplasia before getting into the subject. That's true, but I know that sometimes other visitors pass by so it's good to try to have everybody in the same pace. And, after all, I always try to include a bit of new information in these short introductions (and this is true in this one, too!).

Reading again the more recent articles, I perceived that the language and explanations sometimes might sound hard to understand for the first-time reader. If you think it has been difficult to get into some of the topics we review here, take a look in the first articles of the blog, written in 2012. They may help you to travel more easily across the new ones. For instance, the topic in this article have been reviewed in three articles back in 2012:

22/01/2012. Targeting FGFR3 production to rescue bone growth in achondroplasia
04/02/2012. The RNA World: knocking down FGFR3 in achondroplasia

20/02/2012. The RNA World: knocking down FGFR3 in achondroplasia, part 2 

And, in 2013, I wrote another article about this subject:

16/06/2013. Interfering in the production of FGFR3: a potential strategy to treat achondroplasia

It may be helpful for you to read them before continuing here, because the topic has its complexities. 

Some basic information 

Achondroplasia is caused by a mutation in the gene FGFR3 that encodes (gives instruction for the production of) a protein called fibroblast growth factor receptor 3 (FGFR3) (1,2). FGFR3 has a fundamental role in bone development, which is to reduce bone growth velocity (3). It works like a brake within the thin regions located in both extremities of the long bones called growth plates (Figure 1). Growth plates are the structures responsible for long bone elongation. The cells inside the growth plate, the chondrocytes, drive bone growth through a very complex process, and FGFR3 is part of it.

Figure 1. The growth plate.

In Figures 1 and 2 you can see that the growth plate chondrocytes are organized in a way that, under continuous stimuli from many active growth agents (FGFR3 included), begin from a dormant status (resting zone), run into an organized proliferation frenzy (proliferative zone), undergo massive enlargement (hypertrophic zone) and finally give place to the cells responsible for building the new bone, the osteoblasts. As a piece of detail, some authors in recent reviews have been considering that the size of the hypertrophic zone is determinant for the elongation of the bone. (4)

Figure 2. The chondrocyte lifecycle and many of the agents participating in bone growth development.

As just mentioned above, FGFR3 acts to balance the effect of several other agents that stimulate chondrocytes to proliferate. If there was no FGFR3 at all, bones would grow excessively causing health problems, as we see in reported cases in the literature (5).

However, in achondroplasia, the mutation in the gene FGFR3 makes FGFR3 to be more active than normal, which in turn leads to bone growth arrest. Under effect of the super active FGFR3, chondrocytes reduce their proliferation rate and less of them undergo hypertrophy (Figure 3). With less chondrocytes maturing and growing, the hypertrophic zone does not reach its full potential and we have bone growth impairment. Basically, almost all the clinical features and health consequences of achondroplasia can be explained by this mechanism (4).

Figure 3. Growth plates from a mouse model of achondroplasia (right) and wild type (left).

Compare the lengths of the proliferating zones (PZ) and hypertrophic zones (HZ) between the normal and the affected growth plates. Adapted and reproduced here from Yingcai W et al. PNAS 1999;96:4455-60, for educational purposes only.

FGFR3 activity depends on signals that come from outside the chondrocytes. In figure 4 you can see that FGFR3 is like a TV antenna, placed in the roof of the house (the cell membrane). It receives chemical signals brought by fibroblast growth factors (FGFs) circulating in the vicinity of the chondrocytes inside the growth plate. When a FGF binds the antenna, a chemical signal runs across the antenna pole (the body of FGFR3) and triggers several other chemical cascades inside the chondrocytes (6). Let’s say it is the active FGFR3 “pushing buttons on the cell control panel”.

Figure 4. FGFR3 chemical cascades (from Nature Reviews Cancer 2005)

Scientists now have a good understanding of how FGFR3 exerts its action in the chondrocytes, the “buttons” it pushes to make chondrocytes stop growing, although not infrequently they bring more details to this process.

