Tag Archives: crisprbabies

Could the correction of a deleterious mutation be a disadvantage?

When working on a forthcoming talk about the ethics of correcting gene defects, I asked myself: Are there any empirical examples where the correction of a so called “deleterious” mutation may be a disadvantage? Or in other words: Are there any beneficial side-effects of otherwise deleterious mutations?

(Don’t answer this with the joke that the Y chromosome is a X with a large deletion :-)

Yes, there are some examples of heterozygote advantage

  • HBB-p.E6V leads to sickle cell anemia and malaria resistance
  • CFTR Delta F508 leads to CF and protects against tuberculosis

Maybe I am not asking if there are beneficial/deleterious mutations in a single gene – the question here is more about distant / cis-regulating elements.  And there seems to be a thesis that

deleterious mutations have long been thought to be unimportant, however this view overlooks the pivotal role of epistasis. The unique experiments presented here give new insights into the historical and highly contingent nature of evolution. While evolution frequently finds a well adapted solution in the long-term, evolving populations will frequently climb suboptimal peaks initially. Deleterious mutations become useful because they aide evolution in reconciling short-term and long-term adaptation.

the thesis made it also into a PNAS paper

It might seem obvious that deleterious mutations must impede evolution. However, a later mutation may interact with a deleterious predecessor, facilitating otherwise inaccessible adaptations. … We studied digital organisms—computer programs that replicate and evolve—to compare adaptation in populations where deleterious mutations were disallowed with unrestricted controls. Control populations achieved higher fitness values because some deleterious mutations acted as stepping stones across otherwise impassable fitness valleys. Deleterious mutations can thus sometimes play a constructive role in adaptive evolution.

Looks like humans shouldn’t interfere with their own evolution as long as the rules are not known… The PNAS paper above has been cited many times, it will take some time to scan these for more empirical examples.

Phenotype of the CRISPR CAS babies

CCR5 annotation, including the known delta32 Mutation and the three mutations introduced by He Jiankui. PAM=Protospacer adjacent motif, DSB=putative double strand break.

From the CCR5 sequence we get the following amino acid sequences

>sp|WT.P51681|CCR5_HUMANC-Cchemokinereceptortype5OS=HomosapiensOX=9606GN=CCR5PE=1SV=1
MDYQVSSPIYDINYYTSEPCQKINVKQIAARLLPPLYSLVFIFGFVGNMLVILILINCKR
LKSMTDIYLLNLAISDLFFLLTVPFWAHYAAAQWDFGNTMCQLLTGLYFIGFFSGIFFII
LLTIDRYLAVVHAVFALKARTVTFGVVTSVITWVVAVFASLPGIIFTRSQKEGLHYTCSS
HFPYSQYQFWKNFQTLKIVILGLVLPLLVMVICYSGILKTLLRCRNEKKRHRAVRLIFTI
MIVYFLFWAPYNIVLLLNTFQEFFGLNNCSSSNRLDQAMQVTETLGMTHCCINPIIYAFV
GEKFRNYLLVFFQKHIAKRFCKCCSIFQQEAPERASSVYTRSTGEQEISVGL

>sp|-32.P51681|CCR5_HUMANC-Cchemokinereceptortype5OS=HomosapiensOX=9606GN=CCR5PE=1SV=1
MDYQVSSPIYDINYYTSEPCQKINVKQIAARLLPPLYSLVFIFGFVGNMLVILILINCKR
LKSMTDIYLLNLAISDLFFLLTVPFWAHYAAAQWDFGNTMCQLLTGLYFIGFFSGIFFII
LLTIDRYLAVVHAVFALKARTVTFGVVTSVITWVVAVFASLPGIIFTRSQKEGLHYTCSS
HFPY
IKDSHLGAGPAAACHGHLLLGNPKNSASVSK

>sp|-15.P51681|CCR5_HUMANC-Cchemokinereceptortype5OS=HomosapiensOX=9606GN=CCR5PE=1SV=1
MDYQVSSPIYDINYYTSEPCQKINVKQIAARLLPPLYSLVFIFGFVGNMLVILILINCKR
LKSMTDIYLLNLAISDLFFLLTVPFWAHYAAAQWDFGNTMCQLLTGLYFIGFFSGIFFII
LLTIDRYLAVVHAVFALKARTVTFGVVTSVITWVVAVFASLPGIIFTRSQKEGLHYTCS
SQYQFWKNFQTLKIVILGLVLPLLVMVICYSGILKTLLRCRNEKKRHRAVRLIFTI
MIVYFLFWAPYNIVLLLNTFQEFFGLNNCSSSNRLDQAMQVTETLGMTHCCINPIIYAFV
GEKFRNYLLVFFQKHIAKRFCKCCSIFQQEAPERASSVYTRSTGEQEISVGL

>sp|-4.P51681|CCR5_HUMANC-Cchemokinereceptortype5OS=HomosapiensOX=9606GN=CCR5PE=1SV=1
MDYQVSSPIYDINYYTSEPCQKINVKQIAARLLPPLYSLVFIFGFVGNMLVILILINCKR
LKSMTDIYLLNLAISDLFFLLTVPFWAHYAAAQWDFGNTMCQLLTGLYFIGFFSGIFFII
LLTIDRYLAVVHAVFALKARTVTFGVVTSVITWVVAVFASLPGIIFTRSQKEGLHYTCSS
HFP
YSINSGRISRH

>sp|+1.P51681|CCR5_HUMANC-Cchemokinereceptortype5OS=HomosapiensOX=9606GN=CCR5PE=1SV=1
MDYQVSSPIYDINYYTSEPCQKINVKQIAARLLPPLYSLVFIFGFVGNMLVILILINCKR
LKSMTDIYLLNLAISDLFFLLTVPFWAHYAAAQWDFGNTMCQLLTGLYFIGFFSGIFFII
LLTIDRYLAVVHAVFALKARTVTFGVVTSVITWVVAVFASLPGIIFTRSQKEGLHYTCSS
HFPY
KSVSILEEFPDIKDSHLGAGPAAACHGHLLLGNPKNSASVSK

The -15/WT genotype will have both, a normal and a slightly shortened CCR5 giving no HIV protection at all. The -4/+1 genotype may suffer the fate of non-sense mediated decay. Here are the amino acid predictions for each genotype (I reverted to Topcocns as Protter had some problems in generating correct plots).

Wildtype: There are 7 transmembrane alpha helices I to VII, connected by three extracellular loops (ECL1–3) and three intracellular loops (ICL1–3). ECL2, forms a β-hairpin structure (Tan 2013).
Known delta 32
Nana -4 genotype has a much shorter predicted amino acid sequence with missing G protein coupling. Resembles delta 32.
Nana +1 genotype is somewhat longer but missing helices VI and VII. No analogue known.
Lulu -15: resembles wild type, however, will be sensitive to allosteric changes by small-molecule CCR5 inhibitors .

From the original AJHG paper, however, also delta 32 carriers may be HIV infected. As there exists also non-CCR5 dependent virus replication, also Nana is at risk of HIV infection when being virus exposed.