Comprehensive Evaluation of the Extraocular Muscle Critical Period by Expression Profiling in the Dark-Reared Rat and Monocularly Deprived Monkey

Abstract

purpose. To address the consequences of visual deprivation paradigms in rat (dark rearing) and monkey (monocular deprivation) on extraocular muscle (EOM) development using genome-wide expression profiling.

methods. Serial analysis of gene expression (SAGE) was used to determine alterations in the EOM transcriptome induced by dark rearing of rats from birth to postnatal day 45. Data were compared with previously published normal EOM SAGE library. DNA microarray similarly assessed changes in gene expression patterns of EOMs of monkeys reared from birth to 4 months of age with monocular deprivation.

results. Dark rearing produced changes in expression of 280 transcripts in rat EOM. Of these, 71 were known genes representing functional categories that included energy metabolism/mitochondrial-related (21%), protein synthesis and modification (14%), lipid metabolism (13%), and muscle-related (6%) transcripts. Together, the predominant pattern reflected an energetic shift toward fatty acid β-oxidation and integrated alterations in both myofibers and supportive tissues. The response of monkey rectus muscles to monocular deprivation was considerably less severe.

conclusions. The visual deprivation paradigms used in this study mimic alterations that are associated with the common disorders of strabismus, congenital cataract, and amblyopia. These data show that postnatal EOM maturation is broadly susceptible to changes in activity patterns that are a consequence of visuomotor maldevelopment. The data extend the concept of an EOM-critical period and establish that activity patterns in developing eye movement systems play vital determinant roles in the novel EOM phenotype.

The critical period is a postnatal window during which time patterned retinal activity functions to shape structural and functional properties of the sensory visual system. ^{1 2} Inappropriate patterns of visual experience during this critical period can perturb the formation and/or pruning of neural connections in visual relay and processing centers and irreversibly alter the ability to produce an accurate representation of the world around us. A key correlate of the critical period is its tight temporal constraints, because identically altered sensory experience later in life does not produce a similar degradation of visual acuity. The critical-period concept has influenced strategies for management of congenital or infantile disorders of vision and eye alignment and is not restricted to the afferent visual system. Visual malexperience also produces less well-understood age-dependent alterations in the oculomotor system. ^{3 4 5 6 7 8 9 10} Because skeletal muscle is highly responsive to changes in activation patterns, maldevelopment of visuomotor control systems may, in turn, have deleterious consequences for the extraocular muscles (EOMs). 

Evidence for critical periods in any skeletal muscle is limited. Perturbation of postnatal steroid hormone levels or muscle-activation patterns have consequences for maturation of specialized muscle fiber types in laryngeal or jaw musculature. ^{11 12 13 14} EOM is fundamentally distinct from other skeletal musculature ^{15 16 17 18 19} and there is evidence that both early and late influences shape EOM myofiber phenotypes. The survival of EOM primordia in an organotypic nerve–muscle coculture system is dependent on trophic support that can be provided by oculomotor, but not spinal, motoneurons. ²⁰ In addition, several EOM traits emerge in parallel with visual and motor system maturation, ^{21 22 23 24} so that the complex requirements of the five oculomotor control systems (vestibulo-ocular, optokinetic, pursuit, saccadic, and vergence) may shape novel EOM properties by critical period-dependent mechanisms. We previously extended early findings that visual deprivation paradigms alter oculomotor motor unit properties ^{25 26 27 28} to show that both dark rearing and perinatal compromise of the vestibulo-ocular reflex suppress the postnatal expression of an EOM-specific trait, the EOM myosin heavy chain isoform (Myh13). ^{29 30}  

Here, we used two experimental paradigms to address the EOM critical-period concept: expression profiling with serial analysis of gene expression (SAGE) in dark-reared rats and DNA microarray in monocularly deprived (MD) monkeys. Dark rearing blocks the development of the columnar organization of primary visual cortex. Subsequent analysis of alterations in rat EOMs by SAGE offers the distinct advantage of an unbiased genomic screen, potentially detecting changes in any transcript, that also allows valid comparisons among data sets generated in independent experiments or laboratories. ^{15 31} MD perturbs ocular dominance column development, putting one eye at a competitive advantage in a way that mimics the visual consequences of human strabismus and amblyopia. Analysis of MD monkey EOMs using DNA microarrays (Affymetrix, Santa Clara, CA) provides a broad-based and unbiased assessment of activity-induced changes. Collectively, data from these approaches identify a broad pattern of EOM phenotype determination by visuomotor activity and lend support to the concept of an EOM critical period. 

Materials and Methods

Animals

Forty-five-day-old male Sprague-Dawley rats (Harlan, Indianapolis, IN) were reared in darkness (DR) from birth. All animal maintenance (cage changing and feeding) was conducted making brief use of a low-intensity lamp with a red filter (1A Safelight Filter; Eastman Kodak, Rochester, NY). Rat rod photoreceptors are barely sensitive to the extreme end of the spectrum, and so a dark-adapted eye may be exposed to fairly high luminance levels of deep red light without loss of adaptation. ³²  

Two infant macaque monkeys (Macaca mulatta) were reared from birth to 4 months of age with MD produced by tarsorrhaphy. Two additional normally reared, age-matched monkeys served as the control. The closed eyelids were examined on a daily basis to ensure that opposition remained intact and that no signs of irritation occurred. Monkeys were subjected to visual acuity testing with the forced-choice preferential looking (FPL) method. ³³ Assessments of monocular visual acuity were made with Teller acuity cards using prescribed procedures (Vistech Consultants, Inc., Dayton, OH). Each card has black-and-white stripes of a particular width on one side of the card, surrounded by an isoluminant gray background. The monkey was positioned in such a way that its preferential looking toward stripes could be assessed. Infant monkeys naturally look toward stripes at one end of the card rather then the homogenous gray field on the other end. By varying the number of stripes (cycles) per degree of visual angle, acuity can be estimated reliably. In 4-month-old monkeys, we found no measurable acuity in the MD eye, even at the lowest spatial frequency stimulus (0.32 cyc/deg) tested. This finding is consistent with the profound amblyopia associated with MD. In contrast, the visually experienced eye of MD monkeys evinced acuity development similar to that of the age-matched normal control monkeys (i.e., >9.0 cyc/deg). 

All animal procedures were approved by the Institutional Animal Care and Use Committee at Case Western Reserve University (rat and monkey) and Emory University (monkey) and were in accordance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. 

MicroSAGE Protocol and Data Analysis: Rat EOM

At the end of the DR period, all rectus and oblique EOMs were rapidly isolated and removed from both orbits of each of five rats after asphyxiation with carbon dioxide. These are the same muscles that were used in our prior normative study. Tissues were pooled, flash frozen in liquid nitrogen, and stored at −80°C. A SAGE library then was generated using the microSAGE protocol (version 1.0e; http://www.sagenet.org/ provided in the public domain by the Johns Hopkins Oncology Center, Molecular Genetics Laboratory, Baltimore, MD), ³⁴ with modifications as described previously. ¹⁵ Briefly, muscle tissue poly (A)⁺ RNA was reverse transcribed and 14-bp oligonucleotide sequences (tags) were then cut from cDNA with the anchoring enzyme NlaIII (New England Biolabs, Beverly, MA) and released from the 3′ end by using the tagging enzyme BsmFI (New England Biolabs). After SAGE tag isolation and ligation to form ditags, ditags were amplified by PCR, digested by NlaIII, and ligated to form concatemers. Concatenated ditags were purified and cloned into the SphI site of a vector (pZErO-1; Invitrogen, Carlsbad, CA). 

Two thirds of the SAGE tags reported herein were sequenced from plasmids that were randomly selected from bacterial colonies, and one third were from PCR products amplified with M13 forward and reverse primers from colonies that contained inserts more than 616 bp in length. SAGE tag sequencing was performed by Genome Therapeutics Corp. (Waltham, MA; using BigDye terminator chemistry [Applied Biosystems, Inc., Foster City, CA], on a MegaBace automated sequencer [Amersham Biosciences, Piscataway, NJ]). Tag sequences were extracted with eSAGE 1.2a software. ³⁵ SAGE tag sequences containing base calls with Phred (CodonCode Corp, Dedham, MA) quality values less than 20 were not included in the database. 

After sequencing, sequence and location data for each tag were used to identify specific transcripts based on National Center for Biotechnology Information UniGene build number 115 for rat (http://www.ncbi.nlm.nih.gov/UniGene/ provided in the public domain by NCBI). The DR EOM SAGE data then were compared with our existing light-reared (LR; 45-day old rats reared under a normal 12-hour light–dark cycle) EOM SAGE library (available at http://www.ncbi.nlm.nih.gov/geo/ under accession number GSM581; NCBI), ¹⁵ after reanalysis of the previous library using eSAGE 1.2a software (http://genome.nhgri.nih.gov/eSAGE/ provided in the public domain by NCBI) and the current UniGene build to ensure equivalent decoding of SAGE tags. eSAGE and the statistical method of Audic and Claverie (http://igs-server.cnrs-mrs.fr/∼audic/significance.html/) ³⁶ were used to identify genes that were differentially expressed in EOMs of the DR versus LR animals. 

