Rosa Rubicondior: Unintelligent Design - How a Design Blunder Causes Severe Mental Health Problems

Universität Leipzig: Changes in a single gene can cause mental illness

A recent study from the University of Leipzig, just published, open access, in Molecular Psychiatry highlights the difference between an intelligently designed system and one which evolved naturally. Change a single gene involved in neurotransmission and the human feature that creationists wave as evidence for intelligent design - the human brain - seriously malfunctions.

The gene, GRIN2A, encodes a key subunit of the NMDA receptor — a molecular gateway through the cell membrane of neurones essential for learning, memory, language development, and the ability of the brain to fine-tune its own wiring. When functioning normally, children learn to speak, form memories, and develop the balanced neural circuits that underpin thought and behaviour. When it doesn’t, the result can be epileptic seizures, speech loss, cognitive impairment, and an increased vulnerability to psychiatric illness. In some cases, even sleep becomes a time of neurological storm activity, with continuous spike-wave patterns eroding normal brain development.

For anyone who understands evolution, this fragility makes perfect sense. For those insisting that the human brain is the product of foresight and planning, it presents a serious problem. It is a system built by evolutionary tinkering, not design. The NMDA receptor is one of the pillars of excitatory communication in the brain. Yet it is also a precarious, expensive and failure-prone piece of biological machinery. A single amino acid substitution in the GRIN2A protein can derail synaptic signalling, scramble brain rhythms, or impair the processes that enable children to acquire language.

This is not what robust design looks like.

GRIN2A gene - normal function.

GRIN2A encodes the protein GluN2A (also called NR2A), which is a subunit of the NMDA receptor (N-methyl-D-aspartate receptor), a type of glutamate-gated ion channel in the brain. [1.1]
NMDA receptors that include GluN2A are both ligand-gated and voltage-dependent channels. They open in response to glutamate (and co-agonists) binding + appropriate membrane potential, allowing positively charged ions (cations, including calcium) to flow into neurons. [2.1]
This ion flow underlies excitatory synaptic transmission in the central nervous system, and is essential for synaptic plasticity, including mechanisms like long-term potentiation (LTP) — the kind of neural adaptation widely thought to support memory formation, learning and other higher cognitive functions. [3.1]
More broadly, GRIN2A (via GluN2A-containing NMDA receptors) plays a critical role in brain development, maturation of neural circuits, neuronal excitability regulation, and the capacity for neurons to change their connectivity in response to activity or experience. [4.1]

So in short: GRIN2A is fundamental for the “wiring and rewiring” of the brain — enabling excitatory signalling, synaptic plasticity, learning, memory, and proper development of neural networks.

What goes wrong when GRIN2A is mutated?

What GRIN2A mutations do

Because GRIN2A encodes a key subunit of the NMDA receptor, mutations typically disrupt some combination of the following:

Impaired synaptic signalling

Mutations can:

Reduce the receptor’s ability to respond to glutamate.
Alter calcium influx that is essential for normal neuronal development and communication.
Cause either hypofunction (weaker signalling) or gain-of-function hyperexcitability, depending on the specific variant.

Either way, the finely balanced excitatory signalling required for cognition is thrown off.

Disrupted synaptic plasticity

Faulty GluN2A-containing receptors undermine:

Long-term potentiation (LTP)
Learning and memory
Neuronal circuit refinement during development

This is partly why GRIN2A mutations are linked with developmental delays and cognitive impairment.

Epilepsy and abnormal brain rhythms

The altered excitability of neurons predisposes individuals to:

Epilepsy–aphasia spectrum disorders
Rolandic seizures
Continuous spike-wave during sleep (CSWS)

This is one of the most robust clinical signatures of GRIN2A dysfunction.

Broader neuropsychiatric conditions

Increasing evidence links GRIN2A variants with:

Schizophrenia
Autism spectrum traits
ADHD-like symptoms
Speech and language disorders

The Leipzig article points specifically to psychiatric vulnerability, which fits the mechanistic role of NMDA receptors in cognition, mood regulation, and perception.

