Remote Diagnosis on Upper Respiratory Tract Infections Based on a Neural Network with Few Symptom Words-A Feasibility Study.

This study aims explore the feasibility of using neural network (NNs) and deep learning to diagnose three common respiratory diseases with few symptom words. These three diseases are nasopharyngitis, upper respiratory infection, and bronchitis/bronchiolitis. Through natural language processing, the symptom word vectors are encoded by GPT-2 and classified by the last linear layer of the NN. The experimental results are promising, showing that this model achieves a high performance in predicting all three diseases. They revealed 90% accuracy, which suggests the implications of the developed model, highlighting its potential use in assisting patients' understanding of their conditions via a remote diagnosis. Unlike previous studies that have focused on extracting various categories of information from medical records, this study directly extracts sequential features from unstructured text data, reducing the effort required for data pre-processing.

Bayesian Surprise Predicts Human Event Segmentation in Story Listening.

Event segmentation theory posits that people segment continuous experience into discrete events and that event boundaries occur when there are large transient increases in prediction error. Here, we set out to test this theory in the context of story listening, by using a deep learning language model (GPT-2) to compute the predicted probability distribution of the next word, at each point in the story. For three stories, we used the probability distributions generated by GPT-2 to compute the time series of prediction error. We also asked participants to listen to these stories while marking event boundaries. We used regression models to relate the GPT-2 measures to the human segmentation data. We found that event boundaries are associated with transient increases in Bayesian surprise but not with a simpler measure of prediction error (surprisal) that tracks, for each word in the story, how strongly that word was predicted at the previous time point. These results support the hypothesis that prediction error serves as a control mechanism governing event segmentation and point to important differences between operational definitions of prediction error.

System analysis based on glutamine catabolic-related enzymes identifies GPT2 as a novel immunotherapy target for lung adenocarcinoma.

BACKGROUND: In recent years, targeting glutamine metabolism has gained attention as a promising therapeutic approach. Glutamine catabolic-related enzymes play a crucial role in modulating glutamine metabolism and influencing immune responses in the tumor immune microenvironment (TME). However, current literature on the function of glutamine catabolic enzymes in lung adenocarcinoma (LUAD) is limited. METHODS: We validated the glutamine dependency of LUAD cells in vitro, followed by transcriptome data to identify differentially expressed genes (DEGs), with transcriptome and single-cell data analysis utilized to explore the role of such genes within the tumor immune microenvironment. We performed employed subcutaneous injection of lewis lung carcinoma cells in C57BL/6 mice to confirm the role of candidate genes in tumor growth and anti-tumor immunity. RESULTS: Our study revealed that glutamine is essential for the growth of LUAD cells. Subsequently, we identified four DEGs - glutamate pyruvate transaminase 1 (GPT1), glutamate pyruvate transaminase 2 (GPT2), glutamic-oxaloacetic transaminase 1 (GOT1), and glutamic-oxaloacetic transaminase 2 (GOT2) - in LUAD patients, which were highly expressed in tumor tissue and associated with an immunosuppressive TME. Single-cell sequencing analysis detected high expression levels of GOT1 and GOT2 in immune and stromal cell subpopulations, while GPT1 and GPT2 showed relatively lower expression. Based on the lower immune score and lower expression in immune and stromal cells, we validated the role of GPT2 in vivo for modulating the TME and tumor growth. Inhibition of GPT2 resulted in suppressed tumor growth and increased the expression of CD4 and CD8. Additionally, GPT2 inhibitors induced a stronger antitumor immunity when used in combination with anti-programmed cell death ligand 1. CONCLUSION: This study is the first to show the critical role of glutamine catabolic-related enzymes in the TME, and identified GPT2 as a promising therapeutic target for inhibiting tumor growth and improving anti-tumour immune responses for LUAD. Additional studies will be required to define the roles glutamine catabolic-related enzymes play in LUAD.

Expectations modulate retrieval interference during ellipsis resolution.