DNA Microarray Protocol and Data Analysis: Monkey EOM

Lateral rectus and medial rectus muscles of the MD and control monkeys were independently isolated and compared by using DNA microarray. EOMs from the deprived eye were isolated and processed for DNA microarray, as described previously, ^{17 37} except that human U133 (A and B) array sets (Affymetrix) were used to evaluate monkey EOMs. These arrays interrogate >33,000 well-documented human transcripts. The probe sets on these human arrays have been shown to have sufficient sequence homology for use in monkeys. ³⁸  

Microarray data were scaled to the same target intensity and analyzed by computer (Microarray Suite; MAS ver. 5.0; Affymetrix). Pair-wise comparisons were made between MD and control monkey EOM samples. Transcripts defined as differentially regulated met the criteria of (1) a consistent increased–decreased call versus control in both replicates, based on Wilcoxon’s signed rank test (the algorithm assesses microarray probe pair saturation, calculates a probability [P] and determines increased, decreased, or no-change calls) and (2) absolute value of the average difference in the multiple of change of 1.8 or more. 

Results

Distribution of DR Rat EOM SAGE Tags

SAGE was used to obtain a quantitative gene expression profile from DR EOM that was then compared with our existing LR EOM SAGE library ¹⁵ using eSAGE. Two independent quality-control measurements validate the SAGE library presented herein. First, the average GC content of tags in our DR SAGE library was nearly identical with that of the LR library (DR: 46.4%, LR: 48.0%), indicating the absence of any GC content bias in both (GC bias results from loss of AT-rich regions during tag processing and is defined as GC content ≥55%). Second, linker contamination of our library was very low. The finding of only 0.05% linker sequences was substantially below reports of 0.65% or more in other SAGE libraries. ³⁹  

In SAGE terminology, total SAGE tag number indicates the combined number of copies of all sequenced tags (transcripts). The number of unique tags is equivalent to the total number of distinct transcripts expressed in the library at 1 copy/transcript or more and is the sum of the matched tags (those represented as a known gene or EST in UniGene) and the novel tags (those with identities not recognized by genomics databases). Of the 31,776 total SAGE tags that we accumulated from DR EOM, nearly one-third (10,105) were unique tags and approximately 55% of these were matched tags. Unique tag sequences and counts for the complete DR EOM SAGE library have been deposited to the NCBI Gene Expression Omnibus (GEO) database (http://www.ncbi.nlm.nih.gov/geo/) under accession number GSM3893. Figure 1A summarizes the distribution of matched and novel tags in our DR SAGE library, as a function of copy number. As in most SAGE libraries, the proportion of SAGE tags matching known genes was skewed toward the more abundant tags. Ninety-seven percent of the unique tags in the highest abundance class (≥100 copies/transcript) matched recognized genes or ESTs, whereas only 44.4% of the single-copy tags matched either genes or ESTs in UniGene. Figure 1B shows the percentage of matched genes and ESTs as a function of copy number. 

Comparison of DR and LR Rat EOM SAGE Libraries

The frequency of each tag in our SAGE libraries reflects the relative abundance of the corresponding mRNA in each EOM tissue sample. This feature of SAGE allows digital comparisons between independently generated libraries. The most abundant SAGE tags in DR EOM (detected at a frequency of ≥0.1% of all tags or ≥32 copies/unique tag) also usually were matched tags. Only 5 of the top 96 unique transcripts were novel tags. Similar findings were obtained in the reanalysis of the LR library. Several housekeeping genes (glyceraldehyde-3-phosphate dehydrogenase, β-actin, and lens epithelial protein) were among the most abundant transcripts in both DR and LR libraries, but none was differentially expressed in the two libraries by the conservative criteria used in our study (P ≤ 0.05 and multiple of change ≥3). 

Figure 2 illustrates similarities and differences in unique tag identities detected in the DR and LR EOM SAGE libraries. When single-copy tags were considered, the degree of correspondence in transcripts present in both libraries was low, approximately 20% of unique tags in the combined libraries were shared in DR and LR data. Tags in the intersection of the two libraries were, however, more likely to represent known genes or ESTs (75% matched tags). However, when a threshold of 3 copies per tag or more was applied, 78% of unique tags in the combined data were shared between the two libraries. At this threshold, only 3% (105) of all transcripts represented in both libraries were observed in the DR library only. All 105 of these DR library-specific tags met criteria for significance in comparison with the LR library (P ≤ 0.05). By contrast, 18.6% (564) of unique tags were found only in the larger LR library (only 68 of these differed in expression from the DR library). 

Identity of Differentially Expressed SAGE Tags in DR Versus LR Rat Libraries

Most SAGE tags detected in DR and LR EOMs were represented as single copies; these comprised 71.5% and 72.9% of total tag counts in DR and LR libraries, respectively. Because we used a stringent Phred base-calling threshold, sequencing accuracy was high and most single-copy tags probably represent “real” transcripts. However, at low tag abundance levels there is less confidence in the discrimination capability of the SAGE technique. In identifying differentially regulated transcripts in EOM that result from the dark rearing paradigm, we required a 3-fold change or more and significance level of P ≤ 0.05. By these criteria, 280 unique tags were differentially expressed in DR versus LR EOM, including 172 upregulated and 108 downregulated tags in the DR sample. All 280 unique tags were expressed at three copies or more in at least one of the two libraries. 

Transcripts meeting selection criteria for differential expression in DR EOM were functionally characterized by using gene ontology and other information available from the LocusLink (http://www.ncbi.nlm.nih.gov/LocusLink/ NCBI) and GeneCards (http://bioinfo.weizmann.ac.il/cards/ Weizmann Institute of Science, Rehovot, Israel) databases (Fig. 3 and Table 1 1 1 1 1 1 ). Individual transcripts were assigned to as many as two functional categories. Most of the 280 differentially regulated unique tags represented either ESTs (49%) or uncharacterized novel tags (25%). Of the 71 unique tags representing known genes, the energy metabolism/mitochondrial-related transcript category was the most abundant (21.3% of known genes), followed by the other (15%) protein synthesis/modification (13.8%) and lipid metabolism (12.5%) categories. Muscle-related transcripts constituted only 6.3% of the unique tags representing known genes. Table 1 1 1 1 1 1 lists all unique tags identified as differentially regulated in DR EOM, with corresponding gene identifications and functions. 

Differential Expression in EOMs of MD Monkeys by DNA Microarray

Lateral rectus and medial rectus muscles from control and MD monkeys were independently evaluated (n = 2/experimental group). Because the extent to which MD alters the monocular versus binocular components of conjugate gaze mechanisms is unclear, deprived eye EOMs were compared with those from untreated, age-matched control monkeys. Using DNA microarray, more than 33,000 transcripts were analyzed and data subjected to a strict filter (consistent increase or decrease call across replicates and fold change cutoff of ≥1.8; using Affymetrix MAS 5.0 software). Lateral rectus muscles showed differential regulation of 20 transcripts, with 5 increased and 15 decreased in expression in the muscles of the deprived eye. Twenty transcripts also were differentially regulated in the deprived eye medial rectus, with an equal number up- and downregulated. There was little overlap among the expression patterns in lateral and medial recti, as only glutamine synthase and an uncharacterized mRNA sequence (GenBank BF514079) were differentially regulated in both muscles. Identities and functions of transcripts differentially regulated by MD are shown in Tables 2 and 3 . 

Discussion

Using genome-wide expression profiling with SAGE, we have demonstrated a broad pattern of differential gene expression in rat and monkey EOMs subsequent to experimental perturbations of visual system development. Data show that the altered visuomotor system output associated with dark rearing produced substantial changes in 280 transcripts encompassing a wide range of muscle tissue functions. MD produced alterations in fewer transcripts in monkey EOM. Patterned changes in gene expression in the current study are attributed to the activity dependence of maturing EOM and its associated tissues. These data provide support for the EOM critical-period concept, suggesting that EOM maturation is regulated, at least in part, by the activity patterns of developing visuomotor systems. 

The visual deprivation models and gene profiling techniques used were selected to take full advantage of the strengths of each model. Because the rat has a limited binocular field, we chose dark rearing to globally suppress the development of primary visual cortex. SAGE provides excellent in-depth expression analysis and our previously published, normative rat EOM SAGE library served as a control in our study. By contrast, the substantial binocular field of the macaque monkey allowed us to use the more specific, clinically relevant strategy of MD. A SAGE database is not available for the monkey, and the human database could not be used, because species sequence differences become problematic when gene identity is based on a very short SAGE tag sequence. Thus, human microarrays (Affymetrix), which survey approximately 33,000 genes and ESTs and have been demonstrated to have sufficient sequence homology for use in monkey, ³⁸ were used. Collectively, the visual deprivation paradigms and expression profiling tools used in the study are well accepted and provide considerable insight into transcriptional changes in EOM that result from visual system maldevelopment. 