In truth it bears all the hallmarks of bad, unintelligent design that I highlighted in my book, The Unintelligent Designer: Refuting the Intelligent Design Hoax being unnecessarily complex, wasteful and prone to errors. Intelligently designed systems are built to tolerate faults; evolution has no such luxury. It works with what is already there, building complexity on top of older structures, often to compensate for error-prone compromises, and accepting fragility as the price of incremental adaptation. The NMDA receptor, with its multiple interacting components and fine balance of voltage- and ligand-dependent behaviour, is exactly the sort of clumsy compromise we expect from a process that cannot start again even when the architecture becomes unwieldy. An intelligent designer would scrap a poor design and start again; evolution has a make-do-and-mend approach.

Creationists often describe biological systems as “irreducibly complex”. GRIN2A reminds us that they can also be irreducibly fragile.

One striking feature of GRIN2A is how mutation-sensitive it is. Variants cluster in the receptor’s binding and channel-forming domains — precisely where the physics of ion flow makes the molecule most vulnerable to disruption. This pattern is textbook evolutionary biology: natural selection shapes function but cannot prevent mutations, and it cannot redesign the system wholesale to make it fault-tolerant.

If a designer had set out to build a reliable cognitive system, why place so much of human language development and neural plasticity at the mercy of a single, delicate subunit? Why allow tiny, predictable changes to produce catastrophic developmental outcomes? Why design a learning system that can collapse because of a routine copying error?

These questions have no coherent answer within Intelligent Design, and the repugnant habit of blaming 'sin' thereby implying that victims of biological events utterly beyond their control are somehow to blame for their own condition, avoid the question while betraying the casual inhumanity of fundamentalist religions.

Within evolution, they are entirely expected.

GRIN2A-related disorders do more than embarrass design arguments; they affect real children and families. A gene essential for language, memory and balanced neural activity can malfunction in ways that steal these abilities away. This isn’t the hallmark of a carefully engineered system. It is the hallmark of a system that works well enough to persist across generations — but no better.

Evolution does not produce perfection; it produces populations that function adequately in their ancestral environments. The vulnerabilities exposed by GRIN2A mutations are simply the price of that history.

The Leipzig findings reinforce several points that Intelligent Design cannot address:

Critical systems should not be so failure-prone if they were intentionally built.
Mutation patterns fit evolutionary constraint, not independent foresight.
The architecture of the NMDA receptor is a bricolage of compromises, not a streamlined design.
Real human disease burden arises naturally from ordinary genetic variation, not from intentional “design choices”.

In short, GRIN2A represents what the human brain actually is: an evolved system, powerful but precarious, astonishing but imperfect, brilliant but flawed in ways no competent designer would choose. The more we learn about genes like this, the harder it becomes to claim that our biology is the result of anything but evolution’s slow, constrained, and often inelegant but certainly unintelligent, handiwork. The Leipzig team's work is described in a Universität Leipzig press release:

Changes in a single gene can cause mental illness
Until now, researchers assumed that schizophrenia, anxiety disorders or depression arise from an interplay of many different factors, including genetic ones. An international study led by the Institute of Human Genetics at the University of Leipzig Medical Center has now demonstrated for the first time that changes in a single gene can in fact cause a mental illness. The new research has just been published in the renowned journal Molecular Psychiatry.
According to the World Health Organization (WHO), in 2021 almost one in seven people worldwide was living with a mental illness, with anxiety disorders and depression the most common. Mental disorders typically have complex causes with a major genetic component. Having a close family member who is affected is regarded as one of the greatest known risk factors. Previous studies assumed that mental disorders arise against the background of numerous genetic factors.

Our current findings indicate that GRIN2A is the first known gene that, on its own, can cause a mental illness. This distinguishes it from the polygenic causes of such disorders that have been assumed to date.

Professor Johannes R. Lemke, co-lead author.
Institute of Human Genetics
University of Leipzig Medical Center
Leipzig, Germany.
In the present study, data from 121 individuals with a genetic alteration in the GRIN2A gene were analysed statistically.

We were able to show that certain variants of this gene are associated not only with schizophrenia but also with other mental illnesses. What is striking is that, in the context of a GRIN2A alteration, these disorders already appear in childhood or adolescence – in contrast to the more typical manifestation in adulthood.

Professor Johannes R. Lemke.

What the human geneticist and his research team found particularly noteworthy in the study’s results was that some affected individuals showed exclusively psychiatric symptoms. GRIN2A alterations are otherwise typically associated with conditions like epilepsy or intellectual disability.

