Detection of fresh blood by luminol and DNA after walking over various substrates

ABSTRACT

This preliminary research aimed to determine if a forensically useful human DNA profile could be obtained from latent blood past the point at which the blood could be visualized with luminol. Five participants with differing physical characteristics created shoeprints from fresh human blood over concrete, linoleum and carpet. The amount and distribution of both visible blood and chemiluminescence emanating from latent blood were documented. DNA analysis was performed on samples taken from the fifth last latent shoeprint to show chemiluminescence and the first and tenth steps after which chemiluminescence no longer occurred. DNA from the blood source was obtained past the point of visualization using luminol, with 2 of the 30 samples generating full STR DNA profiles, 18 generating partial profiles of 4 or more alleles, and 3 generating between 1 and 3 alleles. Of the 7 samples which did not generate a profile, 6 came from the concrete. Higher amounts of DNA, and more full and partial profiles, were obtained from linoleum compared to concrete or carpet. Participant effect was not observed. These findings demonstrate that it may be possible to obtain forensically useful DNA from crime scenes even when blood is not able to be visualized with chemiluminescence.

KEYWORDS:

• Secondary transfer
• blood
• DNA analysis
• luminol
• crime scene
• shoeprints

1. Introduction

It is well known that DNA may be transferred either directly or indirectly^1,2. The extent to which there are detectable levels of DNA after a transfer event will depend on several variables including the type and quantity of biological material, its level of dryness, the type of substrate involved, and the manner of contact^2–9.

Bloodstains at crime scenes may not be visible (latent blood), such as after the transfer of very low quantities of blood. For example, an individual may stand on a bloodied surface and then proceed to walk on a surface without blood, transferring diminishing amounts of blood to contacted surfaces with each subsequent step¹⁰. To visualize such stains, crime scene personnel often use chemicals such as luminol which reacts with the haem in blood to generate chemiluminescence^11,12. The luminol test is highly sensitive, reacting with small amounts of blood down to a dilution of 1,000,000:1^12,13.

This preliminary study aims to determine if there is a possibility of obtaining DNA from latent blood past the point of visualization with luminol, in a forensically useful quantity and quality, given a range of floor substrates and persons of different shoe size, weight, and gait.

2. Methods

2.1. Blood

Human blood was collected from a healthy male individual by a trained phlebotomist on five different occasions. Blood was obtained via venepuncture into medical standard BD Vacutainer® Ethylenediaminetetraacetic acid (EDTA) tubes and immediately taken to the laboratory where it was stored at 2°C in a refrigerator until use. Due to the volume of blood required, and the need to transfer it to the laboratory after collection, it was not possible to use blood without an anti-clotting agent. However, as EDTA does not impact the quantity of DNA recovered from blood¹⁴ and is similarly sensitive to luminol as non-treated blood¹⁵, it was not considered a limiting factor in this research. Blood was used within three days of collection to minimize any variation due to differences in storage duration. The blood collected during a single occasion were used in experiments for each of three substrates walked on by the same participant.

2.2. Participants and shoes

Five participants of varying physical characteristics were used in this study (Table 1). The shoes worn by the five participants (Table 1) were brand new pairs of canvas shoes with rubber soles (‘Anko’ brand, Kmart) (Figure 1). Each participant wore a new pair of shoes for each of the different floor types to limit potential differences due to the presence of dried blood on the soles subsequent to use.

Figure 1. The sole tread of the shoes worn (a); an example stain print left by the shoe after stepping in blood and being sprayed with luminol solution (first step of participant 2 on carpet substrate) (b); and step 17 (step point 2) of the same series, from which DNA was sampled (c). The circled area in C represents the approximate area of the stains sampled for DNA (~4.91cm²).

Table 1. Participants’ measurements and characteristics.

