Human semen is an organic fluid that consists of spermatozoa (in some instances leucocytes and epithelial cells) and seminal plasma ^{(1, 2)}. Human semen is routinely analyzed by light microscopy and Computer Assisted Semen Analysis (CASA) in regard to conventional spermatozoa characteristics such as morphology, motility, forward progression, and concentration ⁽³⁾. Other studies use ultrastructural analysis of spermatozoa by Transmission Electron Microscopy (TEM), functional assays to distinguish between fertile and infertile men, and flow cytometry studies for semen samples, etc ^(3–5).

Azoospermia affects 10-15% of infertile men attending infertility clinics. Currently, testicular biopsy is the gold standard procedure for evaluating spermatogenesis status in men with azoospermia. Significant efforts were made to look for a non-invasive method for detection of spermatogenesis ⁽⁶⁾. Therefore, it is revealed that human seminal plasma is an excellent source for diagnosis of azoospermic men.

Metabolomics is defined as the metabolome expressed by the organism ⁽⁷⁾. The metabolome consists of small molecules (metabolites) less than 2 kDa including peptides, oligonucleotides, sugars, nucelosides, organic acids, ketones, aldehydes, amines, amino acids, lipids, steroids, alkaloids, vitamins, and minerals. The metabolome is the closest correlate to phenotype ⁽⁸⁾. Therefore, it is logical to search for a biomarker and a diagnostic test at the metabolome level.

There is a gap of data regarding molecular aspects of male infertility which can be employed as a diagnostic test ⁽⁹⁾. We believe that FT-IR spectroscopy might be the method of choice, as it has become an accepted tool for the characterization of the complex building blocks of biological systems such as proteins, nucleic acids, lipids, carbohydrates, and metabolites. FT-IR spectroscopy has proven to have a diagnostic potential. Several research groups reported its use for identification, differentiation, and classification of microorganisms ^{(10, 11)}. Among its numerous applications, differentiation of normal from cancerous cervical tissue, normal and cancerous colon tissue, normal and malignant lymph cells and tissue, and investigation of brain cancer were reported ^(12–16). To improve FT-IR results, several studies have combined FT-IR results with chemometrics ^(17–21). To our best of knowledge no previous study has been published about the use of optical spectroscopy, ATR-IR and FT-IR to analyze human seminal plasma for the purpose of identifying molecular aspects of male infertility.

In this study, we have first compared the results of ATR-IR obtained from human seminal plasma of normospermic versus azoo-spermic men. In addition, we have improved the results by FT-IR spectroscopy of the metabolome of human seminal plasma in normospermic versus azoospermic men. To our best of knowledge, this is the first study in which significant changes were observed between the metabolome of human seminal plasma in normospermic versus azoospermic men using FT-IR spectroscopy.

Optical spectrospcopy is an excellent technique for diagnosis. Pattern recognition by FT-IR is an easy and a cheap diagnostic test. In the IR different functional groups of peptides, proteins and metabolites can be detected. We applied ATR-IR and FT-IR on whole and extracted metabolites of human seminal plasma.

In the ATR-IR, the sample can be measured in liquid form. Analysis of human seminal plasma in the liquid form showed that there are many components in human seminal plasma in which significant variation can be observed (Figure 1). We decided to focus on the subcomponents. We extracted the metabolome from human seminal plasma and analyzed it by ATR-IR. However, the O-H group had strong signal and suppressed the signal of lower part of spectra which is shown in figure 1. We decided to analyze the extracted dried metabolome by FT-IR. The advantage of dried sample is that methanol and water will evaporate and only dried metabolome will be left (Figure 4).

The dried metabolome was then dissolved in the chloroform. The chloroform signal will not interfere with metabolites signal (600-4000 cm ⁻¹ wave number). As we expected, the fingerprint region of FT-IR spectra showed significant changes. Additionally, it was quite reproducible. We believe that FT-IR could be used for quantification of the extracted metabolome in function groups, if we had access to white diamond liquid sample cell for FT-IR. In order to improve our results, the chemo-metrics method, PCA, was applied on the obtained data. As it is shown in figure 3, the normospermic and azoospermic men are separated quite differently.

One of the problems, which limited sample collection (inability to collect more than 4 samples each time), was that KBr crystal tablet lost its reproducibility. We believe the reason was that metabolites bind to KBr crystal and we were not able to wash it even with several trials. Therefore, we suggested using white diamond liquid holder for FT-IR measurement. We believe using white diamond liquid holder will improve the obtained results and make FT-IR spectra useful for quantification of metabolome. Another instrument which could be used for pattern recognition is Raman spectroscopy ⁽²⁶⁾. In this method, liquid can be used and measured quite easily since water does not interfere with the spectra. It is an ideal instrument for human body fluid measurement. However, it should be emphasized that the currently obtained result is preliminary and larger sample size analysis is needed for validation of the method.

We have shown that chemometrics combined with FT-IR can be used as a simple, fast and cheap screening method for separating human seminal plasma of normospermic versus azoospermic men. Although, this method is a non-invasive way for fast diagnosis of azoospermia, it cannot reveal the details of the problem and biopsy might be needed for having more information on azoospermia.