Natural gene / health

GYS1 (PSSM1)

Type 1 polysaccharide storage myopathy. The muscle stores sugar it cannot burn and ties up under work. Neither combination is lethal.

Crossing two of them

Its gene carrot

There is none. A disorder locus sets hasGeneCarrot() false — forcing one onto somebody’s line is hostile — and SpliceSafety keeps it out of the random splice as well, derived from the fact that it declares a non-informational Condition rather than from any hand-written list.

Gene key
horsegenetics.gys1
Priority
94
Alleles
P N
Outcomes
wild, pssm1, pssm1-severe
Coat effect
none — every outcome is a wild type
Default allele
N
Wild population
2.5% of founders are born P/N — and therefore born affected
Founder draws
1 × nextFloat()
Catalogue entries
1 — it never changes what a horse looks like
It paints nothing, and that is what makes it free

Every combination of this gene is an expression marked wildType. That reads oddly for a gene that can kill a foal, and it is exactly right: wildType means changes nothing about the coat. Because none of the non-coat genes paints, Gene.affectsCoat() is false for all of them, they are left out of a horse’s texture key, and the whole locus collapses into one catalogue entry. What the gene actually does travels on the trait system instead.

The combination table

combination outcome effect
N/Nwild typenothing
P/Npssm1−3 health, −0.015 speed, −0.06 jump
P/Ppssm1-severe−6 health, −0.030 speed, −0.12 jump — still survivable

The one disorder you will actually meet

Everything else on the health layer is rare by construction, because the recessives need two carriers to find each other. This one is dominant, so every copy shows — and at 2.5% of founders it is by a wide margin the disorder a player is most likely to run into. That is true of the real thing too, across draughts, warmbloods, paints and appaloosas.

The frequency is the gene

2.5% is set deliberately low relative to the real population, and that number is what this locus is really about. A dominant disorder at a realistic rate would mean a visible fraction of every paddock is unwell — at which point it stops reading as a disorder and starts reading as the baseline horse being bad, which is a different and much worse game. Rare enough to notice, common enough to matter.

There is no carrier, and founders can be affected

Every other disorder locus in the mod is recessive, and they all lean on two facts: a heterozygote is indistinguishable from normal, and no founder is ever affected. Neither holds here.

The second is the interesting break. A recessive can be absent from its own founder table entirely, because a wild-caught horse is an adult that survived and so carries at most one copy. A dominant has no silent carrier at all — so if founders could never be affected, the allele could never enter the world and the gene would not exist. Founders can therefore be born heterozygous, and a wild-caught horse really can be a sick one. The homozygote is still excluded, which keeps the worst outcome something a breeder made.

That changes what the locus is for. A recessive lethal is something you discover about two horses after the fact; this is something you can see in the animal in front of you and decline to breed from.

No cliff

P/P is worse than P/N rather than fatal, which makes this the mildest locus on the health layer and the only dominant one that is safe to breed from if you accept what you are getting. Unlike HYPP there is no sudden penalty for ignoring it — just a line that gets quietly worse.

The magnitudes are calibrated against the disorders that already ship rather than computed from the reference’s percentages; see HYPP for why.

At a glance

GeneGYS1, glycogen synthase 1 — the principal glycogen-synthesising enzyme of skeletal muscle
ChromosomeEquine chromosome 10, historically the short arm of ECA10; the causal variant is in exon 6
Disease alleleA G→A substitution giving p.Arg309His (R309H)
TraitType 1 polysaccharide storage myopathy (PSSM1)
Functional effectGain of function — the enzyme is abnormally active and poorly switched off
InheritanceAutosomal dominant, but variable and dose-dependent — better described as incompletely dominant or semidominant
CoatNo direct effect known

What the enzyme normally does

Glycogen synthase transfers glucose from UDP-glucose onto a growing glycogen chain, mostly through α-1,4 linkages; glycogen branching enzyme then adds the α-1,6 branch points that make glycogen compact, soluble and easy to mobilise. The two genes are neighbours in the same pathway and fail in opposite directions — GBE1 cannot build the branches, GYS1 builds too much chain.

The normal enzyme is tightly regulated, because glycogen production has to switch on after a meal and down when stores are full. Phosphorylation generally inhibits it; glucose-6-phosphate allosterically activates it; muscle glycogen itself provides feedback. GYS1 is the skeletal-muscle isoform, GYS2 the liver one. PMC5148651

A brake that will not engage

R309H sits in a highly conserved part of the protein, found in exon 6 after PSSM1 was localised to an approximately 3 Mb segment of ECA10. It is a gain of function, not a defect:

  • Substantially greater baseline activity.
  • Much greater affinity for UDP-glucose, its substrate.
  • Strongly active at far lower G6P concentrations than normal enzyme — a half-activation concentration around 0.24 mM against about 3.0 mM for wild type.
  • Inappropriately active even when phosphorylation should be suppressing it.