Participant	Height	Weight	Shoe size*	Step size**
1	165 cm	75 kg	7 women’s	44 cm
2	171 cm	60 kg	9 women’s	55 cm
3	155 cm	58 kg	9 women’s	30 cm
4	187 cm	80 kg	11 ½ men’s	65 cm
5	176 cm	71 kg	11 men’s	30 cm

*Shoe sizes are Australian.

**Step size was measured from the distance of step from the tip of the left foot to the heel of the right foot.

2.3. Substrates

Three floor types were used in these experiments – a coarse concrete floor in an outdoor area with no overhead shade, smooth and non-absorptive linoleum, and a second-hand absorptive carpet (twist pile, wool/synthetic blend) previously used by unknown individuals and stored rolled up for an unknown period. The carpet was 17 mm in height from base to top, with the fibres being 16 mm in height and the base 1 mm in height. Different sections of the same carpet were used for each participant. The carpet was not cleaned prior to the experiment but contact with it was minimized after receipt from the supplier. The concrete and linoleum types were cleaned before and after use by each participant, with a 5% sodium hypochlorite solution which was then washed off with water and the substrate allowed to dry prior to use. The linoleum and carpet samples were inside and not exposed to direct sunlight at any point, but the concrete samples were exposed to direct sunlight for the full time period. During the experimental periods, the maximum outdoor temperatures ranged from 15°C to 20°C and the maximum UV index from 1.5–3.1.

2.4. Luminol

The luminol was prepared by dissolving 0.5 g of luminol powder into 25 g/450 mL sodium carbonate solution (Na₂CO₃ in H₂0), then adding 50 mL of a 30% hydrogen peroxide solution and mixing briefly. The luminol solution was prepared within 30 mins of use.

2.5. Experimental design

The inside base of a flat-bottomed plastic container (23 cm x 39 cm x 12 cm) was covered with 48 mL of fresh human blood. Participants stood with the left shoe in the blood for 5 seconds, enabling the entire bottom of the shoe to be covered in blood. Participants then walked in line with designated steps that were measured to match each participant’s typical walking step size. For each participant, 70 left steps were taken on the linoleum, 50 left steps on the carpet, and 55 left steps on the concrete. Preliminary research (data not shown) showed the numbers indicated above were 10 steps more than where the last chemiluminescence could be seen at the ball of the foot area on each of the floor types. Given this research aimed to determine if DNA could be recovered from latent blood that was unable to be visualized using chemiluminescence, the ball of the foot area (a circular area ~2.5 cm in diameter, Figure 1) was selected as the representative sample, due to this area having no chemiluminescence occurring consistently at earlier steps than other areas of the shoeprint (based on preliminary research), and enabled the research to be conducted within the physical space available.

The resultant transfer bloodstains were photographed immediately after deposition, and then left to dry till nightfall (approximately 7 hours) at which time a luminol solution was sprayed in the area of the steps, and the stains were again photographed.

The fifth last step to show chemiluminescence in the ball of the foot area was designated point 1, the first step to show no chemiluminescence in the ball of the foot area was designated point 2, and the tenth step to show no chemiluminescence in the ball of the foot area was designated point 3. The actual step number at which this occurred varied considerably between participants (Table 2). In all instances, there was some chemiluminescence visible in other parts of the print at all three step points.

Table 2. The number of steps taken by each participant to reach each designated step point (SP).

Participant	Concrete			Carpet			Linoleum
	SP1	SP2	SP3	SP1	SP2	SP3	SP1	SP2	SP3
1	32	37	46	23	28	37	23	28	37
2	12	17	26	30	35	44	15	20	29
3	30	35	44	36	41	50	14	19	28
4	40	45	54	21	26	35	23	28	37
5	15	20	29	32	37	46	16	21	30

SP1 – the fifth last step to show chemiluminescence in the ball of the foot area), SP2 – the first step to show no chemiluminescence at the ball of the foot area, and SP3 – the tenth step to show no chemiluminescence at the ball of the foot area.

The ball of the foot area of the step at each of these points was sampled using wet/dry swabbing¹⁶. The cotton swabs (COPAN, Italy) were left to dry in an aerated swab tube overnight in separate paper envelopes and stored in a freezer at −18°C until DNA analysis was performed.

