Introduction

In forensic molecular genetic analysis, the use of new, sensitive PCR-Multiplex-Kits suitable also for low copy number DNA testing [1–3] as well as sophisticated techniques like single cell laser dissection [4–7] lead to evaluable DNA profiles even in cases that seemed to be hopeless a few years ago [7]. However, this sensitivity is not only an advantage but also leads to problems like contamination of the DNA evidence by anyone working at a crime scene—from the paramedic to the police officer—or scientists doing the analysis [8, 9]. Besides, the impact of the DNA testimony in relation to the crime itself may be questioned, since a person identified by its DNA profile at a crime scene may not have left its DNA directly there. There is also the possibility of a so-called secondary transfer of DNA traces from one surface to another. In this context, the statement of Locard’s Principle, “Wherever he (the culprit) steps, whatever he touches, whatever he leaves, even unconsciously, will serve as a silent witness against him” [10], summarized as “every contact leaves a trace” is as actual as never before [11, 12]. Here, it can be important to establish certain principles of “secondary transfer” possibilities to discriminate relevant evidence from accidentally spread traces. Some reviews have already been published on this topic [2, 3, 13]. However, techniques have further developed and several authors suggest different variables influencing the amount of DNA left on surfaces [14–16]. Some “intrinsic” variables qualifying the person as “good” or “bad shedder” [14] could be individual skin conditions, maybe altered by dermatological diseases [17], age [18], or gender. “Extrinsic” variables also were described such as different surface conditions and different manners of DNA application [19, 20]. Besides nuclear DNA from keratinocytes [17], a recent study suggests “trace DNA” to be derived from sebaceous glands [21]. All these studies investigated scenarios using more or less “dry conditions”. Inspired by questions arising at court in day-to-day forensic genetic routine, this study investigated different scenarios of secondary transfer of epithelial abrasions during laundry by hand and in a washing machine. The aim of our study was to determine whether a transfer of epithelial, blood, or saliva cells between textiles in aqueous detergent solutions is detectable and to which extent this transfer may occur.

Material and methods

Artificial stains

The study comprised 15 individuals (ten women age 39–58 years, five men age 38–58 years.) without any known skin disease to exclude an influence as described by Kamphausen et al. [17]. Samples were collected in 2012 and 2013 in the Institute of Legal Medicine, University Hospital Essen, Germany. All samples were obtained after informed consent and with approval of the Medical Ethics Committee at the University of Duisburg-Essen in accordance with the Declaration of Helsinki and national laws. At first, cotton cloths (10 × 4 cm) were cleaned with DanKlorix® dye 10 % to remove DNA [9], rinsed with clear water thoroughly, and dried. Controls were amplified to assure the lack of DNA contaminations. Different kinds of artificial traces were produced. Individual neck rubs created by close contact (so-called primary transfer) between skin and textile were produced by rubbing the cloth over the neck for approximately 5 s with medium pressure simulating a worn collar. Own surveys in daily casework routine showed that high amounts of DNA can usually be obtained from textile’s collar regions. In addition, pre-test series (data not shown) by rubbing a textile cloth over the neck were performed to verify these findings. Secondary, artificial blood stains were produced by dropping 1.5 ml freshly taken unclotted venous blood on a clean piece of cloth (approximately 8 × 15 cm) and drying for at least 48 h. Furthermore, buccal swabs were taken as donor known to obtain high amounts of epithelial DNA by rubbing a textile cloth over buccal mucosa.

From each individual buccal swabs were taken for comparison purposes.

Different settings of washing conditions

Secondary transfer by hand-washing

One neck rub or textile with blood stain or textile with buccal swab was washed together with one clean cloth and whirled by hand in a previously cleaned (EtOH 70 %) beaker containing 500 ml clear mains water (temp. 39.5–42.5 °C) or 500 ml clear water and standardized No. 77 color-testing detergent for 3 min simulating laundry by hand of sensitive fabrics. Controls were taken previously with DNA-free swabs. Cloths were wringed and dried thoroughly averting further contamination.

Secondary transfer by machine-washing

Laundry in a washing machine under controlled and standardized setting could be performed in the labs of Deutsches Textilforschungszentrum Nord-West e.V., Krefeld, an institute of the University Duisburg-Essen using a standard washing machine ATLAS Linitest plus M228BB (SDL Atlas, Rock Hill, USA) following DIN ISO 105-C06 [22] standard test washing procedure for textiles (Fig. S1). In this setting, standardized No. 77 color-testing detergent and a washing program is used comparable to concentration and composition of commercial washing powders, temperature and duration as used for colored textiles. The stainless steel beakers (Fig. S1) were cleaned with EtOH 70 %; controls were taken with DNA-free swabs and analyzed afterwards.

