Molecular Vision 2007; 13:1866-1872 <http://www.molvis.org/molvis/v13/a209/> Received 5 September 2006 | Accepted 19 September 2007 | Published 3 October 2007

Download Reprint

Characterization of retinal inosine monophosphate dehydrogenase 1 in several mammalian species

Catherine J. Spellicy¹, Stephen P. Daiger^1,2, Lori S. Sullivan¹, Jingya Zhu¹, Qin Liu³, Eric A. Pierce³, Sara J. Bowne¹
 
 

¹Human Genetics Center, School of Public Health, University of Texas Health Science Center, Houston, TX; ²Department of Ophthalmology and Visual Science, University of Texas Health Science Center, Houston, TX; ³FM Kirby Center for Molecular Ophthalmology, University of Pennsylvania School of Medicine, Philadelphia, PA

Correspondence to: Sara J. Bowne, Human Genetics Center, School of Public Health, The University of Texas Health Science Center, 1200 Herman Pressler, Houston, TX 77030; Phone: (713) 500-9836; FAX: (713) 500-0900; email: Sara.J.Bowne@uth.tmc.edu

Abstract

Purpose: The purpose of this study was to characterize the inosine monophosphate dehydrogenase 1 (IMPDH1) protein isoforms in mammalian retinas, in order to determine the species distribution of these variants and identify an optimal animal model for studying IMPDH1-associated retinal diseases. Mutations in IMPDH1 cause the RP10 form of autosomal dominant retinitis pigmentosa, and are a rare cause of Leber congenital amaurosis.

Methods: Retinas from several mammalian species were obtained commercially. Human retinas were isolated by the San Diego Eye Bank and flash frozen within four hours post mortem. Proteins were isolated from retinal tissue using the PARIS^TM protocol. Anti-IMPDH1 antibodies were used to visualize the IMDPH1 proteins on Western blots.

Results: Transcript and protein analyses have shown that IMPDH1 undergoes alternate splicing to produce at least two retinal isoforms in both human and mouse. The relative abundance of these IMPDH1 isoforms is different between mouse and human. This study extends these findings by showing that the two IMPDH1 isoforms are also present in dog, rat, sheep, pig, and cow retina, but that, as with mouse, the relative abundances of these isoforms differ from those found in human retina.

Conclusions: The existence of two major retinal isoforms of the IMPDH1 protein is maintained across all mammalian species tested. The relative abundance of IMPDH1 proteins in human retina is unique in comparison to other mammalian species, indicating an apparent lack of an ideal model organism for human retinal IMPDH1 expression. Pig and/or sheep may prove to be potential model organisms based on the observed retinal isoform abundance in these species. These findings will aid future research in understanding the role of retinal-specific IMPDH1 proteins, and will contribute to research elucidating the pathophysiology associated with IMPDH1 missense mutations.

Introduction

Retinitis pigmentosa (RP) is a heritable retinopathy that is characterized by its vast phenotypic and genetic heterogeneity. RP affects about 100,000 people in the United States alone, and about 1.5 million people worldwide [1]. Many distinct genes cause RP; to date 16 autosomal dominant, 18 autosomal recessive, and 6 X-linked forms have been identified in addition to many other syndromic, systemic, and complex forms of RP (RetNet). Initial symptoms of RP include night blindness and, later, loss of peripheral vision culminating in tunnel vision and legal or complete blindness [2]. Age of onset, rate of progression, penetrance, and ultimate outcome of the disease varies between individuals, both within and between families. The mechanism of disease for some of these genes is explained due to their involvement in well-characterized pathways such as phototransduction or the visual cycle [3]. However, the pathophysiologic mechanisms for many other RP genes are still a mystery.

Inosine monophosphate dehydrogenase 1 (IMPDH1) is an example of an RP gene whose mechanism of action is unclear. Missense mutations in IMPDH1 cause the RP10 form of retinitis pigmentosa [4,5] and are also a rare cause of Leber congenital amaurosis (LCA) [6]. These mutations are dominant acting and tend to exhibit early onset and rapid progression of disease [4,5,7-9]. IMPDH1 is located on chromosome 7q32.1 and codes for the enzyme of the same name.

IMPDH is a highly conserved class of enzymes that is found in prokaryotes and eukaryotes [10,11]. It functions in the de novo purine synthesis pathway; specifically, IMPDH catalyzes the rate-limiting step of guanine nucleotide synthesis by converting inosine 5'-monophosphate (IMP) to xanthosine 5'-monophosphate (XMP) with the concomitant reduction of nicotinamide adenine dinucleotide (NAD).