Its UDP-glucose Km was also much lower — 0.48 mM against 2.4 mM without G6P — so it makes glycogen readily under conditions where normal enzyme would be restrained. Structural modelling suggests why: Arg309 probably takes part in a salt bridge stabilising the low-activity state, and replacing it with histidine is predicted to destabilise that “off” configuration. Well supported biochemically, still partly model-based — there is no equine protein crystal structure showing it directly.

What accumulates: total skeletal-muscle glycogen at roughly 2–4 times normal in some research contexts; abnormal, less-branched glycogen-like material; amylase-resistant polysaccharide inclusions described as polyglucosan or amylopectin-like; and subsarcolemmal vacuoles and cytoplasmic inclusions inside muscle fibres, especially in fast and mixed fast/oxidative fibres including type 2A. Whether the damage comes mainly from mechanical disruption, altered energy metabolism, or something else is unresolved. PMC3409190

Inheritance and dosage

Autosomal — sex does not change the odds. Labs write the alleles N for normal and P1 for R309H; older papers use R for the arginine allele and H for the histidine one.

genotypemeaningtransmission
N/Nno R309H allelecannot pass it on
N/P1heterozygous; can develop disease, from apparently normal to clinically important50% per foal against an N/N mate
P1/P1homozygous; usually greater pathological burdenpasses it to 100% of foals

Calling it “dominant” is right for breeding decisions and misses the biology. There is variable expressivity and evidence of incomplete dominance: two copies generally carry more biochemical and histologic disease than one. In Belgians and Percherons managed under the same conditions, homozygotes had more amylase-resistant inclusions, more vacuolation, higher resting muscle-enzyme activities and higher post-exercise CK than controls — while heterozygotes commonly overlapped with normal horses.

The mod makes every heterozygote sick; reality does not

Gys1Gene gives P/N a guaranteed impairing condition and P/P a worse one. The dose-dependence is exactly right — it is one of the closest matches on the health layer. What it cannot express is that a real N/P1 horse is often clinically quiet, particularly with turnout, regular low-intensity movement and an appropriate ration, and that a positive test is a risk factor rather than a diagnosis. Penetrance is not something the expression model can currently represent (the same limit that flattens TOE1), and it is the single most load-bearing simplification in this locus.

What the disease does

PSSM1 is a skeletal-muscle disease. It is not a colour defect and is not established as a bone, tendon, joint, heart, fertility or neurologic disorder. Reported signs, and expression varies a great deal by genotype, breed background, diet, fitness, management and work type:

  • Muscle stiffness, cramping, soreness, reluctance to go forward.
  • “Tying up” — exertional rhabdomyolysis.
  • Sweating, skin twitching, firm painful muscles, abnormal stance.
  • Exercise intolerance, poor recovery, gait change, resistance to collection.
  • Progressive muscle atrophy or weakness in some horses, particularly draft types.
  • Severe episodes with myoglobinuria or inability to rise.

Blood CK and AST rise with muscle damage but are imperfect screens: some positive horses, especially heterozygotes, sit inside the normal range. In the cited study both were generally higher in homozygotes and the elevation correlated with histologic pathology. The DNA test identifies R309H directly and is more definitive than inferring genotype from enzymes or signs.

Modifiers matter. The RYR1 mutation causing malignant hyperthermia susceptibility can worsen the PSSM1 phenotype in some Quarter Horse-related animals. GYS1 status alone does not fully predict severity, which is why a positive result is a strong causal risk factor for a specific myopathy and not a complete prediction of a horse’s future.

Coat, build, size

No known causal role in coat colour, white patterning, dilution, grey, roan, leopard complex or any pigment trait.

The Noriker colour association, and what it actually shows

An Austrian Noriker study found PSSM1 carrier rates differing substantially between colour-defined breeding subpopulations — higher in chestnut groups, lower in leopard-spotted groups. The authors’ point was that colour-line breeding, genetic bottlenecks and mating patterns created population substructure that moved GYS1 allele frequencies. It does not mean chestnut causes PSSM1, or the reverse. If a stallion line happens to carry both a colour allele and PSSM1, its descendants show an association in that registry — pedigree linkage, not one gene doing two jobs. Anim. Genet. 10.1111/age.12481

Nor is there convincing evidence that R309H makes a horse taller or shorter, heavier or finer, more or less naturally muscled, or better built for any discipline. The allele turns up across Quarter Horse-related breeds, drafts, warmbloods, cobs, ponies and crosses because it is an old founder mutation that predates many modern breed divisions — its spread across body types is history, not a build gene.