SAGE and DNA microarray are formidable tools for assessing the entire transcriptome of cell or tissue types. In an in vivo assessment of the EOM expression profile, we acknowledge that multiple cell types (e.g., muscle, neural, vascular, connective tissue, resident inflammatory cells) contributed to the SAGE and microarray data generated herein, and it is likely that visual deprivation has produced transcriptional changes in supportive tissues as well as myofibers. However, muscle is an integration of a wide array of tissue types and the complex systems biology question of how activity-dependent regulation of the novel EOM phenotype is achieved cannot be addressed only through study of isolated myofibers. In our study, we used stringent controls and data acceptance criteria to provide a conservative estimate of genes that may participate in an EOM critical period. 

In serving as the effector organ for the wide dynamic range of eye movement control systems ranging from pursuit and vergence movements of less than 1 deg/sec to saccades that can exceed 800 deg/sec, while maintaining precise interocular alignment and foveation of visual targets, the EOMs arguably face the most extreme demands of any skeletal muscle. Thus, the baseline morphology and gene expression profile of EOM is fundamentally different from that of other striated muscles, with this muscle group encompassing traits from both cardiac and skeletal musculature in adapting to the eye movement role. ^{15 16 17 18 19} A key gap in EOM biology is the lack of understanding of developmental mechanisms that modulate assembly of the atypical EOM fiber types and their associated supporting tissues. The correlation between the complexity and diversity of eye movement control systems and the novel properties of the EOMs suggested to us that the two were mechanistically linked, with the functional demands placed on EOM acting as direct determinants of the muscle phenotype. 

In prior studies, we demonstrated activity dependence in the postnatal appearance of a key EOM trait, the EOM-specific myosin (Myh13). ^{29 30} That Myh13 is tightly regulated is seen in both its late expression ²⁴ and spatial restriction to only the perijunctional regions of specific EOM fiber types. ^{24 40 41} Interference in maturation of either the visual afferent system ²⁹ or the vestibulo-ocular reflex ³⁰ led to suppression of Myh13 mRNA, presumably as a result of decreased oculomotor motoneuron activity in both paradigms. Neither manipulation was effective in downregulation of Myh13 when applied in adult rats. Collectively, these data established that Myh13 expression by EOM myofibers requires specific signaling, most likely neural activity patterns, during a postnatal temporal window and thereby suggests the existence of an EOM critical period. We did not, however, detect changes in Myh13 here. This may be the result of the very restricted distribution of Myh13, the relatively low level of Myh13 suppression, and/or differences in the sensitivity of the techniques used. In particular, we have taken a very stringent approach toward analysis of the expression profiling data presented herein that may lead to false negatives. 

In this study, we extended the breadth of EOM transcripts that are influenced by the alterations in visuomotor activity that accompany dark rearing. Each differentially regulated transcript is a candidate for the EOM critical period. The 280 distinct transcripts that met selection criteria for differential regulation in DR rat EOM were not restricted to myofiber-specific genes, but included genes shared by myofibers and other cell types and genes usually associated with other tissue types. Several muscle-specific genes (upregulated: sarcosin, triadin 1, calpain 3, and cardiac calsequestrin; downregulated calponin 3) were differentially regulated in DR EOM. EOM is known to express cardiac muscle transcript isoforms, ^{17 19 42} and dark rearing increased the expression of two of these, triadin 1 and cardiac calsequestrin. Several transcripts that function as cytoskeletal elements also were differentially expressed (e.g., upregulated: dynein-associated protein RKM23 and afadin; downregulated: calponin 3). Changes in Myh13 did not reach significance. 

The most substantial change in DR EOM involved transcripts related to intermediary and energy metabolism (>21% of the known genes differentially expressed in DR). A key rate-limiting enzyme of glycolysis (phosphofructokinase-muscle isoform) was downregulated, whereas several transcripts related to lipid metabolism were upregulated (dodecenoyl-CoA delta isomerase, transaldolase 1, malic enzyme 1, enoyl-CoA hydratase short chain 1, and acyl-CoA oxidase) or downregulated (acetoacetyl-CoA synthetase, thyroid hormone responsive protein, and fatty acid-CoA ligase long chain 5). Upregulation of acyl-CoA oxidase and enoyl-CoA hydratase short chain 1 is particularly compelling, because these represent the first two steps of the pathway for β-oxidation of fatty acids. Three ESTs with homology to β-oxidation pathway transcripts (acyl-CoA dehydrogenase, acyl-CoA thioester hydrolase, and carnitine/acylcarnitine translocase) also were upregulated in DR EOM. These data suggest a shift in DR EOM toward an energy mode that is normally predominant in cardiac, but not skeletal, muscle. Consistent with the energetics theme, a total of seven mitochondria-related transcripts with a broad range of functions were induced in DR EOM. Taken together, the shifts in several muscle-specific and energy metabolism protein transcripts are consistent with the alteration of usage patterns of the EOMs. 

DR-induced changes in transcripts that function in protein translation and posttranslational modification were prominent (13.8% of known genes identified in DR EOM), but due to the broad expression of these genes the involved cell types cannot yet be established. Among nonmuscle transcripts differentially expressed in DR EOM were two downregulated collagen genes (collagen type 1 alpha 1 and procollagen C-proteinase enhancer protein). The high percentage of ESTs and novel tags in our library (collectively 74% of all detected transcripts) makes it difficult to assess the full scope of the EOM critical period at this time. Several of the ESTs have sequence homology to genes that fit the altered cytoskeletal and mitochondrial transcript themes. As genomic databases are completed, the EST and novel tag data presented in this report can be reassessed from a broader knowledge of the identities of differentially regulated transcripts. 

The patterned changes in gene expression of MD monkey EOMs were considerably less severe than in DR rats. Although control monkey lateral and medial rectus muscles do not differ in baseline gene expression patterns (unpublished data), they exhibited differential expression responses to MD. Differences in the lateral and medial rectus response are not unexpected, because the loss of binocular vision in MD disrupts disjunctive fusional vergence movements, whereas conjugate eye movements driven by the nondeprived eye appear to be normal. ⁷ Moreover, both visual pursuit and optokinetic eye movements exhibit a temporal–nasal asymmetry that is accentuated in visual maldevelopment. The different affect on horizontal rectus muscles may be a consequence of such differences in conjugate or disjunctive eye movements. As in DR rats, few muscle-specific transcripts were altered in monkey EOMs; increased expression of MEF2C in the MD medial rectus is consistent with myogenesis. Extracellular matrix components were downregulated in both medial and lateral recti, as they were in DR rat EOMs. Two transcripts related to posttranslational protein modification (HSPB1 and PPIF) were upregulated in MD medial rectus, consistent with differential regulation of multiple genes in this class in DR rat EOM. Few transcripts linked to mitochondrial function or energy metabolism were differentially regulated in MD EOMs (PDK4, a glycolysis regulator, was repressed in lateral rectus; TKT, a pentose phosphate pathway enzyme, and OGDH, a tricarboxylic acid cycle enzyme, were induced in medial rectus). Thus, the shift in energy metabolism that was seen in DR rats was not observed in the MD monkey. Genes shared between the medial and lateral recti of MD monkeys and between the rat DR and monkey MD models are indicated in Tables 2 and 3 . Although only two transcripts were shared between MD lateral and medial recti and only one was shared between MD monkey and DR rat, there was sharing of functional categories of differentially regulated transcripts. 

We suggest that the nature and severity of the gene expression changes in the MD monkey versus the DR rat are consequences of inherent differences in the visual deprivation paradigms. Rather than the prevention of primary visual cortex (V1) maturation that occurs in DR, MD alters ocular dominance column maturation in V1, while preserving at least a monocular drive to eye movement control systems from the non-deprived eye. Lennerstrand and Hansen ²⁶ and Lennerstrand ^{27 28} observed decreases in the contraction speed and fatigue resistance in EOMs of MD cats, but saw no accompanying shift in muscle fiber type composition. Similarly, DR rat EOMs show a slowing of contraction speed, ⁴³ but neither DR nor MD altered the expression of many muscle fiber–specific transcripts. Moreover, congenitally strabismic monkeys, which share visual deficits with MD, exhibit modest morphologic changes in specific EOM fiber types. ⁴⁴ Taken together, the differences in expression signatures obtained from DR and MD paradigms then suggest that EOM requires visuomotor-driven activity for proper development, but that any MD-induced functional adaptations in EOM are associated with transcriptional changes that are either modest or largely below the threshold of DNA microarray. 