2.6. Photography

A Nikon D7200 digital camera with an AF-S Nikkor 24 mm f/1.8 G ED lens was used to record the shoeprints. The following settings were used to take photographs during the day: auto on the mode dial, and an ISO sensitivity of 1600. For the photography of the chemiluminescence, the camera was set to manual on the mode dial with an aperture of 2.8, set to bulb setting for long exposure and an ISO sensitivity of 1600.

2.7. DNA analysis

The wet and dry swabs were combined for each sample. DNA was extracted using an organic method to an end volume of 100 µL and quantified using the Quantifiler Trio® (Life Technologies, USA) kit on an ABIPRISM® 7500 (Life Technologies, USA) using HID Real-time PCR analysis software. DNA was amplified using PowerPlex® 21 (Promega, USA) for 30 cycles, as per manufacturer’s instructions, using 15 µl of extract. Amplified product detection and sizing was performed on a 3500xL Genetic Analyser (Life Technologies, USA) with CE injection specifications of 1.2 kV 24s, and genotyped using GeneMapper ID®-X (Life Technologies, USA) software with a baseline threshold of 175 RFU and homozygous threshold of 2000 RFU. The following metrics were determined per sample: total quantity of DNA recovered, degradation index (DI), and number of donor alleles present in the profiles (excluding amelogenin). Profiles were classified as: ‘no profile’ (no alleles present), ‘inconclusive’ (3 or less alleles), ‘partial profile’ (4–39 alleles inclusive), and ‘full profile’ (all alleles). Profile classifications only considered alleles that matched those of the blood donor profile as this study focused on the transfer of the source of blood. Therefore, extraneous alleles present in profiles derived from the carpet (due to the carpet being second-hand) were not reported. Any mixed profiles were deconvoluted using the statistical software STRMix (v2.8.0, ESR, New Zealand) and the contribution percentage attributed to the donor profile was used to ascertain the total amount of DNA contributed by the donor to the sample.

The DI was calculated for each sample by dividing the amount of small autosomal DNA present in the sample by the amount of large autosomal DNA present¹⁷. As per the manufacturer’s guidelines for Quantifiler Trio®, a value of below 1 was considered to be within a normal range of degradation, a value of 1 to 10 indicated slight to moderate degradation, any value over 10 indicated significant degradation¹⁷. The manufacturer states that the DI may be affected by significant degradation of large autosomal DNA or the presence of PCR inhibitors¹⁷. Given that research has indicated luminol does not cause detectable levels of PCR inhibition^18,19, samples with detectable quantities of small autosomal DNA but no large autosomal DNA detected, significant degradation was considered to have occurred.

Due to the small sample size and non-normal distribution of DNA data acquired, non-parametric statistical testing was performed. The quantity of DNA was analysed using the Kruskal-Wallis test (IBM SPSS 27) and the Mann-Whitney U test (IBM SPSS 27) comparing the quantity data from samples between participants, designated points, and floor types. The median data has then been reported.

3. Results

A quantity of donor DNA >0.02 ng was recovered from every step point, from each floor type, walked on by each participant, except for 8 samples collected from concrete (Figure 2). While there was no significant difference between concrete and carpet (U = 3.33, p > 0.05), the median amount of DNA recovered from linoleum was significantly higher than the median amount recovered from concrete and carpet (U = 16.13, p < 0.05). There was no significant difference in median DNA amount recovered between the different participants (H = 1.42, p > 0.05) or points sampled for concrete (H = 0.536, p > 0.05) and carpet (H = 0.125, p > 0.05). Linoleum showed a significant difference between some points (H = 8.57 df = 2). There was no significant difference between points 1 and 2 or points 2 and 3 but the median amount of DNA recovered from point 1 was significantly higher than the median amount recovered from point 3 (U = 10.00, p > 0.05).