There are secondary appearance changes in clinically affected horses: chronic muscle pain, reduced work tolerance, disuse and muscle pathology can give a poor topline, focal atrophy, asymmetry or difficulty holding condition. Disease consequences, not a programmed body type.

Jumping, speed, and the advantage theory

No robust study establishes that R309H improves maximum galloping speed, sprint acceleration, jump height or scope or bascule or technique, race performance, endurance, or training response in healthy sport horses. It is not an accepted performance marker and a PSSM1 test should not be used to pick a horse as more athletic.

The speculation has a real starting point: affected horses can carry more muscle glycogen and may replenish it faster after work, and glycogen is a critical fuel for high-intensity effort. So it has been hypothesised that the allele once helped horses doing frequent daily work on modest feed — the thrifty gene idea. Population-genetic data support historical positive selection around the mutation in Belgian horses, meaning the allele’s frequency and surrounding haplotype are unlikely to be neutral drift alone in that population. PMC3920812

None of which demonstrates modern speed or jumping ability. More stored glycogen is not automatically more usable fuel: PSSM1 muscle has abnormal glycogen regulation and can show an energy deficit during submaximal aerobic exercise, alongside painful cramping, exercise intolerance and muscle injury. A mutation that helped under one historical work-and-feed regime can be harmful under intermittent work and energy-dense feeding.

For jumping the consequence is reliability risk, not a change in scope: an affected horse may resist impulsion, feel stiff behind, fatigue oddly, cramp, or go inconsistently, and pain or subclinical damage can affect take-off commitment, bascule and willingness in a horse with perfectly normal talent. For speed, a quiet carrier under good management may show nothing, and an affected horse loses acceleration, sustained pace, recovery and willingness. The mod’s costs — −0.015 speed and −0.06 jump per copy — encode exactly that reading: a horse that is less reliable, not a horse built differently.

Breed distribution and history

Found across Quarter Horse-related breeds, drafts, some warmbloods, Morgans, Mustangs, Rocky Mountain Horses, cobs, ponies and crosses. A conserved haplotype around the mutation across breeds shows it arose once and was inherited identical-by-descent rather than appearing repeatedly. The original allele-age estimate was roughly 159 generations — an origin about 1,200–1,500 years ago, before most present-day breeds existed. PMC2430182

Frequency varies markedly. Historic studies reported high frequencies in some draft populations; selection analysis found evidence consistent with positive selection in Belgians; and the same signal was not clearly attributable to direct positive selection in Quarter Horses, where founder effects and rapid population expansion may explain persistence better. That history explains the breadth of the spread — and does not mean every horse with draft or Quarter Horse ancestry is positive.

Why the mod’s 2.5% is deliberately below the real rate

This is the one disorder in the mod a player will actually meet. Everything else on the health layer is rare by construction, because a recessive needs two carriers to find each other; this one is dominant, so every copy shows. At WILD_AFFECTED_PERCENT = 2.5 it is by a wide margin the most likely disorder to run into — which is true of the real thing too, across drafts, warmbloods, paints and appaloosas. The number is set low against the real population on purpose: a dominant disorder at a realistic draft-breed rate would mean a visible fraction of every paddock is unwell, which stops reading as a disorder and starts reading as the baseline horse being bad. Rare enough to notice, common enough to matter.

What is still open

Penetrance and modifiers. Not every heterozygote develops obvious disease. Allele dosage is part of it, RYR1 is a documented modifier, and breed-specific muscle-fibre profiles, diet, body condition, turnout, exercise consistency, training load and stress all contribute. Which additional genes matter most is not mapped.

How the inclusions cause symptoms. Abnormal polysaccharide accumulates and more pathology tracks more enzyme leakage; whether the clinical disease comes mainly from direct structural disruption by vacuoles and inclusions, impaired energy regulation during particular exercise conditions, altered signalling and gene expression from excess glycogen, or all three, is unsettled.

The thrifty-gene hypothesis is plausible and indirectly supported by selection signatures and glycogen biology. It is not a demonstration that historical carriers pulled harder or travelled further.

PSSM1 versus “PSSM2”. PSSM1 is genetically defined by the validated R309H mutation. Not every horse once labelled “PSSM” from biopsy or signs carries it — and those cases are not evidence against GYS1 causing PSSM1. Modern veterinary literature treats PSSM1 as the confirmed GYS1 disease and emphasises that no single causative mutation has been validated for the broader historical PSSM2 category. The mod models PSSM1 only, which is the right half to model.

Source: common/genetics/genes/Gys1Gene.java