Learning about the mechanism of action of proteins such as FGFR3 is like creating a route map highlighting their crossroads, checkpoints, train stations, etc. This mapping allows researchers to work on solutions to overcome lack of or excessive activity of these proteins. Many current therapies for a large number of diseases have been created with the aid of these maps since they help to find the right targets to beat a given medical condition.

Thinking in therapies for achondroplasia, as for other many genetic disorders caused by a single overactive protein, the natural move is to find an agent that could stop that protein to exert its function. Theoretically, this could help reverting or minimizing its effects.

Stopping FGFR3 actions can be achieved through several different approaches (look the "train stations" in the signaling cascade map in Figure 3), from blocking the signal reception by the antenna or blocking the buttons in the control panel to counterbalancing the effect of one of those chemical cascades by stimulating an antagonistic one.

Most of these strategies have been already reviewed here in the blog. For instance, you can block the antenna with antibodies designed to bind it. There are at least three antibodies against FGFR3 published in the literature that have been explored in clinical studies for cancers driven by FGFR3 (see this article).

You can block the buttons on the control panel with small molecules called tyrosine kinase inhibitors (TKIs). There are many TKIs available, but so far none good enough to be used in achondroplasia (see this article). And you can also stimulate another cell antenna (or receptor) to reduce the activity of FGFR3, as we have been witnessing with vosoritide, a C-type natriuretic peptide (CNP) analogue currently in phase 2 clinical trial for achondroplasia.Vosoritide imitates CNP by binding its receptor located at the chondrocyte cell membrane (like FGFR3 is). The receptor is activated and its signaling intercepts MAPK, the FGFR3 main signaling cascade, inhibiting it at the level of the "train station" RAF (Figure 5) (see this article).

Figure 5. CNP signaling cascade intercepts the MAPK cascade triggered by FGFR3.

Matsushita M et al. (2013). PLoS ONE 8(12): e81569. doi:10.1371/journal.pone.0081569. Reproduced here for illustration purposes only

These are only some examples among other several strategies already being explored for achondroplasia (see this article). Nevertheless, there is another kind of approach that would be suitable to control or regulate FGFR3 and, after this long introduction, let’s talk about one of them, which would be targeting the production of FGFR3 (it’s in the title, isn’t it?). In fact, we have already reviewed it here in the blog (listed in the top of the article!, did you read them?), but as new information is coming, I thought it would worth to talk about it again. And finally, it may help get into another new potential strategy that also deals with DNA, RNA and the transcription machinery later.

Regulating the regulator 

As you might know, our DNA is a kind of vault where the chemical information needed to create the proteins fundamental for our body is stored. Life is a dynamic process and there is always the need to produce new proteins so the cells, tissues and the body as a whole can function normally (Animation 1, English, 2:41min). 

Animation 1. From DNA to protein (by yourgenome.org).

The way the process by which the DNA is “read” and “copied” to generate proteins often could lead to states where a given gene that encodes the information to generate a protein could be extensively “open” for reading, so the protein could be produced nonstop. This in turn could result in health problems. For example, some kinds of cancer cells, like in breast cancer, are able to start the super production of “antennas” like our FGFR3 to gain advantage of circulating signals that trigger more cancer growth.

In normal conditions, this over production of proteins doesn’t occur because the cell has several quality control measures that work to regulate this process. One of them is comprised by RNA molecules.

But, let’s go slowly here, step by step. I remember in my high school days, when I started to study cell biology, that there were three classical kinds of RNA: the messenger RNA (mRNA), the transfer RNA (tRNA) and the ribosomal RNA (al of them appear in the animation above). Well, since then the list of RNA types has not stopped to grow. Take a look in this Wikipedia page for a list of already identified RNA types. 

Some of these many RNA molecules are involved in the process of controlling the protein production and the focus here is the microRNA, or simply miRNA family (reviewed in those articles mentioned above). These RNA molecules have been shown to control the amount of proteins produced by the cell. Let me invite you to watch the Animation 2 (English, 4:53min) by Katarina Petsche that gives an overall view of miRNA synthesis and how it works:

Animation 2. miRNA synthesis and function (by Katarina Petsche).

