June 2016 Volume 3, Issue 4

Loss of Mannose Phosphate Isomerase Induces p53 Mediated Apoptosis in a Zebrafish Model of Congenital Disorders of Glycosylation

Louis Lin1* and Jaime Chu2
Student1: Sanford H. Calhoun High School, Merrick, New York
Mentor2: Mount Sinai Hospital and School of Medicine, 1 Gustave L. Levy Place, New York, NY 10029
*Corresponding author: loulin@bmchsd.org

Abstract

Congenital disorders of glycosylation (CDG) are under-diagnosed diseases that result from autosomal recessive mutations in genes involved in protein N- glycosylation; a process vital for the function of nearly all secreted proteins. MPI-CDG, a subtype of CDG, occurs from a mutation in the Mannose Phosphate Isomerase (MPI) gene, which depletes the MPI enzyme activity needed for glycosylation. Patients present with multi-systemic symptoms and have high morbidity and mortality, but little is known about the mechanism of disease. To model MPI-CDG in zebrafish embryos, morpholino oligonucleotides were used to knockdown Mpi activity. Acridine orange experiments done at 24 hours post-fertilization demonstrate that MPI loss causes cell death. This cell death can be rescued with the loss of p53, indicating that MPI knockdown leads to p53 mediated apoptosis. Additionally, mannose can drastically reduce MPI morpholino induced cell death in both WT and p53 +/- embryos. This project offers insight into the pathogenesis of MPI-CDG and the role of p53 and mannose metabolism.

Introduction

Congenital Disorders of Glycosylation
N-Glycosylation is a cell-essential process by which carbohydrates are attached to proteins and provide essential signatures for protein function, stability, and localization (1). Nearly every protein that travels through the rough endoplasmic reticulum (ER) and the Golgi apparatus is N-glycosylated. Congenital disorders of glycosylation (CDG) result from autosomal recessive mutations to genes required for N-linked glycosylation of proteins and cause abnormal glycosylation (2). These CDG are under-diagnosed diseases that present with multi- organ manifestations. People with a CDG suffer from neurological, musculoskeletal, gastrointestinal, hepatic, and cardiac pathologies (2, 3). These rare disorders consist of 38 distinct subtypes. Of these 38 subtypes, the only CDG that has a treatment is MPI-CDG: oral mannose supplementation(4).

MPI and MPI-CDG
Mannose phosphate isomerase (MPI) is an essential enzyme that facilitates the isomerization between fructose-6-phosphate to mannose-6-phosphate (2). Fructose-6-phosphate plays an important role in the process of glycolysis. Mannose-6-phosphate plays an important role in glycosylation. Therefore, MPI is a crucial enzyme for two fundamental, cell  essential pathways - glycosylation and glycoysis. MPI-CDG is caused by a mutation to the MPI gene. Human patients only have 3%-18% residual MPI enzyme activity. MPI-CDG (CDG-Ib) is a subtype of Type I CDG and results in deficient MPI production and deficient glycosylation. The glycoproteins are synthesized with unoccupied glycosylation sequons (2). Unlike all other Type I CDG, MPI-CDG is unique in that patients do not present with neurological or other developmental abnormalities. Rather, patients present primarily with gastrointestinal pathologies including hypoglycemia, liver fibrosis, vomiting, diarrhea, failure to thrive, and protein-losing enteropathy (5). These symptoms are similar to other more common gastrointestinal disorders, leading to the under diagnosis of MPI-CDG.

Mannose Treatment
Unlike all other CDG, MPI-CDG has a treatment. Oral mannose can be taken to help alleviate the effects of the disorder (3). Mannose is picked up by a hexokinase transporter and this complementary metabolic pathway is able to bypass the main glycosylation pathway in which MPI is involved. The increase in mannose helps increase mannose-6-phosphate and its conversion to mannose-1-phosphate, improving the faulty protein production (Figure 1). Patients that have taken oral mannose treatments show improved symptoms, except liver disease, which continues to progress (4).