One neck rub or textile with blood stain or textile with buccal swab was washed together with one clean cloth similar to the hand-washing. Each stainless steel beaker contained 150 ml aqueous washing solution (concentration: 4 g/l) made of No. 77 color-testing detergent and demineralised, ultra-filtrated water (Millipore Integral 5 Milli-Q, pro analysi). Temperature was 40.0 °C and duration of the automated washing program 30 min with no change of solution. This washing procedure is used routinely by the Textilforschungszentrum for testing of textiles and detergents; therefore, it resembles a daily washing procedure rather accurately. Cloths were rinsed with 300 ml aqua dest pro analysi subsequently and dried thoroughly averting further contamination.

DNA extraction

After the washing, dried cloths were taped with self-adhesive tape, and cells were collected from the tape with a double swap technique using first a DNA-free swab, moistened with lysis buffer, and then a dry swab. DNA extraction from artificial stains was performed using a modified phenol/chloroform method, published by DeSalle and Bonwich [23] using an extraction volume of 40 μl. DNA of buccal swabs was extracted using 300 μl SwabSolution® (Promega, Mannheim, Germany).

DNA quantification

DNA amount of the artificial stains was measured using the Quantifiler® Human DNA quantification kit according to the manufacturer’s instructions (Applied Biosystems). Two microliter DNA-containing solution per sample were analyzed in triplets with a reliable and reproducible detection threshold down to 25 pg.

DNA amplification and electrophoresis

DNA amplification was done either with multiplex PCR Kit Powerplex® ESX 17 or Powerplex ® S5 following the manufacturer’s instructions (Promega, Mannheim, Germany) with a reduced PCR volume of 12.5 μl in the GeneAmp® PCR system 9700 (Applied Biosystems, Darmstadt, Germany). The employment of this non-standard reaction was done to save money for this study and has been thoroughly and independently tested according to the existing quality managements [24]. In each amplification, a positive control (100 pg 9947A) and a negative control (sterile water) were analyzed. Amplification products were separated and detected on the ABI3130 Genetic Analyzer (Applied Biosystems) in comparison to the allelic ladder which is a component of each kit. Electrophoresis results were analyzed using the GeneMapper® ID Software v3.2. Allele peaks were interpreted when greater than or equal to 50 RFUs.

For the evaluation of the results, three different categories were defined as follows: In a complete profile, every allele in every evaluable STR of the individual in question was found, regardless of any additional peaks. In a partial profile, in ten or more of the 16 loci, but not in each STR, every allele of the individual was found, regardless of any additional peaks. No profile means that in less than ten of the 16 loci, the alleles of the individual were seen (and there may have been other alleles).

Results

Control analyses

Control analyses were done of wash tubs (hand-washing) or steel beakers (machine-washing) for every experimental set-up. In none of these controls, full or partial profiles of the study individuals could be found. Mostly, only zero to three peaks appeared on the electropherograms, thus proving that no significant contamination occurred. Only in two control swabs, more than three peaks/alleles were displayed with no resemblance to alleles of the study individuals or additional alleles found on the textiles after washing.

Textiles with blood stains

Experiments with blood stains were done twice with three to ten of the individuals participating in the study, thus resulting in 20 washing procedures by hand without detergent, six washing procedures with detergent, and another twenty by machine. Macroscopically, no blood could be seen on the cloth after the washing procedure, either by hand or with the machine. Nevertheless, after washing, a full profile of the blood donor could be obtained by molecular genetic analysis from each cloth on which the blood has been applied (donor cloth). In addition, a full or partial profile of the blood donor could be demonstrated on the receiving cloth in 29 and 17 samples, respectively (Table 1, Fig. S2). No significant differences were found between washing by hand and machine-washing or use/no use of detergent, and total DNA amounts varied between 10 and 65 ng. Moreover, no striking differences were found in the repetition of each experiment.

Table 1 Summary of experiments with blood stains and epithelial abrasions
Full size table

Textiles and buccal swabs

Six of our study individuals participated in this part of the investigations. Every set-up was repeated once. A transfer of a sufficient amount of DNA from a buccal swab to the textiles during washing could not be demonstrated, regardless of the washing method. The total DNA amount of cloth extractions varied between 0 and 12 ng, but a full or partial profile of the buccal swab donor could not be obtained in any experiment in contrast to the blood stains.

Textiles with neck rubs

Both in our own casework samples and in recent studies, absorbent and rough surfaces are known to collect more DNA than rather smooth or polished surfaces [1, 13, 19, 25]. Besides the type of surface, often nature of contact is described as a major factor influencing the amount of DNA left on a surface rather than time of contact [1, 13]. Especially from textile’s collar regions, high amounts of DNA can usually be obtained, as seen in daily routine, therefore neck rubs from all 15 study individuals were used as model for transfer of epithelial cells. Every experiment was repeated at least once.