Basically, because of their ability of leading mRNAs to degradation, miRNAs have been classified as "gene silencing" molecules. As the target protein is not produced after all, it is as though the gene was not working (or shut down, silenced). 

There are thousands of different miRNAs and each of them is designed to bind to a limited number of mRNAs. There is already a number of miRNAs identified, like miRNA 100 (or miR-100), that bind to the FGFR3 mRNA, but here we have a problem. Although miRNAs are very specific for their targets, there targets are not unique. When you think in "silencing" a specific gene, let's say FGFR3 in achondroplasia, you don't want to silence other target genes, since this could lead to other health problems.

Among the several miRNAs that have been showed to exert a control function over FGFR3, miR-100 seems to be one of the most relevant, since there is mounting evidence that it does control/silence FGFR3. Several recently published studies explored the connection between miR-100 and FGFR3 in pancreatic cancer (7), glioblastoma (8), sarcoma (9) and  in lung cancer cells (10) showed that in those cells overexpressing (over producing) miR-100 there was less FGFR3 available and vice versa and all concluded that miR-100 was able to inhibit cancer growth in their experiments. Don't feel confused about the fact that FGFR3 is used by cancer cells to grow. This happens because FGFR3 is a brake only for growth plate chondrocytes; in other cells it works like an accelerator... These kind of findings may lead to the development of a strategy where miR-100 expression (production) could be stimulated in those cancer cells which use FGFR3 to grow, to silence the FGFR3 gene and beat the tumor.

So, if such an approach could be used for the treatment of cancer, are miR-100 or other miRNAs suitable to treat achondroplasia?

Probably not, because although the evidence exists that they do inhibit FGFR3, there are also other proteins that may be affected by their actions, which could bring undesired problems. In summary, miRNAs are not specific enough.

Well, then, what is the purpose of this review? The thing is that researchers have been able to create, in the lab, another kind of RNA which does basically the same of what miRNAs do (learning a bit about one, is learning about the other, you see...). However, there is a strong difference: it is made by design, and probably will have much more specificity for FGFR3. This synthetic RNA molecule is called small interference RNA or simply siRNA. Studies already made show that siRNAs use the same cell machinery and cause effects like miRNAs (11). Researchers have to introduce the siRNA inside the cell and after that it works like a miRNA. Today, a "simple" approach to prove that a gene is linked to a determined function is to apply a specific siRNA and watch the effects of silencing that gene. For instance, this has been done in several other disease contexts (11), including achondroplasia, in a study by Pintor and Legeai-Mallet (12).

If you read the 2013 article about RNAs you already know that at least one biotech based in US (Marina Biotech) has patented an RNA interference strategy pledging their utility to treat achondroplasia (13). Now, I have found another patent for the use of nucleic acid molecules to treat achondroplasia (14) from Santaris Pharma (Roche), so it looks like there have been some attempts to target FGFR3 in achondroplasia using the siRNA strategy lately. 

We don't know if these approaches will reach clinical development because one of the key challenges for nucleic acid-based therapeutic strategies (RNA is composed by nucleic acids, as you know) is how to make those agents reach their targets. Because RNA molecules are so powerful, the body has a vast number of protective agents to ensure these molecules will not traffic freely for long. RNA is the genetic code of many causing disease viruses, so the body must be ready to get rid of foreigner nucleic acids to defend itself from infections. Therefore, RNA molecules tend to be degraded rapidly by neutralizing enzymes, so this could partially explain why we don't see fast progress in this area. For achondroplasia, the challenge is even greater because the target cells, the chondrocytes, leave in a very well protected environment, the growth plate cartilage, a place tough to reach (see this previous article of the blog).

Delivering the package

Delivery is really challenging, and for this reason I used the verb "to introduce" when I described the use of siRNA just three paragraphs above. It is not easy to make RNA molecules go inside cells, you have to use a taxi (what is called transport, carrier, vector, etc.) or a disguise. For example, several studies applied appendages to the RNA molecule so it became more stable and resistant to neutralizing enzymes. For diseases where the target cell lies in a tissue with direct blood flow, this could be enough to allow the siRNA to enter the cell, but for a tissue such as the growth plate, where there is no direct blood flow, how can this be achieved?

