More than half a century ago, Laurell and Eriksson, working in their clinical biochemistry
laboratory, noted anomalous findings in some of the samples sent for serum protein
electrophoresis. In a minority of electrophoreses, the alpha-1 protein band was faint
to nonexistent. Such a finding might easily have been dismissed as population variance,
but Laurell and Eriksson left the confines of the lab to determine the clinical phenotype
associated with the anomaly they observed. In doing so, they described what we now
recognize as alpha-1 antitrypsin deficiency (1). Individuals with markedly reduced
serum levels of this protective glycoprotein were predisposed to develop severe emphysema
early in life, often with little or no tobacco smoke exposure. The basal predominant,
panlobular emphysema was distinctive. Moreover, there was a predilection for the development
of cirrhosis. That this disorder was genetic was established early; much later, the
most common severe variant was localized to an abnormality on the SERPINA1 gene. The
initial family kindreds had what we now understand to be the most common genotype
associated with a severe deficiency of alpha-1 antitrypsin, the ZZ genotype.
These bench-to-bedside findings have been of great importance in several ways. Scientifically,
they described the foundations of the protease/antiprotease theory of the of emphysema
pathogenesis. Clinically, they described an illness that became one of the first examples
of personalized medical care with specific targeted therapy to treat a specific abnormality.
More than a quarter-century ago, Wewer and colleagues reported that the infusion of
purified alpha-1 antitrypsin protein in a dosage of 60 mg per kilogram per week could
restore and maintain serum levels to theoretically protective thresholds with infused
protein detectable in BAL fluid (2). Later observational studies confirmed that the
accelerated decline in lung function associated with the deficiency could be slowed
by this treatment and, in those who are most obstructed, loss of life delayed (3,
4). More recently, randomized controlled trials have shown that lung structure, as
measured by computed tomographic scan lung density estimates, is preserved by infusions
(5). Although the treatment of alpha-1 antitrypsin deficiency liver disease has lagged,
there is growing evidence that as well as supportive measures and transplantation,
the use of chaperone proteins holds promise for reducing the impact of damage to the
liver by the retention of alpha-1 antitrypsin polypeptides.
Despite the long history of alpha-1 antitrypsin deficiency, its characteristic clinical
phenotype, and the availability of therapy, the disorder remains underdiagnosed (6).
Simple routine testing could remedy this, and the Global Initiative for Chronic Obstructive
Lung Disease Strategy, for example, recommends routine testing of all newly diagnosed
patients with chronic obstructive pulmonary disease (COPD) (7). However, testing for
the disorder is potentially complex. Not all gene variants result in decreased serum
levels of alpha-1 antitrypsin protein. Some variants produce measurable but dysfunctional
serum proteins, the F variant being the foremost example (8). Thus, the concerned
clinician must consider an array of potential testing strategies. Simplest and least
expensive remains quantifying serum levels of alpha-1 antitrypsin protein. Costing
only a few dollars per test, it is widely available and screens adequately for the
most common of alpha-1 antitrypsin deficiency disorders. But serum level measurements
are inadequate for carrier detection, may be transiently elevated during acute inflammatory
events, and will not detect functionally abnormal variants. If serum levels are low
or the clinician’s index of suspicion is high, serum protein may be phenotyped, the
approach pioneered by Laurell and Eriksson. Although their technology has been updated
or replaced, their legacy has included the nonstandard nomenclature for alpha-1 antitrypsin
abnormalities reflecting the mobility of proteins in electrophoresis gels. Protein
phenotyping has the potential to detect both known and novel variants but may fail
to reveal complex abnormalities. Today, protein phenotyping is typically bypassed
in favor of genetic testing, usually targeting the most common of deficiency and dysfunction
variants Z, S, F, and I. This four-variant PCR panel will characterize the vast majority
of important and known alpha-1 antitrypsin abnormalities, but if a novel abnormality
is suspected, sequencing becomes necessary. When is a novel abnormality suspected?
Such a possibility should be considered when there is a discrepancy between what is
measured and what is expected. Such discordance should not be dismissed as random
variability in test results until the possibility of a novel variant has been considered.
In this issue of the Journal, Matamala and colleagues (pp. 444–451) describe two novel
in cis variants of the SERPINA1 gene that modify the properties of the PI*S allele,
increasing hepatocellular retention and decreasing secretion into the bloodstream.
The result is production of a functional null variant in what by conventional testing
appears to be unremarkable S variant findings, typically a mild deficiency variant
with little risk of liver disease (9). Their elegant investigation of the problem
was sparked by relatively mild discordance among clinical findings, serum alpha-1
antitrypsin quantification and genotype as revealed by targeted gene testing. One
man with COPD had a serum alpha-1 antitrypsin level ∼20% to 25% lower than expected
for what appeared to be SS genotype. Their analysis found one S variant and what they
have termed an S+ variant (S+p.Tyr138Cys), a variant associated with null-like activity.
This additional in cis abnormality modulates the otherwise innocuous S variant to
produce null-like activity with reduction of serum levels by trapping polypeptides
in the hepatocyte. Thus, this patient who might otherwise have been regarded as having
tobacco-related COPD with an unrelated mild alpha-1 antitrypsin deficiency variant
was more correctly identified as having an intermediate deficiency variant potentially
treatable by augmentation therapy. A second patient suffering from liver disease with
the same in cis variant was initially identified as having MS genotype until further
investigation revealed the S+ variant, thereby accounting for lower-than-expected
serum levels of alpha-1 antitrypsin and retention of alpha-1 antitrypsin polypeptides
in the endoplasmic reticulum of hepatocytes. A third patient with pulmonary involvement
but lower-than-expected serum level for genotype MS was found to have type MS+ genotype,
the in cis variant in this example producing null-like activity differently (S+p.Pro391Thr).
These novel findings have implications for all of us in the field of pulmonary medicine.
At the bedside, we must be alert to discordant findings. Although few clinicians could
trace out the testing pathways that led to these complex and novel variants, clinicians
should recognize when simple serum levels are lower than anticipated. Such findings
should not be dismissed without further investigation. Communicating directly with
the testing lab may provide further insights and suggestions for testing. As sequencing
becomes accessible for clinical purposes, we might expect further rare variants to
be described in this way. At the bench, we would do well to emulate the initiatives
of Laurell and Eriksson. Even half a century on, the story of alpha-1 antitrypsin
deficiency remains incomplete, and one cannot help but wonder how many variations
to the story remain untold.