Telomere length kinetics assay (TELKA) sorts the telomere length maintenance (tlm) mutants into functional groups

Genome-wide systematic screens in yeast have uncovered a large gene network (the telomere length maintenance network or TLM), encompassing more than 400 genes, which acts coordinatively to maintain telomere length. Identifying the genes was an important first stage; the next challenge is to decipher their mechanism of action and to organize then into functional groups or pathways. Here we present a new telomere-length measuring program, TelQuant, and a novel assay, telomere length kinetics assay, and use them to organize tlm mutants into functional classes. Our results show that a mutant defective for the relatively unknown MET7 gene has the same telomeric kinetics as mutants defective for the ribonucleotide reductase subunit Rnr1, in charge of the limiting step in dNTP synthesis, or for the Ku heterodimer, a well-established telomere complex. We confirm the epistatic relationship between the mutants and show that physical interactions exist between Rnr1 and Met7. We also show that Met7 and the Ku heterodimer affect dNTP formation, and play a role in non-homologous end joining. Thus, our telomere kinetics assay uncovers new functional groups, as well as complex genetic interactions between tlm mutants.


Supplementary Text 1
TelQuant and Quantification of telomere length: The telomere band (Y' telomeres) signal (median and mean) is measured using the TelQuant software , by comparing its position on a teloblot to that of the control size bands (size 2044 and 779bp). TelQuant software was programmed using the VisualBasic6 programming environment.
TelQuant algorithm: Southern blots are hybridized with radioactive probes and exposed to a sensitive X-ray film. The areas of the X-ray film that were exposed to X-rays are seen as dark signals after the film is developed. Therefore, the Southern blot image is composed of dark areas of the detected signal and a relatively bright background. The TelQuant software scans the Southern blot's image's pixels, gets the RGB Decimal value of each scanned pixel and calculates the intensity for each pixel as 1/RGB Decimal value.
After loading a Southern blot image and manually fixing the 36 running lanes (a line running through the middle of the lane is defined as a Y-line, our blots usually contain 36 lanes, but this number can be adjusted), TelQuant scans each of the 36 Y-lines separately in a criss-cross manner. The scan distance (X) is calculated as (X=the distance between two Y-lines / 2). The scan is preformed horizontally from both sides of the Y-line starting from XY-line -X to XY-line + X. The horizontal scan is defined as X-line, and the vertical Y-line is scanned vertically X-line by X-line from the start of the Y-line to its end.
There are three layers of data processing. The first layer detects the peak signal/density value of the Y-line ('MaxSignal'). The second layer of processing compares each signal point (pixel) to the MaxSignal and grades it: if the point signal is greater than MaxSignal divided by the FilterValue, then it grades it as 1, otherwise, it grades it as 0. Therefore, the FilterValue, which is manually determined, enables TelQuant to process noisy images, requiring stringent filtering. In conclusion, the second processing layer creates a binary array (1,0) of the raw data (can be referred also as black/white filtering). The last processing layer sums up each X-line grade and searches for the longest consecutive segments with high grades. The longest segment is defined as the telomere band. The other 2 segments are defined as the 2044 bp band (the upper size marker) and the 779 bp band (the lower size marker).
After the bands (segments) are located, TelQuant draws a histogram ('Telogram') of the analyzed raw data and finds the median of each band's histogram (MedianLine). The middle of each band (the mean of the band's first detection point to the last detection point) is also calculated (MeanLine) and in most cases its location is very close to the MedianLine. The difference between the locations of both lines is defined as the AreaRatio: AreaRatio=(MedianLine-first detection point)/(last detection point -first detection point) * 100.
AreaRatio different from 50% could be interpreted as a technical matter (unclean samples/images led to a biased detection), but in rare cases it represents a biological issue (e.g. most signal is located preferably in the lower region of the telomere smear, implying that there is a preference for shorter or longer telomeres in the strain examined).
In an optimal Southern blot image, all of the marker bands should be found automatically.
Unfortunately, not all the images are optimal, and bands with weak signals or a strong  Cobalamin-independent methionine synthase, involved in methionine biosynthesis and regeneration.

Eno1
Enolase I, catalyzes the conversion of 2-phosphoglycerate to phosphoenolpyruvate during glycolysis and the reverse reaction during gluconeogenesis.

Fba1
Required for glycolysis and gluconeogenesis; locates to mitochondrial outer surface upon oxidative stress.
Cdc19 Pyruvate kinase, functions as a homotetramer in glycolysis to convert phosphoenolpyruvate to pyruvate.

Mrf1
Mitochondrial translation release factor, involved in stop codon recognition and hydrolysis of the peptidyl-tRNA bond during mitochondrial translation.

Thr1
Homoserine kinase, conserved protein required for threonine biosynthesis. Expression is regulated by the Gcn4 general amino acid pathway.

Pef1
Penta-EF-hand protein required for polar bud growth and cell wall.

Asc1
G-protein beta subunit inhibitor of Gpa2p; ortholog of RACK1 that inhibits translation; core component of the small (40S) ribosomal subunit.
Sec63 Hsp40/DnaJ family that regulates Hsp70 chaperone activity, protein targeting and import into the ER.

Rnr1
Major isoform of the large subunit of the ribonucleotide-diphosphate reductase; the RNR complex catalyzes rate-limiting step in dNTP synthesis.