Mammals have two homologues of IMPDH, namely IMPDH1 and IMPDH2, which are 84% identical at the protein level in humans [12]. The enzymatic activity of the two homologues is indistinguishable and both are active in a homotetramer state [13]. Each IMPDH monomer is composed of an eight-stranded alpha/beta barrel structure that performs the enzymatic function, and a flanking subdomain that is composed of two cystathionine b-synthase-like regions, called CBS domains [14].

Research shows that IMPDH1 and IMPDH2 are differentially expressed between tissues; IMPDH2 is abundant in actively proliferating tissues, especially cancerous ones, whereas IMPDH1 is most abundant in activated blood lymphocytes but present at low levels in most other tissues [15-17]. Our research shows that IMPDH1 is even more abundant in the retina, about three fold more abundant compared to other tissues, while IMPDH2 is scarce in retinal tissues [18]. Recent research also shows that both IMPDH1 and IMPDH2 bind single-stranded polynucleotides with nanomolar affinity via the CBS domains [19], and that the disease-causing RP10 missense mutations in IMPDH1 alter the specificity and and strength of this binding [6,20]. How alteration of the nucleotide binding activity of IMPDH1 may be related to retinal disease is currently under investigation.

The IMPDH1 protein present in most tissues is a 55.6 kDa protein that is 514 amino acids in length. We refer to this protein as "canonical" IMPDH1 (see Figure 1A). The gene structure coding for the canonical protein contains 14 coding exons and up to three alternatively spliced untranslated exons, called exons A, B, and C, present just 5' of exon 1. Although alternative starts of transcription are present that lead to alternate transcripts, all produce canonical IMPDH1 protein upon translation [21].

Our research has recently revealed novel IMPDH1 transcripts and proteins in human and mouse retina [18]. These transcripts/proteins result from both alternate splicing events and alternate starts of translation. The majority of these retinal transcripts contain a novel exon, designated exon 13b, that immediately follows exon 13 and precedes exon 14. Exon A is not only transcribed but also translated in some of these novel retinal proteins. In Bowne et. al. 2006, we show that transcripts and proteins similar to those in human retina are observed in mouse retina; however they exist in different ratios than those observed in human [18]. In fact, the major transcripts and proteins in mouse retina are present in abundances inverse to those observed in human retina. The genomic structure of IMPDH1 and the predominant proteins in both human and mouse retinas are shown in Figure 1. Hereafter, the larger retinal isoform (approximately 65 kDa in human) will be called IMPDH1+A+13b, to indicate addition of residues encoded from exons A and 13b, and the smaller retinal isoform (approximately 56 kDa in human) will be called IMPDH1+13b, indicating that only residues encoded by exon 13b are added in this protein isoform (Figure 1B). The predicted molecular weights of the known IMPDH1 proteins are listed in Table 1.

Our observations raise two questions, first, are the distinct IMPDH1 protein isoforms found in other mammalian retinas and, second, if so, which species is likely to be the best model for human disease? Mouse is the most common model organism in which retinopathies are studied. However, for IMPDH1, mouse may be a less-than-ideal model as the Impdh1 knockout mice (Impdh1^-/-) show slow, mild retinal degeneration and few to no other symptoms [22]. In contrast, dominant mutations in IMPDH1 cause a severe, rapidly progressive phenotype in humans [7]. This indicates that the null allele is not as physiologically detrimental as the missense mutation, and suggests that mutations might cause a gain of function. Given the significant differences in IMPDH1 retinal protein abundance between mouse and human, the use of this mouse model may be limited in its translatability to human retinal IMPDH1 research. Therefore, one aim of this research is to characterize the retinal isoforms of IMPDH1 in other mammalian species and, ideally, to identify a model that is similar to human both in terms of retinal proteins present and in relative abundance of these proteins.

Methods

Tissue acquisition

Wild-type C57BL/6J mouse retinas were obtained from Jackson Laboratories, Bar Harbor, ME, and/or from Dr. Avril Kennan and Dr. Peter Humphries at Trinity College, Dublin. Impdh^-/- mouse retinas were obtained from Dr. Beverly Mitchell at the University of North Carolina and/or from Dr. Avril Kennan and Dr. Peter Humphries at Trinity College, Dublin, Ireland. Mice were between 3 and 5 months of age, therefore in Impdh^-/- retinas the photoreceptors were still intact [22]. Dog, cow, rat, mouse, sheep, and pig retinas were obtained from PelFreez in Rogers, AR. Human retinas from anonymous donors were procured by the San Diego Eye Bank in California and flash frozen within four hours of death. Several studies indicate that tissue procurement within this short postmortem interval ensures little, if any, protein or RNA degradation [23-25]. All retina samples were stored at -80 °C until processed further.