Figure 1. The diagram shows the glycosylation pathway in which MPI is involved. MPI catalyzes the isomerization of Mannose-6-Phosophate into Fructose-6-Phosphate. Mannose is also shown working through hexokinase to help bypass the MPI loss. (image taken from (2))

Zebrafish and MPI-CDG
A new model organism was needed to study MPI-CDG since the Mpi knockout mouse model was embryonic lethal (6) and introduction of mannose only increased the lethality (7). A zebrafish model was then created (2). Zebrafish Mpi and human MPI are 63% identical (2). The model was able to recapitulate similar biochemical and morphological characteristics of human MPI-CDG patients. Furthermore, when the Mpi deficient fish were treated with mannose there was a rescue in the phenotype. This showed that zebrafish were a viable model to study MPI-CDG.

Morpholino Knockdown in Zebrafish
Morpholino Oligonucleotides (MO) are a common technique used for zebrafish models as an anti-sense knockdown of a gene by inhibiting the translation of the RNA transcripts (8). To knockdown mpi, a morpholino was used and injected into the yolk of the embryos at the 1 cell stage. The morpholino mediated knockdown of mpi results in zebrafish with multi-systemic abnormalities. Morphant phenotypes include smaller livers, under utilization of yolk, abnormally shaped jaws, and curved tails. The fish also have a residual Mpi activity of less than 20% as compared to controls, similar to MPI-CDG patients (2).

Cell Death and p53
Cell death is the event when a cell ceases carrying out its functions. It contains a large group of different types of death, from necrosis by injury, to programmed cell death. Programmed cell death (PCD) is mediated by intracellular activities and can help kill cells which are harmful or have deleterious mutations (9). Apoptosis is a regulated form of PCD and controls the death of about 60 billion cells in the body each day. Apoptosis is mediated by many pathways including mitochondrial regulation, caspases, as well as different proteins (10). p53 is a central regulator of apoptosis, controlled by the gene TP53 on chromosome 17. It is known as the “Guardian of the Genome” due to its tumor suppressant abilities. It activates DNA repair when DNA is damaged and induces apoptosis when the genome cannot be repaired. p53 is essential for keeping cells healthy and for preventing cancer. In recent years, the mechanisms and functions of p53 in tumor suppression and its role in sugar metabolism have come to light (11).

Questions and Aims
Currently not much is known about the cellular mechanisms of congenital disorders of glycosylation. Broadly, this research hopes to investigate CDG to gain a greater understanding of the pathophysiology of these disorders as therapies are lacking. The project focuses on MPI-CDG and, using the zebrafish model, the goal was to study how Mpi depletion affects cellular development. In a line of wild type (WT) fish, mpi morpholino was used to knockdown the mpi gene to recapitulate manifestations seen in human MPI-CDG patients. The project studied how the loss of MPI would affect cell death. Additionally, mannose treatment was used to see if it could rescue any cell death.

The second objective was to determine what caused the cell death in the MPI depleted fish. There is a variety of causes of death that range from injury to apoptosis. Given the well documented ability of p53 to induce apoptosis, it was the focus of the project. Using p53-/- and p53+/- fish and comparing their results to WT fish gives insight into how p53 affects cell death due to MPI loss.

Materials and Methods

Zebrafish Maintenance and Husbandry
Adult zebrafish were maintained in a standard fish room facility and taken care by the fish room and lab members. The fish were kept at 28°C on a cycle of 14 hours of light followed by 10 hours of dark. Wild-type (WT; AB and Tab 14) and p53-/- fish were used for mating. Zebrafish were fed three times a day: morning brine shrimp, mid-day dry food (Zeigler Adult Zebrafish Diet), and a final feeding of brine shrimp. All protocols were approved by the institute where the research was conducted.