Memory operations during language comprehension are subject to interference: retrieval is harder when items are linguistically similar to each other. We test how such interference effects might be modulated by linguistic expectations. Theories differ in how these factors might interact; we consider three possibilities: (i) predictability determines the need for retrieval, (ii) predictability affects cue-preference during retrieval, or (iii) word predictability moderates the effect of noise in memory during retrieval. We first demonstrate that expectations for a target word modulate retrieval interference in Mandarin noun-phrase ellipsis in an electroencephalography (EEG) experiment. This result obtains in globally ungrammatical sentences - termed "facilitatory interference." Such a pattern is inconsistent with theories that focus only on the need for retrieval. To tease apart cue-preferences from noisy-memory representations, we operationalize the latter using a Transformer neural network language model. Confronting the model with our stimuli reveals an interference effect, consistent with prior work, but that effect does not interact with predictability in contrast to human EEG results. Together, these data are most consistent with the hypothesis that the predictability of target items affects cue-preferences during retrieval.

Exosomal GPT2 derived from triple-negative breast cancer cells promotes metastasis by activating BTRC.

BACKGROUND: There have been reports of increased glutamate pyruvate transaminase 2 (GPT2) expression in certain cancers including breast cancer. Although the role of GPT2 as a metabolic enzyme is well understood in breast cancer progression, little is known about the other roles of GPT2, especially exosomal GPT2. METHODS: BT549 and BT474 Cells were cultured and their exosomes were isolated by using ultracentrifugation. Cells migrated through the membrane were stained with crystal violet, and then were observed by microscope. Total RNA was extracted from culture cells and transcribed into cDNA, quantitative real-time RT-PCR was used to detect mRNA expression of ICAM1, VCAM1, and MMP9 using SYBR Green qPCR Mix with a 7500 Fast Real-time PCR system. Western blot was used to detect the gene expression of p-lkBa and TSG101 and GPT2 in breast cancer cells. Immunohistochemistry was used to detect the protein expression of GPT2 and BTRC in cancer cells, animal models loaded with metastasis breast cancer cells were established via tail vein injections. Interaction between GPT2 and BTRC in breast cancer cells was investigated via Co-immunoprecipitation. RESULTS: GPT2 was up-regulated in TNBC. Exosomes were isolated effectively from TNBC cells, and confirmed that GPT2 was overexpressed inexosomes. QRT-PCR showed that mRNA expression levels of ICAM1, VCAM1, and MMP9 in TNBC were high. Exosomal GPT2 derived from TNBC enhanced migration and invasion of breast cancer via in vitro cell experiment and in vivo animal model experiment. Exosomal GPT2 binds with BTRC to degrade p-lkBa, and improved metastasis of breast cancer cells. CONCLUSION: We demonstrated that GPT2 was upregulated in TNBC as well as in exosomes derived from triple-negative breast cancer (TNBC) cells. GPT2 expression was associated with the malignancy of breast cancer and promoted metastasis of breast cancer cells. Moreover, exosomal GPT2 derived from TNBC cells was verified to increase the capacity of breast cancer cells to metastasize through activating beta-transducin repeat containing E3 ubiquitin protein ligase (BTRC). This suggested that exosomal GPT2 may be useful for breast cancer patients as a potential biomarker and treatment target.

Theory-Driven Analysis of Natural Language Processing Measures of Thought Disorder Using Generative Language Modeling.