Conclusions

The DR and MD paradigms used in the current study mimic alterations that are associated with the common disorders of strabismus, congenital cataract, and amblyopia. Our data show that postnatal EOM maturation is broadly susceptible to visuomotor maldevelopment, suggesting that postnatal eye movement system activity plays a vital determinant role in the maturation of the novel EOM phenotype. Differentially regulated transcripts identified herein represent new gene candidates for an EOM critical period. Yet, the full nature of this critical period is only partially understood, because many of the transcripts detected in DR rat EOM represent ESTs or novel SAGE tags. Collectively, our findings then extend the concept of an EOM critical period and suggest that the nature of visual deprivation is an important factor in this critical period. 

 Supported by National Eye Institute Grants R01 EY09834 (JDP), R01 EY12779 (JDP), and P30 EY11371, and a Departmental Award and a Walt and Lilly Disney Award for Amblyopia Research (JDP) from Research to Prevent Blindness. JDP holds the Carl F. Asseff, MD, Professorship in Ophthalmology. The DNA microarray core facility is supported by National Institutes of Health Grant P30 CA43703.

 Submitted for publication February 18, 2003; revised April 2, 2003; accepted April 24, 2003.

 Disclosure: G. Cheng, None; M.J. Mustari, None; S. Khanna, None; J.D. Porter, None

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Figure 1.

 

Frequency distribution histograms of SAGE tags derived from DR rat EOM. (A) Semilog plot of the frequency of SAGE tags detected at specific abundance levels. Matched tags are those that are represented in genomics databases as known genes or ESTs. Novel tags represent mRNA sequences not found in these databases. Novel tags were most abundant among the single-copy group and declined as a percentage of all tags with increasing copy number. (B) Plot showing the frequency of matched tags among tag abundance groups.

Figure 2.

 

Venn diagrams representing SAGE tag distribution patterns in the dark-reared (DR) and light-reared (LR) rat EOM libraries. (A) Distribution between the two libraries when all SAGE tag copy number categories were considered. Note overlap represents only 19.7% of the total number of unique tags. (B) Distribution between the two libraries when SAGE tag copy number was restricted to three copies per tag or more. Overlap between the two libraries was 78%.

Figure 3.

 

Functional category distribution of SAGE tags detected as differentially regulated in DR rat EOM. Top: Functional distribution of all 280 unique tags. Bottom: Functional distribution of the 71 matched tags that have been linked to known genes. Mus, muscle-related; Cyto, cytoskeletal; Met, metabolism/mitochondrial related; Lip, lipid metabolism; CS/ECM, cell surface/extracellular matrix; Sig, signaling/cell–cell communication; C/D, cell proliferation/death; Trans, transcription; Pro, protein synthesis/modification; Rec, receptor/transporter/ion channel; Imm, immune response; Oth, other/unknown.

Table 1.

 

Differentially Regulated Transcripts in Dark-Reared Rat EOM

Tag	LR	DR	DR/LR	P	Unigene (RS)	Gene Name	Function
TCACTTTCTA	1	11	18.6	1.2E-04	None	Novel tag	Nov
GCAGTGCAAA	1	9	15.2	0.001	12436	ESTs, highly similar to hypothetical protein; FLJ10055[Homo sapiens]	EST
GTGATAGACA	1	9	15.2	0.001	2944	Peroxiredoxin 6 (Prdx6)	Oth
TGGGAACTGT	1	9	15.2	0.001	17567	ESTs, weakly similar to FZD2_RAT; Frizzled 2 precursor (Frizzled-2) (Fz-2); (rFz2) [Rattus norvegicus]	EST
ACTGACAAAA	0	8	13.5	2.7E-04	12038	ESTs	EST
TACATAGCCA	1	8	13.5	0.002	11363	Pyruvate dehydrogenase kinase 2 (Pdk2)	Met
CTCCTGCAGT	1	7	11.8	0.004	3682	ESTs, moderately similar to ESTD_HUMAN esterase D[H. sapiens]	EST
TGAGCTCTGA	1	7	11.8	0.004	None	Novel tag	Nov
CAACCCCAAA	1	6	10.1	0.011	6541	ESTs, weakly similar to RP3A_RAT Rabphilin-3A[R. norvegicus]	EST
CAGCTATTGT	0	6	10.1	0.002	36855	ESTs, highly similar to JC6161 kinesin-associated protein KAP3 splice form A	EST
CATACGCTCA	1	6	10.1	0.011	None	Novel tag	Nov
CTCGAGCGGC	0	6	10.1	0.002	55376	EST	EST
GCACTTGCTG	1	6	10.1	0.011	3996	ESTs, highly similar to OBRG_RAT; leptin receptor gene-related protein (OB-R gene related protein) (OB-RGRP)[R. norvegicus]	EST
GCATCAGTAC	0	6	10.1	0.002	16830	ESTs	EST
GTGTGGAATC	1	6	10.1	0.011	839	FXYD domain-containing ion transport regulator (Fxyd6)	Rec
AAATAAATGT	0	5	8.4	0.005	9668	Myelin-associated glycoprotein (Mag)	Oth
					8978	ESTs, weakly similar to B34252 acyl-CoA dehydrogenase(EC 1.3.99.3) short-chain-specific precursor, hepatic
CAAAATAGTT	1	5	8.4	0.025	None	Novel tag	Nov
CAGCAATAAA	3	15	8.4	7.2E-05	15530	ESTs	EST
CAGCTATGCA	1	5	8.4	0.025	7258	ESTs, highly similar to male-specific lethal-3 H.log 1 (Drosophila) [Mus musculus]	EST
CCCATAATCC	2	10	8.4	0.001	3380	Synaptic glycoprotein SC2 (SC2)	Oth
CTATTTAATT	1	5	8.4	0.025	8763	ESTs, weakly similar to TXTP_RAT tricarboxylate transport protein, mitochondrial precursor (Citrate transport protein) (CTP) (Tricarboxylate carrier protein) [R. norvegicus]	EST
CTCCTGAAGG	1	5	8.4	0.025	2784	ESTs, highly similar to UB6B_MOUSE; ubiquitin-conjugating enzyme E2-23 kDa; (ubiquitin-protein ligase; ubiquitin carrier protein) [M. musculus]	EST
CTTTTTATAC	0	5	8.4	0.005	13643	ESTs, highly similar to calcium binding atopy-related autoantigen 1; atopy-related autoantigen [H. sapiens]	EST
					1040	ESTs, highly similar to G0S2_MOUSE; putative lymphocyte G0/G1 switch protein 2 (G0S2-like protein) [M. musculus]
GATGCCTCCC	0	5	8.4	0.005	None	Novel tag	Nov
GCAGAGAATG	1	5	8.4	0.025	4077	ESTs, weakly similar to A34866 T-cell surface protein RT6.2 precursor	EST
GCTGGTTCCA	0	5	8.4	0.005	9726	Calpain 3 (Capn3)	Mus
GGCTACGGTT	1	5	8.4	0.025	None	Novel tag	Nov
GGGCTGCCCA	1	5	8.4	0.025	3053	SWI/SNF related, matrix associated, actin dependent regulator of chromatin, subfamily d, member2	Oth
					38063	ESTs, weakly similar to sugar transporter[Caenorhabditis elegans]	EST
GTAGGAAGAT	0	5	8.4	0.005	33378	ESTs, highly similar to feminization 1 H.log a (C. elegans); feminization 1 a H.log [C. elegans; M. musculus]	EST
GTTTCGTGGT	1	5	8.4	0.025	17086	ESTs	EST
TACCCCTTTA	1	5	8.4	0.025	None	Novel tag	Nov
TCACCTTGTT	1	5	8.4	0.025	31796	Acyl-coA oxidase (Acoal)	Met, Lip
TCCTGGCAGT	1	5	8.4	0.025	None	Novel tag	Nov
TGAGCAGACG	1	5	8.4	0.025	35504	ESTs	EST
TGCCATTGCA	0	5	8.4	0.005	32316	ESTs, weakly similar to sialyltransferase 3; sialyltransferase)[R. norvegicus]	EST
TGTTTGCAGA	0	5	8.4	0.005	22183	ESTs, moderately similar to hypothetical protein dJ12208.2[H. sapiens]	EST
TTCACAAAGG	0	5	8.4	0.005	1276	Proteasome (prosome, macropain) subunit, alpha type 5 (Psma5)	Pro
					85470	ESTs, weakly similar to JX0229 multicatalytic endopeptidase complex (EC 3.4.99.46) zeta chain
TTGGAATCCA	0	5	8.4	0.005	20119	ESTs	EST
TTTTGAGGAC	0	5	8.4	0.005	3496	ESTs	EST
ATACTGACAT	2	9	7.6	0.003	None	Novel tag	Nov
							(continues)

Table 1A.