Mating and Culturing
Fish were put into plastic breeding chambers and males and females were separated by a divider. Around 5-11 fish were in each chamber, with roughly a 2:1 female to male ratio. Dividers were removed the next morning to allow for spawning and fertilization. Fertilized embryos were collected following natural spawning and were cultured at 28°C in fish water (0.6g/l Crystal Sea Marinemix; Marine Enterprises International, Baltimore, MD) containing methylene blue (0.002 g/l).

Morpholino Injections
The yolks of embryos were injected at the 1 cell stage with either mpi morpholino (MO) or standard MO (Std; control). The mpi MO (5′-ATGGCGGAAGTGAAAGTGTTTCCTC-3′) targeted the ATG of the mpi transcript, and the Std MO (5′-CCTCTTACCTCAGTTACAATTT ATA-3′) did not target any known zebrafish transcript. The morpholinos were obtained from Gene Tools (Philomath, OR). Needles used for the injections were calibrated to inject 4 nl per embryo using a Narishige IM-300 microinjector; 4 ng of morpholino per embryo was identified as optimal. The embryos were then put into an incubator at 28°C for proper development and taken out for acridine orange staining at 24 hours post fertilization.

Mannose Rescue Treatment
After the mpi MO was injected, half the fish were put in normal egg water and the other half was treated with 50 mM D-mannose (Sigma) in fish water. The mannose treatment is the same form of treatment that human MPI-CDG patients receive.

Acridine Orange
At 24 hours post fertilization (hpf) the zebrafish were dechorionated (releasing the fish from the chorion.) These fish were then put into petri dishes of an acridine orange solution (80 µl of acridine orange diluted in 40 ml of fish water) and stained for 45 minutes. Each petri dish was covered with aluminum foil and put into a dark room since acridine orange is light sensitive. The fish were then washed in five dishes of fresh egg water to reduce excess background staining. Fish were imaged using a light microscope with a Nikon camera. A blue excitation filter was used and the stained dead cells appeared as bright dots.

Quantification and Analysis
Each zebrafish was imaged individually using a light microscope with a camera and pictures were taken. A standardized area of the tail was used for the quantification. A frame of 75 µm by 75 µm was set as a restriction as the area. The frame was then superimposed on the image and positioned over the anal pore of the embryo. Using Nikon's NIS-Elements BR software, cell death was quantified. Dead cells within the standardized area were counted in the total number of dead cells. All data was put in Prism and Excel for statistical analysis and graphs. Unpaired t-tests were used, with a statistically significant result being a p-value of p<0.05.

Results

Wild Type Fish
To study the role of Mpi loss in cell death, the first strain of fish used was normal wild type fish (WT: lines AB and Tab14). WT fish were mated and each clutch of embryos were separated into 3 groups: Std MO injection, mpi MO injection, and mpi MO + 50mM mannose. After acridine orange assays (Figure 2), the fish were imaged and cells were counted (Figure 3). The Std MO control showed little to no cell death with a mean number of dead cells of 1.59 per an area (75 µm by 75 µm). The fish injected with mpi MO to knockdown mpi showed a significant increase of cell death compared to the Std MO, p=<0.0001. The mean number of dead cells was 12.27 per an area.

Mannose was able to rescue the curved tail phenotype in fish with Mpi loss. Furthermore, most, but not all, of the cell death was significantly rescued by mannose. The mean of 5.91 cells per an area, was significantly different from that of mpi MO fish with p=.0002.

 

  • mpi

Figure 2. Photomicrographs of 24 hpf embryo tails after acridine orange staining of WT fish. These images were taken with a bright field and then with a blue excitation filter. Photos are representative of the groups of fish under each type of treatment. Bright dots are dead cells which have been picked up by the staining. The white line is the scale bar for all pictures and is 1 micron long.

  • mpi
    Figure 2. Photomicrographs of 24 hpf embryo tails after acridine orange staining of WT fish. These images were taken with a bright field and then with a blue excitation filter. Photos are representative of the groups of fish under each type of treatment. Bright dots are dead cells which have been picked up by the staining. The white line is the scale bar for all pictures and is 1 micron long.