BACKGROUND: Natural language processing (NLP) holds promise to transform psychiatric research and practice. A pertinent example is the success of NLP in the automatic detection of speech disorganization in formal thought disorder (FTD). However, we lack an understanding of precisely what common NLP metrics measure and how they relate to theoretical accounts of FTD. We propose tackling these questions by using deep generative language models to simulate FTD-like narratives by perturbing computational parameters instantiating theory-based mechanisms of FTD. METHODS: We simulated FTD-like narratives using Generative-Pretrained-Transformer-2 by either increasing word selection stochasticity or limiting the model's memory span. We then examined the sensitivity of common NLP measures of derailment (semantic distance between consecutive words or sentences) and tangentiality (how quickly meaning drifts away from the topic) in detecting and dissociating the 2 underlying impairments. RESULTS: Both parameters led to narratives characterized by greater semantic distance between consecutive sentences. Conversely, semantic distance between words was increased by increasing stochasticity, but decreased by limiting memory span. An NLP measure of tangentiality was uniquely predicted by limited memory span. The effects of limited memory span were nonmonotonic in that forgetting the global context resulted in sentences that were semantically closer to their local, intermediate context. Finally, different methods for encoding the meaning of sentences varied dramatically in performance. CONCLUSIONS: This work validates a simulation-based approach as a valuable tool for hypothesis generation and mechanistic analysis of NLP markers in psychiatry. To facilitate dissemination of this approach, we accompany the paper with a hands-on Python tutorial.

The delta subunit of the GABAA receptor is necessary for the GPT2-promoted breast cancer metastasis.

Objectives: Glutamic pyruvate transaminase (GPT2) catalyzes the reversible transamination between alanine and α-ketoglutarate (α-KG) to generate pyruvate and glutamate during cellular glutamine catabolism. The glutamate could be further converted to γ-aminobutyric acid (GABA). However, the role of GPT2 in tumor metastasis remains unclear. Methods: The wound healing and transwell assays were carried out to analyze breast cancer cell migration and invasion in vitro. Gene ontology analysis was utilized following RNA-sequencing to discover the associated molecule function. The mass spectrometry analysis following phosphoprotein enrichment was performed to discover the associated transcription factors. Most importantly, both the tail vein model and Mammary gland conditional Gpt2-/- spontaneous tumor mouse models were used to evaluate the effect of GPT2 on breast cancer metastasis in vivo. Results: GPT2 overexpression increases the content of GABA and promotes breast cancer metastasis by activating GABAA receptors. The delta subunit GABRD is necessary for the GPT2/GABA-induced breast cancer metastasis in xenograft and transgenic mouse models. Gpt2 knockout reduces the lung metastasis of the genetic Gpt2-/- breast cancer in mice and prolongs the overall survival of tumor burden mice. Mechanistically, GPT2-induced GABAA receptor activation increases Ca2+ influx by turning on its associated calcium channel, and the surged intracellular calcium triggers the PKC-CREB pathway activation. The activated transcription factor CREB accelerates breast cancer metastasis by upregulating metastasis-related gene expressions, such as PODXL, MMP3, and MMP9. Conclusion: In summary, this study demonstrates that GPT2 promotes breast cancer metastasis through up-regulated GABA activation of GABAAR-PKC-CREB signaling, suggesting it is a potential target for breast cancer therapy.

SPTBN1 abrogates renal clear cell carcinoma progression via glycolysis reprogramming in a GPT2-dependent manner.

BACKGROUND: Renal clear cell carcinoma (ccRCC) is the most prevalent tumors worldwide. Discovering effective biomarkers is essential to monitor the prognosis and provide alternative clinical options. SPTBN1 is implicated in various cancerous processes. However, its role in ccRCC remains unelucidated. This study intends to explore the biological function and mechanism of SPTBN1 in ccRCC. METHODS: Single-cell and bulk RNA-seq, tissue microarray, real-time quantitative PCR, and western blotting were applied to verify the expression and predictive value of SPTBN1 in ccRCC. Gain or loss of functional ccRCC cell line models were constructed, and in vitro and in vivo assays were performed to elucidate its tumorigenic phenotypes. Actinomycin D experiment, RNA immunoprecipitation (RIP), specific inhibitors, and rescue experiments were carried out to define the molecular mechanisms. RESULTS: SPTBN1 was down-regulated in ccRCC and knockdown of SPTBN1 displayed a remarkably oncogenic role both in vitro and in vivo; while overexpressing SPTBN1 reversed this effect. SPTBN1 mediated ccRCC progression via the pathway of glutamate pyruvate transaminase 2 (GPT2)-dependent glycolysis. The expression of GPT2 was significantly negatively correlated with that of SPTBN1. As an RNA binding protein SPTBN1, regulated the mRNA stability of GPT2. CONCLUSION: Our research demonstrated that SPTBN1 is significantly down-regulated in ccRCC. SPTBN1 knockdown promotes ccRCC progression via activating GPT2-dependent glycolysis. SPTBN1 may serve as a therapeutic target for the treatment of ccRCC.