Western blot analysis

Total protein lysates were made from the retina of each species via the PARIS protocol from Ambion (Austin, TX). Halt protease inhibitor was added to the cell disruption buffer before homogenization (Pierce, Pittsburg, PA). SDS was added after initial homogenization to a final concentration of 1% and then was further homogenized by drawing it through a 25-gauge syringe. Proteins were quantified using the RC DC Protein Assay (BioRad, Hercules CA) and a Beckman Coulter DV640 spectrophotometer (Fullerton, CA).

Equivalent amounts of protein (about 25 μg of protein) from each species were separated by SDS-PAGE on 8% NuPAGE Novex gels (Invitrogen, Carlsbad, CA) run under reducing conditions. Proteins were transferred electrophoretically to a PVDF membrane using a semi-dry transfer apparatus (BioRad). The antibodies used in this study are affinity purified polyclonal antibodies raised in rabbits against various linear human IMPDH1 epitopes in conjunction with Open Biosystems (Huntsville, AL) [18]. These highly conserved IMPDH1 epitopes were chosen to increase the probability of cross-reactivity of the antibodies between species. Table 2 lists the antibody name, epitope, conservation between species, and reactivity to the various IMPDH proteins. Antibody hybridization was performed as follows: membranes were blocked for 1 h in a 10% nonfat milk solution (PBS-T plus powdered condensed milk, BioRad, Hercules, CA). Primary and secondary antibody hybridizations were done in 5% nonfat milk block solution. Antibody-bound proteins were visualized with the Immun-Star HRP chemiluminescent kit (BioRad) and Kodak Bio-Max Light film.

Results

A series of antibodies were used in this study to identify and differentiate the various IMPDH1 isoforms found in mammalian retina. Historically, creating antibodies to detect IMPDH1 has been a challenge especially creating antibodies that distinguish between IMPDH1 and IMPDH2, due to the high sequence similarity of the two proteins. The antibodies used in this study, initially described in Bowne et al. 2006, were designed in two sets. The first set of antibodies distinguishes between IMPDH1 and IMPDH2, while the second set was designed to identify the two major IMPDH1 retinal isoforms. Upon experimental observation the antibodies are all inherently unique in affinity and specificity (Table 1). This variation presents a challenge in some cases due to differential patterns of IMPDH1 specific binding and background binding. Therefore we believe the antibodies are best used in tandem, comparing the data from each to get an accurate picture of biological variation in IMPDH proteins.

We used a two-fold approach to characterizing the spectrum of IMPDH1 proteins in mammalian retinas. First, antibodies that distinguish IMPDH1 from IMPDH2 were used to determine the complete set of IMPDH1 proteins. Second, antibodies that recognize the unique protein residues encoded by exon A and exon 13b were utilized to determine which of the IMPDH1 retinal isoforms were present.

Initially pan-IMPDH1 antibodies, anti-C-IMPDH1 and anti-N-IMPDH1, were used to identify all IMPDH1 proteins and compare the protein patterns to those from human retinal lysates. Western blot analysis of total retinal lysates with anti-C-IMPDH1 shows at least two major bands in each species examined (Figure 2A). The high molecular weight protein in each species runs at approximately 65 kDa and is similar to the predicted size of the human and mouse proteins which contain translated sequences from both exon A and exon 13b (Table 2). The low molecular weight protein in each species runs at approximately 56 kDa and is similar to the predicted size of the human and mouse IMPDH1 proteins which contain the translated sequence from exon 13b but not from exon A. The high molecular weight protein is more abundant in cow, dog and rat, while the lower molecular weight protein is less abundant in these species. In human retina, the opposite is true: the smaller IMPDH1 isoform is more abundant than the larger IMPDH1 isoform. In sheep and pig the abundance of the two IMPDH1 isoforms are approximately the same. The protein patterns observed in human and mouse retinas are consistent with our previous protein and transcript analyses [18].