This is not a trivial question. One possible strategy to deliver a drug to the target is to find something that only that target has (a marker), which is not an easy task, but still feasible. In the case of the growth plate chondrocyte, possibly one of these markers is a cell receptor (another antenna) called CD44. The issue here is that CD44 is expressed by many different cells. So, is it useless? Maybe not, because most of the cells that have CD44 do not express FGFR3 in a significant level, so some influence of a siRNA against FGFR3 on them possibly would not cause any trouble. The idea would be to attach or cover the RNA molecule to another molecule that can bind to CD44. Once connected, the cell system that manage the good shape of the antennas would bring the busy CD44 inside the cell where the siRNA could be released to find its target and exert its expected function. This is just a personal idea, I never found any work exploring this approach in this specific area.

Conclusion

It could sound that this review didn't help much in terms of bringing another new spectacular potential therapy for achondroplasia. It would be more like a general review of the disorder, since the blog already has reviews about miRNAs and siRNAs. However, we are not wasting time: the processes briefly mentioned here will serve to help us to travel across an even more complex one, which is gene editing, a potential therapeutic strategy that is under the spotlight now.

Could it be used to treat achondroplasia?

From what I have already learned, I would say yes, but let's see in the next article. I am still studying this topic...


References

1.Bellus GA et al. Achondroplasia is defined by recurrent G380R mutations of FGFR3. Am J Hum Genet 1995;56(2):368-73. Free access.

2.Rousseau F et al. Mutations in the gene encoding fibroblast growth factor receptor-3 in
achondroplasia. Nature 1994;371(6494):252-4. Free access.

3. Laederich MB and Horton WA. FGFR3 targeting strategies for achondroplasia.Expert Rev Mol Med 2012;14:e11.

4. Narayana J and Horton WA. FGFR3 biology and skeletal disease. Connect Tissue Res 2015;56(6):427-33.

5. Makrythanasis P et al. A novel homozygous mutation in FGFR3 causes tall stature, severe lateral tibial deviation, scoliosis, hearing impairment, camptodactyly, and arachnodactyly. Hum Mutat 2014;35(8):959-63.

6. Ornitz DM and Marie PJ. Fibroblast growth factor signaling in skeletal development and disease. Genes Dev. 2015 Jul 15;29(14):1463-86. Free access.7. Li Z et al. MicroRNA-100 regulates pancreatic cancer cells growth and sensitivity to chemotherapy through targeting FGFR3. Tumour Biol 2014;35(12):11751-9.

8. Luan Y et al. Overexpression of miR-100 inhibits cell proliferation, migration, and chemosensitivity in human glioblastoma through FGFR3. Onco Targets Ther 2015;8: 3391-400. Free access.

9. Bi Y et al. Overexpression of miR-100 inhibits growth of osteosarcoma through FGFR3.Tumour Biol 2015;36(11):8405-11.

10. Luo J et al. Overexpression of miR-100 inhibits cancer growth, migration, and chemosensitivity in human NSCLC cells through fibroblast growth factor receptor 3.Tumour Biol 2015; published online Aug 28. DOI 10.1007/s13277-015-3850-z. 

11.Videira M et al. Preclinical development of siRNA therapeutics: towards the match between fundamental science and engineered systems.Nanomedicine 2014;10(4):689-702.
 
12. Guzman-Aránguez A, Legeai‐Mallet L, Pintor J. Fibroblast growth factor receptor 3 inhibition by small interfering RNAs in achondroplasia. Anales Real Acad Nac Farm 2011; 77(1).

13. Marina Biotech Patent WO 2011139842 A2. Nucleic acid compounds for inhibiting fgfr3 gene expression and uses thereof. 

14. Santaris Pharma WO 2014080004 A1. Compositions and methods for modulation of fgfr3 expression.