Context-based facilitation of semantic access follows both logarithmic and linear functions of stimulus probability.

Stimuli are easier to process when context makes them predictable, but does context-based facilitation arise from preactivation of a limited set of relatively probable upcoming stimuli (with facilitation then linearly related to probability) or, instead, because the system maintains and updates a probability distribution across all items (with facilitation logarithmically related to probability)? We measured the N400, an index of semantic access, to words of varying probability, including unpredictable words. Word predictability was measured using both cloze probabilities and a state-of-the-art machine learning language model (GPT-2). We reanalyzed five datasets (n = 138) to demonstrate and then replicate that context-based facilitation on the N400 is graded, even among unpredictable words. Furthermore, we established that the relationship between word predictability and context-based facilitation combines linear and logarithmic functions. We argue that this composite function reveals properties of the mapping between words and semantic features and how feature- and word-related information is activated on-line.

Smartphone Sensor-Based Human Motion Characterization with Neural Stochastic Differential Equations and Transformer Model.

With many conveniences afforded by advances in smartphone technology, developing advanced data analysis methods for health-related information from smartphone users has become a fast-growing research topic in the healthcare field. Along these lines, this paper addresses smartphone sensor-based characterization of human motions with neural stochastic differential equations (NSDEs) and a Transformer model. NSDEs and modeling via Transformer networks are two of the most prominent deep learning-based modeling approaches, with significant performance yields in many applications. For the problem of modeling dynamical features, stochastic differential equations and deep neural networks are frequently used paradigms in science and engineering, respectively. Combining these two paradigms in one unified framework has drawn significant interest in the deep learning community, and NSDEs are among the leading technologies for combining these efforts. The use of attention has also become a widely adopted strategy in many deep learning applications, and a Transformer is a deep learning model that uses the mechanism of self-attention. This concept of a self-attention based Transformer was originally introduced for tasks of natural language processing (NLP), and due to its excellent performance and versatility, the scope of its applications is rapidly expanding. By utilizing the techniques of neural stochastic differential equations and a Transformer model along with data obtained from smartphone sensors, we present a deep learning method capable of efficiently characterizing human motions. For characterizing human motions, we encode the high-dimensional sequential data from smartphone sensors into latent variables in a low-dimensional latent space. The concept of the latent variable is particularly useful because it can not only carry condensed information concerning motion data, but also learn their low-dimensional representations. More precisely, we use neural stochastic differential equations for modeling transitions of human motion in a latent space, and rely on a Generative Pre-trained Transformer 2 (GPT2)-based Transformer model for approximating the intractable posterior of conditional latent variables. Our experiments show that the proposed method can yield promising results for the problem of characterizing human motion patterns and some related tasks including user identification.

The m6A reader IGF2BP2 regulates glutamine metabolism and represents a therapeutic target in acute myeloid leukemia.

N6-Methyladenosine (m6A) modification and its modulators play critical roles and show promise as therapeutic targets in human cancers, including acute myeloid leukemia (AML). IGF2BP2 was recently reported as an m6A binding protein that enhances mRNA stability and translation. However, its function in AML remains largely elusive. Here we report the oncogenic role and the therapeutic targeting of IGF2BP2 in AML. High expression of IGF2BP2 is observed in AML and associates with unfavorable prognosis. IGF2BP2 promotes AML development and self-renewal of leukemia stem/initiation cells by regulating expression of critical targets (e.g., MYC, GPT2, and SLC1A5) in the glutamine metabolism pathways in an m6A-dependent manner. Inhibiting IGF2BP2 with our recently identified small-molecule compound (CWI1-2) shows promising anti-leukemia effects in vitro and in vivo. Collectively, our results reveal a role of IGF2BP2 and m6A modification in amino acid metabolism and highlight the potential of targeting IGF2BP2 as a promising therapeutic strategy in AML.

