Injection amounts and column loading

Sparse identifications and a ruined blank can be the same column, loaded at the wrong amount. Too little peptide and the spectra are noisy. Too much and the peaks broaden, the detector saturates, and the next injection inherits the previous sample. Bottom-up proteomics in plain language assumes the chromatography presented the instrument with separated peptides. Injection amount is how you keep that assumption true. The troubleshooting decision is which failure you are looking at, and which knob, dilution or wash, you turn first.

Too little peptide

A very small load produces weak precursors. Data-dependent methods then fragment noise or fragment nothing of use, and the identification list thins even when the false discovery threshold is unchanged. Label-free intensities become unstable because the peaks sit near the detection floor. The repair is not a looser score. The repair is more peptide on the column, within the load the column can accept, or a method built for low input on purpose, with the handling that low input demands.

Before you concentrate a sample to the last microlitre, ask whether the peptides were lost earlier. A pellet that was washed away, a bead elution that stayed on the beads, or a digest that never left the desalting tip will not be fixed by injecting a larger volume of an empty tube. A peptide assay or an ultraviolet trace at peptide wavelengths, used as your method describes, tells you whether the vial has material. Promega protocols include assay classes for protein and peptide amount. Use the assay that matches the buffer you actually have. Detergent will lie to some colorimetric assays.

Vials and inserts matter at low volume. A sample sitting in a large vial with a needle that cannot reach the bottom is an apparent low load. When to use plastic tubes, tips and plates is the consumables question. Adsorption of scarce peptides onto a large plastic surface is real. Low-bind plastics are a class, not a slogan, and they do not replace a sensible volume.

Too much peptide

An overloaded reversed-phase column stops separating. Peaks widen, tail into their neighbours, and sometimes flatten at the top when the detector cannot count higher. Spectra taken across a flat-top are mixtures, so the search either misses them or assigns them with a score you should not trust. Pressure may rise if the load includes particulates or material that precipitates in the aqueous start of the gradient.

Saturation also wrecks quantification. A flat-top peak has lost the relationship between amount and height. Two samples that both saturate look similar even when they differ. Dilute, and confirm that the peak has a real apex again, before you report a ratio.

The column maker publishes a loading range that depends on diameter, length and particle type. A narrow nano column and a wider analytical column are not loaded with the same peptide mass. Treat the published range as the planning window and the chromatogram as the confirmation. This article deliberately gives no universal mass and no universal volume. Concentration times volume is the load. Without the concentration, a volume copied from a paper is a guess.

The trap is not the analytical column

In a trap-elute plumbing, the sample is loaded onto a trap, salts are washed away, and a valve sends the peptides onto the analytical column and then the emitter. The trap is often the piece that can accept more dirt and more mass. The analytical column is the piece that sets the peak shape. Overloading the analytical column while "following the trap's capacity" is a common way to get broad peaks from a system that seemed in specification.

Direct injection skips the trap. Everything in the vial meets the analytical column. The same nominal load can be too high, and salts that a trap would have sent to waste now sit on the separation column. Know which plumbing you have before you compare your load with someone else's.

Carryover hides in this plumbing. Peptide retained on the trap, in the valve grooves, in the needle, or on the analytical column appears in the next run at a similar retention time. If that next run is a biological replicate, carryover looks like confirmation. If it is a negative control, carryover looks like a failed control. The distinguishing experiment is a blank injection immediately after the strong sample, using the same method. Peptide peaks in that blank are carryover until proved otherwise.

A path that starts with the blank

Inject a blank after the strongest sample in a pilot, not after you have already interpreted the queue. If the blank is clean, your load is at least not leaving a ghost of itself. If the blank shows the same peaks at a few percent of the height, or at a large fraction of the height, you have carryover. Dilute the strong sample. Extend the wash. Use the autosampler needle-wash class the instrument method allows, typically a solvent strong enough to dissolve hydrophobic peptides, and confirm the blank again.

If dilution makes identifications sparse, you have found the edges of the useful window: the amount that is visible and the amount that is still clean. Stay inside it. If dilution does not cure carryover, the column or the needle may be genuinely fouled. A long wash recommended by the column supplier, or a new column, comes before another biological set. Do not "run a blank and ignore it" as a way to scrub the system while pretending the blank is data.

Pressure rise is the stop sign. A climbing backpressure during injection means particulates, precipitation or a blocked frit. Abort before a fitting leaks solvent into the source. Filter or spin the sample as the method allows, and do not keep injecting the same vial to see if the pressure "settles".

Symptom, and the first check

NIST writing on measurement ranges is a useful attitude: a detector has a window, and numbers outside it are not small versions of numbers inside it. ProteomeXchange methods that state the column dimensions and the load class are the ones you can compare with your own. A method section that gives only an injection volume, with no concentration and no diameter, is not a recipe to copy.

Research limits and high-pressure plumbing

You are troubleshooting a research chromatograph. A cleaner blank does not turn the study into a diagnostic assay. High-pressure liquid chromatography can leak solvent, and acetonitrile and acid in the mobile phase are the hazards on the safety data sheet. If pressure rises, stop and depressurise as the instrument training says. Do not tighten fittings on a live high-pressure line. Biosafety of what was digested remains an institutional decision. Overloading a column with an infectious sample's peptides is still a question for that decision, not a reason to skip containment because the proteins were "already digested".

A queue that stops halfway

A power cut mid-sequence leaves an autosampler in an unknown state: a needle that may still hold sample, a valve mid-switch, a column with a partial gradient on it. When the power returns, do not resume the queue as if the next vial were the next biological sample. Equilibrate, inject a blank, then a known digest, and only then decide whether the remaining vials are still comparable. In a shared facility, write that restart rule into the booking so the next user's run is not your carryover. Heat in the room also changes viscosity and backpressure. If the column oven and the room disagree with the method's assumed temperature, peak shape and pressure will drift even at a fair load. Follow the column's temperature and pressure limits rather than the weather.

What a load enquiry should contain

Column diameter class, flow regime, trap-elute or direct injection, approximate peptide amount class if you know it, and whether samples are dirty digests or cleaned standards. Say how many injections you need the column to survive, and whether a blank between samples is part of the design. Ask for the loading window the column supplier describes. Do not ask for a single microlitre figure detached from concentration.

The protein identification by LC-MS/MS reference, the shotgun discovery proteomics reference and the differential abundance reference are the method context. Send the column class and the symptom with the quote request. A method can be discussed from the chromatogram you are actually seeing. "Inject more" is not yet a diagnosis.