Despite the rather high amount of DNA left on neck rubs (prior studies displayed total DNA amounts between 15 and 189 ng) in comparison to textiles just touched by hand, full DNA profiles (in regard to the abovementioned criteria) could not be amplified after washing, neither for cloths washed by hand (with or without detergent) nor for machine-washing (Table 1). Drop-in artifacts, drop-out artifacts, and ladder amplification artifacts were observed repeatedly. Nevertheless, there were some differences between hand and machine-washing. Textiles washed by hand showed some residual DNA on the donor cloth (total DNA amount 670 pg to 52 ng), a partial profile could be obtained for 11 individuals (seven women, four men) in at least one experiment (Fig. S3), but never a full profile. Regarding the other four individuals, all donor textiles showed no profile. In all acceptor textiles (total DNA amount 210 pg to 2 ng), no profile was found. Regarding machine-washed clothes, neither on donor nor on acceptor textiles (total DNA amount 140 pg to 870 pg), full or partial profiles of the 15 cell donors could be detected with the exception of two experiments with partial profiles. Here, the lowest number of analyzable loci per electropherogram was found; only zero to eight loci showed peaks at all.

Discussion

Direct and indirect transfer of cellular material containing low amounts of DNA from dried blood or saliva spots has been proven by several studies in the past (e.g., [1, 13, 19, 26]), but none of these investigated a transfer by washing. cDNA may be transferred from one textile item to another by washing, especially in a washing machine, often emerges during a trial at court and has to be answered by the forensic expert. The transfer of cells from one cloth to another in a washing machine can happen in two different ways. First, the cells can be suspended in the solution and afterwards passed onto the acceptor cloth. Second, the transfer may occur by accidental direct contact between donor and acceptor cloth.

This study investigated potential cell transfers of different sources in “wet conditions”, i.e., either using a washing machine or hand-washing techniques. Test conditions were comparable to a careful hand-washing of touchy clothes and to a daily used machine-washing by employing a standardized DIN ISO standard washing procedure.

Not surprisingly, the results differed considerably for the different cell types. The transfer of blood cells from one item to another has been shown repeatedly (e.g., [19, 26, 27]) and was also frequently possible in our study. Dried blood stains could not only be transferred directly but also indirectly by transfer of tiny dried blood particles as shown by Wiegand et al. [26]. This transfer of tiny blood particles seems also possible under wet conditions. The amount of detergent used in our study (resembling that of “normal” machine-washing) was obviously not high enough to hinder the cells or the naked DNA from clinging to the textiles. Goray et al. [19] showed that wet blood samples could be transferred more easily than dry ones; this may also be the possibility with samples turned from dry to wet in a washing solution. On the other hand, washing procedures are often employed by culprits to remove blood (and DNA derived from blood cells) from stained cloths. Here, our results clearly demonstrate that this is very difficult to achieve. In fact, neither washing by hand nor machine-washing succeeded in eliminating this kind of DNA from donor cloths.

Saliva samples are less likely to produce full STR profiles after transfer [19, 26], especially using cotton textiles as source. This is in line with our results, since no full STR profiles could be obtained after washing buccal swabs together with clothes. Here, the chance for transfer is also limited by the contact possibilities—a close contact of textile to textile is more probable than between swab and textile.

The main part of this study was the investigation of transfer of cells or DNA from items with epithelial abrasions under wet conditions, because this is the question asked most in court. In general, clothes are a very good source for the DNA of the wearer (own studies, data not shown; [13, 28]), but the transfer of this DNA or epithelial cells to another (textile) item has only rarely been investigated [25]. Goray et al. performed primary transfer of epithelial cells by rubbing the test person’s palm either on a plastic surface or a cotton surface. Secondary transfer either on a plastic surface or cotton cloth was done by passive contact, pressure or friction, fresh or after drying. As a result, the transfer rates of fresh and dried skin cell stains differed only insignificantly, whereas manner of contact (slight/passive contact vs. pressure/friction) and manner of donor and acceptor surface were identified as significant factors with highest deposit of DNA found in secondary transfer from non-absorbent surface (plastic) to absorbent (cotton cloth) acceptor surface. Pressure and friction were factors linked to higher transfer rates than passive, slight contact [25]. Our results demonstrated that the transfer of epithelial skin cells from one textile to another during a washing process in a sufficient amount for reliable STR analysis (i.e., a full profile) is nearly impossible. Moreover, both hand-washing and machine-washing proved to be quite efficient in removing DNA derived from epithelial cells from the clothes. Here, dilution of the traces in the aqueous solution or solving the cells from the fabric’s surface by the detergent possibly decreases the amount of DNA containing abrasions remaining on the textiles after the laundry. Moreover, the composition of the detergent, containing bleaches or aggressive oxygen radicals degrading DNA could be partly responsible for the results. These conclusions are even more valid in a set-up with a real washing machine, in which the volume of the aqueous solution is not only much higher, but will also be changed at least once during the washing process thus reducing the probability of cell transfer through the solution.

Conclusions

The transfer of DNA or cells between textiles or fabrics in aqueous solutions is principally possible, if mostly restricted to blood. Using cells or DNA from buccal swabs or epithelial abrasions, no transfer could be established suitable for a successful and reliable genetic analysis leading to a full or at least partial profile to stand up in court. Despite the rather low number of experiments and despite the fact that our experimental set-up is only an approximate portray of the huge reaction chamber in a normal washing machine, the results of this study propose that neither a relevant nor a significant transfer of epithelial trace DNA between textiles occurs while being washed together using current multiplex STR-kits for forensic analysis.