GPT2 Is Induced by Hypoxia-Inducible Factor (HIF)-2 and Promotes Glioblastoma Growth.

Hypoxia-inducible factor (HIF) directly activates the transcription of metabolic enzymes in response to hypoxia to reprogram cellular metabolism required for tumor cell proliferation. Through analyzing glutamate-linked aminotransferases, we here identified glutamate pyruvate transaminase 2 (GPT2) as a direct HIF-2 target gene in human glioblastoma (GBM). Hypoxia upregulated GPT2 mRNA and protein levels in GBM cells, which required HIF-2 but not HIF-1. HIF-2 directly bound to the hypoxia response element of the human GPT2 gene, leading to its transcription in hypoxic GBM cells. GPT2 located at the nucleus and mitochondria and reduced α-ketoglutarate levels in GBM cells. Genetic or pharmacological inhibition of GPT2 decreased GBM cell growth and migration under normoxia and hypoxia. Knockout of GPT2 inhibited GBM tumor growth in mice. Collectively, these findings uncover a hypoxia-inducible aminotransferase GPT2 required for GBM progression.

Paternal uniparental disomy of chromosome 16 resulting in homozygosity of a GPT2 mutation causes intellectual and developmental disability.

Recessive mutations in glutamate pyruvate transaminase 2 (GPT2) have recently been found to be associated with intellectual and developmental disability (IDD). In this study, we discovered a homozygous missense variant, NM_133443: [c.1172C > T, p. Pro391Leu], of GPT2 on chromosome 16 in a proband diagnosed with IDD through trio whole-exome sequencing (WES). The pathogenicity of the variant was further verified by bioinformatics analysis and functional studies in vitro. This autosomal recessive disease was caused by paternal uniparental disomy (UPD) which was further proven by single nucleotide polymorphism array (SNP array). In past literature, recessive diseases in chromosome 16 were usually due to maternal UPD where Mendel's law of inheritance was not applicable. However, in our case we found that paternal UPD can cause recessive diseases related to the GPT2 gene on chromosome 16. Our study provides an important line of evidence for the diagnosis of GPT2-related intellectual developmental disorders.

Loss of mitochondrial enzyme GPT2 causes early neurodegeneration in locus coeruleus.

Locus coeruleus (LC) is among the first brain areas to degenerate in Alzheimer's disease and Parkinson's disease; however, the underlying causes for the vulnerability of LC neurons are not well defined. Here we report a novel mechanism of degeneration of LC neurons caused by loss of the mitochondrial enzyme glutamate pyruvate transaminase 2 (GPT2). GPT2 Deficiency is a newly-recognized childhood neurometabolic disorder. The GPT2 enzyme regulates cell growth through replenishment of tricarboxylic acid (TCA) cycle intermediates and modulation of amino acid metabolism. In Gpt2-null mice, we observe an early loss of tyrosine hydroxylase (TH)-positive neurons in LC and reduced soma size at postnatal day 18. Gpt2-null LC shows selective positive Fluoro-Jade C staining. Neuron loss is accompanied by selective, prominent microgliosis and astrogliosis in LC. We observe reduced noradrenergic projections to and norepinephrine levels in hippocampus and spinal cord. Whole cell recordings in Gpt2-null LC slices show reduced soma size and abnormal action potentials with altered firing kinetics. Strikingly, we observe early decreases in phosphorylated S6 in Gpt2-null LC, preceding prominent p62 aggregation, increased LC3B-II to LC3B-I ratio, and neuronal loss. These data are consistent with a possible mechanism involving deficiency in protein synthesis and cell growth, associated subsequently with abnormal autophagy and neurodegeneration. As compared to the few genetic animal models with LC degeneration, loss of LC neurons in Gpt2-null mice is developmentally the earliest. Early neuron loss in LC in a model of human neurometabolic disease provides important clues regarding the metabolic vulnerability of LC and may lead to new therapeutic targets.