The GRIN2A gene plays a central role in regulating the electrical excitability of nerve cells. In the present study, certain variants led to reduced activity of the NMDA receptor, a key molecule in signal transmission in the brain. Together with Dr Steffen Syrbe, Professor at the Heidelberg Medical Faculty and paediatric neurologist at Heidelberg University Hospital, the clinicians showed that this aspect could also be therapeutically relevant: in an initial treatment series, patients showed marked improvements in their psychiatric symptoms following therapy with L-serine – a dietary supplement that activates the NMDA receptor.

Professors Johannes Lemke and Steffen Syrbe have been working together for almost 15 years, in both research and clinical practice, on disorders of the glutamate receptor in the brain in children with neurological diseases. During this time, Professor Lemke has established an international registry that comprises the world’s largest cohort of GRIN2A patients, which formed the basis for the current publication.

Publication:
Lemke, J.R., Eoli, A., Krey, I. et al.
GRIN2A null variants confer a high risk for early-onset schizophrenia and other mental disorders and potentially enable precision therapy.
Mol Psychiatry (2025). https://doi.org/10.1038/s41380-025-03279-4

Abstract
Rare genetic factors have been shown to substantially contribute to mental illness, but so far, no precision treatments for mental disorders have been described. It was recently identified that rare variants in GRIN2A encoding the GluN2A subunit of the N-methyl-D-aspartate receptor (NMDAR) confer a substantial risk for schizophrenia. To determine the prevalence of mental disorders among individuals with GRIN2A-related disorders, we enquired the presence of psychiatric symptoms in 235 individuals with pathogenic variants in GRIN2A who had previously enrolled in our global GRIN registry. We identified null variants in GRIN2A (GRIN2A_null) to be significantly associated with a broad spectrum of mental disorders including schizophrenia compared to a longitudinal population cohort (FinRegistry) as well as missense variants (GRIN2A_missense). In our cohort, GRIN2A_null-related mental disorders manifest in early childhood or adolescence, which is substantially earlier than the average adult onset in the general population. In 68% of co-incident epilepsy and mental disorder, mental disorders start after epilepsy offset and the age of epilepsy offset correlated with mental disorder onset. GRIN2A_null-related phenotypes appear to occasionally even manifest as isolated mental disorder, i.e. as schizophrenia or mood disorder without further GRIN2A-specific symptoms, such as intellectual disability and/or epilepsy. As L-serine is known to mediate co-agonistic effects on the NMDAR, we applied it to four individuals with GRIN2A_null-related mental disorders, all of whom experienced improvements of their neuropsychiatric phenotype. GRIN2A_null appears to be the first monogenic cause of early-onset and even isolated mental disorders, such as early-onset schizophrenia. Genetic testing should be considered in the diagnostic work-up of affected individuals to improve diagnosis and potentially offer personalized treatment as increasing brain concentrations of NMDAR co-agonists appears to be a promising precision treatment approach successfully targeting deficient glutamatergic signaling in individuals with mental disorders, i.e. due to GRIN2A_null. Introduction
Mental disorders pose a substantial burden to the affected individual and society, but treatment advances have been limited by our incomplete understanding of their molecular etiology. [1] According to the World Health Organization, in 2019 one in every eight people was living with a mental disorder worldwide, such as anxiety, depression, eating disorder or schizophrenia (https://www.who.int/news-room/fact-sheets/detail/mental-disorders).

Mental disorders have complex causes with a substantial genetic contribution. Having an affected close family member has been considered one of the highest known risk factors. [2] Previous studies considered mental disorders to be highly polygenic [3] and identified thousands of common genetic variants, each contributing a small risk increment. [4,5,6] A shared pathophysiological etiology and the implication of similar biological pathways have been suggested due to a high degree of shared genetic effects across different types of mental disorders [7, 8] as well as overlapping clinical presentations.