 

Figure 3. The graph shows the means and standard deviations for the quantified number of cells in the standardized area of all the fish. Std has 38 fish from 4 clutches, mpi MO has 34 fish from 3 clutches, and mpi MO+Mannose has 47 fish from 5 clutches.


p53-/- Mutant Fish
To determine if the cells went through p53-dependent apoptosis, the next strain of fish used were p53-/- MT fish that were missing both alleles of p53. If there was a rescue in cell death it would not only mean that the cells had gone though apoptosis but also mean that p53 was the mediator. This would indicate that there is an inverse relationship between MPI and p53: as MPI activity decreases, p53 activity increases.

After the acridine orange assays (Figure 4) the fish were imaged and cells were counted (Fig. 5A). Once again the Std MO showed little to no cell death with a mean of 0.78 per an area. The mpi MO fish had a mean of 2.0 dead cells per an area. Though this was statistically significant compared to the Std MO, p=.0183, there was still very little cell death. The mpi MO fish treated with mannose were similar to the mpi MO with a mean of 1.89 per an area. The difference between mpi MO and mpi MO+Mannose was not significant, p=.8342. To statistically prove that p53 is the mediator of cell death and helps rescue cell death, the p53-/- fish were compared to the WT fish (Fig. 5B). When the means and values of the p53-/- fish were compared to the WT fish, the p-values showed that p53 was the mediator of the cell death in mpi MO. The t-test between wild type mpi MO and p53-/- mpi MO showed p=<.0001. The t-test between the mannose treated fish of WT and p53-/- was also significant with p=<.0001. These two experiments show that MPI loss causes apoptosis and that p53 is the mediator of the apoptosis.

 

  • Figure-4

Figure 4. Photomicrographs of 24 hpf embryo tails after acridine orange staining of p53-/- fish. The images were taken with a bright field and then with a blue excitation filter. Photos are representative of the groups of fish under each type of treatment. Bright dots are dead cells which have been picked up by the staining. The white line is the scale bar for all pictures and is 1 micron long.

  • Figure-4
    Figure 4. Photomicrographs of 24 hpf embryo tails after acridine orange staining of p53-/- fish. The images were taken with a bright field and then with a blue excitation filter. Photos are representative of the groups of fish under each type of treatment. Bright dots are dead cells which have been picked up by the staining. The white line is the white line is the scale bar for all pictures and is 1 micron long.

 

  • Figure-5-A
  • Figure-5-B

Figure 5. (A) The graph shows the means and standard deviations for the quantified number of cell. Std has 23 fish from 5 clutches, mpi MO has 27 fish from 4 clutches, and mpi MO+Mannose (Man.) has 37 fish from 4 clutches. (B) The graph compares the WT and p53 -/- fish.

  • Figure-5-A
    Figure 5. (A) The graph shows the means and standard deviations for the quantified number of cell. Std has 23 fish from 5 clutches, mpi MO has 27 fish from 4 clutches, and mpi MO+Mannose (Man.) has 37 fish from 4 clutches.
  • Figure-5-B
    Figure 5. (B) The graph compares the WT and p53 -/- fish.


Since mannose and p53 both helped rescue cell death in the mpi MO fish, the next step was to see if the two were working in the same pathway or if they were working separately. Not having both p53 alleles was enough to rescue the cell death, so the mannose did not help in the p53-/- fish since the amount of cell death was already low. This caused mpi MO and mpi MO+Mannose not to be statistically significant in p53-/- fish. So to test the relationship between mannose and p53, acridine orange assays were done with p53+/- heterozygotes (Hets).

p53+/- Heterozygous Fish
To get the p53 heterozygous fish, the WT and p53-/- lines of fish were mated. This meant that resulting embryos had only one p53 MT allele. This experiment was meant to see if mannose and p53 could together rescue cell death in the same fish. In the mpi MO fish, the test was to see if both alleles of p53 are necessary for rescue. If having one allele produced cell death it would mean that one allele of p53 is enough to recognize MPI loss. In the mpi MO+Mannose fish, the test was to see if mannose and p53 could both rescue at the same time.