Prediction as a basis for skilled reading: insights from modern language models.

Reading is not an inborn human capability, and yet, English-speaking adults read with impressive speed. This study considered how predictions of upcoming words impact on this skilled behaviour. We used a powerful language model (GPT-2) to derive predictions of upcoming words in text passages. These predictions were highly accurate and showed a tight relationship to fine-grained aspects of eye-movement behaviour when adults read those same passages, including whether to skip the next word and how long to spend on it. Strong predictions that were incorrect resulted in a prediction error cost on fixation durations. Our findings suggest that predictions for upcoming words can be made based on the analysis of text statistics and that these predictions guide how our eyes interrogate text at very short timescales. These findings open new perspectives on reading and language comprehension and illustrate the capability of modern language models to inform understanding of human language processing.

Silencing alanine transaminase 2 in diabetic liver attenuates hyperglycemia by reducing gluconeogenesis from amino acids.

Hepatic gluconeogenesis from amino acids contributes significantly to diabetic hyperglycemia, but the molecular mechanisms involved are incompletely understood. Alanine transaminases (ALT1 and ALT2) catalyze the interconversion of alanine and pyruvate, which is required for gluconeogenesis from alanine. We find that ALT2 is overexpressed in the liver of diet-induced obese and db/db mice and that the expression of the gene encoding ALT2 (GPT2) is downregulated following bariatric surgery in people with obesity. The increased hepatic expression of Gpt2 in db/db liver is mediated by activating transcription factor 4, an endoplasmic reticulum stress-activated transcription factor. Hepatocyte-specific knockout of Gpt2 attenuates incorporation of 13C-alanine into newly synthesized glucose by hepatocytes. In vivo Gpt2 knockdown or knockout in liver has no effect on glucose concentrations in lean mice, but Gpt2 suppression alleviates hyperglycemia in db/db mice. These data suggest that ALT2 plays a significant role in hepatic gluconeogenesis from amino acids in diabetes.

Thyroid hormone regulates glutamine metabolism and anaplerotic fluxes by inducing mitochondrial glutamate aminotransferase GPT2.

Thyroid hormones (THs) are key metabolic regulators coordinating short- and long-term energy needs. In skeletal muscle, THs modulate energy metabolism in pathophysiological conditions. Indeed, hypo- and hyperthyroidism are leading causes of muscle weakness and strength; however, the metabolic pathways underlying these effects are still poorly understood. Using molecular, biochemical, and isotope-tracing approaches combined with mass spectrometry and denervation experiments, we find that THs regulate glutamine metabolism and anaplerotic fluxes by up-regulating the glutamate pyruvate transaminase 2 (GPT2) gene. In humans, GPT2 autosomal recessive mutations cause a neurological syndrome characterized by intellectual disability, microcephaly, and progressive motor symptoms. Here, we demonstrate a role of the TH/GPT2 axis in skeletal muscle in which it regulates muscle weight and fiber diameter in resting and atrophic conditions and results in protection from muscle loss during atrophy. These results describe an anabolic route by which THs rewire glutamine metabolism toward the maintenance of muscle mass.

Abrogating GPT2 in triple-negative breast cancer inhibits tumor growth and promotes autophagy.