 

(continued). Differentially Regulated Transcripts in Dark-Reared Rat EOM

Tag	LR	DR	DR/LR	P	Unigene (RS)	Gene Name	Function
AAAAAAGAAA	0	4	6.8	0.014	19930	ESTs, moderately similar to RIKEN cDNA 2310079N02[M. musculus]	EST
AAGGTCTTTA	0	4	6.8	0.014	13244	ESTs	EST
ACCATCCCTT	0	4	6.8	0.014	None	Novel tag	Nov
ATAAAACCAG	0	4	6.8	0.014	None	Novel tag	Nov
CACAAGCGGT	0	4	6.8	0.014	None	Novel tag	Nov
CCCCCAATTC	0	4	6.8	0.014	31977	Vesicle-associated membrane protein 2 (Vamp2)	Oth
					6430	ESTs
					3748	ESTs, highly similar to gamma tubulin complex component; KIAA1669 protein; similar to Xenopus gamma tubulin interacting protein (yeast SPC98 H.log) [H. sapiens]
CTGAGGGAGG	0	4	6.8	0.014	24472	ESTs, weakly similar to hypothetical protein MGC13119[H. sapiens]	EST
CTGATCCAGT	0	4	6.8	0.014	11946	Ubiquitin fusion degradation I-like	Pro
GAAGTTGTGC	0	4	6.8	0.014	12833	ESTs, highly similar to T43442 hypothetical protein DKFZp43p1514	EST
GCAACCAAAA	0	4	6.8	0.014	37542	Suppressor of K+ transport defect 3	Oth
GCTGTTCTGT	0	4	6.8	0.014	3628	Glutamate oxaloacetate transaminase 2; mitochondrial (aspartate aminotransferase 2) (Got2)	Met
GGCGGGTGGA	0	4	6.8	0.014	20220	ESTs	EST
GTATGCCCCC	0	4	6.8	0.014	None	Novel tag	Nov
GTCCAAGATT	0	4	6.8	0.014	72495	ESTs	EST
GTGATGCCAC	0	4	6.8	0.014	22278	ESTs, highly similar to CGE0_HUMAN; protein CGI-140 (protein PTD008) [H. sapiens]	EST
GTGCTAACAA	0	4	6.8	0.014	44114	ESTs, highly similar to AT10_MOUSE ADAMTS-10(a disintegrin and metalloproteinase with thrombospondin motifs 10; ADAM-TS 10, ADAM-TS10) [M. musculus]	EST
GTTAGGTAGG	0	4	6.8	0.014	3421	Adenylate kinase 2 (Ak2)	Met
TAACCAATCA	0	4	6.8	0.014	3548	ESTs, highly similar to small GTP-binding protein rab5[R. norvegicus]	EST
TAGGCCACCA	0	4	6.8	0.014	None	Novel tag	Nov
TATGTGGAAT	0	4	6.8	0.014	7983	ESTs, highly similar to catenin alpha-like 1; alpha-catenin related protein [M. musculus]	EST
TGCCAAGCCC	0	4	6.8	0.014	19513	ESTs	EST
TGGATCTGAG	0	4	6.8	0.014	72471	ESTs	EST
TGTGACAGTG	0	4	6.8	0.014	13976	ESTs	EST
TTAATTCATT	2	8	6.8	0.007	2230	Translocase of inner mitochondrial membrane 23 H.log(yeast; Timm23)	Met
GACGCCCCCC	3	11	6.2	0.002	None	Novel tag	Nov
CTGTGCTCGA	2	7	5.9	0.015	26731	ESTs	EST
GGAAGTCAGG	2	7	5.9	0.015	None	Novel tag	Nov
GCGAAAGAAT	3	10	5.6	0.004	6847	Enoyl Coenzyme A hydratase, short chain 1 (Echs1)	Met, Lip
GCCCCGGTGC	4	13	5.5	0.001	38221	ESTs, highly similar to cardiac abnormality/abnormal facies (CATCH22); microdeletion syndrome [M. musculus]	EST
AAAATCAAGT	0	3	5.1	0.038	19102	ESTs, weakly similar to S21976; probable RNA-directed DNA polymerase (EC 2.7.7.49; clone MH2C)	EST
AAGGAAATAA	2	6	5.1	0.032	29754	Prohibitin	C/D
ACAGACACTT	0	3	5.1	0.038	8129	ESTs	EST
ACATCGTGCG	0	3	5.1	0.038	53919	ESTs, highly similar to human SLC21A6	EST
ACCGTTATAA	0	3	5.1	0.038	1722	Lysosomal-associated membrane protein 2 (Lamp2)	Oth
AGAGAAGGGT	0	3	5.1	0.038	44362	Inositol 1,4,5-trisphosphate 3-kinase B (ltpkb)	Sig
AGGGAGCCGC	0	3	5.1	0.038	None	Novel tag	Nov
ATAAGCACAT	0	3	5.1	0.038	None	Novel tag	Nov
ATAAGTCATA	0	3	5.1	0.038	None	Novel tag	Nov
ATGATTTAAC	0	3	5.1	0.038	41151	ESTs	EST
ATGGGTGGAA	0	3	5.1	0.038	36164	ESTs, highly similar to hypothetical protein FLJ20211[H. sapiens]	EST
ATGTGTATGT	0	3	5.1	0.038	2560	ESTs	EST
CAAAGACAAT	0	3	5.1	0.038	40403	ESTs, highly similar to HSPC038 protein [H. sapiens]	EST
CATAACACAT	0	3	5.1	0.038	None	Novel tag	Nov
CCGGGGTGAT	0	3	5.1	0.038	None	Novel tag	Nov
CTATGTAGAC	0	3	5.1	0.038	23858	ESTs, moderately similar to HCDI protein [H. sapiens]	EST
CTGCTGTAAT	0	3	5.1	0.038	55106	Glutamate dehydrogenase (Glud1)	Oth
CTGTGTGATC	2	6	5.1	0.032	22047	ESTs, moderately similar to T46271 hypothetical protein DKFZp564P2163.1	EST
CTTCCCACTG	0	3	5.1	0.038	2274	Ubiquitin conjugating enzyme E2I (Ube2i)	Pro
CTTGCTTTTT	0	3	5.1	0.038	44161	Lamin A (Lmna)	Oth
CTTTGAAAGG	0	3	5.1	0.038	None	Novel tag	Nov
							(continues)

Table 1B.

 