Figure 2. Amount of DNA (ng) recovered from each step point (SP) for each floor type.

No degradation was detected in the 32 samples for which a degradation index (DI) could be calculated. Large autosomal DNA was not detected in one linoleum sample and two carpet and concrete samples, indicating degradation and/or inhibition had occurred, and no DNA was detected in 7 concrete samples. It is unlikely the lack of detection was the result of PCR inhibition as the Internal PCR Control did not indicate inhibition had occurred in any samples, with the Ct for all samples ranging from 26.70 to 28.33. One sample from the concrete floor was not assigned a DI value as the amount of small autosomal DNA was below the threshold of detection, despite large autosomal DNA being detected.

DNA profiles derived from the blood were obtained at the first and tenth step beyond where chemiluminescence was visible (step point 2 and 3 respectively) (Table 3). The extent to which this was possible was dependent on the type of flooring walked on, with higher quantity DNA and higher quality profiles collected more frequently from linoleum than from other floor types at the same step points (Figure 2, Table 3). Of the 12 samples from which no profile was obtained, 9 were from the concrete. Of the remaining 6 concrete samples, 1 profile was inconclusive, and 5 were partial profiles (ranging from 17 to 39 alleles). Of the 15 carpet samples, no profile could be obtained from 3 samples, 1 profile was inconclusive and 11 were partial profiles (4–39 alleles). Of the 15 linoleum samples, 1 was inconclusive, 8 were partial profiles (6–39 alleles) and 6 were full profiles.

Table 3. The number of donor alleles (excluding amelogenin) within the profiles at step-points 1, 2 and 3 per floor-type for all 5 participants.

		Number of alleles
Substrate	Point	0	1–3	4–14	15–39	40
	Profile	None	Inconclusive	Partial profile		Full
Concrete	1	4			1
	2	2			3
	3	3	1		1
Carpet	1	1		2	2
	2	1		2	2
	3	1	1	2	1
Linoleum	1				1	4
	2				4	1
	3		1	1	2	1

4. Discussion

This research aimed to determine if there is a possibility of obtaining DNA from latent blood past the point of visualization with luminol, and if so, after how many steps beyond this point DNA can still be retrieved at a quantity and quality that would be forensically useful. The data demonstrated that it is possible, but how far past visualization, and the quantity and quality of the DNA recovered, varied depending on the floor type.

The results showed the median amount of DNA collected from the linoleum was significantly higher than the amount collected from concrete, and also between step points 1 and 3 on the linoleum. While not significantly different, the median amounts of DNA collected from carpet were generally higher than the median amounts collected from concrete. These results align with that of Schmidt et al¹⁰ who found DNA was lost from the soles of shoes at a faster rate when walking on concrete than on a forest underground, with the inference being that less DNA on the shoe would mean there would be less to transfer to concrete over subsequent steps. The differences between carpet and concrete samples are likely to be due to the different properties of the floor types. Blood applied to concrete can soak beneath the area available for swabbing and can therefore not be collected as readily as blood applied to carpet and linoleum. While carpet fibres absorbed blood, it was still possible to recover most of this blood as pressure applied during swabbing would release some of the blood from the fibres. However, the non-porous nature of linoleum meant blood loss was minimal. There is also the possibility of the DNA sampling method not being as efficient for the concrete and carpet floors compared to the linoleum, due to the higher degree of roughness and porosity of these floor types, as research has demonstrated that the cotton swabs used in this research retrieves DNA in lower amounts from porous surfaces than non-porous surfaces^20,21.

Degradation of DNA was observed in samples from all substrates as evidenced by both the DI and the quality of the profiles. Of the 12 samples which generated no profile, five were from samples in which large autosomal DNA was not detected. The remaining seven were from samples in which neither large nor small autosomal DNA could be detected so it is inconclusive as to whether the lack of detectable DNA in these samples is due to initial absence of DNA or degradation. Samples for which no profile was generated were predominantly recovered from concrete rather than from linoleum and carpet.

