Questions: Frameshift Mutations and Insertions/Deletions

Question 1

A gene has a 1-nucleotide deletion 50 bp after the start codon, and then 150 bp further downstream, a 1-nucleotide insertion. A student predicts the protein is probably functional from the insertion site onward, since these two changes together restore the original reading frame. What is the most important problem with this prediction?

• A. The prediction is correct — when the reading frame is restored, downstream amino acids are normal
• B. The reading frame is restored after the insertion, but the 150-nucleotide region between the two mutations is still completely frameshifted — encoding wrong amino acids and almost certainly containing a premature stop codon that terminates translation before the insertion site
• C. Single-nucleotide deletions and insertions can never cancel each other out in any circumstance
• D. The insertion would need to be in the same codon as the deletion to have any compensating effect

Answer: (B)

The student correctly identifies that a -1 deletion followed by a +1 insertion yields a net frameshift of zero — downstream codons after the insertion are back in the correct reading frame. But this ignores what happens in between. The 150-nucleotide stretch between the two mutations is read in the wrong frame, producing a completely garbled amino acid sequence. Critically, in a randomly frameshifted sequence, a stop codon (UAA, UAG, or UGA) appears on average every ~20 codons — so the ribosome almost certainly terminates translation before ever reaching the 'restored' region downstream of the insertion. Compensating mutations only produce near-normal proteins when they are very close together.

Question 2

Why is a 3-nucleotide deletion in a coding sequence typically much less damaging than a 1-nucleotide deletion?

• A. Three-nucleotide deletions are repaired more efficiently by cellular proofreading mechanisms
• B. A 3-nucleotide deletion removes one complete codon without shifting the reading frame of any downstream codon, potentially producing a protein missing just one amino acid; a 1-nucleotide deletion shifts every downstream codon, garbling the entire rest of the protein
• C. Three-nucleotide deletions only occur in non-coding introns, so they cannot affect protein sequence
• D. A 1-nucleotide deletion is more likely to occur in an essential gene, making it more harmful by coincidence

Answer: (B)

The genetic code is read in strict non-overlapping triplets. A deletion of exactly 3 nucleotides (or any multiple of 3) removes whole codons but leaves the reading frame intact — the ribosome resumes reading correctly immediately after the deletion. The protein may be missing one or a few amino acids and may lose function if those residues are critical, but the rest of the sequence is unaffected. A 1-nucleotide deletion shifts the triplet boundaries for every codon that follows, producing a completely different and usually nonfunctional amino acid sequence from that point on.

Question 3

A frameshift mutation in the middle of a long protein-coding gene almost always produces a nonfunctional protein, even if the mutation occurs far from the active site, partly because the scrambled downstream codons frequently contain a premature stop codon.

• T. True
• F. False

Answer: True

In a randomly scrambled sequence (i.e., the wrong reading frame), stop codons (UAA, UAG, UGA) occur with the frequency expected by chance — approximately once every 20 codons. So even if the frameshift occurs hundreds of nucleotides upstream of the active site, the ribosome will almost certainly encounter a premature stop codon well before completing translation of the full-length protein. The truncated, garbled protein that results is typically nonfunctional and is often targeted for degradation.

Question 4

An insertion of 4 nucleotides into a protein-coding sequence is less damaging to protein function than an insertion of 3 nucleotides, because 4 is larger and therefore removes more of the downstream coding sequence.

• T. True
• F. False

Answer: False

The key variable is not the size of the indel but whether the size is divisible by three. A 3-nucleotide insertion adds exactly one codon without disturbing the reading frame of any downstream codon — it is an in-frame indel that may add one amino acid but leaves the rest of the protein intact. A 4-nucleotide insertion shifts the reading frame (4 mod 3 = 1), corrupting every downstream codon. By the same logic, a 6-nt insertion is less damaging than a 4-nt insertion, and a 9-nt insertion less damaging than a 7-nt insertion. Size is misleading; divisibility by three is what matters.

Question 5

Explain in terms of how the ribosome reads mRNA why a single-nucleotide deletion corrupts every codon downstream, while a deletion of exactly 3 nucleotides can be tolerated much better.

Model answer: The ribosome reads mRNA in consecutive, non-overlapping triplets starting from the AUG start codon — there are no punctuation marks or spaces between codons. The reading frame is defined entirely by where counting begins and is maintained by always advancing exactly 3 nucleotides per codon. If one nucleotide is deleted, the ribosome still advances 3 at a time, but every triplet after the deletion is now shifted by one position relative to the original sequence. What was codon 10 is now read as the last 2 nucleotides of what was codon 10 plus the first nucleotide of what was codon 11 — a completely different codon. This shift propagates through the rest of the mRNA. A 3-nucleotide deletion, by contrast, removes exactly one triplet. The ribosome skips that codon but then resumes reading the original sequence in the original frame — because 3 nucleotides gone means the next nucleotide after the deletion is still the first nucleotide of the next original codon.

This is why the divisibility-by-three rule is the central principle for predicting frameshift severity. It also explains why even very short frameshifted regions (between two compensating mutations) are likely to be lethal to protein function — the scrambled codons don't just change amino acids, they almost certainly introduce a premature stop that prevents translation from reaching the corrected region.