(continued). Differentially Regulated Transcripts in Dark-Reared Rat EOM

Tag	LR	DR	DR/LR	P	Unigene (RS)	Gene Name	Function
GAAAAGTCGA	0	3	5.1	0.038	20571	ESTs	EST
GACACAGAAA	0	3	5.1	0.038	41880	ESTs	EST
GAGACTCTAC	0	3	5.1	0.038	9091	ESTs, highly similar to KIAA0710 gene product [H. sapiens]	EST
GAGGATGCTG	0	3	5.1	0.038	71588	EST	EST
GATAAACACA	0	3	5.1	0.038	43517	ESTs	EST
GATCACCTGT	2	6	5.1	0.032	17012	ESTs, highly similar to MGPI_MOUSE Microfibril-associated glycoprotein precursor (MAGP) (MAGP-1) [M. musculus]	EST
GATGAGGGTC	0	3	5.1	0.038	18199	ESTs, highly similar to A41784 tumor necrosis factor-alpha-induced protein B12	EST
GATTCTGCCC	0	3	5.1	0.038	41894	ESTs	EST
GCACACACAT	0	3	5.1	0.038	None	Novel tag	Nov
GCACGAACAT	0	3	5.1	0.038	7295	EST	EST
GCTGAGACTG	0	3	5.1	0.038	11534	ESTs, highly similar to TYRO protein tyrosine kinase binding protein; killer cell activating receptor associated protein[M. musculus]	EST
GGACACCAAA	3	9	5.1	0.008	2938	Mitochondrial ribosomal protein L 17	Pro
GGAGTAGATT	0	3	5.1	0.038	3519	Malic enzyme 1, soluble (Mel)	Met, Lip
GGCTATTTTA	0	3	5.1	0.038	3988	EST	EST
GGCTCAGCTC	0	3	5.1	0.038	None	Novel tag	Nov
GGGCCATTAG	3	9	5.1	0.008	58	Afadin (AF-6)	Sig, Cyto
GGTCCTACCC	0	3	5.1	0.038	33467	ESTs	EST
GTCGCTTCTG	0	3	5.1	0.038	6199	ESTs, moderately similar to U2AG_MOUSE Splicing factor U2AF 35 kDa subunit (U2 auxiliary factor 35 kDa subunit) (U2 snRNP auxiliary factor small subunit) [M. musculus]	EST
GTGACGGCCT	0	3	5.1	0.038	None	Novel tag	Nov
GTGCGGGCTC	0	3	5.1	0.038	39333	Endosulfine alpha (Ensa)	Rec
GTTTGAGCAT	0	3	5.1	0.038	17490	ESTs	EST
TAAAATGTAG	0	3	5.1	0.038	None	Novel tag	Nov
TAACGTCTAG	0	3	5.1	0.038	59806	EST	EST
TAGGTCACAG	0	3	5.1	0.038	None	Novel tag	Nov
TATTAGTTAT	0	3	5.1	0.038	26749	ESTs	EST
TCCAACTTCT	0	3	5.1	0.038	None	Novel tag	Nov
TCTCTGCCTG	0	3	5.1	0.038	11272	SH3 domain binding protein CR16	Sig
TCTGCCCTCC	0	3	5.1	0.038	25440	ESTs	EST
TCTTCTGTGG	0	3	5.1	0.038	3616	ESTs	EST
TGAACAAAAA	0	3	5.1	0.038	8839	EST	EST
TGACATACGT	0	3	5.1	0.038	7910	ESTs	EST
TGATACGATT	0	3	5.1	0.038	3694	ESTs, highly similar to TRAP/Mediator complex component TRAP25 [H. sapiens]	EST
TGCACGCGTC	0	3	5.1	0.038	2315	Translin-associated factor X	Oth
TGCTGAATCA	0	3	5.1	0.038	25036	ESTs	EST
TGGAAAAAAA	0	3	5.1	0.038	35208	ESTs, moderately similar to UCRX_HUMAN ubiquinol-cytochrome C reductase complex 7.2 kDa protein (cytochrome C1, nonheme 7 kDa protein; complex III subunit X; 7.2 kDa cytochrome c1-associated protein subunit; HSPC119) [H. sapiens]	EST
TGGGGTCTCA	0	3	5.1	0.038	35476	ESTs	EST
TGGGGTTTAT	0	3	5.1	0.038	4107	ESTs	EST
TGTGCACCAG	0	3	5.1	0.038	12034	ESTs	EST
TGTGTAATGT	0	3	5.1	0.038	758	ESTs, moderately similar to RIKEN cDNA 4833415N24[M. musculus]	EST
TGTTTTGGAA	0	3	5.1	0.038	8157	ESTs, highly similar to T08778 hypothetical protein DKFZp58611520.1	EST
TTAAGTGTTG	0	3	5.1	0.038	37781	ESTs, moderately similar to G protein-coupled receptor kinase-associated ADP ribosylation factor GTPase-activating protein (GIT1) [R. norvegicus]	EST
TTCTGCTCCT	0	3	5.1	0.038	21976	ESTs	EST
TTGAAGTGGT	0	3	5.1	0.038	73073	ESTs, weakly similar to intersectin (SH3 domain protein 1A) [R. norvegicus]	EST
TTGACGGTGG	0	3	5.1	0.038	6834	ESTs	EST
TTGATGTTGA	0	3	5.1	0.038	2197	ESTs, moderately similar to erythroblast macrophage protein [M. musculus]	EST
TTGTCTGTAA	0	3	5.1	0.038	28239	ESTs	EST
TTTAAAGGAA	0	3	5.1	0.038	None	Novel tag	Nov
TTTACTGGGT	0	3	5.1	0.038	2469	ESTs, highly similar to ZK652.3.p [C. elegans]	EST
TTTATAAGTT	0	3	5.1	0.038	17999	ESTs, highly similar to OM40_MOUSE; Probable mitochondrial import receptor subunit TOM40 H.log (translocase of outer membrane 40 kDa subunit H.log) [M. musculus]	EST
							(continues)

Table 1C.

 

(continued). Differentially Regulated Transcripts in Dark-Reared Rat EOM

Tag	LR	DR	DR/LR	P	Unigene (RS)	Gene Name	Function
TTTGCATATT	0	3	5.1	0.038	19104	ESTs, weakly similar to ACTB_HUMAN; actin, cytoplasmic 1 (beta-actin) [R. norvegicus]	EST
TTTTGAATGC	0	3	5.1	0.038	15531	ESTs	EST
CTGAGTGTGG	7	19	4.6	2.2E-04	3902	ESTs, highly similar to B32394 succinate dehydrogenase (ubiquinone) (EC 1.3.5.1) 27K iron-sulfur protein	EST
TTAATAAAAG	4	11	4.6	0.005	3393	T-cell activation protein-related (PGR1)	Imm
					39	ESTs, highly similar to Y184_HUMAN HYPOTHETICAL PROTEIN KIAA0184 [H. sapiens]
CTACACTGGC	3	8	4.5	0.018	41863	Triadin 1	Mus
GGGCACAGGA	3	8	4.5	0.018	7680	ESTs, highly similar to Fas-activated serine/threonine kinase [M. musculus]	EST
GTCCCAAGGA	3	8	4.5	0.018	4005	Dodecenoyl-coenzyme A delta isomerase (Dci)	Met, Lip
					80835	Rat mRNA for delta3, delta2-enoyl-CoA isomerase
GTTCCGACAG	3	8	4.5	0.018	8104	ESTs	EST
TGTGGTGGAG	3	8	4.5	0.018	24190	ESTs	EST
CAGATCAGAA	5	13	4.4	0.003	14050	ESTs	EST
CAGATCTTTG	4	10	4.2	0.010	3761	Ubiquitin C (UBA52)	Pro
					4300	Ubiquitin A-52 residue ribosomal protein fusion product 1
GACTTGGTCA	4	10	4.2	0.010	None	Novel tag	Nov
TGGGTTAGAC	5	12	4.1	0.005	919	ESTs, highly similar to prefoldin 1; prefoldin subunit 1[H. sapiens]	EST
CTGAGAAATA	3	7	3.9	0.037	8792	ESTs, highly similar to AC48_MOUSE 48; kDa acyl-CoA thioester hydrolase, mitochondrial precursor(p48; Mt-ACT48; protein U8) [M. musculus]	EST
GTGAAGAGTT	3	7	3.9	0.037	18515	ESTs, weakly similar to ubiquitin specific protease 2[R. norvegicus]	EST
TAAACCAGGT	3	7	3.9	0.037	7349	ESTs	EST
TGAAATTTTG	3	7	3.9	0.037	3136	Transaldolase 1	Met, Lip
TAGGGTTACA	4	9	3.8	0.020	28875	Sarcomeric muscle protein (sarcosin)	Mus
TTGTGATTAC	4	9	3.8	0.020	21198	MSI	Mus
AGCAGAGAAT	5	11	3.7	0.011	4180	ESTs	EST
ATACTGACCT	13	27	3.5	1.1E-04	None	Novel tag	Nov
GTAACACATA	10	21	3.5	0.001	61080	ESTs, weakly similar to JC6197 stromelysin 3 (EC 3.4.24)	EST
CAGACCTTGG	5	10	3.4	0.021	3560	Dynein-associated protein RKM23 (Km23)	Cyto
CGCTGGGATG	5	10	3.4	0.021	8526	ESTs, weakly similar to ribosomal protein S23 [R. norvegicus]	EST
GAGGTCAGTG	4	8	3.4	0.039	None	Novel tag	Nov
TCTGGGTCAT	4	8	3.4	0.039	2633	Isocitrate dehydrogenase 3 (NAD+) alpha (Idh3a)	Met
TGCTGCTGCC	5	10	3.4	0.021	3850	ESTs, moderately similar to C chain C, human glyoxalase I complexed with S-P-nitrobenzyloxycarbonylglutathione[H. sapiens]	EST
ACGCAATAAA	11	21	3.2	0.001	19158	ESTs, highly similar to PUA1_mouse adenylosuccinate synthetase, muscle isozyme (IMP–aspartate ligase) (ADSS) (AMPSASE) [M. musculus]	EST
CACATCTGGT	6	11	3.1	0.021	11088	Heat shock 70kD protein 5 (Hspa5)	Pro
TGGCACTATC	17	31	3.1	1.2E-04	7279	ESTs	EST
TTGAAATCAA	6	11	3.1	0.021	22882	NADH ubiquinone oxidoreductase subunit B13	Met
					86600	ESTs, highly similar to vacuolar protein sorting 29 (S. pombe); vacuolar protein sorting 29 (yeast); vacuolar sorting protein 29 [M. musculus]
GCCAAGCCAT	5	9	3.0	0.040	36770	ESTs, weakly similar to CLK3_RAT protein kinase CLK3(CDC-like kinase 3) [R. norvegicus]	EST
TAACGCTGTT	5	9	3.0	0.040	19224	ESTs, weakly similar to solute carrier family 25 (carnitine/acylcarnitine translocase), member 20[R. norvegicus]	EST
ACATTTCCCC	15	3	−3.0	0.047	8524	ESTs, highly similar to putative protein kinase C inhibitor[R. norvegicus]	EST
CCTAGCCCCT	15	3	−3.0	0.047	7961	ESTs, weakly similar to transforming growth factor-beta (TGF-beta) masking protein large subunit [R. norvegicus]	EST
CCTTTAATAA	15	3	−3.0	0.047	8405	ESTs, highly similar to CRP1_MOUSE cysteine-rich protein 1 (Cysteine-rich intestinal protein) (CRIP) [R. norvegicus]	EST
TGGGTTGCTA	15	3	−3.0	0.047	None	Novel tag	Nov
CCCGTGTGCT	16	3	−3.2	0.034	3381	Ribosomal protein S9 (Rps9)	Pro
ATTCTGATGA	23	4	−3.4	0.009	9448	ESTs	EST
TGGTGCCTCT	23	4	−3.4	0.009	3061	Fatty acid coenzyme A ligase, long chain 5 (Facl5)	Met, Lip
AACACTACGG	6	0	−3.6	0.046	4120	ESTs, weakly similar to S11349 nonhistone chromosomal protein HMG-17	EST
AAGAAAGGAG	6	0	−3.6	0.046	2910	Procollagen C-proteinase enhancer protein (Pcolce)	CS/ECM
							(continues)

Table 1D.