Among mental disorders, schizophrenia was shown to have a particularly high degree of heritability. [4, 9] Rare protein-truncating null variants in numerous genes have been considered to contribute to the heritability of schizophrenia and other mental disorders. [10, 11] These typically cause a much larger increase in disease risk than individual common variants, but are present in a smaller proportion of affected individuals. [12] The reduced reproductive rate of individuals with schizophrenia reduces the transmission rate of any large heritable risk factors in the population. [13] Risk variants with strong effects thus usually arise de novo, not being present in either of the parents. Few genetic factors conferring considerable risk increments for schizophrenia have been identified – in particular copy number variants (CNV) involving large chromosomal segments usually covering multiple genes. [14,15,16] The majority of individuals with mental disorders due to such de novo CNV display a recognizable syndromic phenotype comprising intellectual disability (ID) and/or other syndrome-associated features, including dysmorphism and congenital malformations. [17] The most prevalent example is 22q11.2 deletion syndrome, where up to 24% of affected adults are known to develop schizophrenia [18], and vice versa, up to 1% of individuals with schizophrenia are described to have a 22q11.2 deletion. [19,20,21] Another CNV on 16p13.2 containing GRIN2A, a gene encoding the GluN2A subunit of the N-methyl-D-aspartate receptor (NMDAR) and previously associated with various epilepsy phenotypes [22,23], is one of the very few loci that is recurrently associated with various mental disorders, i.e. bipolar disorders [24] and schizophrenia [25]. Still, CNV analysis or even exome- or genome-sequencing is not part of the recommended routine diagnostic work-up in individuals with mental disorders (https://ispg.net/genetic-testing-statement/).

While rare de novo variants affecting individual genes have been identified to contribute to schizophrenia globally [26, 27], only a single gene, SETD1A, has been robustly associated with schizophrenia, previously. [28] This is, however, in the context of a syndromic and severe neurodevelopmental disorder as well as an adult-onset psychiatric condition. Isolated mental disorders, particularly non-syndromic psychiatric phenotypes, have not yet been associated with alterations of a single gene conferring a large risk with a Mendelian inheritance pattern. [29, 30]

A recent exome sequencing study identified rare coding variants in 10 genes conferring substantial risk for schizophrenia, including protein-truncating null variants in GRIN2A (GRIN2A_null; overall p-value 7×10⁻⁷, Odds Ratio = 24.1). [16] This is in agreement with our previous report of three individuals with GRIN2A_null from our GRIN registry (https://grin-portal.broadinstitute.org/) with either psychotic disorders (n = 2) or anxiety disorder (n = 1). [23] Our registry comprises 235 individuals with GRIN2A-related disorders that were not ascertained for mental disorders, and thus provides a unique opportunity to systematically investigate the incidence of mental disorders across the lifetime and its association with other symptoms of GRIN2A-related disorders, such as epilepsy. This is of particular relevance as the presence of epilepsy alone significantly increases the prevalence of mental illness with respective prevalence rates of up to ~35% for mood disorders, ~25% for anxiety disorders, ~5% for psychotic disorders (and 1.7% for schizophrenia in particular). [31,32,33]

The orally available non-essential amino acid L-serine mediates co-agonistic effects on the NMDAR in neurons via its enantiomer D-serine. [34] We recently reported on improved behavior, electroencephalography (EEG) and seizure frequency in individuals with GRIN2A_null treated with L-serine. [35] We therefore also collected retrospective observational data on treatment responses to L-serine in this cohort of individuals with GRIN2A_null-related mental disorders.

This study highlights just how precarious the human brain really is. The gene GRIN2A, which encodes a crucial component of the NMDA receptor, plays a central role in learning, memory, language development and the maturation of neural circuits. When this gene functions normally, children develop speech, acquire skills and form stable patterns of thought. But even modest mutations can derail these processes, leading to epilepsy, disrupted brain rhythms, language loss, cognitive impairments and a heightened risk of psychiatric illness. A single faulty subunit in a single receptor can compromise some of the most fundamental aspects of human mental life.

This is exactly what we expect from evolution — and precisely what we should not expect from intelligent design. A system built with foresight would not place the foundations of cognition and language at the mercy of one fragile molecular component. Nor would it allow tiny, predictable copying errors to produce devastating developmental consequences. The NMDA receptor is a classic product of evolutionary tinkering: complex, delicate, functionally vital and prone to catastrophic failure when disturbed.

For an Intelligent Design advocate willing to look squarely at the biology, GRIN2A poses an unavoidable challenge. If the human brain were engineered to be robust and reliable, it would not collapse because of minor alterations in a single gene. But if the brain is the outcome of evolutionary processes — makeshift, constrained and built from whatever happened to work well enough — then this vulnerability makes perfect sense. The Leipzig findings are not merely a medical insight; they are a reminder that our biology bears all the hallmarks of natural evolution, and none of the signs of deliberate, competent design.

Advertisement

Amazon