Uncontrolled proliferation and altered metabolic reprogramming are hallmarks of cancer. Active glycolysis and glutaminolysis are characteristic features of these hallmarks and required for tumorigenesis. A fine balance between cancer metabolism and autophagy is a prerequisite of homeostasis within cancer cells. Here we show that glutamate pyruvate transaminase 2 (GPT2), which serves as a pivot between glycolysis and glutaminolysis, is highly upregulated in aggressive breast cancers, particularly the triple-negative breast cancer subtype. Abrogation of this enzyme results in decreased tricarboxylic acid cycle intermediates, which promotes the rewiring of glucose carbon atoms and alterations in nutrient levels. Concordantly, loss of GPT2 results in an impairment of mechanistic target of rapamycin complex 1 activity as well as the induction of autophagy. Furthermore, in vivo xenograft studies have shown that autophagy induction correlates with decreased tumor growth and that markers of induced autophagy correlate with low GPT2 levels in patient samples. Taken together, these findings indicate that cancer cells have a close network between metabolic and nutrient sensing pathways necessary to sustain tumorigenesis and that aminotransferase reactions play an important role in maintaining this balance.

A novel missense variant in GPT2 causes non-syndromic autosomal recessive intellectual disability in a consanguineous Iranian family.

Intellectual disability (ID) affects 1-3% of the general population worldwide. Genetic factors play an undeniable role in the etiology of Non-Syndromic Intellectual disability (NS-ID). Nowadays, whole-exome sequencing (WES) technique is used frequently to identify the causative genes in such heterogeneous diseases. Herein, we subjected four patients with initial diagnostics of NS-ID in a consanguineous Iranian family. To find the possible genetic cause(s), Trio-WES was performed on the proband and his both healthy parents. Sanger sequencing was performed to confirm the identified variant by WES and also investigate whether it co-segregates with the patients' phenotype in the family. Using several online in-silico predictors, the probable impacts of the variant on structure and function of GPT2 protein were predicted. A novel variant, c.266A>G; p.(Glu89Gly), in exon 3 of GPT2 (NM_133443.3) was identified using Trio-WES. The candidate variant was also verified by Sanger sequencing. All affected members showed the common clinical features suffering from a non-progressive mild-to-severe ID. Also, different clinical observations compared to previously reported cases such as no facial features, no obvious structural malformations, ability to speak but with difficulty, and lack of any morphological defects were noted for the first time in this family. The c.266A>G; p.(Glu89Gly) variant reported here is the sixth variant identified up to now in the GPT2 gene, to be associated with NS-ID. Our data support the potential malfunction of the substituted GPT2 protein resulted from the novel variant, however, we strongly suggest confirming this finding more by doing functional analysis.

Transcriptional Regulation of the Glucose-6-Phosphate/Phosphate Translocator 2 Is Related to Carbon Exchange Across the Chloroplast Envelope.

The exchange of reduced carbon across the inner chloroplast envelope has a large impact on photosynthesis and growth. Under steady-state conditions it is thought that glucose 6-phosphate (G6P) does not cross the chloroplast membrane. However, growth at high CO2, or disruption of starch metabolism can result in the GPT2 gene for a G6P/Pi translocator to be expressed presumably allowing G6P exchange across the chloroplast envelope. We found that after an increase in light, the transcript for GPT2 transiently increases several 100-fold within 2 h in both the Col-0 and WS ecotypes of Arabidopsis thaliana. The increase in transcript for GPT2 is preceded by an increase in transcript for many transcription factors including Redox Responsive Transcription Factor 1 (RRTF1). The increase in GPT2 transcript after exposure to high light is suppressed in a mutant lacking the RRTF1 transcription factor. The GPT2 response was also suppressed in a mutant with a T-DNA insert in the gene for the triose-phosphate/Pi translocator (TPT). However, plants lacking TPT still had a robust rise in RRTF1 transcript in response to high light. From this, we conclude that both RRTF1 (and possibly other transcription factors) and high amounts of cytosolic triose phosphate are required for induction of the expression of GPT2. We hypothesize that transient GPT2 expression and subsequent translation is adaptive, allowing G6P to move into the chloroplast from the cytosol. The imported G6P can be used for starch synthesis or may flow directly into the Calvin-Benson cycle via an alternative pathway (the G6P shunt), which could be important for regulating and stabilizing photosynthetic electron transport and carbon metabolism.