 

(continued). Differentially Regulated Transcripts in Dark-Reared Rat EOM

Tag	LR	DR	DR/LR	P	Unigene (RS)	Gene Name	Function
ACGCCCTTCA	6	0	−3.6	0.046	3292	ESTs, moderately similar to mitochondrial ribosomal protein S17 [M. musculus]	EST
ACTACCGGGC	6	0	−3.6	0.046	37686	ESTs, moderately similar to S11276 alpha-adaptin c	EST
AGGTCCTTGT	6	0	−3.6	0.046	None	Novel tag	Nov
CACGCCTTTC	6	0	−3.6	0.046	9215	Acetoacetyl-CoA synthetase (Hba1)	Met, Lip
CACGCCTTTC	6	0	−3.6	0.046	11229	Hemoglobin, alpha 1 (Hba1)	Oth
CATTGCGTGG	6	0	−3.6	0.046	2880	ESTs	EST
CCCTGGTCAC	6	0	−3.6	0.046	None	Novel tag	Nov
CGCGGAGGCC	6	0	−3.6	0.046	11207	Thrombospondin 4 (Thbs4)	CS/ECM
CGTGGATCCC	6	0	−3.6	0.046	12256	ESTs, moderately similar to CDNC_MOUSECYCLIN-DEPENDENT KINASE INHIBITOR 1C(CYCLIN-DEPENDENT KINASE INHIBITOR P57)(P57KIP2) [M. musculus]	EST
CTTCAGCTAT	6	0	−3.6	0.046	22479	ESTs	EST
GAAAGAAACT	6	0	−3.6	0.046	31991	Secreted acidic cystein-rich glycoprotein (osteonectin)	C/D
GAACAATGGA	6	0	−3.6	0.046	8706	pR-ET2 encoded oncodevelopmental protein	Imm
GACCAACAGA	6	0	−3.6	0.046	2953	Collagen, type 1, alpha 1 (Colla1)	CS/ECM
GATATCAACT	6	0	−3.6	0.046	None	Novel tag	Nov
GATTCCCCCC	6	0	−3.6	0.046	16597	ESTs	EST
GATTCTTCAG	6	0	−3.6	0.046	7262	Defender against cell death 1	C/D
GCACCGGGAA	6	0	−3.6	0.046	None	Novel tag	Nov
GCATATTTGA	6	0	−3.6	0.046	2026	ESTs, moderately similar to cytochrome c oxidase subunit VIIb; 1100001F07Rik [M. musculus]	EST
GCTTTGGCTT	6	0	−3.6	0.046	31976	Branched chain keto acid dehydrogenase kinase (Bekdk)	Met, Lip
GTTTCCCCTC	6	0	−3.6	0.046	None	Novel tag	Nov
TAAATTGTAG	6	0	−3.6	0.046	None	Novel tag	Nov
TCCCCTACAT	6	0	−3.6	0.046	None	Novel tag	Nov
TCCTCAGCAC	6	0	−3.6	0.046	6120	ESTs, highly similar to C54819 actin-capping protein beta chain, splice form 2	EST
TGATTCGGTT	6	0	−3.6	0.046	40162	ESTs	EST
TGGACTGCTG	6	0	−3.6	0.046	11301	Mannosidase, alpha, class 2C, member (Man2c1)	Oth
TGGCCAGGGC	6	0	−3.6	0.046	None	Novel tag	Nov
TGGGCCCCTG	6	0	−3.6	0.046	None	Novel tag	Nov
TTCAACCTCA	6	0	−3.6	0.046	16833	ESTs, weakly similar to ROD_RAT; heterogeneous nuclear ribonucleoprotein D0 (hnRNP D0; AU-rich element RNA-binding protein 1) [R. norvegicus]	EST
TTTGGGAAAA	6	0	−3.6	0.046	52763	Ankyrin 3 (G)	Cyto
TTTTTACTGA	6	0	−3.6	0.046	8450	ESTs, highly similar to integral membrane protein 3[M. musculus]	EST
AAATAAAGAT	25	4	−3.7	0.005	4958	Sodium channel, voltage-gated, type 1, beta polypeptide (Scn1b)	Rec
					65438	ESTs
AAGACAGCTG	19	3	−3.8	0.013	39743	RT1 class Ib gene (RT1-R)	Imm
					37183	ESTs
					83611	MHC class I RT1.O type 149 processed pseudogene
ACAGTGGGGA	13	2	−3.8	0.037	22995	ESTs, highly similar to pleckstrin 2 [M. musculus]	EST
GTGTTCTATA	32	5	−3.8	0.001	None	Novel tag	Nov
CTTTATTCCA	33	5	−3.9	0.001	2953	Collagen, type 1, alpha 1 (Collal)	CS/ECM
TCTTCTCACA	20	3	−3.9	0.009	6172	Ribosomal protein L39 (Rpl39)	Pro
AAATGCTTGG	7	0	−4.1	0.029	74273	ESTs, weakly similar to S11661 talin	EST
AAGTCCTTTT	7	0	−4.1	0.029	6211	Glucocorticoid-induced leucine zipper (Gilz)	Trans
					35608	ESTs
AATGGGAGGC	7	0	−4.1	0.029	None	Novel tag	Nov
ACCCCCAGTC	14	2	−4.1	0.026	81140	Thyroid hormone responsive protein (spot14) (Thrsp)	Met, Lip
AGATGTATTT	7	0	−4.1	0.029	4178	ESTs, highly similar to RIKEN cDNA 1010001C05[M. musculus]	EST
AGGCTGGTGA	7	0	−4.1	0.029	6016	Proteasome (prosome, macropain) subunit, beta type 1 (Psmb1)	Imm, Pro
					25100	ESTs
ATAACAACAT	7	0	−4.1	0.029	81261	EST	EST
ATACACATAA	35	5	−4.1	4.3E-04	None	Novel tag	Nov
ATGAAGCCAG	14	2	−4.1	0.026	54594	Voltage-dependent anion channel 1 (Vdac1)	Rec
ATGTAGCCAG	7	0	−4.1	0.029	20160	tRNA selenocysteine associated protein	Pro
CAAACAATCA	7	0	−4.1	0.029	None	Novel tag	Nov
CACGGCCTCT	7	0	−4.1	0.029	None	Novel tag	Nov
CATTCGGAGA	7	0	−4.1	0.029	None	Novel tag	Nov
CCCAGCACTT	7	0	−4.1	0.029	22230	ESTs	EST
GATCAAGTCA	7	0	−4.1	0.029	None	Novel tag	Nov
							(continues)

Table 1E.

 