Results from the anti-N-IMPDH1 antibody were similar to the anti-C-IMPDH1 results in dog and mouse. The anti-N-IMPDH1 antibody does not seem to recognize the IMPDH1 proteins in cow, sheep, or pig (see Figure 2B). This is likely due to the variant residue found in the epitope used for anti-N-IMPDH1 (see Table 2). That anti-N-IMPDH1 does not recognize IMPDH1 in sheep is a mystery given the complete conservation of the epitope.

Anti-C-IMPDH1+2 showed four bands in each species tested except the knock-out mouse. This antibody binds both IMPDH1 and IMPDH2 concurrently. In Figure 2C, from higher molecular weight to lower molecular weight, the arrows represent the human IMPDH1 isoforms, indicating exons A and 13b (IMPDH1+A+13b) in green, and including exon 13b only (IMPDH1+13b) in red. Human IMPDH2 is indicated by the blue arrow. The lowest molecular weight band (in yellow) - observed in all species except knockout mouse - is thought to be nonspecific binding of the antibody. It is possible that this band may be canonical IMPDH1 given that it is not observed in the IMPDH1 knockout mouse, but this is unlikely given that neither the anti-C-IMPDH1 nor the anti-N-IMPDH1 specific antibodies detect the band. Nonetheless this requires further investigation.

Additional antibodies were used to confirm that the IMPDH1 proteins identified using the pan-IMPDH1 antibodies contained exon A and/or exon 13b encoded residues. One antibody was specific to protein sequences in exon A only (anti-exon-A-IMPDH1), while the other was specific to protein sequences in only exon 13b (anti-exon-13b-IMPDH1) [18]. Analysis of retinal cell lysates with the anti-exon-A-IMPDH1 antibody shows that there is one major protein that contains residues resulting from translation of exon A in mouse, dog, pig, sheep, and rat (see Figure 2D, green arrow). This is consistent with our previous findings that both mouse and human have a retinal protein that includes residues encoded by the exon A genomic sequence [18]. This protein corresponds to the higher molecular weight proteins seen with anti-C-IMPDH1.

Western blot analyses using the anti-exon 13b-IMPDH1 antibody consistently show two bands in each of the species analyzed (see Figure 2E). These bands correspond to the IMPDH1 isoforms identified with anti-C-IMPDH1. This, too, accords with our previous findings using human and mouse retinal IMPDH1 protein and message RNA [18]. The molecular weight of the two IMPDH1 proteins identified with anti-exon-13b-IMPDH1 varies slightly between the species studied. Table 1 lists the predicted molecular weights of the IMPDH1 protein isoforms in human and mouse [18].

These results show that most mammalian species have two predominant IMPDH1 retinal proteins: a larger protein that includes translated sequence from both exons A and 13b, and a smaller protein that includes translated sequence from exon 13b. This finding is consistent with our previous data described in Bowne et. al., 2006 [18]. In Bowne et. al., 2006, we determined the range of IMPDH1 transcripts found in human and mouse retina and examined the relative transcript/protein abundance using both semiquantitative PCR to gauge relative transcript abundance, and Western blotting to visualize relative protein abundance [18].

In dog, cow, mouse, and rat the larger IMPDH1 protein containing both exon A and exon 13b translated sequence is more abundant. In human retina the smaller IMPDH1 protein containing exon 13b translated sequence is the most abundant isoform. Sheep and pig are distinct from the other mammalian species studied in that they have relatively equal amounts of the two retinal IMPDH1 proteins. These analyses suggest that human retina has a unique pattern of IMPDH1 protein abundances.

Discussion

Missense mutations in IMPDH1 cause retinal-specific degeneration. There are two homologs of IMPDH present in mammals, specifically IMPDH1 and IMPDH2. In retinal tissue it has been shown that IMPDH2 expression is very low or absent in comparison to IMPDH1 expression [22]. We recently confirmed this observation using serial analysis of gene expression (SAGE) to measure gene expression in the retina [18]. This indicates that IMPDH1 is of particular importance in photoreceptor cells and may be primarily responsible for the production of guanine nucleotides in these cells [22].

Recently, two novel isoforms of IMPDH1 have been discovered in human and mouse retina. One includes translation of both exons A and 13b resulting in an approximately 65 kDa protein which we refer to here as IMPDH1+A+13b. The second novel isoform in retina is a protein including translation of exon 13b only, excluding exon A, resulting in an approximately 56 kDa protein which we refer to as IMPDH1+13b [18].