After the preliminary acridine orange assays (Figure 6) the fish were imaged and cells were counted (Fig. 7A,B). The p53+/- Std MO showed little to no cell death with a mean of .62. Interesting enough, the p53+/- mpi MO had a mean of 5.6, statistically significant to the p53+/- Std MO with p=<.0001. This means that having 1 p53 allele is enough to recognize MPI loss and induce apoptosis. Furthermore, the p53+/- mpi MO was significant to the p53-/- mpi MO with p=.0015. This further supports that one p53 allele is enough to induce apoptosis and also means that one p53 allele alone is not enough to completely rescue cell death from MPI loss. Comparing p53+/- mpi MO and mpi MO+Mannose showed even more promising results. The mpi MO+Mannose had a mean of 1.29 per an area. Unlike the p53-/- fish, the heterozygotes had statistically significant mannose rescue, p=.0014. This proves that mannose treatment is able and necessary to rescue cell death when there is a p53 mutation or deletion.

  • Figure-6

Figure 6. Photomicrographs of 24 hpf embryo tails after acridine orange staining of p53+/- fish. These images were taken with a bright field and then with a blue excitation filter. Photos are representative of the groups of fish under each type of treatment. Bright dots are dead cells which have been picked up by the staining. The white line is the scale bar for all pictures and is 1 micron long.

  • Figure-6
    Figure 6. Photomicrographs of 24 hpf embryo tails after acridine orange staining of p53+/- fish. These images were taken with a bright field and then with a blue excitation filter. Photos are representative of the groups of fish under each type of treatment. Bright dots are dead cells which have been picked up by the staining. The white line is the scale bar for all pictures and is 1 micron long.

  • Figure-7-A
  • Figure-7-B

Figure 7. (A) The graph shows the means and standard deviations for the quantified number of cell. Std has 13 fish from 1 clutch, mpi MO has 5 fish from 1 clutch, and mpi MO+Mannose (Man.) has 7 fish from 1 clutches. (B) The graph compares the p53 -/- and p53+/- fish.

  • Figure-7-A
    Figure 7. (A) The graph shows the means and standard deviations for the quantified number of cell. Std has 13 fish from 1 clutch, mpi MO has 5 fish from 1 clutch, and mpi MO+Mannose (Man.) has 7 fish from 1 clutches.
  • Figure-7-B
    Figure 7. (B) The graph compares the p53 -/- and p53+/- fish.


Discussion

Congenital disorders of glycosylation are under-diagnosed disorders that arise from autosomal recessive mutations. These mutations occur in genes that are essential for protein glycosylation, proper protein function, and localization. Mutations in this pathway cause faulty production of glycoproteins. CDG patients present multi-systemic pathologies.

Of the many subtypes of CDG, only MPI-CDG has a treatment. MPI-CDG occurs when the MPI gene has a mutation and leads to decreased MPI activity and poor conversion between mannose-6-phosphate and fructose-6-phosphate. The treatment of oral mannose can be used to bypass the MPI loss and increase proper glycosylation of proteins. This project uses zebrafish to study if MPI loss leads to cell death and if the cell death could be rescued.

Figure 8 shows a final graphical and table representation of all the acridine orange experiments. The mpi MO results clearly show a decrease of cell death as the number of functional p53 alleles decreases, showing that p53 helps rescue cell death in the presence of MPI loss. The mpi MO fish treated with mannose show that mannose helps rescue cell death as well, in the presence of MPI loss.

 

  • Figure-8-A
  • Figure-8-B

Figure 8. (A) The bar graph shows combined comparisons of all the acridine orange experiments. (B) The table shows the mean number of dead cells for each treatment.

  • Figure-8-A
    Figure 8. (A) The bar graph shows combined comparisons of all the acridine orange experiments.
  • Figure-8-B
    Figure 8. (B) The table shows the mean number of dead cells for each treatment.