There has been research to show that luminol can cause degradation of DNA, but it is typically minimal and with large amounts of time between the application of luminol and the DNA analysis of the samples^22,23. However, given the same luminol reagent was applied to all substrates, and sampled after the same amount of time, it is likely that other factors were impacting the integrity of the DNA. The concrete was the only substrate type located outdoors and environmental insults may have compromised the DNA. However, while the bloody shoeprints were exposed to approximately 7 hours of direct and indirect solar radiation, this is unlikely to have been the sole cause of DNA degradation in these samples. Research has found that whole human blood exposed for 2 hours to direct solar radiation with a UV index of 10 and temperatures ranging between 33.9°C and 36.7°C showed no appreciable degradation²⁴. While the concrete samples were exposed to the environment for a longer period, the UV index was much lower and temperature substantially cooler. Therefore, unknown degradative properties of concrete may have had an impact, either alone, or in conjunction with environmental exposure.

The results of this research demonstrate there is a possibility of getting DNA from past the point of visualization in a sufficient quantity to get partial and full profiles. It is important to note that partial profiles are forensically useful, despite being incomplete, as they may still be used to screen DNA databases for potential matches or to exclude a suspect from being a contributor to a profile. Furthermore, with luminol able to react with blood in a 1,000,000:1 dilution^12,13, it would theoretically be possible to have 0.04 ng of human DNA in a sample at the lowest levels of detection. Given a forensically useful profile can be obtained from 0.04 ng and less of DNA, DNA can be detected beyond the point at which luminol detects blood.

This was a preliminary study and was looking at a set volume of fresh blood. It could be expected that the results of transfer would differ with smaller or larger amounts of blood and if there was differing levels of dryness⁶. This study also included only one shoe sole type; further research into other shoe soles may show different results. Furthermore, a relatively small target area was sampled in this study and more DNA may be collected from a larger area, and at a greater distance from the source of the blood, as blood from the remainder of the shoe would be transferred over a greater number of steps.

It is important to note that there were limitations to this experiment, such as the small sample size and the fact that only the ball of the foot area was targeted instead of the whole shoe area. Different data may also have been observed if the right foot had been used instead of the left foot, due to potential variations in an individual’s gait and walking style. Another limitation was that the steps were made under controlled conditions and sampled within 8 hours of the blood being deposited, which may not reflect real-world scenarios.

Also, as this study only sampled up to the tenth shoeprint beyond which chemiluminescence was no longer visible, further research is necessary to ascertain the point at which DNA is no longer able to be obtained given the type and quantity of biological material transferred, the type of shoe sole worn, and the substrate being stepped on.

This research lends itself to police casework by demonstrating the ability of current DNA analysis methods to obtain DNA from blood past the point of visualization by chemiluminescence after application of a luminol solution, in a quantity large enough to generate an informative DNA profile. Therefore, if there is an area where suspected blood may have been, it may still be useful to swab the area to try and obtain DNA, even if chemiluminescence did not result with the application of luminol. For example, if an offender absconds from a crime scene in a vehicle after walking in the victim’s blood, latent blood which is not able to be visualized may be present on the control pedals of the vehicle and DNA from the victim could therefore potentially be obtained by swabbing these. Furthermore, these findings may provide an explanation for the unexpected presence of an individual’s DNA in the background DNA of a targeted sample.

This emphasizes that samples could still be taken from target areas where DNA may be expected to have been deposited, given relevant other information, even in situations where there is no visible biological material present. While the lack of a positive result from the presumptive testing for blood may impede the use of the DNA in court, where proof may be required regarding the biological source from which the profile was derived, the evidence may still provide an important investigative lead. This would include linking people and/or items to a crime scene or each other or providing a reasonable explanation as to the presence of the DNA on a surface without an obvious source.

Acknowledgments

The authors would like to acknowledge Lotte Smeulder from the Avans University of Applied Science for her preliminary work towards this study.

Disclosure statement

No potential conflict of interest was reported by the author(s).