(continued). Differentially Regulated Transcripts in Dark-Reared Rat EOM

Tag	LR	DR	DR/LR	P	Unigene (RS)	Gene Name	Function
GCAATTATTG	7	0	−4.1	0.029	9197	ESTs	EST
GCCGCAAAGG	7	0	−4.1	0.029	None	Novel tag	Nov
GGCAGAAACT	7	0	−4.1	0.029	None	Novel tag	Nov
GGGATGGACG	7	0	−4.1	0.029	30176	Complement component 4	Imm
TAGGCCACCC	7	0	−4.1	0.029	None	Novel tag	Nov
TAGGTGTGGG	7	0	−4.1	0.029	871	Calponin 3, acidic (Cnn3)	Cyto, Mus
					28172	ESTs
TCTATTCTCA	14	2	−4.1	0.026	None	Novel tag	Nov
GCTTCTCAGT	23	3	−4.5	0.003	11004	Phosphofructokinase, muscle (Pfkm)	Met
AATCGAAGGC	8	0	−4.7	0.018	None	Novel tag	Nov
ATACCCATAA	8	0	−4.7	0.018	None	Novel tag	Nov
ATTTATTACA	8	0	−4.7	0.018	None	Novel tag	Nov
CCCTGAGTCC	24	3	−4.7	0.002	69	Ribosomal protein L14	Pro
CGCAAGGCCC	8	0	−4.7	0.018	49290	ESTs, highly similar to HTGN29 protein [H. sapiens]	EST
GACACCTTGA	8	0	−4.7	0.018	35336	ESTs, weakly similar to hypothetical protein FLJ20559[H. sapiens]	EST
GATCCCCCCA	8	0	−4.7	0.018	None	Novel tag	Nov
GCACTAGCTG	8	0	−4.7	0.018	37311	Progesterone receptor membrane component 1 (Pgrmc1)	Rec
GCGCCCTGAG	8	0	−4.7	0.018	None	Novel tag	Nov
ACTCGGATGC	33	4	−4.9	2.8E-04	18013	ESTs, moderately similar to RIKEN cDNA 1810011O01[R. norvegicus]	EST
AGCACAGTTA	33	4	−4.9	2.8E-04	13092	ESTs, weakly similar to RET1_RAT retinol-binding protein I; cellular (Cellular retinol-binding protein; CRBP)[R. norvegicus]	EST
AGGTCCACCA	9	0	−5.3	0.011	None	Novel tag	Nov
ATACTGCACT	18	2	−5.3	0.006	None	Novel tag	Nov
ATACTGCCAC	9	0	−5.3	0.011	None	Novel tag	Nov
GCCTCCAAGA	9	0	−5.3	0.011	1002	Plasminogen activator, tissue (Plat)	CS/ECM
GCGGTGGCAG	9	0	−5.3	0.011	38807	ESTs	EST
GGTGAGGTTT	9	0	−5.3	0.011	3870	ESTs, moderately similar to hypothetical protein MNCb-0169 [M. musculus]	EST
TGAGGTCTTG	9	0	−5.3	0.011	8476	ESTs	EST
CAGCTCTGGG	10	1	−5.9	0.036	15228	ESTs, highly similar to toll-associated serine protease[M. musculus]	EST
CTCCGAGAGG	10	1	−5.9	0.036	4287	ESTs, weakly similar to A43932 mucin 2 precursor, intestinal	EST
CTCTCTAAAC	10	1	−5.9	0.036	947	ESTs, weakly similar to B26423 serin proteinase inhibitor 2.2	EST
GATGTGACCA	10	1	−5.9	0.036	125	ESTs, highly similar to eukaryotic translation initiation factor 3, subunit 2 (beta, 36kD); TGF-beta receptor binding protein; DNA segment, Chr 4, ERATO Doi 632, expressed [M. musculus]	EST
TAGGTACAGC	10	0	−5.9	0.007	None	Novel tag	Nov
TCTGAATCTT	10	0	−5.9	0.007	None	Novel tag	Nov
TTATTTGGTG	10	1	−5.9	0.036	34356	ThyM.cell surface antigen	Imm, CS/ECM
AAAACACATA	11	1	−6.5	0.024	None	Novel tag	Nov
GCCTCTGCTG	11	1	−6.5	0.024	3777	ESTs, moderately similar to RIKEN cDNA 2310040G17; expressed sequence AI425883 [M. musculus]	EST
AATGCCCCCC	71	6	−7.0	2.7E-09	None	Novel tag	Nov
AAACTGACAC	12	1	−7.1	0.016	None	Novel tag	Nov
CTGAAAAAAA	12	1	−7.1	0.016	81140	Thyroid hormone responsive protein (spot14) (Thrsp)	Met, Lip
					6510	ESTs, highly similar to mannosidase, beta A, lysosomal-like[H. sapiens]
GCTCCCAGGG	12	1	−7.1	0.016	23659	ESTs	EST
GGGTCAACTG	12	0	−7.1	0.003	4083	S-100 related protein, clone 42C (S100a10)	Sig, C/D
TTTGGGAGAA	12	0	−7.1	0.003	None	Novel tag	Nov
CTGGAGCATC	13	1	−7.7	0.011	68036	ESTs	EST
ATAACCCATA	14	1	−8.3	0.007	None	Novel tag	Nov
ATACCACATA	14	1	−8.3	0.007	None	Novel tag	Nov
ATATTGACAC	15	1	−8.9	0.005	None	Novel tag	Nov
ATAACACAAT	16	0	−9.5	4.4E-04	None	Novel tag	Nov
CCCTGAGCGG	17	1	−10.1	0.002	2514	Transferrin (Tf)	Rec
ATACTTGACA	61	3	−12.0	8.9E-10	25364	ESTs	EST

Table 2.

 

Differentially Regulated Transcripts in Monocularly Deprived Monkey Lateral Rectus

GenBank Acc. No.	Gene Name	Symbol	Change (×)	Function
NM_018563	Hypothetical protein PR00758	PRO0758	48.5	Unknown
AF094508	Dentin sialophosphoprotein	DSPP	4.1	Extracellular matrix
NM_015978	Putative protein-tyrosine kinase	LOC51086	2.7	Unknown
NM_025012	Hypothetical protein FLJ13769;	FLJ13769	2.4	Unknown
BG252666	ATPase, Class I, type 8B; member 1	ATP8B1	2.3	Aminophospholipid transport
NM_000393	Collagen, type V, alpha 2	COL5A2	−1.8	Extracellular matrix
AU144167	Collagen, type III, alpha 1 (Ehlers-Danlos syndrome type IV; autosomal dominant)	COL3A1	−1.8	Extracellular matrix
NM_002048	Growth arrest-specific 1	GAS1	−1.9	Cell proliferation and growth
NM_000089	Collagen, type I, alpha 2	COL1A2	−1.9	Extracellular matrix
M97935	Signal transducer and activator of transcription 1, 91kDa	STAT1	−2.0	Signal transduction; cell proliferation and growth
NM_000088	Collagen, type I, alpha 1	COL1A1	−2.0	Extracellular matrix
AF138300	Decorin	DCN	−2.1	Extracellular matrix
AF130082	Hypothetical protein PRO3121	PRO3121	−2.1	Unknown
AV753392	Heterogeneous nuclear ribonucleoprotein	HNRPH1	−2.2	RNA processing
AF138302	Decorin	DCN	−2.3	Extracellular matrix
BF514079	cDNA FLJ38575 fis, clone HCHON2007046, mRNA sequence	—	−2.4	Unknown
NM_002065	Glutamate-ammonia ligase (glutamine synthase)	GLUL	−2.7	Amino acid metabolism and NO biosynthesis
A1813758	Collagen, type III, alpha 1 (Ehlers-Danlos syndrome type IV, autosomal dominant)	COL3A1	−2.7	Extracellular matrix
NM_002612	Pyruvate dehydrogenase kinase, isoenzyme 4	PDK4	−4.7	Glycolysis
NM_022831	Hypothetical protein FLJ12806	FLJ12806	−7.9	Unknown

Table 3.

 

Differentially Regulated Transcripts in Monocularly Deprived Monkey Medial Rectus

GenBank Acc. No.	Gene Name	Symbol	Change (×)	Function
L12711	Transketolase (Wernicke-Korsakoff syndrome)	TKT	7.1	Pentose phosphate pathway
U63131	CDC37 cell division cycle 37 homologue S. cerevisiae)	CDC37	3.6	Cell cycle regulation; cell proliferation and growth
J03068	N-acylaminoacyl-peptide hydrolase	APEH	2.8	Amino acid metabolism
NM_000703	ATPase, Na+/K+ transporting, alpha 3 polypeptide	ATP1A3	2.8	Cation transport
NM_001540	Heat shock 27kD protein 1	HSPB1	2.7	Protein modification; stress response
NM_002541	Oxoglutarate (alpha-ketoglutarate) dehydrogenase (lipoamide)	OGDH	2.3	Energy metabolism
NM_002869	RAB6A, member RAS oncogene family	RAB6A	2.3	GTPase; membrane transport
AF172268	KIAA0551 protein	KIAA0551	2.1	Unknown
NM_002397	MADS box transcription enhancer factor 2, polypeptide C (myocyte enhancer factor 2C)	MEF2C	1.9	Myogenesis
NM_005729	Peptidylprolyl isomerase F (cyclophilin F)	PPIF	1.9	Protein modification
NM_005952	Metallothionein IX	MTIX	−1.8	Heavy metal binding
BC001971	Cyclin-dependent kinase inhibitor 1B (p27, Kip1)	CDKN1B	−1.8	Cell cycle regulation; cell proliferation and growth
NM_001186	BTB and CNC homology 1, basic leucine zipper transcription factor 1	BACH1	−1.8	Transcription
AL161952	Glutamate-ammonia ligase (glutamine synthase)	GLUL	−1.9	Amino acid metabolism
NM_001855	Collagen, type XV, alpha 1	COL15A1	−1.9	Extracellular matrix
M83772	Flavin-containing monooxygenase 3	FMO3	−1.9	Microsome
BF791738	Hypothetical protein PRO2751	PRO2751	−2.0	Unknown
AF153330	Solute carrier family 19 (thiamin transporter), member 2	SLC19A2	−2.1	Transporter
NM_000161	GTP cyclohydrolase 1 (dopa-responsive dystonia)	GCH1	−2.1	Amino acid metabolism; biosynthesis of NO
BF514079	cDNA FLJ38575 fis, clone HCHON2007046, mRNA sequence	—	−2.4	Unknown

The authors thank Xiaohua Zhou, Sriram Kasturi, Anita Merriam, Patrick Leahy, Francisco Andrade, and Bendi Gong for technical support.