The use of the mpi morpholino allowed for the knockdown of zebrafish Mpi and created a model to study the cell death. Through acridine orange staining, cell death was able to be quantified in each fish. Acridine orange assays of a line of wild type fish injected with the mpi MO showed that the loss of MPI leads to significant cell death. To test the effects of the oral mannose treatment, mpi MO injected embryos were also put in mannose fish water. The acridine orange assay of these fish showed a significant rescue of cell death. This proves that early treatment of mannose is able to help minimize the amount of cell death a fish experienced. The use of a p53-/- line of zebrafish allowed for the study of how p53 played a role in the cell death. Among the large amount of cell death mediators, p53 was chosen for its role in apoptosis. Acridine orange assays of p53-/- fish injected with the mpi MO showed a significant decrease in cell death when compared to their wild type counterparts. This indicates that cell death due to MPI loss is caused by p53-mediated apoptosis.

However, because p53 already rescued the cell death, the mannose did very little in the p53-/- fish. Therefore, a p53 heterozygous model was used next. By mating the WT and p53-/- fish, p53+/- fish were created and then subjected to acridine orange staining after MO injections. The mpi MO injection was able to create a mean number of cell death between that of WT mpi MO and p53-/- mpi MO. This shows that one p53 allele is enough to recognize the Mpi loss and induce apoptosis and also enough to partially rescue some of the cell death. When treated with mannose, these p53 heterozygous fish showed further cell rescue. This means that the mannose is necessary for the rescue of the remaining dead cells.

The role of p53 in protein N-glycosylation is a mechanism that has never been documented until now. Furthermore, although it was previously assumed that the defective glycosylation and protein synthesis in MPI- CDG were the main proponents of the disease (3, 4), this project shows how p53 and apoptosis play a crucial factor in the cellular mechanisms and pathologies of MPI-CDG.

References

1. Jenkins, Nigel. "Getting the Glycosylation Right: Implications for the Biotechnology Industry." Nature Biotechnology 14.8 (1996): 975-81. Nature.

2. Chu, Jaime et al. “A Zebrafish Model of Congenital Disorders of Glycosylation with Phosphomannose Isomerase Deficiency Reveals an Early Opportunity for Corrective Mannose Supplementation.” Disease Models & Mechanisms 6.1 (2013): 95–105. PMC.

3. Freeze, Hudson. "Update and Perspectives on Congenital Disorders of Glycosylation." Glycobiology 11.12 (2001): 129-43. Oxford Journals.

4. Mention, K. "Genetic Disorders of Glycosylation." Chapter 42. Essentials of Glycobiology. 2nd ed. Np.: n.p., 2009. PMC.

5. Sparks, Susan E. et al. "Congenital Disorders of N-linked Glycosylation Pathway Overview." GeneReviews: n. pag. PMC.

6. De Rossi, Charles et al. “Ablation of mouse phosphomannose isomerase (Mpi) causes mannose 6-phosphate accumulation, toxicity, and embryonic lethality.” J Biol Chem. (2006): 281(9):5916-27. PMC.

7. Sharma, Vandana et al. “Mannose Supplements Induce Embryonic Lethality and Blindness in Phosphomannose Isomerase Hypomorphic Mice.” The FASEB Journal 28.4 (2014): 1854– 1869. PMC.

8. Bill, Brent R. et al. “A Primer for Morpholino Use in Zebrafish.” Zebrafish 6.1 (2009): 69–77.
PMC.

9. Kroemer, G et al. “Classification of Cell Death: Recommendations of the Nomenclature Committee on Cell Death 2009.” Cell Death and Differentiation16.1 (2009): 3–11. PMC.

10. Elmore, Susan. “Apoptosis: A Review of Programmed Cell Death.” Toxicologic Pathology 35.4 (2007): 495–516. PMC.

11. Puzio-Kuter, Anna M. “The Role of p53 in Metabolic Regulation.” Ed. Arnold J. Levine. Genes & Cancer 2.4 (2011): 385–391. PMC.