All mammals tested in this study, sheep, mouse, human, rat, pig, and cow, have the same two, predominant, retinal IMPDH1 proteins most likely corresponding to the 56 kDa and 65 kDa proteins described previously [18]. However, the relative abundance of these two predominant proteins varies between species. Thus, production of two retinal protein isoforms is an evolutionarily conserved feature of IMPDH1, suggesting possible important physiologic roles for the different isoforms. While none of the species studied share the same ratios of retinal IMPDH1 proteins seen in human retina, pig and sheep retina come the closest in that they have equal abundances of the IMPDH1+A+13b and IMPDH1+13b proteins. Therefore, pig or sheep may be better animals in which to study the biology of IMPDH1 in the retina.

In contrast to the qualitative conservation of IMPDH1 isoforms, the observed difference in abundances of retinal IMPDH1 proteins suggests that despite the high genomic conservation of this gene, its expression varies significantly between closely-related species. This could be caused by a sequence change in promoter or enhancer regions or in some other important regulatory sequence.

Results show that retinal IMPDH1 proteins differ not only in relative abundance but also differ slightly in molecular weight between species. The differences in molecular weight suggest posttranslational modification varies between species.

Further experiments are needed to characterize the enzymatic activity of these retinal isoforms, if there is any, and to test the effect of the autosomal dominant retinitis pigmentosa and LCA-associated mutations on both the structure and activity of the proteins. IMPDH1 has been shown to bind nucleic acids [19,20], therefore this function of the IMPDH1 retinal isoforms needs to be investigated, as well as the effect of the disease-causing mutations on binding function. These studies are currently underway.

The findings reported in this study are useful in advancing our knowledge and research capability relating to retinal IMPDH1 and the pathophysiology of missense mutations in IMPDH1 gene.

Acknowledgements

This research was supported by The Foundation Fighting Blindness; The William Stamps Farish Fund; The Hermann Eye Fund; NIH-NEI grants EY07142, EY14170, and EY12910; the F.M. Kirby Foundation; Houston Area NEI Vision Training Program, EY007024; and a Bridging Grant from the University of Texas Health Science Center, Houston. We thank Dr. Avril Kennan and Dr. Pete Humphries at Trinity College Dublin, and Dr. Beverly Mitchell at the University of North Carolina for the gift of the mouse retinas.

References

1. Haim M. Epidemiology of retinitis pigmentosa in Denmark. Acta Ophthalmol Scand Suppl 2002; 233:1-34.

2. Heckenlively JR, Daiger SP. Hereditary retinal and choroidal degenerations. In: Rimoin DL, Connor M, Pyeritz RE, Korf BR, Emery AE, editors. Emery & Rimoin's principals and practice of medical genetics. Vol 3. 4th ed. New York: Churchill Livingstone; 2002. p. 3555-3593.

3. Kennan A, Aherne A, Humphries P. Light in retinitis pigmentosa. Trends Genet 2005; 21:103-10.

4. Bowne SJ, Sullivan LS, Blanton SH, Cepko CL, Blackshaw S, Birch DG, Hughbanks-Wheaton D, Heckenlively JR, Daiger SP. Mutations in the inosine monophosphate dehydrogenase 1 gene (IMPDH1) cause the RP10 form of autosomal dominant retinitis pigmentosa. Hum Mol Genet 2002; 11:559-68.

5. Kennan A, Aherne A, Palfi A, Humphries M, McKee A, Stitt A, Simpson DA, Demtroder K, Orntoft T, Ayuso C, Kenna PF, Farrar GJ, Humphries P. Identification of an IMPDH1 mutation in autosomal dominant retinitis pigmentosa (RP10) revealed following comparative microarray analysis of transcripts derived from retinas of wild-type and Rho(-/-) mice. Hum Mol Genet 2002; 11:547-57.

6. Bowne SJ, Sullivan LS, Mortimer SE, Hedstrom L, Zhu J, Spellicy CJ, Gire AI, Hughbanks-Wheaton D, Birch DG, Lewis RA, Heckenlively JR, Daiger SP. Spectrum and frequency of mutations in IMPDH1 associated with autosomal dominant retinitis pigmentosa and leber congenital amaurosis. Invest Ophthalmol Vis Sci 2006; 47:34-42.

7. Kozma P, Hughbanks-Wheaton DK, Locke KG, Fish GE, Gire AI, Spellicy CJ, Sullivan LS, Bowne SJ, Daiger SP, Birch DG. Phenotypic characterization of a large family with RP10 autosomal-dominant retinitis pigmentosa: an Asp226Asn mutation in the IMPDH1 gene. Am J Ophthalmol 2005; 140:858-867.

8. Wada Y, Sandberg MA, McGee TL, Stillberger MA, Berson EL, Dryja TP. Screen of the IMPDH1 gene among patients with dominant retinitis pigmentosa and clinical features associated with the most common mutation, Asp226Asn. Invest Ophthalmol Vis Sci 2005; 46:1735-41.

9. Schatz P, Ponjavic V, Andreasson S, McGee TL, Dryja TP, Abrahamson M. Clinical phenotype in a Swedish family with a mutation in the IMPDH1 gene. Ophthalmic Genet 2005; 26:119-24.

10. Zhang R, Evans G, Rotella F, Westbrook E, Huberman E, Joachimiak A, Collart FR. Differential signatures of bacterial and mammalian IMP dehydrogenase enzymes. Curr Med Chem 1999; 6:537-43.

11. Hedstrom L. IMP dehydrogenase: mechanism of action and inhibition. Curr Med Chem 1999; 6:545-60.

12. Natsumeda Y, Ohno S, Kawasaki H, Konno Y, Weber G, Suzuki K. Two distinct cDNAs for human IMP dehydrogenase. J Biol Chem 1990; 265:5292-5.

13. Carr SF, Papp E, Wu JC, Natsumeda Y. Characterization of human type I and type II IMP dehydrogenases. J Biol Chem 1993; 268:27286-90.

14. Sintchak MD, Fleming MA, Futer O, Raybuck SA, Chambers SP, Caron PR, Murcko MA, Wilson KP. Structure and mechanism of inosine monophosphate dehydrogenase in complex with the immunosuppressant mycophenolic acid. Cell 1996; 85:921-30.

15. Senda M, Natsumeda Y. Tissue-differential expression of two distinct genes for human IMP dehydrogenase (E.C.1.1.1.205). Life Sci 1994; 54:1917-26.

16. Nagai M, Natsumeda Y, Konno Y, Hoffman R, Irino S, Weber G. Selective up-regulation of type II inosine 5'-monophosphate dehydrogenase messenger RNA expression in human leukemias. Cancer Res 1991; 51:3886-90.

17. Jain J, Almquist SJ, Ford PJ, Shlyakhter D, Wang Y, Nimmesgern E, Germann UA. Regulation of inosine monophosphate dehydrogenase type I and type II isoforms in human lymphocytes. Biochem Pharmacol 2004; 67:767-76.

18. Bowne SJ, Liu Q, Sullivan LS, Zhu J, Spellicy CJ, Rickman CB, Pierce EA, Daiger SP. Why do mutations in the ubiquitously expressed housekeeping gene IMPDH1 cause retina-specific photoreceptor degeneration? Invest Ophthalmol Vis Sci 2006; 47:3754-65.

19. McLean JE, Hamaguchi N, Belenky P, Mortimer SE, Stanton M, Hedstrom L. Inosine 5'-monophosphate dehydrogenase binds nucleic acids in vitro and in vivo. Biochem J 2004; 379:243-51.

20. Mortimer SE, Hedstrom L. Autosomal dominant retinitis pigmentosa mutations in inosine 5'-monophosphate dehydrogenase type I disrupt nucleic acid binding. Biochem J 2005; 390:41-7.

21. Gu JJ, Spychala J, Mitchell BS. Regulation of the human inosine monophosphate dehydrogenase type I gene. Utilization of alternative promoters. J Biol Chem 1997; 272:4458-66.

22. Aherne A, Kennan A, Kenna PF, McNally N, Lloyd DG, Alberts IL, Kiang AS, Humphries MM, Ayuso C, Engel PC, Gu JJ, Mitchell BS, Farrar GJ, Humphries P. On the molecular pathology of neurodegeneration in IMPDH1-based retinitis pigmentosa. Hum Mol Genet 2004; 13:641-50.

23. Stan AD, Ghose S, Gao XM, Roberts RC, Lewis-Amezcua K, Hatanpaa KJ, Tamminga CA. Human postmortem tissue: what quality markers matter? Brain Res 2006; 1123:1-11.

24. De Paepe ME, Mao Q, Huang C, Zhu D, Jackson CL, Hansen K. Postmortem RNA and protein stability in perinatal human lungs. Diagn Mol Pathol 2002; 11:170-6.

25. Malik KJ, Chen CD, Olsen TW. Stability of RNA from the retina and retinal pigment epithelium in a porcine model simulating human eye bank conditions. Invest Ophthalmol Vis Sci 2003; 44:2730-5.